File: saturne1.tex

package info (click to toggle)
code-saturne 4.3.3+repack-1
  • links: PTS, VCS
  • area: main
  • in suites: stretch
  • size: 77,992 kB
  • sloc: ansic: 281,257; f90: 122,305; python: 56,490; makefile: 3,915; xml: 3,285; cpp: 3,183; sh: 1,139; lex: 176; yacc: 101; sed: 16
file content (8808 lines) | stat: -rw-r--r-- 446,618 bytes parent folder | download
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
2628
2629
2630
2631
2632
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
2654
2655
2656
2657
2658
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689
2690
2691
2692
2693
2694
2695
2696
2697
2698
2699
2700
2701
2702
2703
2704
2705
2706
2707
2708
2709
2710
2711
2712
2713
2714
2715
2716
2717
2718
2719
2720
2721
2722
2723
2724
2725
2726
2727
2728
2729
2730
2731
2732
2733
2734
2735
2736
2737
2738
2739
2740
2741
2742
2743
2744
2745
2746
2747
2748
2749
2750
2751
2752
2753
2754
2755
2756
2757
2758
2759
2760
2761
2762
2763
2764
2765
2766
2767
2768
2769
2770
2771
2772
2773
2774
2775
2776
2777
2778
2779
2780
2781
2782
2783
2784
2785
2786
2787
2788
2789
2790
2791
2792
2793
2794
2795
2796
2797
2798
2799
2800
2801
2802
2803
2804
2805
2806
2807
2808
2809
2810
2811
2812
2813
2814
2815
2816
2817
2818
2819
2820
2821
2822
2823
2824
2825
2826
2827
2828
2829
2830
2831
2832
2833
2834
2835
2836
2837
2838
2839
2840
2841
2842
2843
2844
2845
2846
2847
2848
2849
2850
2851
2852
2853
2854
2855
2856
2857
2858
2859
2860
2861
2862
2863
2864
2865
2866
2867
2868
2869
2870
2871
2872
2873
2874
2875
2876
2877
2878
2879
2880
2881
2882
2883
2884
2885
2886
2887
2888
2889
2890
2891
2892
2893
2894
2895
2896
2897
2898
2899
2900
2901
2902
2903
2904
2905
2906
2907
2908
2909
2910
2911
2912
2913
2914
2915
2916
2917
2918
2919
2920
2921
2922
2923
2924
2925
2926
2927
2928
2929
2930
2931
2932
2933
2934
2935
2936
2937
2938
2939
2940
2941
2942
2943
2944
2945
2946
2947
2948
2949
2950
2951
2952
2953
2954
2955
2956
2957
2958
2959
2960
2961
2962
2963
2964
2965
2966
2967
2968
2969
2970
2971
2972
2973
2974
2975
2976
2977
2978
2979
2980
2981
2982
2983
2984
2985
2986
2987
2988
2989
2990
2991
2992
2993
2994
2995
2996
2997
2998
2999
3000
3001
3002
3003
3004
3005
3006
3007
3008
3009
3010
3011
3012
3013
3014
3015
3016
3017
3018
3019
3020
3021
3022
3023
3024
3025
3026
3027
3028
3029
3030
3031
3032
3033
3034
3035
3036
3037
3038
3039
3040
3041
3042
3043
3044
3045
3046
3047
3048
3049
3050
3051
3052
3053
3054
3055
3056
3057
3058
3059
3060
3061
3062
3063
3064
3065
3066
3067
3068
3069
3070
3071
3072
3073
3074
3075
3076
3077
3078
3079
3080
3081
3082
3083
3084
3085
3086
3087
3088
3089
3090
3091
3092
3093
3094
3095
3096
3097
3098
3099
3100
3101
3102
3103
3104
3105
3106
3107
3108
3109
3110
3111
3112
3113
3114
3115
3116
3117
3118
3119
3120
3121
3122
3123
3124
3125
3126
3127
3128
3129
3130
3131
3132
3133
3134
3135
3136
3137
3138
3139
3140
3141
3142
3143
3144
3145
3146
3147
3148
3149
3150
3151
3152
3153
3154
3155
3156
3157
3158
3159
3160
3161
3162
3163
3164
3165
3166
3167
3168
3169
3170
3171
3172
3173
3174
3175
3176
3177
3178
3179
3180
3181
3182
3183
3184
3185
3186
3187
3188
3189
3190
3191
3192
3193
3194
3195
3196
3197
3198
3199
3200
3201
3202
3203
3204
3205
3206
3207
3208
3209
3210
3211
3212
3213
3214
3215
3216
3217
3218
3219
3220
3221
3222
3223
3224
3225
3226
3227
3228
3229
3230
3231
3232
3233
3234
3235
3236
3237
3238
3239
3240
3241
3242
3243
3244
3245
3246
3247
3248
3249
3250
3251
3252
3253
3254
3255
3256
3257
3258
3259
3260
3261
3262
3263
3264
3265
3266
3267
3268
3269
3270
3271
3272
3273
3274
3275
3276
3277
3278
3279
3280
3281
3282
3283
3284
3285
3286
3287
3288
3289
3290
3291
3292
3293
3294
3295
3296
3297
3298
3299
3300
3301
3302
3303
3304
3305
3306
3307
3308
3309
3310
3311
3312
3313
3314
3315
3316
3317
3318
3319
3320
3321
3322
3323
3324
3325
3326
3327
3328
3329
3330
3331
3332
3333
3334
3335
3336
3337
3338
3339
3340
3341
3342
3343
3344
3345
3346
3347
3348
3349
3350
3351
3352
3353
3354
3355
3356
3357
3358
3359
3360
3361
3362
3363
3364
3365
3366
3367
3368
3369
3370
3371
3372
3373
3374
3375
3376
3377
3378
3379
3380
3381
3382
3383
3384
3385
3386
3387
3388
3389
3390
3391
3392
3393
3394
3395
3396
3397
3398
3399
3400
3401
3402
3403
3404
3405
3406
3407
3408
3409
3410
3411
3412
3413
3414
3415
3416
3417
3418
3419
3420
3421
3422
3423
3424
3425
3426
3427
3428
3429
3430
3431
3432
3433
3434
3435
3436
3437
3438
3439
3440
3441
3442
3443
3444
3445
3446
3447
3448
3449
3450
3451
3452
3453
3454
3455
3456
3457
3458
3459
3460
3461
3462
3463
3464
3465
3466
3467
3468
3469
3470
3471
3472
3473
3474
3475
3476
3477
3478
3479
3480
3481
3482
3483
3484
3485
3486
3487
3488
3489
3490
3491
3492
3493
3494
3495
3496
3497
3498
3499
3500
3501
3502
3503
3504
3505
3506
3507
3508
3509
3510
3511
3512
3513
3514
3515
3516
3517
3518
3519
3520
3521
3522
3523
3524
3525
3526
3527
3528
3529
3530
3531
3532
3533
3534
3535
3536
3537
3538
3539
3540
3541
3542
3543
3544
3545
3546
3547
3548
3549
3550
3551
3552
3553
3554
3555
3556
3557
3558
3559
3560
3561
3562
3563
3564
3565
3566
3567
3568
3569
3570
3571
3572
3573
3574
3575
3576
3577
3578
3579
3580
3581
3582
3583
3584
3585
3586
3587
3588
3589
3590
3591
3592
3593
3594
3595
3596
3597
3598
3599
3600
3601
3602
3603
3604
3605
3606
3607
3608
3609
3610
3611
3612
3613
3614
3615
3616
3617
3618
3619
3620
3621
3622
3623
3624
3625
3626
3627
3628
3629
3630
3631
3632
3633
3634
3635
3636
3637
3638
3639
3640
3641
3642
3643
3644
3645
3646
3647
3648
3649
3650
3651
3652
3653
3654
3655
3656
3657
3658
3659
3660
3661
3662
3663
3664
3665
3666
3667
3668
3669
3670
3671
3672
3673
3674
3675
3676
3677
3678
3679
3680
3681
3682
3683
3684
3685
3686
3687
3688
3689
3690
3691
3692
3693
3694
3695
3696
3697
3698
3699
3700
3701
3702
3703
3704
3705
3706
3707
3708
3709
3710
3711
3712
3713
3714
3715
3716
3717
3718
3719
3720
3721
3722
3723
3724
3725
3726
3727
3728
3729
3730
3731
3732
3733
3734
3735
3736
3737
3738
3739
3740
3741
3742
3743
3744
3745
3746
3747
3748
3749
3750
3751
3752
3753
3754
3755
3756
3757
3758
3759
3760
3761
3762
3763
3764
3765
3766
3767
3768
3769
3770
3771
3772
3773
3774
3775
3776
3777
3778
3779
3780
3781
3782
3783
3784
3785
3786
3787
3788
3789
3790
3791
3792
3793
3794
3795
3796
3797
3798
3799
3800
3801
3802
3803
3804
3805
3806
3807
3808
3809
3810
3811
3812
3813
3814
3815
3816
3817
3818
3819
3820
3821
3822
3823
3824
3825
3826
3827
3828
3829
3830
3831
3832
3833
3834
3835
3836
3837
3838
3839
3840
3841
3842
3843
3844
3845
3846
3847
3848
3849
3850
3851
3852
3853
3854
3855
3856
3857
3858
3859
3860
3861
3862
3863
3864
3865
3866
3867
3868
3869
3870
3871
3872
3873
3874
3875
3876
3877
3878
3879
3880
3881
3882
3883
3884
3885
3886
3887
3888
3889
3890
3891
3892
3893
3894
3895
3896
3897
3898
3899
3900
3901
3902
3903
3904
3905
3906
3907
3908
3909
3910
3911
3912
3913
3914
3915
3916
3917
3918
3919
3920
3921
3922
3923
3924
3925
3926
3927
3928
3929
3930
3931
3932
3933
3934
3935
3936
3937
3938
3939
3940
3941
3942
3943
3944
3945
3946
3947
3948
3949
3950
3951
3952
3953
3954
3955
3956
3957
3958
3959
3960
3961
3962
3963
3964
3965
3966
3967
3968
3969
3970
3971
3972
3973
3974
3975
3976
3977
3978
3979
3980
3981
3982
3983
3984
3985
3986
3987
3988
3989
3990
3991
3992
3993
3994
3995
3996
3997
3998
3999
4000
4001
4002
4003
4004
4005
4006
4007
4008
4009
4010
4011
4012
4013
4014
4015
4016
4017
4018
4019
4020
4021
4022
4023
4024
4025
4026
4027
4028
4029
4030
4031
4032
4033
4034
4035
4036
4037
4038
4039
4040
4041
4042
4043
4044
4045
4046
4047
4048
4049
4050
4051
4052
4053
4054
4055
4056
4057
4058
4059
4060
4061
4062
4063
4064
4065
4066
4067
4068
4069
4070
4071
4072
4073
4074
4075
4076
4077
4078
4079
4080
4081
4082
4083
4084
4085
4086
4087
4088
4089
4090
4091
4092
4093
4094
4095
4096
4097
4098
4099
4100
4101
4102
4103
4104
4105
4106
4107
4108
4109
4110
4111
4112
4113
4114
4115
4116
4117
4118
4119
4120
4121
4122
4123
4124
4125
4126
4127
4128
4129
4130
4131
4132
4133
4134
4135
4136
4137
4138
4139
4140
4141
4142
4143
4144
4145
4146
4147
4148
4149
4150
4151
4152
4153
4154
4155
4156
4157
4158
4159
4160
4161
4162
4163
4164
4165
4166
4167
4168
4169
4170
4171
4172
4173
4174
4175
4176
4177
4178
4179
4180
4181
4182
4183
4184
4185
4186
4187
4188
4189
4190
4191
4192
4193
4194
4195
4196
4197
4198
4199
4200
4201
4202
4203
4204
4205
4206
4207
4208
4209
4210
4211
4212
4213
4214
4215
4216
4217
4218
4219
4220
4221
4222
4223
4224
4225
4226
4227
4228
4229
4230
4231
4232
4233
4234
4235
4236
4237
4238
4239
4240
4241
4242
4243
4244
4245
4246
4247
4248
4249
4250
4251
4252
4253
4254
4255
4256
4257
4258
4259
4260
4261
4262
4263
4264
4265
4266
4267
4268
4269
4270
4271
4272
4273
4274
4275
4276
4277
4278
4279
4280
4281
4282
4283
4284
4285
4286
4287
4288
4289
4290
4291
4292
4293
4294
4295
4296
4297
4298
4299
4300
4301
4302
4303
4304
4305
4306
4307
4308
4309
4310
4311
4312
4313
4314
4315
4316
4317
4318
4319
4320
4321
4322
4323
4324
4325
4326
4327
4328
4329
4330
4331
4332
4333
4334
4335
4336
4337
4338
4339
4340
4341
4342
4343
4344
4345
4346
4347
4348
4349
4350
4351
4352
4353
4354
4355
4356
4357
4358
4359
4360
4361
4362
4363
4364
4365
4366
4367
4368
4369
4370
4371
4372
4373
4374
4375
4376
4377
4378
4379
4380
4381
4382
4383
4384
4385
4386
4387
4388
4389
4390
4391
4392
4393
4394
4395
4396
4397
4398
4399
4400
4401
4402
4403
4404
4405
4406
4407
4408
4409
4410
4411
4412
4413
4414
4415
4416
4417
4418
4419
4420
4421
4422
4423
4424
4425
4426
4427
4428
4429
4430
4431
4432
4433
4434
4435
4436
4437
4438
4439
4440
4441
4442
4443
4444
4445
4446
4447
4448
4449
4450
4451
4452
4453
4454
4455
4456
4457
4458
4459
4460
4461
4462
4463
4464
4465
4466
4467
4468
4469
4470
4471
4472
4473
4474
4475
4476
4477
4478
4479
4480
4481
4482
4483
4484
4485
4486
4487
4488
4489
4490
4491
4492
4493
4494
4495
4496
4497
4498
4499
4500
4501
4502
4503
4504
4505
4506
4507
4508
4509
4510
4511
4512
4513
4514
4515
4516
4517
4518
4519
4520
4521
4522
4523
4524
4525
4526
4527
4528
4529
4530
4531
4532
4533
4534
4535
4536
4537
4538
4539
4540
4541
4542
4543
4544
4545
4546
4547
4548
4549
4550
4551
4552
4553
4554
4555
4556
4557
4558
4559
4560
4561
4562
4563
4564
4565
4566
4567
4568
4569
4570
4571
4572
4573
4574
4575
4576
4577
4578
4579
4580
4581
4582
4583
4584
4585
4586
4587
4588
4589
4590
4591
4592
4593
4594
4595
4596
4597
4598
4599
4600
4601
4602
4603
4604
4605
4606
4607
4608
4609
4610
4611
4612
4613
4614
4615
4616
4617
4618
4619
4620
4621
4622
4623
4624
4625
4626
4627
4628
4629
4630
4631
4632
4633
4634
4635
4636
4637
4638
4639
4640
4641
4642
4643
4644
4645
4646
4647
4648
4649
4650
4651
4652
4653
4654
4655
4656
4657
4658
4659
4660
4661
4662
4663
4664
4665
4666
4667
4668
4669
4670
4671
4672
4673
4674
4675
4676
4677
4678
4679
4680
4681
4682
4683
4684
4685
4686
4687
4688
4689
4690
4691
4692
4693
4694
4695
4696
4697
4698
4699
4700
4701
4702
4703
4704
4705
4706
4707
4708
4709
4710
4711
4712
4713
4714
4715
4716
4717
4718
4719
4720
4721
4722
4723
4724
4725
4726
4727
4728
4729
4730
4731
4732
4733
4734
4735
4736
4737
4738
4739
4740
4741
4742
4743
4744
4745
4746
4747
4748
4749
4750
4751
4752
4753
4754
4755
4756
4757
4758
4759
4760
4761
4762
4763
4764
4765
4766
4767
4768
4769
4770
4771
4772
4773
4774
4775
4776
4777
4778
4779
4780
4781
4782
4783
4784
4785
4786
4787
4788
4789
4790
4791
4792
4793
4794
4795
4796
4797
4798
4799
4800
4801
4802
4803
4804
4805
4806
4807
4808
4809
4810
4811
4812
4813
4814
4815
4816
4817
4818
4819
4820
4821
4822
4823
4824
4825
4826
4827
4828
4829
4830
4831
4832
4833
4834
4835
4836
4837
4838
4839
4840
4841
4842
4843
4844
4845
4846
4847
4848
4849
4850
4851
4852
4853
4854
4855
4856
4857
4858
4859
4860
4861
4862
4863
4864
4865
4866
4867
4868
4869
4870
4871
4872
4873
4874
4875
4876
4877
4878
4879
4880
4881
4882
4883
4884
4885
4886
4887
4888
4889
4890
4891
4892
4893
4894
4895
4896
4897
4898
4899
4900
4901
4902
4903
4904
4905
4906
4907
4908
4909
4910
4911
4912
4913
4914
4915
4916
4917
4918
4919
4920
4921
4922
4923
4924
4925
4926
4927
4928
4929
4930
4931
4932
4933
4934
4935
4936
4937
4938
4939
4940
4941
4942
4943
4944
4945
4946
4947
4948
4949
4950
4951
4952
4953
4954
4955
4956
4957
4958
4959
4960
4961
4962
4963
4964
4965
4966
4967
4968
4969
4970
4971
4972
4973
4974
4975
4976
4977
4978
4979
4980
4981
4982
4983
4984
4985
4986
4987
4988
4989
4990
4991
4992
4993
4994
4995
4996
4997
4998
4999
5000
5001
5002
5003
5004
5005
5006
5007
5008
5009
5010
5011
5012
5013
5014
5015
5016
5017
5018
5019
5020
5021
5022
5023
5024
5025
5026
5027
5028
5029
5030
5031
5032
5033
5034
5035
5036
5037
5038
5039
5040
5041
5042
5043
5044
5045
5046
5047
5048
5049
5050
5051
5052
5053
5054
5055
5056
5057
5058
5059
5060
5061
5062
5063
5064
5065
5066
5067
5068
5069
5070
5071
5072
5073
5074
5075
5076
5077
5078
5079
5080
5081
5082
5083
5084
5085
5086
5087
5088
5089
5090
5091
5092
5093
5094
5095
5096
5097
5098
5099
5100
5101
5102
5103
5104
5105
5106
5107
5108
5109
5110
5111
5112
5113
5114
5115
5116
5117
5118
5119
5120
5121
5122
5123
5124
5125
5126
5127
5128
5129
5130
5131
5132
5133
5134
5135
5136
5137
5138
5139
5140
5141
5142
5143
5144
5145
5146
5147
5148
5149
5150
5151
5152
5153
5154
5155
5156
5157
5158
5159
5160
5161
5162
5163
5164
5165
5166
5167
5168
5169
5170
5171
5172
5173
5174
5175
5176
5177
5178
5179
5180
5181
5182
5183
5184
5185
5186
5187
5188
5189
5190
5191
5192
5193
5194
5195
5196
5197
5198
5199
5200
5201
5202
5203
5204
5205
5206
5207
5208
5209
5210
5211
5212
5213
5214
5215
5216
5217
5218
5219
5220
5221
5222
5223
5224
5225
5226
5227
5228
5229
5230
5231
5232
5233
5234
5235
5236
5237
5238
5239
5240
5241
5242
5243
5244
5245
5246
5247
5248
5249
5250
5251
5252
5253
5254
5255
5256
5257
5258
5259
5260
5261
5262
5263
5264
5265
5266
5267
5268
5269
5270
5271
5272
5273
5274
5275
5276
5277
5278
5279
5280
5281
5282
5283
5284
5285
5286
5287
5288
5289
5290
5291
5292
5293
5294
5295
5296
5297
5298
5299
5300
5301
5302
5303
5304
5305
5306
5307
5308
5309
5310
5311
5312
5313
5314
5315
5316
5317
5318
5319
5320
5321
5322
5323
5324
5325
5326
5327
5328
5329
5330
5331
5332
5333
5334
5335
5336
5337
5338
5339
5340
5341
5342
5343
5344
5345
5346
5347
5348
5349
5350
5351
5352
5353
5354
5355
5356
5357
5358
5359
5360
5361
5362
5363
5364
5365
5366
5367
5368
5369
5370
5371
5372
5373
5374
5375
5376
5377
5378
5379
5380
5381
5382
5383
5384
5385
5386
5387
5388
5389
5390
5391
5392
5393
5394
5395
5396
5397
5398
5399
5400
5401
5402
5403
5404
5405
5406
5407
5408
5409
5410
5411
5412
5413
5414
5415
5416
5417
5418
5419
5420
5421
5422
5423
5424
5425
5426
5427
5428
5429
5430
5431
5432
5433
5434
5435
5436
5437
5438
5439
5440
5441
5442
5443
5444
5445
5446
5447
5448
5449
5450
5451
5452
5453
5454
5455
5456
5457
5458
5459
5460
5461
5462
5463
5464
5465
5466
5467
5468
5469
5470
5471
5472
5473
5474
5475
5476
5477
5478
5479
5480
5481
5482
5483
5484
5485
5486
5487
5488
5489
5490
5491
5492
5493
5494
5495
5496
5497
5498
5499
5500
5501
5502
5503
5504
5505
5506
5507
5508
5509
5510
5511
5512
5513
5514
5515
5516
5517
5518
5519
5520
5521
5522
5523
5524
5525
5526
5527
5528
5529
5530
5531
5532
5533
5534
5535
5536
5537
5538
5539
5540
5541
5542
5543
5544
5545
5546
5547
5548
5549
5550
5551
5552
5553
5554
5555
5556
5557
5558
5559
5560
5561
5562
5563
5564
5565
5566
5567
5568
5569
5570
5571
5572
5573
5574
5575
5576
5577
5578
5579
5580
5581
5582
5583
5584
5585
5586
5587
5588
5589
5590
5591
5592
5593
5594
5595
5596
5597
5598
5599
5600
5601
5602
5603
5604
5605
5606
5607
5608
5609
5610
5611
5612
5613
5614
5615
5616
5617
5618
5619
5620
5621
5622
5623
5624
5625
5626
5627
5628
5629
5630
5631
5632
5633
5634
5635
5636
5637
5638
5639
5640
5641
5642
5643
5644
5645
5646
5647
5648
5649
5650
5651
5652
5653
5654
5655
5656
5657
5658
5659
5660
5661
5662
5663
5664
5665
5666
5667
5668
5669
5670
5671
5672
5673
5674
5675
5676
5677
5678
5679
5680
5681
5682
5683
5684
5685
5686
5687
5688
5689
5690
5691
5692
5693
5694
5695
5696
5697
5698
5699
5700
5701
5702
5703
5704
5705
5706
5707
5708
5709
5710
5711
5712
5713
5714
5715
5716
5717
5718
5719
5720
5721
5722
5723
5724
5725
5726
5727
5728
5729
5730
5731
5732
5733
5734
5735
5736
5737
5738
5739
5740
5741
5742
5743
5744
5745
5746
5747
5748
5749
5750
5751
5752
5753
5754
5755
5756
5757
5758
5759
5760
5761
5762
5763
5764
5765
5766
5767
5768
5769
5770
5771
5772
5773
5774
5775
5776
5777
5778
5779
5780
5781
5782
5783
5784
5785
5786
5787
5788
5789
5790
5791
5792
5793
5794
5795
5796
5797
5798
5799
5800
5801
5802
5803
5804
5805
5806
5807
5808
5809
5810
5811
5812
5813
5814
5815
5816
5817
5818
5819
5820
5821
5822
5823
5824
5825
5826
5827
5828
5829
5830
5831
5832
5833
5834
5835
5836
5837
5838
5839
5840
5841
5842
5843
5844
5845
5846
5847
5848
5849
5850
5851
5852
5853
5854
5855
5856
5857
5858
5859
5860
5861
5862
5863
5864
5865
5866
5867
5868
5869
5870
5871
5872
5873
5874
5875
5876
5877
5878
5879
5880
5881
5882
5883
5884
5885
5886
5887
5888
5889
5890
5891
5892
5893
5894
5895
5896
5897
5898
5899
5900
5901
5902
5903
5904
5905
5906
5907
5908
5909
5910
5911
5912
5913
5914
5915
5916
5917
5918
5919
5920
5921
5922
5923
5924
5925
5926
5927
5928
5929
5930
5931
5932
5933
5934
5935
5936
5937
5938
5939
5940
5941
5942
5943
5944
5945
5946
5947
5948
5949
5950
5951
5952
5953
5954
5955
5956
5957
5958
5959
5960
5961
5962
5963
5964
5965
5966
5967
5968
5969
5970
5971
5972
5973
5974
5975
5976
5977
5978
5979
5980
5981
5982
5983
5984
5985
5986
5987
5988
5989
5990
5991
5992
5993
5994
5995
5996
5997
5998
5999
6000
6001
6002
6003
6004
6005
6006
6007
6008
6009
6010
6011
6012
6013
6014
6015
6016
6017
6018
6019
6020
6021
6022
6023
6024
6025
6026
6027
6028
6029
6030
6031
6032
6033
6034
6035
6036
6037
6038
6039
6040
6041
6042
6043
6044
6045
6046
6047
6048
6049
6050
6051
6052
6053
6054
6055
6056
6057
6058
6059
6060
6061
6062
6063
6064
6065
6066
6067
6068
6069
6070
6071
6072
6073
6074
6075
6076
6077
6078
6079
6080
6081
6082
6083
6084
6085
6086
6087
6088
6089
6090
6091
6092
6093
6094
6095
6096
6097
6098
6099
6100
6101
6102
6103
6104
6105
6106
6107
6108
6109
6110
6111
6112
6113
6114
6115
6116
6117
6118
6119
6120
6121
6122
6123
6124
6125
6126
6127
6128
6129
6130
6131
6132
6133
6134
6135
6136
6137
6138
6139
6140
6141
6142
6143
6144
6145
6146
6147
6148
6149
6150
6151
6152
6153
6154
6155
6156
6157
6158
6159
6160
6161
6162
6163
6164
6165
6166
6167
6168
6169
6170
6171
6172
6173
6174
6175
6176
6177
6178
6179
6180
6181
6182
6183
6184
6185
6186
6187
6188
6189
6190
6191
6192
6193
6194
6195
6196
6197
6198
6199
6200
6201
6202
6203
6204
6205
6206
6207
6208
6209
6210
6211
6212
6213
6214
6215
6216
6217
6218
6219
6220
6221
6222
6223
6224
6225
6226
6227
6228
6229
6230
6231
6232
6233
6234
6235
6236
6237
6238
6239
6240
6241
6242
6243
6244
6245
6246
6247
6248
6249
6250
6251
6252
6253
6254
6255
6256
6257
6258
6259
6260
6261
6262
6263
6264
6265
6266
6267
6268
6269
6270
6271
6272
6273
6274
6275
6276
6277
6278
6279
6280
6281
6282
6283
6284
6285
6286
6287
6288
6289
6290
6291
6292
6293
6294
6295
6296
6297
6298
6299
6300
6301
6302
6303
6304
6305
6306
6307
6308
6309
6310
6311
6312
6313
6314
6315
6316
6317
6318
6319
6320
6321
6322
6323
6324
6325
6326
6327
6328
6329
6330
6331
6332
6333
6334
6335
6336
6337
6338
6339
6340
6341
6342
6343
6344
6345
6346
6347
6348
6349
6350
6351
6352
6353
6354
6355
6356
6357
6358
6359
6360
6361
6362
6363
6364
6365
6366
6367
6368
6369
6370
6371
6372
6373
6374
6375
6376
6377
6378
6379
6380
6381
6382
6383
6384
6385
6386
6387
6388
6389
6390
6391
6392
6393
6394
6395
6396
6397
6398
6399
6400
6401
6402
6403
6404
6405
6406
6407
6408
6409
6410
6411
6412
6413
6414
6415
6416
6417
6418
6419
6420
6421
6422
6423
6424
6425
6426
6427
6428
6429
6430
6431
6432
6433
6434
6435
6436
6437
6438
6439
6440
6441
6442
6443
6444
6445
6446
6447
6448
6449
6450
6451
6452
6453
6454
6455
6456
6457
6458
6459
6460
6461
6462
6463
6464
6465
6466
6467
6468
6469
6470
6471
6472
6473
6474
6475
6476
6477
6478
6479
6480
6481
6482
6483
6484
6485
6486
6487
6488
6489
6490
6491
6492
6493
6494
6495
6496
6497
6498
6499
6500
6501
6502
6503
6504
6505
6506
6507
6508
6509
6510
6511
6512
6513
6514
6515
6516
6517
6518
6519
6520
6521
6522
6523
6524
6525
6526
6527
6528
6529
6530
6531
6532
6533
6534
6535
6536
6537
6538
6539
6540
6541
6542
6543
6544
6545
6546
6547
6548
6549
6550
6551
6552
6553
6554
6555
6556
6557
6558
6559
6560
6561
6562
6563
6564
6565
6566
6567
6568
6569
6570
6571
6572
6573
6574
6575
6576
6577
6578
6579
6580
6581
6582
6583
6584
6585
6586
6587
6588
6589
6590
6591
6592
6593
6594
6595
6596
6597
6598
6599
6600
6601
6602
6603
6604
6605
6606
6607
6608
6609
6610
6611
6612
6613
6614
6615
6616
6617
6618
6619
6620
6621
6622
6623
6624
6625
6626
6627
6628
6629
6630
6631
6632
6633
6634
6635
6636
6637
6638
6639
6640
6641
6642
6643
6644
6645
6646
6647
6648
6649
6650
6651
6652
6653
6654
6655
6656
6657
6658
6659
6660
6661
6662
6663
6664
6665
6666
6667
6668
6669
6670
6671
6672
6673
6674
6675
6676
6677
6678
6679
6680
6681
6682
6683
6684
6685
6686
6687
6688
6689
6690
6691
6692
6693
6694
6695
6696
6697
6698
6699
6700
6701
6702
6703
6704
6705
6706
6707
6708
6709
6710
6711
6712
6713
6714
6715
6716
6717
6718
6719
6720
6721
6722
6723
6724
6725
6726
6727
6728
6729
6730
6731
6732
6733
6734
6735
6736
6737
6738
6739
6740
6741
6742
6743
6744
6745
6746
6747
6748
6749
6750
6751
6752
6753
6754
6755
6756
6757
6758
6759
6760
6761
6762
6763
6764
6765
6766
6767
6768
6769
6770
6771
6772
6773
6774
6775
6776
6777
6778
6779
6780
6781
6782
6783
6784
6785
6786
6787
6788
6789
6790
6791
6792
6793
6794
6795
6796
6797
6798
6799
6800
6801
6802
6803
6804
6805
6806
6807
6808
6809
6810
6811
6812
6813
6814
6815
6816
6817
6818
6819
6820
6821
6822
6823
6824
6825
6826
6827
6828
6829
6830
6831
6832
6833
6834
6835
6836
6837
6838
6839
6840
6841
6842
6843
6844
6845
6846
6847
6848
6849
6850
6851
6852
6853
6854
6855
6856
6857
6858
6859
6860
6861
6862
6863
6864
6865
6866
6867
6868
6869
6870
6871
6872
6873
6874
6875
6876
6877
6878
6879
6880
6881
6882
6883
6884
6885
6886
6887
6888
6889
6890
6891
6892
6893
6894
6895
6896
6897
6898
6899
6900
6901
6902
6903
6904
6905
6906
6907
6908
6909
6910
6911
6912
6913
6914
6915
6916
6917
6918
6919
6920
6921
6922
6923
6924
6925
6926
6927
6928
6929
6930
6931
6932
6933
6934
6935
6936
6937
6938
6939
6940
6941
6942
6943
6944
6945
6946
6947
6948
6949
6950
6951
6952
6953
6954
6955
6956
6957
6958
6959
6960
6961
6962
6963
6964
6965
6966
6967
6968
6969
6970
6971
6972
6973
6974
6975
6976
6977
6978
6979
6980
6981
6982
6983
6984
6985
6986
6987
6988
6989
6990
6991
6992
6993
6994
6995
6996
6997
6998
6999
7000
7001
7002
7003
7004
7005
7006
7007
7008
7009
7010
7011
7012
7013
7014
7015
7016
7017
7018
7019
7020
7021
7022
7023
7024
7025
7026
7027
7028
7029
7030
7031
7032
7033
7034
7035
7036
7037
7038
7039
7040
7041
7042
7043
7044
7045
7046
7047
7048
7049
7050
7051
7052
7053
7054
7055
7056
7057
7058
7059
7060
7061
7062
7063
7064
7065
7066
7067
7068
7069
7070
7071
7072
7073
7074
7075
7076
7077
7078
7079
7080
7081
7082
7083
7084
7085
7086
7087
7088
7089
7090
7091
7092
7093
7094
7095
7096
7097
7098
7099
7100
7101
7102
7103
7104
7105
7106
7107
7108
7109
7110
7111
7112
7113
7114
7115
7116
7117
7118
7119
7120
7121
7122
7123
7124
7125
7126
7127
7128
7129
7130
7131
7132
7133
7134
7135
7136
7137
7138
7139
7140
7141
7142
7143
7144
7145
7146
7147
7148
7149
7150
7151
7152
7153
7154
7155
7156
7157
7158
7159
7160
7161
7162
7163
7164
7165
7166
7167
7168
7169
7170
7171
7172
7173
7174
7175
7176
7177
7178
7179
7180
7181
7182
7183
7184
7185
7186
7187
7188
7189
7190
7191
7192
7193
7194
7195
7196
7197
7198
7199
7200
7201
7202
7203
7204
7205
7206
7207
7208
7209
7210
7211
7212
7213
7214
7215
7216
7217
7218
7219
7220
7221
7222
7223
7224
7225
7226
7227
7228
7229
7230
7231
7232
7233
7234
7235
7236
7237
7238
7239
7240
7241
7242
7243
7244
7245
7246
7247
7248
7249
7250
7251
7252
7253
7254
7255
7256
7257
7258
7259
7260
7261
7262
7263
7264
7265
7266
7267
7268
7269
7270
7271
7272
7273
7274
7275
7276
7277
7278
7279
7280
7281
7282
7283
7284
7285
7286
7287
7288
7289
7290
7291
7292
7293
7294
7295
7296
7297
7298
7299
7300
7301
7302
7303
7304
7305
7306
7307
7308
7309
7310
7311
7312
7313
7314
7315
7316
7317
7318
7319
7320
7321
7322
7323
7324
7325
7326
7327
7328
7329
7330
7331
7332
7333
7334
7335
7336
7337
7338
7339
7340
7341
7342
7343
7344
7345
7346
7347
7348
7349
7350
7351
7352
7353
7354
7355
7356
7357
7358
7359
7360
7361
7362
7363
7364
7365
7366
7367
7368
7369
7370
7371
7372
7373
7374
7375
7376
7377
7378
7379
7380
7381
7382
7383
7384
7385
7386
7387
7388
7389
7390
7391
7392
7393
7394
7395
7396
7397
7398
7399
7400
7401
7402
7403
7404
7405
7406
7407
7408
7409
7410
7411
7412
7413
7414
7415
7416
7417
7418
7419
7420
7421
7422
7423
7424
7425
7426
7427
7428
7429
7430
7431
7432
7433
7434
7435
7436
7437
7438
7439
7440
7441
7442
7443
7444
7445
7446
7447
7448
7449
7450
7451
7452
7453
7454
7455
7456
7457
7458
7459
7460
7461
7462
7463
7464
7465
7466
7467
7468
7469
7470
7471
7472
7473
7474
7475
7476
7477
7478
7479
7480
7481
7482
7483
7484
7485
7486
7487
7488
7489
7490
7491
7492
7493
7494
7495
7496
7497
7498
7499
7500
7501
7502
7503
7504
7505
7506
7507
7508
7509
7510
7511
7512
7513
7514
7515
7516
7517
7518
7519
7520
7521
7522
7523
7524
7525
7526
7527
7528
7529
7530
7531
7532
7533
7534
7535
7536
7537
7538
7539
7540
7541
7542
7543
7544
7545
7546
7547
7548
7549
7550
7551
7552
7553
7554
7555
7556
7557
7558
7559
7560
7561
7562
7563
7564
7565
7566
7567
7568
7569
7570
7571
7572
7573
7574
7575
7576
7577
7578
7579
7580
7581
7582
7583
7584
7585
7586
7587
7588
7589
7590
7591
7592
7593
7594
7595
7596
7597
7598
7599
7600
7601
7602
7603
7604
7605
7606
7607
7608
7609
7610
7611
7612
7613
7614
7615
7616
7617
7618
7619
7620
7621
7622
7623
7624
7625
7626
7627
7628
7629
7630
7631
7632
7633
7634
7635
7636
7637
7638
7639
7640
7641
7642
7643
7644
7645
7646
7647
7648
7649
7650
7651
7652
7653
7654
7655
7656
7657
7658
7659
7660
7661
7662
7663
7664
7665
7666
7667
7668
7669
7670
7671
7672
7673
7674
7675
7676
7677
7678
7679
7680
7681
7682
7683
7684
7685
7686
7687
7688
7689
7690
7691
7692
7693
7694
7695
7696
7697
7698
7699
7700
7701
7702
7703
7704
7705
7706
7707
7708
7709
7710
7711
7712
7713
7714
7715
7716
7717
7718
7719
7720
7721
7722
7723
7724
7725
7726
7727
7728
7729
7730
7731
7732
7733
7734
7735
7736
7737
7738
7739
7740
7741
7742
7743
7744
7745
7746
7747
7748
7749
7750
7751
7752
7753
7754
7755
7756
7757
7758
7759
7760
7761
7762
7763
7764
7765
7766
7767
7768
7769
7770
7771
7772
7773
7774
7775
7776
7777
7778
7779
7780
7781
7782
7783
7784
7785
7786
7787
7788
7789
7790
7791
7792
7793
7794
7795
7796
7797
7798
7799
7800
7801
7802
7803
7804
7805
7806
7807
7808
7809
7810
7811
7812
7813
7814
7815
7816
7817
7818
7819
7820
7821
7822
7823
7824
7825
7826
7827
7828
7829
7830
7831
7832
7833
7834
7835
7836
7837
7838
7839
7840
7841
7842
7843
7844
7845
7846
7847
7848
7849
7850
7851
7852
7853
7854
7855
7856
7857
7858
7859
7860
7861
7862
7863
7864
7865
7866
7867
7868
7869
7870
7871
7872
7873
7874
7875
7876
7877
7878
7879
7880
7881
7882
7883
7884
7885
7886
7887
7888
7889
7890
7891
7892
7893
7894
7895
7896
7897
7898
7899
7900
7901
7902
7903
7904
7905
7906
7907
7908
7909
7910
7911
7912
7913
7914
7915
7916
7917
7918
7919
7920
7921
7922
7923
7924
7925
7926
7927
7928
7929
7930
7931
7932
7933
7934
7935
7936
7937
7938
7939
7940
7941
7942
7943
7944
7945
7946
7947
7948
7949
7950
7951
7952
7953
7954
7955
7956
7957
7958
7959
7960
7961
7962
7963
7964
7965
7966
7967
7968
7969
7970
7971
7972
7973
7974
7975
7976
7977
7978
7979
7980
7981
7982
7983
7984
7985
7986
7987
7988
7989
7990
7991
7992
7993
7994
7995
7996
7997
7998
7999
8000
8001
8002
8003
8004
8005
8006
8007
8008
8009
8010
8011
8012
8013
8014
8015
8016
8017
8018
8019
8020
8021
8022
8023
8024
8025
8026
8027
8028
8029
8030
8031
8032
8033
8034
8035
8036
8037
8038
8039
8040
8041
8042
8043
8044
8045
8046
8047
8048
8049
8050
8051
8052
8053
8054
8055
8056
8057
8058
8059
8060
8061
8062
8063
8064
8065
8066
8067
8068
8069
8070
8071
8072
8073
8074
8075
8076
8077
8078
8079
8080
8081
8082
8083
8084
8085
8086
8087
8088
8089
8090
8091
8092
8093
8094
8095
8096
8097
8098
8099
8100
8101
8102
8103
8104
8105
8106
8107
8108
8109
8110
8111
8112
8113
8114
8115
8116
8117
8118
8119
8120
8121
8122
8123
8124
8125
8126
8127
8128
8129
8130
8131
8132
8133
8134
8135
8136
8137
8138
8139
8140
8141
8142
8143
8144
8145
8146
8147
8148
8149
8150
8151
8152
8153
8154
8155
8156
8157
8158
8159
8160
8161
8162
8163
8164
8165
8166
8167
8168
8169
8170
8171
8172
8173
8174
8175
8176
8177
8178
8179
8180
8181
8182
8183
8184
8185
8186
8187
8188
8189
8190
8191
8192
8193
8194
8195
8196
8197
8198
8199
8200
8201
8202
8203
8204
8205
8206
8207
8208
8209
8210
8211
8212
8213
8214
8215
8216
8217
8218
8219
8220
8221
8222
8223
8224
8225
8226
8227
8228
8229
8230
8231
8232
8233
8234
8235
8236
8237
8238
8239
8240
8241
8242
8243
8244
8245
8246
8247
8248
8249
8250
8251
8252
8253
8254
8255
8256
8257
8258
8259
8260
8261
8262
8263
8264
8265
8266
8267
8268
8269
8270
8271
8272
8273
8274
8275
8276
8277
8278
8279
8280
8281
8282
8283
8284
8285
8286
8287
8288
8289
8290
8291
8292
8293
8294
8295
8296
8297
8298
8299
8300
8301
8302
8303
8304
8305
8306
8307
8308
8309
8310
8311
8312
8313
8314
8315
8316
8317
8318
8319
8320
8321
8322
8323
8324
8325
8326
8327
8328
8329
8330
8331
8332
8333
8334
8335
8336
8337
8338
8339
8340
8341
8342
8343
8344
8345
8346
8347
8348
8349
8350
8351
8352
8353
8354
8355
8356
8357
8358
8359
8360
8361
8362
8363
8364
8365
8366
8367
8368
8369
8370
8371
8372
8373
8374
8375
8376
8377
8378
8379
8380
8381
8382
8383
8384
8385
8386
8387
8388
8389
8390
8391
8392
8393
8394
8395
8396
8397
8398
8399
8400
8401
8402
8403
8404
8405
8406
8407
8408
8409
8410
8411
8412
8413
8414
8415
8416
8417
8418
8419
8420
8421
8422
8423
8424
8425
8426
8427
8428
8429
8430
8431
8432
8433
8434
8435
8436
8437
8438
8439
8440
8441
8442
8443
8444
8445
8446
8447
8448
8449
8450
8451
8452
8453
8454
8455
8456
8457
8458
8459
8460
8461
8462
8463
8464
8465
8466
8467
8468
8469
8470
8471
8472
8473
8474
8475
8476
8477
8478
8479
8480
8481
8482
8483
8484
8485
8486
8487
8488
8489
8490
8491
8492
8493
8494
8495
8496
8497
8498
8499
8500
8501
8502
8503
8504
8505
8506
8507
8508
8509
8510
8511
8512
8513
8514
8515
8516
8517
8518
8519
8520
8521
8522
8523
8524
8525
8526
8527
8528
8529
8530
8531
8532
8533
8534
8535
8536
8537
8538
8539
8540
8541
8542
8543
8544
8545
8546
8547
8548
8549
8550
8551
8552
8553
8554
8555
8556
8557
8558
8559
8560
8561
8562
8563
8564
8565
8566
8567
8568
8569
8570
8571
8572
8573
8574
8575
8576
8577
8578
8579
8580
8581
8582
8583
8584
8585
8586
8587
8588
8589
8590
8591
8592
8593
8594
8595
8596
8597
8598
8599
8600
8601
8602
8603
8604
8605
8606
8607
8608
8609
8610
8611
8612
8613
8614
8615
8616
8617
8618
8619
8620
8621
8622
8623
8624
8625
8626
8627
8628
8629
8630
8631
8632
8633
8634
8635
8636
8637
8638
8639
8640
8641
8642
8643
8644
8645
8646
8647
8648
8649
8650
8651
8652
8653
8654
8655
8656
8657
8658
8659
8660
8661
8662
8663
8664
8665
8666
8667
8668
8669
8670
8671
8672
8673
8674
8675
8676
8677
8678
8679
8680
8681
8682
8683
8684
8685
8686
8687
8688
8689
8690
8691
8692
8693
8694
8695
8696
8697
8698
8699
8700
8701
8702
8703
8704
8705
8706
8707
8708
8709
8710
8711
8712
8713
8714
8715
8716
8717
8718
8719
8720
8721
8722
8723
8724
8725
8726
8727
8728
8729
8730
8731
8732
8733
8734
8735
8736
8737
8738
8739
8740
8741
8742
8743
8744
8745
8746
8747
8748
8749
8750
8751
8752
8753
8754
8755
8756
8757
8758
8759
8760
8761
8762
8763
8764
8765
8766
8767
8768
8769
8770
8771
8772
8773
8774
8775
8776
8777
8778
8779
8780
8781
8782
8783
8784
8785
8786
8787
8788
8789
8790
8791
8792
8793
8794
8795
8796
8797
8798
8799
8800
8801
8802
8803
8804
8805
8806
8807
8808
%-------------------------------------------------------------------------------

% This file is part of Code_Saturne, a general-purpose CFD tool.
%
% Copyright (C) 1998-2016 EDF S.A.
%
% This program is free software; you can redistribute it and/or modify it under
% the terms of the GNU General Public License as published by the Free Software
% Foundation; either version 2 of the License, or (at your option) any later
% version.
%
% This program is distributed in the hope that it will be useful, but WITHOUT
% ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
% FOR A PARTICULAR PURPOSE.  See the GNU General Public License for more
% details.
%
% You should have received a copy of the GNU General Public License along with
% this program; if not, write to the Free Software Foundation, Inc., 51 Franklin
% Street, Fifth Floor, Boston, MA 02110-1301, USA.

%-------------------------------------------------------------------------------

\nopagebreak
%==================================
%==================================
\section{Introduction}
%==================================
%==================================

\CS is an application designed to solve the Navier-Stokes
equations in the cases of 2D, 2D axi-symmetric and 3D flows. Its main module is
designed for the simulation of flows which may be steady or
unsteady, laminar or turbulent, incompressible or potentially dilatable,
isothermal or not. Scalars and turbulent fluctuations of scalars can be taken into
account. The code includes specific modules, referred to as ``specific physics'',
for the treatment of Lagrangian particle tracking, semi-transparent radiative transfer,
gas combustion, pulverised coal combustion,
electricity effects (Joule effect and electric arcs) and compressible flows.

\CS is free software; you can redistribute it
and/or modify it under the terms of the GNU General Public License
as published by the Free Software Foundation; either version 2 of
the License, or (at your option) any later version.
\CS is distributed in the hope that it will be
useful, but WITHOUT ANY WARRANTY; without even the implied warranty
of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.\footnote{You should have
received a copy of the GNU General Public License
along with \CS; if not, write to the
Free Software Foundation, Inc.,
51 Franklin St, Fifth Floor,
Boston, MA  02110-1301  USA}


\CS relies on a finite volume discretisation and allows the use of
various mesh types which may be hybrid (containing several kinds of
elements) and may have structural non-conformities (hanging nodes).


\CS is composed of two main elements and an optional GUI,
as shown on \figurename~\ref{fig:elements}:
\begin{itemize}
\item the Kernel module is the numerical solver
\item the Preprocessor module is in charge of mesh import\\
\end{itemize}

\begin{figure}[!h]
\centerline{
\includegraphics*[width=14cm]{cs_components}}
\caption{\CS elements}\label{fig:elements}
\end{figure}

\indent\CS also relies on the PLE (Parallel Location and Exchange) library (developed by
the same team, under LGPL license) for the management of code coupling;
this library can also be used independently.

This document is a practical user guide for \CS version \verscs.
It is the result of the joint effort of
all the members in the development team.

This document provides practical information for the usage of \CS.
For more details about the algorithms and their numerical implementation,
please refer to the reports
 \cite{ijvf}, \cite{boucker00} and \cite{mechitoua98},
and to the theoretical documentation \cite{theory}.

The latest updated version of this document is available on-line with the version of \CS
and accessible through the command
\texttt{code\_saturne info --guide theory}.

This document first
presents all the necessary elements to run a calculation
with \CS version \verscs. It then lists all the variables of the code
which may be useful for more advanced users.
The user subroutines of all the modules within the code are then documented.
Eventually, for each keyword and user-modifiable parameter in the code,
their definition, allowed values, default values and conditions for use are given.
These keywords and parameters are grouped under headings
based on their function. An alphabetical index is also given at the end of
the document for easier reference.

%==================================
\section{Quick start}
%==================================
%==================================
\subsection{Running a calculation}
%==================================
We assume in this section that the user has at his disposal the calculation data file (calculation set up) or already prepared it following for instance the step-by-step guidance provided in \CS tutorial. The steps described below are intended to provide the user a way to run quickly on a workstation a calculation through the Graphical User Interface (GUI).

The first thing to do before running \CS is to define an alias to the \texttt{code\_saturne} script
(see \S\ref{sec:prg_environementCS}), for example:
\begin{center}
\texttt{alias cs='\$\{prefix\}/bin/code\_saturne'}.
\end{center}
When using the \emph{bash} shell, a completion file may be sourced so as to
allow for syntax auto-completion:
\begin{center}
\texttt{source \$\{prefix\}/etc/bash\_completion.d/code\_saturne'}.
\end{center}
The second thing is to prepare the computation directories. For instance, the study directory \texttt{T\_JUNCTION}, containing a single calculation directory CASE1, will be created by typing the command (see \S\ref{sec:prg_cscreate}):\
\begin{center}
\texttt{code\_saturne create -s T\_JUNCTION}\
\end{center}
The mesh files should be copied in the directory \texttt{MESH} (though they may also be selected from another directory, see \S\ref{sec:prg_stepbystepcalculation}),
and the Fortran user files necessary for the calculation in the directory \texttt{CASE1/SRC}.  Finally, the calculation data file \texttt{case\_name.xml} read by the GUI should be copied to the directory \texttt{CASE1/DATA}.
Once these steps completed, the user should go in the directory \texttt{CASE1/DATA} and type de command line \texttt{./SaturneGUI case\_name.xml} to load the calculation file into the interface. A window similar to \figurename\ref{fig:3_e1} will appear. Click on the heading ``Calculation management'', select the heading ``Prepare batch calculation'', see \figurename~\ref{fig:43_e1}. After having chosen the number of processors, press ``start calculation'' to run the calculation.

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=0.9\textwidth]{gui_case_dir}
\caption{Identity and paths}
\label{fig:3_e1}
\end{center}
\end{figure}

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=0.9\textwidth]{gui_prepare_execution}
\caption{Prepare execution}
\label{fig:43_e1}
\end{center}
\end{figure}

If no problem arises, the simulation results can be found in the directory \texttt{CASE1/RESU} and be read directly by {\em ParaView} or {\em EnSight} in \texttt{CASE1/RESU/<YYYYMMDD-hhmm>/postprocessing}. Calculation history can be found in the file \texttt{<YYYYMMDD-hhmm>/listing}.

%==================================
\subsection{Troubleshooting}
%==================================
If the calculation does not run properly, the user is advised to check the
following points in\\
\texttt{CASE1/RESU/<YYYYMMDD-hhmm>}:
\begin{list}{$\bullet$}{}
\item  if the calculation stops in the pre-processor, the user should check for error messages in the file \texttt{preprocessor*.log}.
\item if the problem is related to boundary conditions, the user should visualise the file \texttt{error.ensight} with {\em EnSight} or {\em ParaView},
\item if the calculation stops in the \CS core, the user should look for messages at the end of the files \texttt{listing} and \texttt{error*}. In addition, the user can track the following keywords in the listing; these are specific error signals:
  \begin{list}{-}{}
  \item  \texttt{SIGFPE}: a floating point exception occurred. It happens when there is a division by 0, when the calculation did not converge, or when a real number reached a value over $10^{300}$. Depending on the architecture \CS is running
on, this type of exception may be caught or ignored.
  \item  \texttt{SIGSEGV}: a memory error such as a segmentation violation occurred. An array may have exceeded its allocated memory size and a memory location in use was overwritten.
  \end{list}
In order to easily find the problem, it is also advised to use a debug version of \CS (see the installation documentation) in combination with the use of the valgrind tool (if it is installed). The use of valgrind can be specified in the GUI in the advanced options of the item ``Prepare batch calculation'' under the heading ``Calculation management'' or without the GUI, in the \texttt{cs\_user\_scripts.py} file (this file can be found in \texttt{DATA/REFERENCE} and should be copied in \texttt{DATA}, see \S\ref{sec:prg_stepbystepcalculation}).
\end{list}
%==================================
%==================================
\section{Practical information about \CS}
%==================================
%==================================

%==================================
\subsection{System Environment for \CS}
%==================================

%==================================
\subsubsection{Preliminary settings}
%==================================
\label{sec:prg_environementCS}

In order to use \CS, the user should define the following alias (in their \texttt{.bashrc},
or equivalent, or \texttt{.alias} file, depending on the environment):
\begin{center}
\texttt{alias cs='\$\{install\_directory\}/bin/code\_saturne'}
\end{center}
where \texttt{install\_directory} is the base directory where
\CS and its components have been installed\footnote{Without this step, using the absolute path is still possible}.

This step may be skipped if \texttt{\$\{install\_directory\}} is in a standard location (such as \texttt{/usr} or \texttt{/usr/local}.


%==================================
\subsubsection{Configuration file}
%==================================
A configuration file for \CS is available in \texttt{\$\{install\_directory\}/etc}. This file can be useful as a post-install step
for computing environments using a batch system, for separate front-end and compute systems (such as Blue Gene systems),
or for coupling with \syrthes 4 or \CA (see the installation documentation for more details).\\

A user may define a local configuration, by copying
\texttt{\$\{install\_directory\}/etc/code\_saturne.cfg} (if present)
or \texttt{\$\{install\_directory\}/etc/code\_saturne.cfg.template} to\\
\texttt{\$HOME/.code\_saturne.cfg}, then uncomment and define the applicable sections.\\

Note that this user configuration file's settings usually apply to all installed \CS versions.

Two options in the \texttt{.code\_saturne.cfg} file could be useful for the user:
\begin{list}{$\bullet$}{}
\item Set the temporary directory (see \S\ref{sec:prg_temporarydirectory} for more details on the temporary execution directory).
\item Set the mesh database directory: it is possible to indicate a path where meshes are stored.
In this case, the GUI will propose this directory automatically for mesh selection. Without the GUI, it is
then possible to fill in the \texttt{cs\_user\_scripts.py} file (see \S\ref{sec:prg_stepbystepcalculation})
with the name of the desired mesh of the database directory and the code will
find it automatically (be careful if you have the same name for a mesh in the database directory
and in the \texttt{MESH} directory, the mesh in \texttt{MESH} will be used).
\end{list}

%==================================
\subsubsection{Standard directory hierarchy}
%==================================
\label{sec:prg_architecture}%
The standard architecture for the simulation studies is:

\noindent
An optional study directory containing:
\begin{list}{$\bullet$}{}
\item A directory \texttt{MESH} containing the mesh(es)
      necessary for the study
\item A directory \texttt{POST} for the potential post-processing scripts (not
used directly by the code)
\item One or several calculation directories
\end{list}

\noindent
Every calculation directory contains:
\begin{list}{$\bullet$}{}
\item A directory \texttt{SRC} for the potential user subroutines
      necessary for the calculation
\item A directory \texttt{DATA} for the calculation data (data
      file from the interface, input profiles, thermo-chemical data, ...), the user script and the XML file.
\item A directory \texttt{SCRIPTS} for the launch script
\item A directory \texttt{RESU} for the results\\
To improve the calculation traceability, the files and directories
sent to \texttt{RESU} after a calculation are  placed in a subdirectory
named after that run's ``id'', which is by default based on the run date
and time, using the format: \texttt{YYYYMMDD-hhmm}.
It is also possible to force a specific run id, using the \texttt{--id}
option of \texttt{code\_saturne run}.
\end{list}

\noindent
In the standard cases, \texttt{RESU/<run\_id>} contains a
\texttt{postprocessing} directory with the post-processing
(visualization) files, a \texttt{restart} directory for the calculation
restart files, a \texttt{monitoring} directory for the files of chronological
record of the results at specific locations (probes),\\
\texttt{preprocessor.log} and \texttt{listing} files reporting the
Preprocessor and the Kernel execution. All files from the \texttt{DATA}
directory not in subdirectories are also copied. For a tracing of
the modifications in prior calculations, the user-subroutines used in
a calculation are stored in a \texttt{src\_saturne} subdirectory. The data files
(such as the XML Interface data file and thermo-chemical data files) and
launch script are also copied into the results directory. \texttt{compil.log} and
\texttt{summary} are respectively reports of the compilation stage and
general information on the calculation (type of machine, user,
version of the code, ...).

\begin{table}[h!t]
\begin{tabular}{lll}
\multicolumn{3}{l}{Below are typical contents of a case directory CASE1 in a study STUDY} \\
\multicolumn{2}{l}{\texttt{STUDY/CASE1/DATA:}}&{\bf \CS data}\\
&        \texttt{SaturneGUI}        &Graphical User Interface launch script\\
&        \texttt{study.xml}         &Graphical User Interface parameter file\\
&        \texttt{REFERENCE}         &Example of user scripts and meteorological\\
&                                   &  or thermochemical date files (used with the\\
&                                   & specific physics modules)\\
\multicolumn{2}{l}{\texttt{STUDY/CASE1/SRC:}}&{\bf \CS user subroutines }\\
&        \texttt{REFERENCE}         &  Available user subroutines\\
&        \texttt{EXAMPLES}          &  Examples of user subroutines\\
&        \texttt{cs\_user\_boundary\_conditions.f90}  &  User subroutines used for the present calculation\\
&        \texttt{cs\_user\_parameters.f90} &\\
\multicolumn{2}{l}{\texttt{STUDY/CASE1/RESU/}\emph{\texttt{YYYYMMDD-hhmm:}}}&{\bf Results} for the
                                                             calculation YYYYMMDD-hhmm\\
&        \texttt{postprocessing}    &Directory containing the \CS post-processing output\\
&                                   &in the {\em EnSight}, \med, or CGNS format (both volume and boundary);\\
&        \texttt{src\_saturne}      &copy of the \CS user subroutines used for the calculation\\
&        \texttt{monitoring}        &Directory containing the chronological records for \CS\\
&        \texttt{checkpoint}        &Directory containing the \CS restart files \\
&        \texttt{compile.log}       &Compilation log\\
&     \texttt{study.xml}            &Graphical User Interface parameter file used for the\\
&                                   &calculation\\
&        \texttt{runcase}           &Copy of the launch script used for the calculation\\
&        \texttt{preprocessor.log}  &Execution report for the \CS Preprocessor\\
&        \texttt{listing}           &Execution report for the Kernel module of \CS\\
&        \texttt{summary}           &General information (machine, user, version, ...)\\
\multicolumn{2}{l}{\texttt{STUDY/CASE1/SCRIPTS:}}&{\bf Launch script}\\
&        \texttt{runcase}           &Launch script (which may contain batch
                                     system keywords)\\
\end{tabular}
\end{table}

When running, the code
may use additional files or directories inside its execution directory, set
by the execution script, which include a \texttt{mesh\_input} file or directory,
as well as a \texttt{restart} directory (which is a link or copy of a previous
run's \texttt{checkpoint} directory), as well as a \texttt{run\_solver.sh}
script.

For coupled calculations, whether with \CS itself or \syrthes, each coupled
calculation domain is defined by its own directory (bearing the same name
as the domain), but results are placed in a \texttt{RESU\_COUPLING}
directory, with a subdirectory for each run, itself containing one
subdirectory per coupled domain. Coupled cases are run through
the standard the {\texttt{code\_saturne run} command, but require a coupling
parameters file (\texttt{coupling\_parameters.py}) specified using the
{\texttt{--coupling option}. The run command must be called from the toplevel
({\texttt{STUDY}) directory, so an additional {\texttt{STUDY/runcase} launch
script is used in this case. Note that case-local scripts (such as
{\texttt{STUDY/CASE1/SCRIPTS/runcase}) are still used by the master script to
determine which parameter file to use.

So in the coupled case, calculation results would not be placed in
\texttt{STUDY/CASE1/RESU/}\emph{\texttt{YYYYMMDD-hhmm}}, but in
\texttt{STUDY/RESU\_COUPLING/}\emph{\texttt{YYYYMMDD-hhmm}}\texttt{/CASE1}, with the \texttt{summary}
file being directly placed in \texttt{STUDY/RESU\_COUPLING/}\emph{\texttt{YYYYMMDD-hhmm}}
(as it references all coupled domains).

\begin{table}[h!t]
\begin{tabular}{lll}
\multicolumn{3}{l}{Below are typical additional contents with a coupled \syrthes
  case SOLID1 in a study STUDY} \\
\multicolumn{2}{l}{\texttt{STUDY/runcase\_coupling}}&{Coupled launch script}\\
\multicolumn{2}{l}{\texttt{STUDY/SOLID1/DATA:}}&{\bf \syrthes data}\\
&        \texttt{syrthes\_data.syd}          &\syrthes data file \\
&        \texttt{syrthes.py}                 &\syrthes script\\
&        \texttt{usr\_examples}              &\syrthes user subroutine examples\\
\multicolumn{2}{l}{\texttt{STUDY/RESU\_COUPLING/}\emph{\texttt{YYYYMMDD-hhmm}}\texttt{/SOLID1:}}&{\bf results
 (file names defined in \texttt{syrthes.env})}\\
&        \texttt{src}                &\syrthes user subroutines
                                     used in the calculation\\
&        \texttt{compile.log}        &\syrthes compilation report\\
&        \texttt{listsyr}            &Execution log\\
&        \texttt{geoms}              &\syrthes \ solid geometry file\\
&        \texttt{histos1}            &\syrthes chronological records at
                                               specified monitoring points\\
&        \texttt{resus1}             &\syrthes calculation restart file (1 time step)\\
&        \texttt{resusc1}            &\syrthes chronological solid
                                      post-processing file\\
&                                    &(may be transformed into the {\em EnSight}\\
&                                    & or {\em MED} format with the {\em syrthes4ensight}\\
&                                    & or {\em syrthes4med30} utility)\\
\end{tabular}
\end{table}

%==================================
\subsubsection{\CS Kernel library files}
%==================================
\label{sec:prg_library}%
Information about the content of the \CS base directories is given below. It
is not of vital interest for the user, but given only as general
information. Indeed, the case preparer command \texttt{code\_saturne~create}
automatically extracts the necessary files and prepares the launch script
without the user having to go directly into the \CS base directories
(see \S\ref{sec:prg_cscreate}).
The \texttt{code\_saturne~info} command gives direct
access to the most needed information (especially the user's and programmer's
guides and the tutorial) without the user having to look for them in the \CS
directories.

The subdirectories \texttt{\{install\_directory\}/lib} and \texttt{\{install\_directory\}/bin }
contain the libraries and compiled executables respectively.

The data files (for instance thermochemical data) are located in the
directory \texttt{data}.

The user subroutines are available in the directory \texttt{users},
under subdirectories corresponding to each module:
\texttt{base} (general routines),
\texttt{cfbl} (compressible flows),
\texttt{cogz} (gas combustion),
\texttt{cplv} (pulverised coal combustion),
\texttt{ctwr} (cooling towers modelling),
\texttt{elec} (electric module),
\texttt{fuel} (heavy fuel oil combustion module),
\texttt{lagr} (Lagrangian module,
\texttt{pprt} (general specific physics routines) and
\texttt{rayt} (radiative heat transfer).
The case preparer command \texttt{code\_saturne~create} copies all these files
in the user directory \texttt{SRC/REFERENCE} during the case preparation.

The directory \texttt{bin} contains an example of the launch script, the
compilation parameter files and various utility programs.

%==================================
\subsection{Setting up and running a calculation}
%==================================

%==================================
\subsubsection{Step by step calculation}
%==================================
\label{sec:prg_stepbystepcalculation}%

This paragraph summarises the different steps which are necessary to
prepare and run a standard case:

\begin{list}{$\bullet$}{}

\item Check the version of \CS set for use in the environment variables
(\texttt{code\_saturne~info --version}). If it does not correspond to
the desired version, update the user profile or aliases to get the
required version, logging out of the session and in again if necessary (cf.
\S\ref{sec:prg_environementCS}).

\item Prepare the different directories using the \texttt{code\_saturne~create}
command (see \S\ref{sec:prg_cscreate}).

\item It is recommended to place the mesh(es) in the directory \texttt{MESH},
but they may be selected from other directories, either with the Graphical User Interface (GUI)
 or the \texttt{cs\_user\_scripts.py} file (see below). Make sure they are
in a format compliant with \CS (see \S\ref{sec:prg_meshes}). There can be
several meshes in case of mesh joining or coupling with
\syrthes\footnote{\syrthes 4 uses meshes composed of 4-node tetrahedra}.

\item Go to the directory \texttt{DATA} and launch the
      GUI using the command \texttt{./SaturneGUI}.

\item If not using the GUI, copy the
  \texttt{DATA/REFERENCE/cs\_user\_scripts.py} file to \texttt{DATA} and
  edit it, so that the correct run options and paths may be set. For advanced
  uses, this file may also be used in conjunction with the GUI. Just as with
  user Fortran subroutines below, settings defined in this file have priority
  over those defined in the GUI.

\item Place the necessary user subroutines in the directory \texttt{SRC} (see
\S\ref{sec:prg_ssprgutilis}). When not using the Interface, some subroutines are
compulsory.

\begin{list}{}{}

\item {\bf For all physics:}

    \begin{list}{}{}
        \item {\em compulsory without Graphical User Interface:}
        \begin{list}{-}{}
            \item \texttt{usipph} (in \texttt{cs\_user\_parameters.f90}) to
              specify the turbulence and temperature models

            \item \texttt{usipsu} (in \texttt{cs\_user\_parameters.f90}) to
              define most user parameters

            \item \texttt{cs\_user\_boundary\_conditions} to manage the boundary conditions
        \end{list}

        \item {\em  very useful without Graphical User Interface:}
        \begin{list}{-}{}
            \item \texttt{\texttt{cs\_user\_model.c}}
              (in \texttt{cs\_user\_parameters.c}) to
              define user scalars (species)

            \item \texttt{usipes} (in \texttt{cs\_user\_parameters.f90}) to
              define monitoring points and additional parameters for results outputs
        \end{list}

        \item {\em  very useful:}
        \begin{list}{-}{}
            \item \texttt{usphyv} (in
              \texttt{cs\_user\_physical\_properties.f90}) to manage variable physical
                  properties (fluid density, viscosity ...)

            \item \texttt{cs\_user\_initialization} to manage the non-standard initialisations
        \end{list}
    \end{list}

  \item{\bf For the ``gas combustion'' specific physics:}

    \begin{list}{}{}
        \item {\em compulsory without Graphical User Interface:}
        \begin{list}{-}{}
            \item \texttt{usppmo} (in \texttt{cs\_user\_parameters.f90})
               to select a specific physics module and combustion model
        \end{list}

        \item {\em very useful:}
        \begin{list}{-}{}
            \item  \texttt{cs\_user\_combustion}
                   (in \texttt{cs\_user\_parameters.f90}),
                   depending on the selected combustion model,
                   to specify the calculation options
                   for the variables
                   corresponding to combustion model
        \end{list}
    \end{list}

  \item{\bf For the ``pulverized fuel combustion'' specific physics:}

    \begin{list}{}{}
        \item {\em compulsory without Graphical User Interface:}
        \begin{list}{-}{}
            \item \texttt{usppmo} (in \texttt{cs\_user\_parameters.f90})
               to select the specific physics module
        \end{list}

        \item {\em very useful:}
        \begin{list}{-}{}
            \item  \texttt{cs\_user\_combustion}
               (in \texttt{cs\_user\_parameters.f90})
               to specify the calculation options
                   for the variables
                   corresponding to pulverized fuel combustion
        \end{list}
    \end{list}

or \texttt{cs\_user\_combustion}
  \item{\bf For the ``heavy fuel combustion'' specific physics:}

 (not accessible through the Graphical User Interface in version \verscs)
    \begin{list}{}{}
        \item {\em compulsory:}
        \begin{list}{-}{}
            \item \texttt{usppmo} (in \texttt{cs\_user\_parameters.f90})
               to select the specific physics module
            \item \texttt{cs\_user\_combustion}
               (in \texttt{cs\_user\_parameters.f90})
               to specify the calculation options
                   for the variables
                   corresponding to heavy fuel combustion
        \end{list}
    \end{list}


     \item{\bf For the ``atmospheric module'' specific physics:}

    \begin{list}{}{}
       \item {\em compulsory without Graphical User Interface:}
        \begin{list}{-}{}
            \item \texttt{usppmo} (in \texttt{cs\_user\_parameters.f90})
                  to select the specific physics module
        \end{list}

        \item {\em very useful:}
        \begin{list}{-}{}
            \item  \texttt{usati1} (in \texttt{cs\_user\_parameters.f90})
                   to manage the reading of the meteo file
           \item  \texttt{usadtv}or \texttt{usatsoil} (in \texttt{cs\_user\_atmospheric\_model.f90})
                   to manage the options to the specific physics

        \end{list}
    \end{list}


     \item{\bf For the ``electric module'' specific physics
      (Joule effect and electric arcs):}

    \begin{list}{}{}
       \item {\em compulsory without Graphical User Interface:}
        \begin{list}{-}{}
            \item \texttt{usppmo} (in \texttt{cs\_user\_parameters.f90})
                  to select the specific physics module

            \item \texttt{cs\_user\_initialization} to initialise the enthalpy in
                  case of Joule effect

            \item \texttt{cs\_user\_physical\_properties.c}
                  to define the physical
                  properties in case of Joule effect
        \end{list}

        \item {\em very useful:}
        \begin{list}{-}{}
            \item  \texttt{cs\_user\_model} and \texttt{cs\_user\_parameters} (in \texttt{cs\_user\_parameters.c})
                   to manage the options related
                   to the variables corresponding to the electric module

        \end{list}
    \end{list}

%     \item{\bf For the ``heavy fuel oil combustion module'' specific physics:}
%(not accessible through the Graphical User Interface in version \verscs)
%    \begin{list}{}{}
%        \item {\em compulsory:}
%        \begin{list}{-}{}
%            \item \texttt{usppmo} to select the specific physics module
%        \end{list}
%    \end{list}


    \item{\bf For the ``Lagrangian module'' (dispersed phase):}

(the continuous phase is managed in the same way as for a case of standard
physics)
    \begin{list}{}{}
        \item {\em compulsory without Graphical User Interface:}
        \begin{list}{-}{}
            \item \texttt{uslag1} to manage the calculation conditions

            \item \texttt{uslag2} to manage the boundary conditions for the
             dispersed phase

        \end{list}

    \end{list}

   \item {\bf For the ``compressible module'':}

    \begin{list}{}{}
        \item {\em compulsory without Graphical User Interface:}
        \begin{list}{-}{}
            \item \texttt{usppmo} (in \texttt{cs\_user\_parameters.f90})
                  to select the specific physics module
        \end{list}
        \item {\em very useful:}
        \begin{list}{-}{}
            \item \texttt{uscfx1} and \texttt{uscfx2} (in \texttt{cs\_user\_parameters.f90})
                  to manage the calculation parameters

%            \item \texttt{cs\_user\_boundary\_conditions} to manage the
%                  boundary conditions

           \item \texttt{usphyv} (in \texttt{cs\_user\_physical\_properties}
                  to manage the variable physical properties

        \end{list}
    \end{list}

\end{list}


A comprehensive list of the user subroutines and their instructions
      for use are given in \S\ref{sec:prg_ssprgutilis}.

\item If necessary, place in the directory \texttt{DATA} the different
      external data (input profiles, thermochemical data files, ...)

\item Prepare the launch script \texttt{runcase}, directly or through the
      Graphical Interface (see \S\ref{sec:prg_runcase}), or prepare the
      \texttt{DATA/cs\_user\_scripts.py} file.

\item Run the calculation and analyse the results

\item If necessary, purge the temporary files (in \texttt{RESU/<run\_id>} or
      \texttt{<scratch>/<run\_id>} directory) (see \S\ref{sec:prg_temporarydirectory}).
\end{list}


%==================================
\subsubsection{Temporary execution directory}
%==================================
\label{sec:prg_temporarydirectory}%
During a calculation, \CS may use a temporary directory for the compilation and
the execution if such a ``scratch'' directory is defined in the GUI, by setting the
\texttt{CS\_SCRATCHDIR} environment variable, or in the \texttt{code\_saturne.cfg} file.
In this case, it is only at the end of the compilation that the result files are only copied at the end in the directory
\texttt{RESU}. This is recommended if the compute environment includes different
file-systems, some better suited to data storage, others to intensive I/O.
If this is not the case, there is no point in running in a scratch directory
rather than the results directory, as this incurs additional file copies.

If the environment variable \texttt{CS\_SCRATCHDIR} is defined,
its value has priority over that defined in the preference file
so if necessary, it
is possible to define a setting specific to a given run using this mechanism.

\noindent
{\em WARNING: in case of an error, the temporary directories are not deleted
after a calculation, so that they may be used for debugging. They may then
accumulate and may hinder the correct operation of the machine.\\
\centerline{\bf It is therefore essential to remove them regularly.}}


%==================================
\subsubsection{Execution modes}
%==================================
\label{sec:prg_executionmodes}%
As explained before, \CS is composed of two main modules, the Preprocessor and the
Kernel (solver). The Preprocessor reads the meshes.
The resulting data is transferred to the Kernel through specific
files, named \texttt{mesh\_input}, or placed in a directory of that name when multiple meshes are imported. In a standard calculation, the files
are not copied from the temporary execution directory to the results directory,
as they have no interest for data analysis, and are considered ``internal''
files, whose format or contents is not guaranteed not to change between \CS versions.

Yet, the Preprocessor does not run in parallel and may require a
large amount of memory. The launch scripts therefore allows specifically
choosing which modules to run, either through the GUI or through the
\texttt{cs\_user\_scripts.py} file:

\hspace*{0.5cm} If a {mesh\_input} file or directory is defined (which may be
either a {mesh\_input} from a previous Preprocessor run or a {mesh\_output}
from a previous solver run), the script will copy or link it to
the execution directory, and the Preprocessor will not be rerun.

\hspace*{0.5cm} If \texttt{domain.exec\_kernel = False}, the Kernel will not
be run. This is useful when only the mesh import stage is required.

In a similar manner, the Kernel accepts several command-line options relative to execution mode, notably \texttt{domain.solver\_args\_kernel = '--preprocess'} or \texttt{'--quality'}, restricting the run to the preprocessing stages, or preprocessing stages augmented by mesh quality criteria computation. Whenever the preprocessing stages defined lead to an effective mesh modification, a \texttt{mesh\_output} file is produced, which can be used directly as an input for a successive calculation.

The GUI presents the range of options in the form of four execution modes:

\begin{list}{$\bullet$}{}

\item {\bf mesh import}: the Preprocessor is run to transform one or more meshes into an internal \texttt{mesh\_input} file (or directory in case of multiple meshes).

\item {\bf mesh preprocessing}: the Kernel is run in preprocessing mode, so as to handle all mesh modification operations, such as joining, periodicity, smoothing, \emph{etc.} If a \texttt{mesh\_input} file or directory is provided, it is used directly; otherwise, mesh import is run first.

\item {\bf mesh quality criteria}: similar to preprocessing, with the addition of mesh quality criteria computation, and post-processing output of those criteria. Some additional mesh consistency checks are also run.

\item {\bf standard}: this includes preprocessing, followed by a standard computation.

\end{list}

Note that to allow preprocessing in multiple passes, all defined preprocessing operations are run even on previously preprocessed meshes. In most cases, those will not produce additional changes (such as joining already joined meshes), but in the case of mesh smoothing, they might lead to small changes. So when using a previously preprocessed mesh it is recommended
not to define any preprocessing operations, so as to skip the preprocessing stage.

It is encouraged to separate the preprocessing and calculation runs, as
this not only speeds up calculations, but also ensures that the mesh is identical, regardless of the architecture or number of processors it is run on. Indeed, when running the same pre-processing stages such as mesh joining on a different machine or a different number of processors, very minor floating-point truncation errors may lead to very slightly different preprocessed meshes.

Note also that mesh partitioning is done directly by the Kernel.
Depending on the partitioning algorithm used, a partition map
(\texttt{partition\_output/domain\_number\_*}) may be output,
allowing the use of the same partitioning in future calculations.
By default, this file is output when using graph-based partitioners, which may
use randomization and do not guarantee a reproducible output, and is not output
when using a deterministic space-filling curve based partitioning.

If the code was built only with a serial partitioning library,
graph-based partitioning may best be run in a serial pre-processing stage.
In some cases, serial partitioning might also provide better partitioning
quality than parallel partitioning, so if both are available, comparing
the performance of the code may be worthwhile, at least for calculations
expected to run for many iterations.

%==================================
\subsubsection{Environment variables\label{sec:envcs}}
%==================================

Setting a few environment variables specific to \CS allows modifying
its default behaviour. The environment variables used by \CS
are described here:

\envvar{CS\_SCRATCHDIR}

Allows defining the execution directory (see
\S\ref{sec:prg_temporarydirectory}),
overriding the default path or settings from the global
or user \texttt{code\_saturne.cfg}.

\envvar{CS\_MPIEXEC\_OPTIONS}

This variable allows defining extra arguments to be passed to
the MPI execution command by the run scripts.
If this option is defined, it will have priority over the value defined
in the preference file (or by computed defaults), so if necessary, it
is possible to define a setting specific to a given run using this
mechanism. This may be useful when tuning the installation to a given
machine, for example experimenting MPI mapping and ``bind to core''
features.

%==================================
\subsubsection{Interactive modification of selected parameters}
%==================================
\label{sec:prg_control_file}%
During a calculation, it is possible to change the limit time step number
(\texttt{ntmabs}) specified through the GUI or in \texttt{cs\_user\_parameters.f90}.
To do so, a file named \texttt{control\_file} must be placed in the
execution directory (see \S\ref{sec:prg_temporarydirectory}).
The existence of this file is checked at the beginning of each time step.

To change the maximum number of time steps, this file must contain a line
indicating the value of the new limit number of time steps.\\
If this new limit has already been reached,
 \CS will stop
properly at the end of the current time step (the results and restart files
will be written correctly).\\
This procedure allows the user to stop a calculation in a clean and interactive
way whenever they wish.

The \texttt{control\_file} may also contain a few other commands,
allowing the user to force checkpointing or postprocessing at a given time step
or physical time, or to force an update of log and \texttt{listing} files.
The following commands are available (using the common notations $< >$ to
indicate a required argument, $[ ]$ to indicate an optional argument).

\begin{tabular}[top]{|p{6.5cm}%
                     |>{\PreserveBackslash\raggedright\hspace{0pt}}p{8.5cm}|}
\hline
\texttt{max\_time\_step}                   & $<$time\_step\_number$>$ \\
\texttt{max\_time\_value}                  & $<$time\_value$>$ \\
\hline
\texttt{checkpoint\_time\_step}            & $<$time\_step\_number$>$ \\
\texttt{checkpoint\_time\_value}           & $<$time\_value$>$ \\
\texttt{checkpoint\_wall\_time}            & $<$wall\_clock\_time$>$ \\
\texttt{checkpoint\_time\_step\_interval}   & $<$time\_step\_interval$>$ \\
\texttt{checkpoint\_time\_value\_interval}  & $<$time\_interval$>$  \\
\texttt{checkpoint\_wall\_time\_interval}   & $<$wall\_time\_interval$>$ \\
\hline
\texttt{control\_file\_wtime\_interval}     & $<$wall\_time\_interval$>$ \\
\hline
\texttt{flush}                             & $[$time\_step\_number$]$ \\
\hline
\texttt{postprocess\_time\_step}           & $<$time\_step\_number$>$ [writer\_id] \\
\texttt{postprocess\_time\_value}          & $<$time\_step\_value$>$ [writer\_id] \\
\hline
\texttt{time\_step\_limit}                 & $<$time\_step\_count$>$ \\
\hline
\end{tabular}

The \texttt{time\_step\_limit} differs from the \texttt{max\_time\_step} command,
in the sense that it allows reducing the maximum number of time steps,
but not increasing it. Also, in the case of a restart, it refers to the
number of additional time steps, not to the number of absolute time steps.

Note that for the \texttt{postprocess\_time\_*} options, the last argument
(\texttt{writer\_id}} is optional. If not defined, or 0, postprocessing
is activated for all writers; if specified, only the writer with the specified
id is affected. Also, postprocessing output by one ore more writers at a
future time step may be cancelled using the negative value of that time step.

For the \texttt{flush} option, the time step is also optional. If not
specified, logs and time plots are updated at the beginning of the
next time step. Also, if the \texttt{control\_file} is empty (such as
when created by the \texttt{touch control\_file} command on Unix/Linux
systems, a \texttt{flush} request for the next time step.

Multiple entries may be defined in this file, with one line per entry.

%==================================
\subsection{Case preparer}
%==================================
\label{sec:prg_cscreate}%
The case preparer command \texttt{code\_saturne~create} automatically creates a
study directory according to the typical architecture and copies and
pre-fills an example of calculation launch script.

The syntax of \texttt{code\_saturne~create} is as follows:

\noindent
\texttt{code\_saturne~create --study STUDY CASE\_NAME1 CASE\_NAME2...}\\
creates a study directory \texttt{STUDY} with case subdirectories
\texttt{CASE\_NAME1} and \texttt{CASE\_NAME2}...
If no case name is given, a default case directory called \texttt{CASE1} is
created.

\noindent
\texttt{code\_saturne~create --case Flow3 --case Flow4}\\
executed in the directory \texttt{STUDY} adds the case directories
\texttt{Flow3} and \texttt{Flow4}. Whenever multiple cases are created simultaneously, it is assumed they may be coupled, so a \texttt{runcase\_coupling} file and \texttt{RESU\_COUPLING} directory are also created.

In the directory \texttt{DATA}, the \texttt{code\_saturne~create} command
places a subdirectory \texttt{REFERENCE} containing examples of thermochemical
data files used for pulverised coal combustion, gas combustion, electric arcs, or
a meteo profile. The file to be used for the calculation must be copied directly
in the \texttt{DATA} directory and its name may either be unchanged, or
be referenced using the GUI or using the \texttt{usppmo} subroutine in
\texttt{cs\_user\_parameters.f90}. As a rule of thumb, all files in \texttt{DATA}
except for \texttt{SaturneGUI} are copied, but subdirectories are not.\\
The \texttt{code\_saturne~create} command also places in the directory
\texttt{DATA} the launch script for the Graphical User Interface:
\texttt{SaturneGUI}.

In the directory \texttt{SRC}, the \texttt{code\_saturne~create} command creates a
subdirectory \texttt{REFERENCE} containing all the available user subroutines,
and the subdirectory \texttt{EXAMPLES} containing examples of user subroutines.
Only the user subroutines placed directly under
the directory \texttt{SRC} will be considered. The others will be ignored.

In the directory \texttt{SCRIPTS}, the \texttt{code\_saturne~create} command copies an example of the launch script: \texttt{runcase}.
The XML file may be specified in the script (see \S\ref{sec:prg_runcase}),
and using the GUI sets it automatically.

\smallskip \noindent

%==================================
\subsection{Supported mesh and post-processing output formats
\label{sec:formats}}
%==================================

\CS supports multiple mesh formats, all of these having been requested
at some time by users or projects based on their meshing or post-processing
tools. All of these formats have advantages and disadvantages (in terms
of simplicity, functionality, longevity, and popularity) when compared to
each other. The following formats are currently supported by \CS:

\begin{list}{-}{}

\item \hyperref[sec:fmtdesc_des]{\simail (NOPO)}
\item \hyperref[sec:fmtdesc_unv]{\ideas universal}
\item \hyperref[sec:fmtdesc_med]{\med}
\item \hyperref[sec:fmtdesc_cgns]{CGNS}
\item \hyperref[sec:fmtdesc_ensight6]{\ensight 6}
\item \hyperref[sec:fmtdesc_ensightg]{\ensightg}
\item \hyperref[sec:fmtdesc_neu]{\gambit neutral}
\item \hyperref[sec:fmtdesc_gmsh]{\gmsh}
\item \hyperref[sec:fmtdesc:ccm]{STAR-CCM+}
\item \hyperref[sec:fmtdesc_catalyst]{\catalyst (co-processing)}
\end{list}

These formats are described in greater detail in the following sections.
Unless a specific option is used, the \pcs determines the mesh format directly
from the file suffix: %
{\em``\texttt{.case}''} for \ensight (6 or Gold),
{\em``\texttt{.ccm}''} for STAR-CCM+,
{\em``\texttt{.cgns}''} for CGNS,
{\em``\texttt{.des}''} for \simail,
{\em``\texttt{.med}''} for \med,
{\em``\texttt{.msh}''} for \gmsh,
{\em``\texttt{.neu}''} for \gambit neutral,
{\em``\texttt{.unv}''} for I-deas universal.

Note that the preprocessor can read gzipped mesh files directly (for Formats
other than MED or CGNS, which use specific external libraries) on most machines.

%==================================
\subsubsection{Formats supported for input\label{sec:formats_in}}
%==================================

\subsubsubsection{NOPO/\simail (INRIA/Distene)%
\label{sec:fmtdesc_des}}

This format is output by \simail, which was used heavily at EDF until
a few years ago. \CS does not
currently handle cylindrical or spherical coordinates, but it seems that
\simail always outputs meshes in Cartesian coordinates, even if points
have been defined in another system. Most ``classical'' element types
are usable, except for pyramids.

Note that depending on the architecture on which a file was
produced by \simail\footnote{``little endian'' on Intel or AMD processors, or
``big endian'' on most others, and starting with \simail 7, 32-bit or 64-bit
 integer and floating-point numbers depending on architecture},
it may not be directly readable by \simail on a different machine, while
this is not a problem for the \pcs, which automatically detects the
byte ordering and the 32/64 bit variant and adjusts accordingly.

\smallskip \noindent
\begin{tabular}[top]{|p{4.5cm}%
                     |>{\PreserveBackslash\raggedright\hspace{0pt}}p{10.5cm}|}
\hline
Default extension: & {\tt .des}\\
\hline
File type:         & semi-portable ``Fortran'' binary (IEEE integer and
                     floating-point numbers on 4 or 8 bytes, depending on
                     32 or 64 bit \simail version, bytes also ordered based
                     on the architecture)\\
\hline
Surface elements:  & triangles, quadrangles
                     (+ volume element face references)\\
\hline
Volume elements:   & tetrahedra, prisms, hexahedra\\
\hline
Zone selection:    & element face references and volume sub-domains\\
                   & (interpreted as numbered groups)\\
\hline
Compatibility:     & all files of this type as long as the coordinate
                     system used is Cartesian and not cylindrical or
                     spherical\\
\hline
Documentation:     & Simail user documentation and release notes or
                     MODULEF documentation:
                     \url{http://www-rocq.inria.fr/modulef} \par
                     Especially: \par
                     \url{http://www-rocq.inria.fr/modulef/Doc/FR/Guide2-14/node49.html} \\
\hline
\end{tabular}

\subsubsubsection{\ideas universal file%
\label{sec:fmtdesc_unv}}

This format was very popular in the 1990's and early 2000's, and though
the I-deas tool has not focused on the CFD (or even meshing) market since
many years, it is handled (at least in part) by many tools, and may
be considered as a major ``legacy'' format. It may contain many different
datasets, relative to CAD, meshing, materials, calculation results,
or part representation. Most of these datasets are ignored by \CS,
and only those relative to vertex, element, group, and coordinate system
definitions are handled.

This format's definition evolves with \ideas versions, albeit in a limited
manner: some datasets are declared obsolete, and are replaced by others,
but the definition of a given dataset type is never modified. Element and
Vertex definitions have not changed for many years, but group definitions
have gone through several dataset variants through the same period,
usually adding minor additional group types not relevant to meshing.
If one were to read a file generated with a more recent version of \ideas
for which this definitions would have changed with no update in the \pcs,
as the new dataset would be unknown, it would simply be ignored.

Note that this is a text format. Most element types are handled, except
for pyramids.

\smallskip \noindent
\begin{tabular}[top]{|p{4.5cm}%
                     |>{\PreserveBackslash\raggedright\hspace{0pt}}p{10.5cm}|}
\hline
Default extension: & {\tt .unv}\\
\hline
File type:         & text\\
\hline
Surface elements:  & triangles, quadrangles\\
\hline
Volume elements:   & tetrahedra, prisms, hexahedra\\
\hline
Zone selection:    & colors (always) and named groups\\
\hline
Compatibility:     & \ideas (\emph{Master Series} 5 to 9, \emph{NX Series} 10 to 12)
                     at least\\
\hline
Documentation:     & Online I-deas NX Series documentation\\
\hline
\end{tabular}

\subsubsubsection{\gambit neutral%
\label{sec:fmtdesc_neu}}

This format may be produced by Ansys \fluent's GAMBIT meshing tool.
As this tool does not export meshes to other formats directly handled
by the \pcs (though \fluent itself may export files to the CGNS or
\ideas universal formats), it was deemed useful to enable the \pcs
to directly read files in \gambit neutral format.

Note that this is a text format. ``Classical'' element types are usable.

\smallskip \noindent
\begin{tabular}[top]{|p{4.5cm}%
                     |>{\PreserveBackslash\raggedright\hspace{0pt}}p{10.5cm}|}
\hline
Default extension: & {\tt .neu}\\
\hline
File type:         & text\\
\hline
Surface elements:  & triangles, quadrangles\\
\hline
Volume elements:   & tetrahedra, pyramids, prisms, hexahedra\\
\hline
Zone selection:    & boundary conditions for faces, element groups for cells\\
                   & (interpreted as named groups)\\
\hline
Documentation:     & GAMBIT on-line documentation\\
\hline
\end{tabular}

\subsubsubsection{\ensight 6%
\label{sec:fmtdesc_ensight6}}

This format is used for output by the \harpoon meshing tool, developed
by Sharc Ltd (also the distributor of \ensight for the United Kingdom).
This format may represent all ``classical'' element types.

Designed for post processing, it does not explicitly handle the definition
of surface patches or volume zones, but allows the use of many \emph{parts}
(i.e. groups of elements) which use a common vertex list.
A possible convention (used at least by \harpoon) is to add surface
elements to the volume mesh, using one \emph{part} per group. The volume
mesh may also be separated into several \emph{parts} so as to identify
different zones. As \emph{part} names may contain up to 80 characters,
we do not transform them into groups (whose names could be unwieldy),
so we simply convert their numbers to group names.

Also note that files produced by \harpoon may contain badly oriented
prisms, so the \pcs orientation correction option
(\texttt{--reorient}) may must be used. Meshes built by this tool also
contain hanging nodes, with non-conforming elements sharing some vertices.
Mesh joining must thus also be used, and is not activated automatically,
as the user may prefer to specify which surfaces should be joined,
and which ones should not (\textit{i.e.} to conserve thin walls).

\smallskip \noindent
\begin{tabular}[top]{|p{4.5cm}%
                     |>{\PreserveBackslash\raggedright\hspace{0pt}}p{10.5cm}|}
\hline
Default extension: & {\tt .case}\\
\hline
File type:         & text file (extension \emph{.case}), and text,
                     binary, or Fortran binary file with
                     (\emph{.geo} extension), describing integers and floats in the IEEE format,
                     using 32 bits\\
\hline
Surface elements:  & triangles, quadrangles\\
\hline
Volume elements:   & tetrahedra, pyramids, prisms, hexahedra\\
\hline
Zone selection:    & part numbers interpreted as numbered groups\\
\hline
Compatibility:     & All files of this type\\
\hline
Documentation:     & on-line documentation, also available at:
                     \url{www3.ensight.com/EnSight10_Docs/UserManual.pdf}\\
\hline
\end{tabular}

\subsubsubsection{\gmsh%
\label{sec:fmtdesc_gmsh}}

This format is used by the free \href{http://www.geuz.org/gmsh}{\gmsh}
tool. This tool has both meshing and post-processing functionality,
but \CS only imports the meshes.

Note that some meshes produced by \gmsh man contain some badly oriented
elements, so the \pcs's \texttt{-reorient} option may be necessary.

The \pcs handles versions 1 and 2 of this array. In version 1,
two labels are associated with each element: the first defines the
element's physical entity  number, the second defines its elementary
entity number. Using version 2, it is possible to associate an
arbitrary number of labels with each element, but files produced
by \gmsh use 2 labels, with the same meanings as with version 1.

The decision was taken to convert physical entity numbers to groups. It is possible
to build a mesh using \gmsh without defining any  physical entities
(in which case all elements will belong to the same group, but the \gmsh
documentation clearly says that geometric entities are to be used
so as to group elementary entities having similar ``physical'' meanings.

To obtain distinct groups with a mesh generated by \gmsh, it
is thus necessary for the user to define physical entities.
This requires an extra step, but allows for fine-grained control
over the groups associated with the mesh, while using only elementary
entities could lead to a high number of groups.

\smallskip \noindent
\begin{tabular}[top]{|p{4.5cm}%
                     |>{\PreserveBackslash\raggedright\hspace{0pt}}p{10.5cm}|}
\hline
Default extension: & {\tt .msh}\\
\hline
File type:         & text or binary file\\
\hline
Surface elements:  & triangles, quadrangles\\
\hline
Volume elements:   & tetrahedra, pyramids, prisms, hexahedra\\
\hline
Zone selection:    & physical entity numbers interpreted as numbered groups\\
\hline
Compatibility:     & all files of this type\\
\hline
Documentation:     & included documentation, also available at:
                     \href{http://www.geuz.org/gmsh}
                          {http://www.geuz.org/gmsh}\\
\hline
\end{tabular}

%==================================
\subsubsection{Formats supported for input or output\label{sec:formats_inout}}
%==================================

\subsubsubsection{\ensightg%
\label{sec:fmtdesc_ensightg}}

This format may represent all ``classical'' element types, as well as
arbitrary polygons and convex polyhedra.

This format evolves slightly from one \ensight version to another, keeping
backwards compatibility. For example, polygons could not be used in the
same \emph{part} as other element types prior to version 7.4, which removed
this restriction and added support for polyhedra. Version 7.6 added support
for material type definitions.

This format offers many possibilities not used by \CS, such as defining
values on part of a mesh only (using ``undefined'' marker values or
partial values), assigning materials to elements, defining rigid
motion, or defining per-processor mesh parts with ghost cells for
parallel runs. Note that some libraries allowing direct \ensightg support
do not necessarily support the whole format specification.
Especially, VTK does not support material types.
Also, both \ensightg (8.2 and above) and VTK allow for automatic distribution,
reducing the usefulness of pre-distributed meshes with per-processor files.

Note than when using \paraview, if multiple parts (i.e. meshes) are
present in a give case, using the ``Extract Blocks'' filter is
required to separate those parts and obtain a proper visualization,
unless the \texttt{separate\_meshes} writer option is used.
The VisIt software does not seem to handle multiple parts in an \ensight case,
so different meshes must be assigned to different \emph{writers},
or the \texttt{separate\_meshes} option must be used
(see \S\ref{sec:prg_definitionpostprocess}) when using this tool.

This format may be used as an input format, similar to \ensight 6.
Compared to the latter, each \emph{part} has its own coordinates and vertex
connectivity; hence as a convention, we consider that surface or
volume zones may only be considered to be part of the same mesh
if the file defines vertex IDs (which we consider to be
unique vertex labels). In this case, \emph{part} numbers
are interpreted as group names. Without vertex IDs, only one part is read,
and no groups are assigned.

\smallskip \noindent
\begin{tabular}[top]{|p{4.5cm}%
                     |>{\PreserveBackslash\raggedright\hspace{0pt}}p{10.5cm}|}
\hline
Default extension: & {directory {\tt{{\it \{case\_name\}}.ensight}},
                     containing a file with the \tt .case} extension\\
\hline
File type:         & multiple binary or text files\\
\hline
Surface elements:  & triangles, quadrangles, polygons\\
\hline
Volume elements:   & tetrahedra, pyramids, prisms, hexahedra, convex polyhedra\\
\hline
Zone selection:    & possibility of defining element materials (not used), or
                     interpret part number as group name if vertex IDs are
                     given\\
\hline
Compatibility:     & files readable by \ensight 7.4 to 10.0, as well as tools
                     based on the \href{http://www.vtk.org}{\vtk} library,
                     especially \paraview\ (\url{http://www.paraview.org})\\
\hline
Documentation:     & online documentation, also available at:
                     \url{www3.ensight.com/EnSight10_Docs/UserManual.pdf}\\
\hline
\end{tabular}

\subsubsubsection{\med}\hyperdef{sec}{med}{}
\label{sec:fmtdesc_med}

Initially defined by EDF R\&D, this format (\emph{Mod\`ele d'\'echanges de Donn\'ees},
or \emph{Model for Exchange of Data}) has been defined and maintained through
a \med working group comprising members of EDF R\&D and CEA (the \CS team
being represented). This is the reference format for the
\href{http://www.opencascade.org/SALOME/Salome.html}{\emph SALOME} environment.
This format is quite complete, allowing the definition of all ``classical''
element types, in nodal or descending connectivity.
It may handle polygonal faces and polyhedral cells,
as well as the definition of structured meshes.

This format, which requires a library also depending on the free HDF5 library,
allows both for reading and writing meshes with their attributes (``families'' of
color/attribute and group combinations), as well as handling calculation data,
with the possibility (unused by \CS) of defining variables only on a subset
(``profile'') of a mesh.

The \med library is available under a \href{http://www.gnu.org}{LGPL} license,
and is even packaged in some Linux distributions
(at least Debian and Ubuntu). \CS requires at least \med 3.0.2, which in turn
requires HDF5 1.8. This format is upwards-compatible with \med 2.3,
so files in that version of the format may be read, though not output.

\smallskip \noindent
\begin{tabular}[top]{|p{4.5cm}%
                     |>{\PreserveBackslash\raggedright\hspace{0pt}}p{10.5cm}|}
\hline
Default extension:    & {\tt .med}\\
\hline
File type:            & portable binary, based on the HDF5 library
                       (\url{http://www.hdfgroup.org/HDF5/index.html})\\
\hline
Surface elements:     & triangles, quadrangles, simple polygons\\
\hline
Volume elements:      & tetrahedra, pyramids, prisms, hexahedra, simple polyhedra\\
\hline
Zone selection:       & element families ({\it i.e.} colors and groups)\\
\hline
Input compatibility:  & \med 2.3, 3.0, 3.1, or 3.2
                     (only unstructured nodal connectivity is supported)\\
\hline
Output compatibility: & \med 3.0 and above \\
\hline
Documentation:        & on-line documentation. Download link at \url{http://files.salome-platform.org/Salome/other/med-3.2.0.tar.gz}\\
\hline
\end{tabular}

\subsubsubsection{CGNS}\hyperdef{sec}{cgns}{}
\label{sec:fmtdesc_cgns}

Promoted especially by the AIAA, NASA, Boeing Commercial, and ANSYS ICEM CFD (as well as ONERA in France),
this format(\emph{CFD General Notation System}) is quite well established in
the world of CFD. The concept is similar to that of \med, with a bigger
emphasis on normalization of variable names or calculation information, and
even richer possibilities. Contrary to \med, the first version of this format
was limited to multi-bloc structured meshes, unstructured meshes having been
added in CGNS 2.

Slightly older than \med, this library was free from the start, with a good
English documentation, and is thus much better known. It is more focused
on CFD, where \med is more generic. A certain number of tools accompany
the CGNS distribution, including a mesh visualizer (which does not handle
polygonal faces although the format defines them), and an interpolation
tool.

\CS should be able to read almost any mesh written in this format, though
meshes with over-set interfaces may not be usable for a calculation
(calculations with over-set interfaces may be possible in the context of coupling \CS
with itself but with two separate meshes).
Other (abutting) interfaces are not handled automatically (as there are
at least 3 or 4 ways of defining them, and some mesh tools do not export
them\footnote{For example, \icemcfd can join non-conforming meshes, but it
exports joining surfaces as simple boundary faces with user-defined boundary
conditions.}), so the user is simply informed of their existence in the
\pcs's log file, with a suggestion to use an appropriate conformal joining
option. Structured zones are converted to unstructured zones immediately after
being read.

Boundary condition information is interpreted as groups with the same
name. The format does not yet provide for selection of volume elements,
as only boundary conditions are defined in the model (and can be assigned to
faces in the case of unstructured meshes, or vertices in any case).
Note that boundary conditions defined at vertices are not ignored by
the \pcs, but are assigned to the faces of which all vertices bear
the same condition.\footnote{If one of a face's vertices does not bear
a boundary condition, that condition is not transferred to the face.}

The \pcs also has the capability of building additional volume or surface groups,
based on the mesh sections to which cells or faces belong. This may be
activated using a sub-option of the mesh selection, and allows obtaining
zone selection information from meshes that do not have explicit
boundary condition information but that are subdivided in appropriate zones or
sections (which depends on the tool used to build the mesh).

When outputting to CGNS, an unstructured connectivity is used for the calculation
domain, with no face joining information or face boundary condition
information.\footnote{Older versions of the documentation specified that
a field must be defined on all elements of a zone, so that adding faces
on which to base boundary conditions to a volume mesh would have required
also defining volume fields on these faces. More recent versions of the
documentation make it clear that a field must be defined on all elements
of maximum dimension in a zone, not on all elements.}

Though many tools support CGNS, that support is often quite disappointing,
at least for unstructured meshes. Thus, some editors seem to use
different means to mark zones to associate with boundary conditions than
the ones recommended in the CGNS documentation, and some behaviours
are worse. Also, many readers do not allow the user to choose between multiple
CGNS bases (meshes in the \CS sense), so when outputting to CGNS, it may
be necessary to output each post-processing mesh using a separate writer.
VisIt 2.4.2 may fail to read a volume mesh output by \CS, but read a surface
mesh correctly, while the same volume mesh may be read with no problems
by \ensight 10.
The support of polygons (\emph{ngons} in the CGNS standard), is even worse,
and even the verification tools published alongside the CGNS library were
recently unable to handle them, and reported errors in valid files containing
such elements. It seems support of polyhedral elements has appeared in ParaView's
CGNS reader mid 2016, so the situation may be improving.
For mesh input, some ICEM CFD versions used a CGNS 3 beta library,
which led to some issues. CGNS 3 output from ICEM CFD 13 is known to work well with
\CS, but that same version is unable to read files generated by \CS, as
it seems to ``cheat'' with CGNS version numbers and to confuse CGNS 3 and 3.1 specs.
ICEM 14 seems to have fixed that bug, so there should be no issues with up-to
date versions of that software.

\smallskip \noindent
\begin{tabular}[top]{|p{4.5cm}%
                     |>{\PreserveBackslash\raggedright\hspace{0pt}}p{10.5cm}|}
\hline
Default extension:    & {\tt .cgns}\\
\hline
File type:            & portable binary (uses the ADF library specific to CGNS, or HDF5)\\
\hline
Surface elements:     & triangles, quadrangles, simple polygons\\
\hline
Volume elements:      & tetrahedra, pyramids, prisms, hexahedra
                        (simple polyhedra allowed by CGNS 3 but not supported yet)\\
\hline
Zone selection:       & Surface zone selection using boundary conditions, no volume zone
                        selection, but the \pcs allows creation of groups associated to
                        zones or sections in the mesh using mesh selection sub-options\\
\hline
Input compatibility:  & CGNS 2.5 or CGNS 3.1\\
\hline
Output compatibility: & CGNS 3.1 and above\\
\hline
Documentation:        & See CGNS site: \url{http://www.cgns.org}\\
\hline
\end{tabular}

\subsubsubsection{\starccmp%
\label{sec:fmtdesc:ccm}}

This polyhedral format is the current CD-Adapco format, and is based on
CD-Adapco's libccmio, which is based on ADF (the low-level file format
used by CGNS prior to the shift to HDF-5). libccmio comes with a version
of ADF modified for performance, but also works with a standard version
from CGNS.

Currently, geometric entity numbers are converted to numbered groups,
with the corresponding names printed to the \pcs log. Depending on whether
the names were generated automatically or set by the user, it would be
preferable to use the original group names rather than base their
names on their numbers.

This format may also be used for output, though its limitations
make this a less general solution than other output formats:
only 3D meshes are handled, though values can be output on boundary
face regions (which may not overlap). As such, to ensure consistency,
output using this format is limited as follows:

\begin{itemize}
\item output of the full volume mesh and cell or vertex data on that
      mesh is handled normally.
\item output of the full surface mesh and per face data on that mesh
      handled normally, only if output of the full volume mesh to
      this format is also enabled. It is ignored otherwise.
\item output of sub-meshes or meshes built during the preprocessing
      stage and all other data is ignored.
\end{itemize}

As such, this formal may be useful for interoperability of data
with a CCMIO-based tool-chain, but simultaneously using another output
format to visualize possible error output is recommended.

The CCMIO library is distributed freely by CD-Adapco upon demand.

\smallskip \noindent
\begin{tabular}[top]{|p{4.5cm}%
                     |>{\PreserveBackslash\raggedright\hspace{0pt}}p{10.5cm}|}
\hline
Default extension: & {\tt .ccm}\\
\hline
File type:         & binary file using modified ADF library.\\
\hline
Surface elements:  & polygons\\
\hline
Volume elements:   & polyhedra\\
\hline
Zone selection:    & named face and cell sets\\
                   & (interpreted as numbered groups, with names appearing in log)\\
\hline
Compatibility:     & all files of this type?\\
\hline
Documentation:     & documentation and source code provided by CD-Adapco\\
\hline
\end{tabular}

%==================================
\subsubsection{Formats supported for output only\label{sec:formats_out}}
%==================================

\subsubsubsection{\catalyst%
\label{sec:fmtdesc_catalyst}}

This is not a ``true'' output format in the sense that output is not written
directly to file, but is exported to the \catalyst co-processor.
In turn, this co-processor will execute operations based on a
special \paraview Python script, and directly generate output such
as images or movies.

Co-processing scripts may be generated using \paraview 4, with the
CoProcessing plugin, using initial output in another format (such as \ensightg).
A \CS postprocessing writer will try to read a script named
\texttt{\emph{<writer\_name>}.py}, which should be places in a case's
\texttt{DATA} directory.

Note that this output is still at an experimental stage, and is heavily
dependent on \paraview. Some operations may work very well, while other,
similar operations may fail.

\smallskip \noindent
\begin{tabular}[top]{|p{4.5cm}%
                     |>{\PreserveBackslash\raggedright\hspace{0pt}}p{10.5cm}|}
\hline
Default extension: & not applicable\\
\hline
File type:         & co-processing\\
\hline
Surface elements:  & triangles, quadrangles, polygons\\
\hline
Volume elements:   & tetrahedra, pyramids, prisms, hexahedra, convex polyhedra\\
\hline
Compatibility:     & \catalyst from \href{http://paraview.org}{\paraview} 4.0\\
\hline
Documentation:     & online documentation and Wiki,  at:
                     \url{http://paraview.org/Wiki/Main_Page}\\
\hline
\end{tabular}

%==================================
\subsubsection{Meshing tools and associated formats}
%==================================

Most often, the choice of a mesh format is linked to the choice of
a meshing tool. Still, some tools allow exporting a mesh under several
formats handled by \CS. This is the case of \fluent and \icemcfd,
which can export meshes to both the \ideas universal and CGNS formats
(\fluent's \gambit is also able to export to \ideas universal format).

Traditionally, users exported files to the \ideas universal format,
but it does not handle pyramid elements, which are often used by these
tools to transition from hexahedral to tetrahedral cells in the case
of hybrid meshes. The user is encouraged to export to CGNS, which
does not have this limitation.

Tools related to the \salome platform should preferably use
\salome{}'s native \med format (export to \ideas universal is
also possible, but has some limitations).

%==================================
\subsubsection{Meshing remarks}
%==================================
\label{sec:prg_meshes}%

{\em WARNING: }
Some turbulence models ($k-\varepsilon$, $R_{ij}-\varepsilon$ SSG, ...) used in
\CS are ``High-Reynolds'' models. Therefore the size of the cells
neighbouring the wall must be greater than the thickness of the viscous
sub-layer (at the wall, $y^+>2.5$ is required, and $30<y^+<100$ is
preferable). If the mesh does not match this constraint, the results may
be false (particularly if thermal phenomena are involved). For more details
on these constraints, see the keyword \texttt{iturb}.

%==================================
\subsection{Preprocessor command line options}
%==================================
\label{sec:prg_optappelecs}%
The main options are:
\begin{list}{$\bullet$}{}
\item \texttt{--help}: provides a summary of the different command line options

\item \texttt{<mesh>}: the last argument is used to specify the name of the mesh file.
The launch script automatically calls the Preprocessor for every
mesh in the \texttt{MESHES[]} list specified by the user.

\item \texttt{--reorient}: attempts to re-orient badly-oriented cells
if necessary to compensate for mesh-generation software
whose output does not conform to the format specifications.

\end{list}

%==================================
\subsection{Kernel command line options}
%==================================
\label{sec:prg_optappelnoy}%
In the standard cases, the compilation of \CS and its execution are entirely
controlled by the launch script. The potential command line options are passed
through user modifiable variables at the beginning of the \texttt{cs\_user\_scripts.py} file
(this file may be copied from the \texttt{DATA/REFERENCE} to the \texttt{DATA} and edited).
This way, the user only has to fill these variables and doesn't need
to search deep in the script for the Kernel command line. For more advanced
usage, the main options are described below:

\begin{list}{$\bullet$}{}
\item \texttt{--app-name}: specifies the application name. This is
useful only in the case of code coupling, where the application name
is used to distinguish between different code instances launched together.

\item \texttt{--mpi}: specifies that the calculation is running
with MPI communications. The number of processors used will be determined
automatically by the Kernel. With most MPI implementations, the
code will detect the presence of an MPI environment automatically, and
this option is redundant. It is only kept for the rare case in which the
MPI environment might not be detected.

\item \texttt{--preprocess}: triggers the preprocessing-only mode.
The code may run without any Interface parameter file or any user subroutine.
Only the initial operations such as mesh joining and modification are
executed.

\item \texttt{-q} or \texttt{--quality}: triggers the verification mode.
The code may run without any Interface parameter file or any user subroutine.
This mode includes the preprocessing stages, and adds elementary tests:\\
\begin{list}{-}{}
\item the quality criteria of the mesh are calculated (non-orthogonality angles,
internal faces off-set, \ldots) and corresponding
visualizable post-processing output is generated.\\
\item a few additional mesh consistency tests are run.\\
\end{list}

\item \texttt{--benchmark}: triggers the benchmark mode, for a timing
of elementary operations on the machine. A secondary option
\texttt{--mpitrace} can be added. It is to be activated when the benchmark mode
is used in association with an MPI trace utility. It restricts the elementary
operations to those implying MPI communications and does only one of each
elementary operation, to avoid overfilling the MPI trace report.\\
This command is to be placed in the \\texttt{domain.solver\_args} variable
in the \texttt{cs\_user\_scripts.py} file to be added automatically to the
Kernel command line.

\item \texttt{--log n}: specifies the destination of the output for a
single-processor calculation or for the processor of rank 0 in a parallel
calculation.\\
\hspace*{0.5cm}\texttt{n=0}: output directed towards the standard output\\
\hspace*{0.5cm}\texttt{n=1}: output redirected towards a file \texttt{listing}
(default behaviour)\\
This option can be specified in the \texttt{domain.logging\_args} field
of the user script.

\item \texttt{--logp n}: specifies the destination of the output for the
processors of rank 1 to $N-1$ in a calculation in parallel on $N$ processors
({\em i.e.} the redirection of all but the first processor).\\
\hspace*{0.5cm}\texttt{n=-1}: no output for the processors of rank 1 to $N-1$
(default behaviour).\\
\hspace*{0.5cm}\texttt{n=0}: no redirection. Every processor will write to the
standard output. This might be useful in case a debugger is used, with separate
terminals for each processor.\\
\hspace*{0.5cm}\texttt{n=1}: one file for the output of each processor. The
output of the processors of rank 1 to $N-1$ are directed to the files
\texttt{listing\_r0001} to \texttt{listing\_r$N-1$}.
This option can be specified in the \texttt{domain.logging\_args} field
of the user script.

\item \texttt{-p <filename>} or \texttt{--param <filename>}: specifies the name of the GUI
parameter file to use for the calculation.\\
The value of \texttt{<filename>} is to be defined by the \texttt{--param} option
of \texttt{code\_saturne run}, either directly or in the standard \texttt{runcase}
script (the file will be searched for in the \texttt{data} directory, though
an absolute path name may also be defined).

\item \texttt{-h} or \texttt{--help}: displays a summary of the different
command line options.
\end{list}

%==================================
\subsection{Launch scripts}
%==================================
\label{sec:prg_runcase}%

The case preparer command \texttt{code\_saturne~create} places an example of launch script,
\texttt{runcase}, in the \texttt{SCRIPTS} directory. This script is quite minimalist and is known to work on every architecture \CS has been tested on.
If a batch system is available, this script will contain options
for batch submission.
The script will then contain a line setting the proper \texttt{PYTHONPATH}
variable for \CS to run.
Finally, it simply contains the \texttt{code\_saturne run}  command,
possible with a \texttt{--param} option when a parameters file
defined by the GUI is used. Other options recognized by
\texttt{code\_saturne run} may be added.

In the case of a coupled calculation, this script also exists, and
may be used for preprocessing stages, but an additional
\texttt{runcase\_coupling} is added in the directory above the coupled case
directories, and may be used to define the list of coupled cases,
as well as global options, such as MPI options of the temporary
execution directory. An additional \texttt{runcase\_batch} file will
contain batch submission options when a batch system is available
(and is the file that should be submitted when using a batch system).

When not using the GUI, or if additional options must be accessed,
the \texttt{cs\_user\_scripts.py} file may be copied from
the \texttt{DATA/REFERENCE} to the \texttt{DATA} and edited.
This file contains several Python functions:

\begin{list}{$\bullet$}{}

\item \texttt{define\_domain\_parameter\_file} allows defining
 the choice of a parameters file produced by the GUI. This is
 generally not useful, as the parameters file may be directly
 defined in \texttt{runcase} or \texttt{runcase\_coupling}, or passed
 as an option to \texttt{code\_saturne run}, but could be useful
 when running more complex parametric scripts, and is provided for
 the sake of completeness.
\item \texttt{define\_domain\_parameters} allows defining
 most parameters relative to case execution for the current
 domain, including advanced options not accessible
 through the GUI. This function is the most important one in the user
 scripts file, and contains descriptions of the various options.
 Note that in most examples, setting of options is preceded by
 a \texttt{if domain.param == None:} line, ensuring the settings
 are only active if no GUI-defined parameters file is present.
 This is used to prevent accidental override of parameters defined
 by the GUI: parameters defined through the user script have priority
 over the GUI parameters file, so if both are used, these tests
 may be removed for parameters which should be defined through
 user scripts.
\item \texttt{define\_case\_parameters} allows defining
 most parameters relative to the global calculation, such as
 the number of processors or the execution directory.
 To avoid potentially conflicting definitions, this function is ignored
 for coupled calculations, where the corresponding parameters
 may be defined in the \texttt{runcase\_coupling} script.
\item \texttt{define\_mpi\_environment} allows defining
 advanced MPI parameters or redefining MPI options if the automatic
 settings are incorrect, and its use should only rarely be necessary.
 To avoid potentially conflicting definitions, this function is ignored
 for coupled calculations, where the corresponding parameters
 may be defined in the \texttt{runcase\_coupling} script.
\end{list}

%==================================
\subsection{Graphical User Interface}
%==================================
\label{sec:prg_gui}%
A Graphical User Interface is available with \CS.
This Interface creates or reads an XML file according to
a specific \CS schema which is then interpreted by the code.

In version \verscs, the Graphical Interface manages calculation parameters,
standard initialisation values and boundary
conditions for standard physics, pulverised fuel combustion, gas combustion,
atmospheric flows, Lagrangian module,  electrical model, compressible model and radiative
transfers (user subroutines can still be completed though).

The Interface is optional. Every data that can be specified through the
Interface can also be specified in the user subroutines. In case of
conflict, all calculation parameters, initialisation value or boundary condition
set directly in the user subroutines will prevail over what is defined by the
Interface. However, it is no longer necessary to redefine everything in the
user subroutines. Only what was not set or could not be set using the Graphical
Interface should be specified.

{\em WARNING: }
There are some limitations to the changes that can be made between the Interface
and the user routines. In particular, it is not possible to specify a certain
number of solved variables in the Interface and change it in the user routines
(for example, it is not possible to specify the use of a $k-\varepsilon$ model
in the Interface and change it to $R_{ij}-\varepsilon$ in \texttt{cs\_user\_parameters.f90}, or
to define additional scalars in \texttt{cs\_user\_parameters.f90} with respect to the
Interface). Also, all boundaries should be referenced in the Interface, even if
the associated conditions are intended to be modified in \texttt{cs\_user\_boundary\_conditions}, and
their nature (entry, outlet, wall\footnote{Smooth and rough walls are considered
to have the same nature}, symmetry) should not be changed.

For example, in order to set the boundary conditions of a calculation
corresponding to a channel flow with a given inlet velocity profile, one
should:\\
- set the boundary conditions corresponding to the wall and the output
using the Graphical Interface\\
- set a dummy boundary condition for the inlet (uniform velocity for instance)
- set the proper velocity profile at inlet in \texttt{cs\_user\_boundary\_conditions}. The wall and
output areas must not appear in \texttt{cs\_user\_boundary\_conditions}. The dummy velocity
entered in the Interface will not be taken into account.

The Graphical User Interface is launched with the \texttt{./SaturneGUI} command
in the directory \texttt{DATA}. The first step is
then to load an existing parameter file (in order to modify it) or to
open a new one. The headings to be filled for a standard calculation are the
following:

\begin{list}{-}{}
\item Identity and paths: definition of the calculation directories
      (STUDY, CASE, DATA, SRC, SCRIPTS, MESH).

\item Calculation environment: definition of the mesh file(s),
      stand-alone execution of the Preprocessor module
      (used by the Interface to get the groups of the boundary
      faces).

\item Thermophysical models: physical model, ALE mobile mesh features,
      turbulence model, thermal model, coupling with \syrthes.

\item Additional scalars: definition, initialisation of the scalars,
      and physical characteristics.

\item Physical properties: reference pressure, fluid characteristics, gravity.
      It is also possible to write user laws for the density, the viscosity,
      the specific heat and the thermal conductivity in the interface through
      the use of a formulae interpreter.

\item Volume conditions: initialisation of the variables, and definition of
      the zones where to apply head losses or source terms.

\item Boundary conditions: definition of the boundary conditions for
      each variable. The colors of the boundary faces may be read
      directly from a ``preprocessor.log*'' files created by the \pcs
      or a ``listing'' file from a previous Kernel run.

\item Numerical parameters: number and type of time step, advanced parameters
      for the numerical solution of the equations.

\item Calculation control: parameters concerning the time averages, time step,
      location of
      the probes where some variables will be monitored over time,
      definition of the frequency of the outputs in the calculation
      listing and in the chronological records and of the EnSight outputs.
      The item {\itshape Profiles} allows to save, with a  given frequency,
      1D profiles on an axis defined from two points provided by the user.

\item Calculation management: management of the calculation restarts,
      updating of the launch script (temporary execution directory, parallel
      computing, user data or result files, ...) and interactive launch of the
      calculation.

\end{list}

The \CS tutorial \cite{tutorial} offers a step-by-step guidance to the setting
up of some simple calculations with the \CS Interface.

To launch \CS using an XML parameter file,
the name of the file must
be given using the \texttt{--param} option of \texttt{code\_saturne run} in
the launch script (see \S\ref{sec:prg_runcase}). When the launch
script is edited from the Interface (Calculation management $\rightarrow$
Prepare batch analysis), this option is set automatically.

%==================================
\subsection{User subroutines}
%==================================
%==================================
\label{sec:prg_ssprgutilis}
%==================================
\subsubsection{Preliminary comments}
%==================================

The user can run the calculations with or without an interface, with or
 without the user subroutines. Without interface, some user subroutines
 are needed (see \S\ref{sec:prg_stepbystepcalculation}). With interface,
all the user subroutines are optional.

The parameters can be read in the interface and then in the user
subroutines. In the case that a parameter is specified in the interface
 and in a user subroutine, it is the value in the user subroutine that
 is taken into account. For this reason, all the examples of
 user subroutines are placed in the \texttt{EXAMPLES} directory by the
 case setup \texttt{code\_saturne~create} (and available subroutines in the
 directory \texttt{REFERENCE}).

%==================================
\subsubsection{Example routines}
%==================================

Some user subroutines may be used for many different user definitions. As
including enough examples in those subroutines would make them very
difficult to read, these routines provided as templates only, with
separate examples in a case's \texttt{EXAMPLES} subdirectory of its
\texttt{SRC} directory.

Example file names are defined by inserting the name of the matching example
in the file name. For example, a basic example for
\texttt{cs\_user\_boundary\_conditions.f90} is provided in \\
\texttt{cs\_user\_boundary\_conditions-base.f90}, while an example dedicated
to atmospheric flows is provided in
{cs\_user\_boundary\_conditions-atmospheric.f90}.

The user is encouraged to check what examples are available, and to study
those that are relevant to a given setup.

Template user subroutines contain three sections the user may define,
marked by the following strings:

\begin{itemize}
\item \texttt{INSERT\_VARIABLE\_DEFINITIONS\_HERE}
\item \texttt{INSERT\_ADDITIONAL\_INITIALIZATION\_CODE\_HERE}
\item \texttt{INSERT\_MAIN\_CODE\_HERE}
\end{itemize}

Comparing template and example files with a graphical file comparison tool
should help the user highlights the matching sections from the examples,
so it is recommended as good practice for those not already very familiar
with those user subroutines.

%==================================
\subsubsection{Main variables}
%==================================

This section presents a non-exhaustive list of the main variables that
may be encountered by the user. Most of them should not be modified by the
user. They are calculated automatically from the data. However it may be
useful to know what they represent.
Developers can also refer to \cite{theory}.

These variables are listed in the alphabetical index at the end of this
document (see \S~\ref{sec:prg_motscles}).

The type of each variable is given: integer [i], real number [r],
integer array [ia], real array [ra].

%==================================
\subsubsubsection{Array sizes}
%==================================
\label{sec:prg_dimensions}

\variabsize{ndim}{Space dimension (ndim=3)}

\bigskip

\variabsize{ncel}{Number of real cells in the mesh}

\variabsize{ncelet}{Number of cells in the mesh, including the
ghost cells of the ``halos'' (see note 1)}

\variabsize{nfac}{Number of internal faces (see note 2)}

\variabsize{nfabor}{Number of boundary faces (see note 2)}

\bigskip

\variabsize{lndfac}{Size of the array \texttt{nodfac}\index{nodfac} of
internal faces - nodes connectivity (see note 3)}

\variabsize{lndfbr}{Size of the array \texttt{nodfbr}\index{nodfbr} of
boundary faces - nodes connectivity (see note 3)}

\variabsize{nnod}{Number of vertices in the mesh}

\bigskip

\variabsize{nfml}{Number of referenced families of entities (boundary
faces, elements, ...)}

\variabsize{nprfml}{Number of properties per referenced entity family}

\bigskip

\variabsize{nvar}{Number of solved variables (must be lower than
\texttt{nvrmax})}

\variabsize{nscamx}{Maximum number of scalars solutions of an
advection equation, apart from the variables of the turbulence model ($k$,
$\varepsilon$, $R_{ij}$, $\omega$, $\varphi$, $\overline{f}$, $\alpha$, $\nu_t$), that is
to say the temperature and other scalars (passive or not, user-defined or not)}

\variabsize{nscal}{Effective number of scalars solutions of an
advection equation, apart from the variables of the turbulence model ($k$,
$\varepsilon$, $R_{ij}$, $\omega$, $\varphi$, $\overline{f}$, $\alpha$, $\nu_t$), that is
to say the temperature and other scalars (passive or not, user-defined or
not). These scalars can be divided into two distinct groups: \texttt{nscaus}
user-defined scalars and \texttt{nscapp} scalars related to a ``specific
physics''. \texttt{nscal=nscaus+nscapp}, and \texttt{nscal}
must be less than or equal to \texttt{nscamx}}

\variabsize{nscapp}{Effective number of scalars related to a
``specific physics''. These scalars are solutions of an advection
equation and distinct from the scalars of the turbulence model ($k$,
$\varepsilon$, $R_{ij}$, $\omega$, $\varphi$, $\overline{f}$, $\alpha$, $\nu_t$). They are
automatically defined by the choice of the selected specific physics
model (gas combustion with Eddy Break-Up model, pulverised coal
combustion, ...). For example: mass fractions, enthalpy, ...}

\variabsize{nscaus}{Effective number of user-defined scalars. These
scalars are solutions of an advection equation and distinct from the
scalars of the turbulence model ($k$, $\varepsilon$, $R_{ij}$, $\omega$,
$\varphi$, $\overline{f}$, $\alpha$, $\nu_t$) and from the \texttt{nscapp} scalars related to the
``specific physics''. For example: passive tracers, temperature (when no
specific physics model is selected), ...}

\variabsize{nestmx}{Maximum number of error estimators for
Navier-Stokes}

\bigskip

\variabsize{npromx}{Maximum number of physical properties. These will
be stored in the arrays \texttt{propce}}

\variabsize{nproce}{Number of properties defined at the
cells. These will be stored in the array \texttt{propce}}

\variabsize{nvisls}{Number of scalars with variable diffusivity}


\bigskip

\variabsize{nushmx}{Maximum number of user chronological files
(in the case where \texttt{ushist} is used)}

\bigskip

\variabsize{nclacp}{Number of coal classes for the pulverised
coal combustion module. It is the total number of classes, {\it i.e.}
the sum of the number of classes for every represented coal. \texttt{nclacp} must
be less than or equal to \texttt{nclcpm}}

\variabsize{nclcpm}{Maximum number of coal classes for the
pulverised coal combustion module}


\minititre{Note 1: ghost cells - ``halos''}
A cell (real cell) is an elementary mesh element of the spatial
discretisation of the calculation domain. The mesh is made of \texttt{ncel} cells.\\
When using periodicity and parallelism, extra ``ghost'' cells
(called ``halo'' cells) are defined for temporary storage of some information
(on a given processor).
The total number of real and ghost cells is \texttt{ncelet}. \\
\hspace*{1cm} Indeed, when periodicity is enabled, the cells with
periodic faces do not have any real neighbouring cell across these
particular faces. Their neighbouring cell is elsewhere in the calculation
domain (its position is determined by the periodicity). In order to
temporarily store the information coming from this ``distant''
neighbouring cell, a ghost cell (``halo'') is created. \\
\hspace*{1cm} The same kind of problem exists in the case of a
calculation on parallel machines: due to the decomposition of the
calculation domain, some cells no longer have access to all
their neighbouring cells, some of them being treated by another processor. The
creation of ghost cells allows to temporarily store the information
coming from real neighbouring cells treated by other processors.\\
The variables are generally arrays of size \texttt{ncelet} (number of real and
fictitious cells). The calculations (loops) are made on \texttt{ncel} cells (only
the real cells, the fictitious cells are only used to store information).

\minititre{Note 2: internal faces}
An internal face is an interface shared by two cells (real or ghost
ones) of the mesh. A boundary face is a face which has only one real
neighbouring cell. In the case of periodic calculations, a periodic face
is an internal face. In the case of parallel running calculations, the
faces situated at the boundary of a partition may be internal faces or
boundary faces (of the whole mesh);

\minititre{Note 3: face-node connectivity}\label{sec:prg_nodfac}
The faces - nodes connectivity is stored by
means of four integer arrays: \texttt{ipnfac} and \texttt{nodfac} for the
internal faces, \texttt{ipnfbr} and \texttt{nodfbr} for the boundary faces.
\texttt{nodfac} (size \texttt{lndfac})
contains the list of all the nodes of all the internal faces; first the nodes of
the first face, then the nodes of the second face, and so on.
\texttt{ipnfac} (size: \texttt{nfac+1}) gives the position \texttt{ipnfac(ifac)}
in \texttt{nodfac} of the first node of each internal face \texttt{ifac}.
Therefore, the reference numbers of all
the nodes of the internal face \texttt{ifac} are: \texttt{nodfac(ipnfac(ifac))},
\texttt{nodfac(ipnfac(ifac)+1)}, ..., \texttt{nodfac(ipnfac(ifac+1)-1)}.
In order for this last formula to be valid even for \texttt{ifac=nfac},
\texttt{ipnfac} is of size \texttt{nfac+1} and \texttt{ipnfac(nfac+1)}
is equal to \texttt{lndfac+1}.\\
The composition of the arrays \texttt{nodfbr} and \texttt{ipnfbr} is similar.

\minititre{Note 4: modules}
{\bf The user must not modify the existing modules.} This would require the
recompilation of the complete version, operation which is not allowed in
standard use.

%==================================
\subsubsubsection{Geometric variables}
%==================================

The main geometric variables are available in most of the
subroutines and directly accessible through the following arrays,
defined in the \texttt{mesh} module (i.e. \texttt{use mesh}).

\variab{cdgfac}{cdgfac(ndim,nfac)}{ra}{Coordinates of the
centres of the internal faces}

\variab{cdgfbo}{cdgfbo(ndim,nfabor)}{ra}{Coordinates of the centres of the
boundary face}

\variab{ifacel}{ifacel(2,nfac)}{ia}{Index-numbers of the (only) two neighbouring
cells for each internal face}

\variab{ifabor}{ifabor(nfabor)}{ia}{Index-number of the (unique) neighbouring
cell for each boundary face}

\variab{ipnfac}{ipnfac(nfac+1)}{ia}{Position of the first node of the each internal
face in the array \texttt{nodfac} (see note 3 in paragraph \ref{sec:prg_dimensions})}

\variab{ipnfbr}{ipnfbr(nfabor+1)}{ia}{Position of the first node of the each boundary
face in the array \texttt{nodfbr} (see note 3 in paragraph \ref{sec:prg_dimensions})}

\variab{nodfac}{nodfac(lndfac)}{ia} {Index-numbers of the nodes of each
internal face (see note 3 in paragraph \ref{sec:prg_dimensions})}

\variab{nodfbr}{nodfbr(lndfbr)}{ia}{Index-numbers of the nodes of each
boundary face (see note 3 in paragraph \ref{sec:prg_dimensions})}

\variab{surfac}{surfac(ndim,nfac)}{ra}{Surface vector of the internal
faces. Its norm is the surface of the face and it is oriented from \texttt{ifacel(1,.)}
to \texttt{ifacel(2,.)}}

\variab{surfbo}{surfbo(ndim,nfabor)}{ra}{Surface vector of the boundary
faces. Its norm is the surface of the face and it is oriented outwards}

\variab{volume}{volume(ncelet)}{ra}{Volume of each cell}

\variab{xyzcen}{xyzcen(ndim,ncelet)}{ra}{Coordinates of the cell centres}

\variab{xyznod}{xyznod(ndim,nnod)}{ra}{Coordinates of the mesh vertices}

In addition, other geometric variables are useful for gradients
reconstruction.
The main variables of this type are the following:

\variab{dijpf}{dijpf(ndim,nfac)}{ra}{For every internal face,
the three components of the vector  $\vect{I'J'}$, where I' and J' are
respectively the orthogonal projections of the neighbouring cell centres I and J
on a straight line orthogonal to the face and passing through its center}

\variab{diipb}{diipb(ndim,nfabor)}{ra}{For every boundary
face, the three components of the vector $\vect{II'}$. I' is the
orthogonal projection of I, center of the neighbouring cell, on the
straight line perpendicular to the face and passing through its center}

\variab{idist}{dist(nfac)}{ra}{For every internal face,
dot product of the vectors $\vect{IJ}$ and $\vect{n}$.
I and J are respectively the centres
of the first and the second neighbouring cell. The vector $\vect{n}$ is
the unit vector normal to the face and oriented from the first to the
second cell}

\variab{distb}{distb(nfabor)}{ra}{For every boundary face,
dot product between the vectors $\vect{IF}$ and $\vect{n}$.
I is the center of the neighbouring cell. F is the face center.
The vector $\vect{n}$ is the unit vector normal to the face and
oriented to the exterior of the domain}

\variab{dofij}{dofij(ndim,nfac)}{ra}{For every internal
face, the three components of the vector $\vect{OF}$. O is the intersection
point between the face and the straight line joining the centres
of the two neighbouring cells. F is the face center}

\variab{pond}{pond(nfac)}{ra}{For every internal face,
$\displaystyle\frac{\vect{FJ}.\vect{n}}{\vect{IJ}.\vect{n}}$.
Regarding the quality of mesh, its ideal value is 0.5}

\variab{surfan}{surfan(nfac)}{ra}{Norm of the surface vector of
the internal faces}

\variab{surfbn}{surfbn(nfabor)}{ra}{Norm of the surface vector of the boundary faces}

%====================================
\subsubsubsection{Physical variables}
%====================================

Almost all physical variables\footnote{except some of the properties defined at
the cell centers} can be accessed via the \texttt{cs\_field} API and are
available in all the subroutines as fields (either through their name or their
id). The previous system, which used multidimensional arrays, has been
progressively replaced by the \texttt{cs\_field} API and this process is now
still on-going only for the properties defined at cell centers. These
properties are therefore the last ones to be additionally accessible through a
deprecated multidimensional array:

\variab{propce}{propce(ncelet,nproce)}{ra}{Properties defined at the
cell centres. For instance: viscosity, ...}

For a thorough description of the user management of all physical
variables as well as the corresponding syntaxes between the \texttt{cs\_field}
API (both in C and Fortran) and the previous system, please refer to
\doxygenfile{field.html}{the dedicated \texttt{doxygen} documentation}.

Note that local arrays of values of physical variables, retrieved via the
\texttt{cs\_field} API, follow a naming convention, fully described at
\doxygenfile{local.html}{this page} of the \texttt{doxygen} documentation. It
is highly recommended to follow this convention to ease the comprehension.

\underline{About the solved variables}

The indexes allowing marking out the different solved variables (from 1 to
\texttt{nvar}) are integers available in a ``module'' called
\texttt{numvar}.

For example, \texttt{ipr} refers to the variable ``pressure''.

The list of integers referring to solved variables is given below. These
variable index-numbers can be used to retrieve the corresponding field indices
(for instance, \texttt{ivarfl(ipr)} is the field index for the pressure),
but also for some arrays of variable associated options (for instance,
\texttt{blencv(ik)} is the percentage of second-order convective scheme for the
turbulent energy when a corresponding turbulent model is used).

\begin{list}{$\bullet$}{}
\item \texttt{ipr}\index{\texttt{ipr}}: pressure
\footnote{\texttt{ipr} corresponds to a
reduced pressure, from which the standard hydrostatic pressure has been
deducted. The total solved pressure is a field of type property.}.
\item \texttt{iu}\index{\texttt{iu}}: velocity along the X axis.
\item \texttt{iv}\index{\texttt{iv}}: velocity along the Y axis.
\item \texttt{iw}\index{\texttt{iw}}: velocity along the Z axis.
\item \texttt{ik}\index{\texttt{ik}}: turbulent energy, in $k-\varepsilon$,
$k-\omega$ modelling or v2f ($\varphi$-model and BL-v2/k model) modelling.
\item \texttt{ir11}\index{\texttt{ir11}}: Reynolds stress R11, in
      $R_{ij}-\varepsilon$ or SSG modelling.
\item \texttt{ir22}\index{\texttt{ir22}}: Reynolds stress R22, in
      $R_{ij}-\varepsilon$ or SSG modelling.
\item \texttt{ir33}\index{\texttt{ir33}}: Reynolds stress R33, in
      $R_{ij}-\varepsilon$ modelling.
\item \texttt{ir12}\index{\texttt{ir12}}: Reynolds stress R12, in
      $R_{ij}-\varepsilon$ modelling.
\item \texttt{ir13}\index{\texttt{ir13}}: Reynolds stress R13, in
      $R_{ij}-\varepsilon$ modelling.
\item \texttt{ir23}\index{\texttt{ir23}}: Reynolds stress R23, in
      $R_{ij}-\varepsilon$ modelling.
\item \texttt{iep}\index{\texttt{iep}}: turbulent dissipation in $k-\varepsilon$,
$R_{ij}-\varepsilon$ or v2f ($\varphi$-model and BL-v2/k model) modelling.
\item \texttt{iomg}\index{\texttt{iomg}}: Specific dissipation rate $\omega$, in
$k-\omega$ SST modelling.
\item \texttt{iphi}\index{\texttt{iphi}}: variable $\varphi=\overline{v^2}/k$ in v2f ($\varphi$-model and BL-v2/k model).
\item \texttt{ifb}\index{\texttt{ifb}}: variable $\overline{f}$ in v2f ($\varphi$-model).
\item \texttt{ial}\index{\texttt{ifb}}: variable $\alpha$ in elliptic blending models (BL-v2/k and EBRSM).
\item \texttt{inusa}\index{\texttt{ifb}}: variable $\\tilde{nu}_t$ in Spalart-Allmaras model.
\item \texttt{isca(j)}\index{\texttt{isca}}: scalar j(1$\leqslant$j$\leqslant$nscal).
\end{list}

\bigskip

Concerning the solved scalar variables (apart from the variables
pressure, $k$, $\varepsilon$, $R_{ij}$, $\omega$, $\varphi$,
$\overline{f}$, $\alpha$, $\nu_t$), the following is very important:
\begin{list}{-}{}
\item The designation ``scalar'' refers to scalar variables which are
      solution of an advection equation, apart from the variables of the
      turbulence model  ($k$, $\varepsilon$, $R_{ij}$, $\omega$,
      $\varphi$, $\overline{f}$, $\alpha$, $\nu_t$): for instance the
      temperature, scalars which may be passive or not, ``user'' or not.
      The mean value of the square of the fluctuations of a ``scalar'' is a
      ``scalar'', too. The scalars may be divided into two groups:
      \texttt{nscaus} ``user'' scalars and \texttt{nscapp}
      ``specific physics'' scalars, with \texttt{nscal=nscaus+nscapp}.
      \texttt{nscal} must be less than or equal to \texttt{nscamx}.
\item The \texttt{j}$^{\text{th}}$ user scalar is, in
      the whole list of the \texttt{nscal} scalars, the scalar number
      \texttt{j}. In the list of the \texttt{nvar} solved variables, it
      corresponds to the variable number \texttt{isca(j)}.
\item The \texttt{j}$^{\text{th}}$ scalar related to a specific physics is, in
      the whole list of the \texttt{nscal} scalars, the scalar number
      \texttt{iscapp(j)}. In the list of the \texttt{nvar} solved variables, it
      corresponds to the variable number
      \texttt{isca(iscapp(j)})\index{\texttt{iscapp}}.

\item Apart from specific physics, the temperature (or the enthalpy) is the
      scalar number \texttt{iscalt}\index{iscalt} in the list of the
      \texttt{nscal} scalars. It corresponds to the variable number
      \texttt{isca(iscalt)}. if there is no thermal scalar,
      \texttt{iscalt} is equal to -1.
\item A ``user'' scalar number \texttt{j} may represent the mean of the
      square of the fluctuations of a scalar \texttt{k} ({\em i.e.} the average
      $\overline{\varphi^\prime\varphi^\prime}$ for a fluctuating scalar
      $\varphi$ ). This can be made either {\em via} the
      interface or by declaring that scalar using
      \texttt{cs\_parameters\_add\_variable\_variance} in
      \texttt{cs\_user\_parameters.c} (if the scalar in question is not a ``user''
      scalar, the selection is made automatically). For instance, if \texttt{j}
      and \texttt{k} are ``user'' scalars, the variable $\varphi$ corresponding
      to \texttt{k} is the variable number
      \texttt{isca(k)=isca(iscavr(j))}.\footnote{It is really
      $\overline{\varphi^\prime\varphi^\prime}$, and not
      $\displaystyle\sqrt{\overline{\varphi^\prime\varphi^\prime}}$}.
\end{list}

\bigskip

\underline{About the physical properties at the cell centers}\\
In \CS, the physical properties\footnote{Other variables are stored in the
arrays \texttt{propce}. They are not, strictly speaking,
``physical properties'' but it is convenient to have them
in the same array as the proper physical properties} are stored in the
\texttt{propce} arrays.
Some properties, like the density, are only stored for cells and boundary
faces. To avoid having different index numbers for a physical property,
depending on the array it is used in, the following structure is used in \CS:

\begin{list}{-}{}

\item All the properties (used or not) have a unique and distinct index-number,
given automatically by the code and stored in an integer or an integer array
(its size may be the maximum number of scalars
or the maximum number of variables).

\item The indexes referring to the different properties stored in the
\texttt{prop{\bf xx}} arrays are given respectively by the following integer arrays:

\variab{ipproc}{ipproc(npromx)}{ia}{Rank \texttt{i} in \texttt{propce(.,i)} of the
properties defined at the cell centres}

\end{list}

For instance, the index number corresponding to the density is \texttt{irom}.\\
In the list of the properties defined at the cell center, the density
is therefore the \texttt{ipproc(irom)}$^{\text{th}}$
property: its value at the center of the cell \texttt{iel} is given by \texttt{prop{\bf ce}(iel,ippro{\bf c}(irom))}.
}

The list of properties accessible in the \texttt{propxx} arrays is given below (this does
not include the properties linked to the specific physics modules):

\variab{irom}{irom}{ia}{Property number
corresponding to the density ({\em i.e.} $\rho$ in $kg.m^{-3}$).\\
Stored at the cells and the boundary faces}

\variab{iroma}{iroma}{ia}{Property number
corresponding to the density ({\em i.e.} $\rho$ in $kg.m^{-3}$) at the
previous time step, in the case of a second-order extrapolation in time.\\
Stored at the cells and the boundary faces}

\variab{iviscl}{iviscl}{ia}{Property number
corresponding to the fluid molecular dynamic viscosity ({\em i.e.} $\mu$ in
$kg.m^{-1}.s^{-1}$).\\
Stored at the cells}

\variab{ivisla}{ivisla}{ia}{Property number
corresponding to the fluid molecular dynamic viscosity ({\em i.e.} $\mu$
in $kg.m^{-1}.s^{-1}$) at the previous time step, in the case of a
second-order extrapolation in time.\\
Stored at the cells}

\variab{ivisct}{ivisct}{ia}{Property number
corresponding to the fluid turbulent dynamic viscosity ({\em i.e.}
$\mu_t$ in $kg.m^{-1}.s^{-1}$).\\
Stored at the cells}

\variab{ivista}{ivista}{ia}{Property number
corresponding to the fluid turbulent dynamic viscosity ({\em i.e.}
$\mu_t$ in $kg.m^{-1}.s^{-1}$) at the previous time step, in the case of a
second-order extrapolation in time.\\
Stored at the cells}

\variab{icp}{icp}{ia}{Property number
corresponding to the specific heat, when it is variable
({\em i.e.} $C_p$ in $m^2.s^{-2}.K^{-1}$). See note below.\\
Stored at the cells}

\variab{icpa}{icpa}{ia}{Property number
corresponding to the specific heat, when it is variable
({\em i.e.} $C_p$ in $m^2.s^{-2}.K^{-1}$), at the previous time step,
in the case of a second-order extrapolation in time. See note below.\\
Stored at the cells}

\variab{iestim}{iestim(nestmx)}{ia}{Property
number for the \texttt{nestmx} error estimators for Navier-Stokes. The estimators
currently available are \texttt{iestim(iespre\index{iespre})},\\
\texttt{iestim(iesder\index{iesder})},
\texttt{iestim(iescor\index{iescor})},
\texttt{iestim(iestot\index{iestot})}
stored at the cells}

\variab{ifluaa}{ifluaa(nvarmx)}{ia}{Property number corresponding to the
mass flow associated with each variable at
the previous time step, in the case of a second-order extrapolation in time.\\
Stored at the internal faces and boundary faces}

\variab{ismago}{ismago}{i}{Property number
corresponding to the variable $C$ of the dynamic model, {\em i.e}
so that $\mu_t=\rho C\overline{\Delta}^2\sqrt{2S_{ij}S_{ij}}$ (with the
notations of \cite{benhamadouche01}). $C$ corresponds to $C_s^2$ in the
classical model of Smagorinsky.\\
Stored at the cells}

\variab{icour}{icour}{i}{CFL number in each cell at the
present time step.\\
Stored at the cells}

\variab{ifour}{ifour}{i}{Fourier number in each cell at
the present time step.\\
Stored at the cells}

\variab{iprtot}{iprtot}{i}{Total pressure in each cell.\\
Stored at the cells}

\variab{ivisma}{ivisma(1 or 3)}{ia}{When the ALE method for deformable meshes is
activated, \texttt{ivisma} corresponds to the ``mesh viscosity'', allowing to limit
the deformation in certain areas. This mesh viscosity can be isotropic or be
taken as a diagonal tensor (depending on the value of the parameter
\texttt{iortvm}\index{\texttt{iortvm}}.\\
Stored at the cells}

\minititre{Note: Variable physical properties}\label{provar}
Some physical properties such as specific heat or diffusivity are often
constant (choice made by the user).
In that case, in order to limit the necessary memory, these
properties are stored as a simple real number rather than in a domain-sized
array of reals.
\begin{list}{$\bullet$}{}
\item This is the case for the specific heat $C_p$.
\begin{list}{-}{}
\item If $C_p$ is constant, it can be specified in
      the interface or by indicating \texttt{icp=0} in \\
      \texttt{cs\_user\_parameters.f90}, and the property will be stored in
      the real number \texttt{cp0}.
\item If $C_p$ is variable, it can be specified in the interface or by
      indicating \texttt{icp=1} in \\
      \texttt{cs\_user\_parameters.f90}. The code will then
      modify this value to make \texttt{icp} refer to the effective
      property number corresponding to the specific heat,
      in a way which is transparent for the user. For each cell
      \texttt{iel}, the value of $C_p$ is then given in \texttt{usphyv} and
      stored in the array \texttt{propce(iel,ipproc(icp))}.
\end{list}
\item This is the same for the diffusivity $K$ of each scalar \texttt{iscal}.
\begin{list}{-}{}
\item If $k$ is constant, it can be specified in the interface or by
      calling \texttt{field\_set\_key\_int(ivarfl(isca(iscal)), kivisl, -1)}
      in \texttt{cs\_user\_parameters.f90}, (in \texttt{usipsu}) and the property
      will be stored in the real number \texttt{visls0(iscal)}.
\item If $k$ is variable, it can be specified in the interface or by
      calling \texttt{field\_set\_key\_int(ivarfl(isca(iscal)), kivisl, 0)}
      in \texttt{cs\_user\_parameters.f90}, (in \texttt{usipsu}). The code will then
      modify this key value to make it refer to the effective
      field id corresponding to the diffusivity of the scalar
      \texttt{iscal}, in a way which is transparent for the user. For each cell
      \texttt{iel}, the value of $k$ is then given in \texttt{usphyv} and stored
      in the field whose id is given by calling
      \texttt{field\_set\_key\_int(ivarfl(isca(iscal)), kivisl, ...)}.
\end{list}
\end{list}

\bigskip

Two other variables, \texttt{hbord} and \texttt{tbord}, should be noted here,
although they are relatively local (they appear only in the treatment of the
boundary conditions) and are used only by developers.

\variab{hbord}{hbord(nfabor)}{ra}{Array of the exchange coefficient for
temperature (or enthalpy) at the boundary faces. The table is allocated only if
\texttt{isvhb}\index{\texttt{isvhb}} is set to 1 in the subroutine \texttt{tridim}
(which is note a user subroutine), which is done automatically, but only if the coupling
with \syrthes or the 1D thermal wall module are activated.}

\variab{tbord}{tbord(nfabor)}{ra}{Temperature (or enthalpy) at the boundary
faces\footnote{It is the physical temperature at the boundary faces, not the
boundary condition for temperature. See \cite{theory} for more details on
boundary conditions}. The table is allocated only if
\texttt{isvtb}\index{\texttt{isvtb}} is set to 1 in the subroutine \texttt{tridim}
 (which is note a user subroutine), which is done automatically but only if the
coupling with \syrthes or the 1D thermal wall module are activated.}

Tables \texttt{hbord} and \texttt{tbord} are of size \texttt{nfabor},
although they concern only the wall boundary faces.


%==================================
\subsubsubsection{Variables related to the numerical methods}
%==================================

The main numerical variables and ``pointers'' are displayed below.

\minititre{Boundary conditions}

\variab{ifmfbr}{ifmfbr(nfabor)}{ia}{Family number of the boundary
faces. See note 1}

\variab{iprfml}{iprfml(nfml,nprfml)}{ia}{Properties of the
families of referenced entities. See note 1}

\variab{isympa}{isympa(nfabor)}{i}{Integer to mark out the %FIXME``wall'' (\texttt{itypfb=iparoi} or \texttt{iparug}) or
``symmetry'' (\texttt{itypfb=isymet}) boundary faces where the
mass flow has to be canceled when the ALE module is switched off (these faces are impermeable). For instance,
if the face \texttt{ifac} is %FIXME a wall or
symmetry face,
\texttt{isympa(ifac)=0}, otherwise \texttt{isympa(ifac)=1}}

\variab{itrifb}{itrifb(nfabor)}{ia}{Indirection array allowing to
sort the boundary faces according to their boundary condition type \texttt{itypfb}}

\variab{itypfb}{itypfb(nfabor)}{ia}{Boundary condition type at the
boundary face \texttt{ifac} (see user subroutine \texttt{cs\_user\_boundary\_conditions})}

\variab{uetbor}{uetbor(nfabor)}{ra}{Friction velocity at the wall,
in the case of a LES calculation with van Driest-wall damping}

\minititre{Distance to the wall}

\variab{dispar}{dispar(ncelet)}{ra}{Distance between the center of
a given volume and the closest wall, when it is necessary
($R_{ij}-\varepsilon$ with wall echo,
LES with van Driest-wall damping, or $k-\omega$ (SST) turbulence model)
and when \texttt{icdpar=1}. The distance between the center of the cell
\texttt{iel} and the closest wall is \texttt{dispar(iel)}}

\variab{yplpar}{yplpar}{ra}{Non-dimensional distance $y^+$ between a given
volume and the closest wall, when it is necessary (LES with van
Driest-wall damping) and when \texttt{icdpar=1}. The dimensionless wall distance $y^+$
between the center of the cell \texttt{iel} and the closest wall is therefore
\texttt{yplpar(iel1)}}

\minititre{Pressure drops and porosity}

\variab{icepdc}{icepdc(ncepdc)}{ia}{Number of the \texttt{ncepdc}
cells in which a pressure drop is imposed. See \texttt{iicepd} and the user
subroutine \texttt{cs\_user\_head\_losses.f90}}

\variab{ckupdc}{ckupdc(ncepdc,6)}{ra}{Value of the
coefficients of the pressure drop tensor of the
\texttt{ncepdc} cells in which a pressure drop is imposed. Note the 6 values are
sorted as follows: (k11, k22, k33, k12, k23, k33).
See \texttt{ickpdc} and the user subroutine \texttt{cs\_user\_head\_losses.f90}}

\variab{ncepdc}{ncepdc}{ia}{Number of cells in
which a pressure drop is imposed. See the user subroutine \texttt{cs\_user\_head\_losses.f90}}

\variab{porosi}{porosi(ncelet)}{ra}{Value of the porosity}

\minititre{Mass sources}

\variab{icetsm}{icetsm(ncetsm)}{ia}{Number of the \texttt{ncetsm}
cells in which a mass source term is imposed. See \texttt{iicesm} and the user
subroutine \texttt{cs\_user\_mass\_source\_terms}}

\variab{itypsm}{itypsm(ncetsm,nvar)}{ia}{Type of mass source term
for each variable (0 for an injection at ambient value, 1 for an
injection at imposed value). See the user subroutine \texttt{cs\_user\_mass\_source\_terms}}

\variab{ncetsm}{ncetsm}{i}{Number of cells with
mass sources. See the user subroutine \texttt{cs\_user\_mass\_source\_terms}}

\variab{smacel}{smacel(ncetsm,nvar)}{ra}{Value of the mass source
term for pressure. For the other variables, eventual imposed injection
value. See the user subroutine \texttt{cs\_user\_mass\_source\_terms}}

\minititre{Wall 1D thermal module}

\variab{nfpt1d}{nfpt1d}{i}{Number of boundary faces which are coupled
with a wall 1D thermal module. See the user subroutine \texttt{uspt1d}}

\variab{ifpt1d}{ifpt1d}{ia} {Array allowing marking out the numbers of
the \texttt{nfpt1d} boundary faces which are coupled with a wall 1D
thermal module. The numbers of these boundary faces are
given by \texttt{ifpt1d(ii)}, with
1$\leqslant$\texttt{ii}$\leqslant$\texttt{nfpt1d}.
See the user subroutine \texttt{uspt1d}}

\variab{nppt1d}{nppt1d}{ia}{Number of discretisation cells in the 1D wall for the
\texttt{nfpt1d} boundary faces which are coupled with a 1D wall thermal module. The
number of cells for these boundary faces is given by
\texttt{nppt1d(ii)}, with
1$\leqslant$\texttt{ii}$\leqslant$\texttt{nfpt1d}. See
the user subroutine \texttt{uspt1d}}

\variab{eppt1d}{eppt1d}{ia}{Thickness of the 1D wall for the
\texttt{nfpt1d} boundary faces which are coupled with a 1D wall thermal
module. The wall thickness for these boundary faces is therefore given
by \texttt{eppt1d(ii)}, with
1$\leqslant$\texttt{ii}$\leqslant$\texttt{nfpt1d}. See the user subroutine
\texttt{uspt1d}}

\minititre{Others}

\variab{dt}{dt(ncelet)}{ra}{Value of the time step}

\variab{ifmcel}{ifmcel(ncelet)}{ia}{Family number of the elements. See note 1}

\variab{s2kw}{s2kw(ncelet)}{ra}{Square of the norm of the deviatoric part of the deformation
rate tensor ($S^2=2S_{ij}^D S_{ij}^D$). This array is defined only
with the $k-\omega$ (SST) turbulence model}

\variab{divukw}{divukw}{ia}{Divergence of the velocity. More precisely it is the trace of the velocity gradient (and not a finite volume divergence term). In the
cell \texttt{iel},  $div(\vect{u})$ is given by \texttt{divukw(iel1)}.
This array is defined only with the $k-\omega$ SST turbulence model
(because in this case it may be calculated at the same time as $S^2$)}.

\minititre{Note: boundary conditions}
The \textbf{gradient} boundary conditions in \CS boil down to determine a value for the
current variable $\varia$ at the boundary faces $\fib$, that is to say $\varia_\fib$,
value expressed as a function of $\varia_{\centip}$, value of $\varia$ in $\centip$,
projection of the center of the adjacent cell on the straight line
perpendicular to the boundary face and crossing its center:
\begin{equation}
\varia_\fib=A_{\fib}^g +B_{\fib}^g \varia_{\centip}.
\end{equation}

For a face \texttt{ifac}, the pair of coefficients $A_{\fib}^g , \, B_{\fib}^g$ is
may be accessed using the \texttt{field\_get\_coefa\_s} and
\texttt{field\_get\_coefb\_s} functions, replacing \texttt{s} with \texttt{v}
for a vector.

The \textbf{flux} boundary conditions in \CS boil down to determine the value of the diffusive
flux of the
current variable $\varia$ at the boundary faces $\fib$, that is to say
 $D_{\ib} \left(K_\fib, \, \varia \right)$,
value expressed as a function of $\varia_{\centip}$, value of $\varia$ in $\centip$,
projection of the center of the adjacent cell on the straight line
perpendicular to the boundary face and crossing its center:
\begin{equation}
D_{\ib} \left(K_\fib, \, \varia \right) = A_{\fib}^f +B_{\fib}^f \varia_{\centip}.
\end{equation}

For a face \texttt{ifac}, the pair of coefficients $A_{\fib}^f , \, B_{\fib}^f$
may be accessed using the \texttt{field\_get\_coefaf\_s} and
\texttt{field\_get\_coefbf\_s} functions, replacing \texttt{s} with \texttt{v}
for a vector.

The \textbf{divergence} boundary conditions in \CS boil down to determine a value for the
current variable $\varia$ (mainly the Reynolds stress components, the divergence $\divv \left(\tens{R} \right)$ used in the calculation of the momentum equation) at the boundary faces $\fib$,
that is to say $\varia_\fib$,
value expressed as a function of $\varia_{\centip}$, value of $\varia$ in $\centip$,
projection of the center of the adjacent cell on the straight line
perpendicular to the boundary face and crossing its center:
\begin{equation}
\varia_\fib=A_{\fib}^d +B_{\fib}^d \varia_{\centip}.
\end{equation}

For a face \texttt{ifac}, the pair of coefficients $A_{\fib}^d , \, B_{\fib}^d$
may be accessed using the \texttt{field\_get\_coefad\_s} and
\texttt{field\_get\_coefbd\_s} functions, replacing \texttt{s} with \texttt{v}
for a vector.

\clearpage
%==================================
\subsubsubsection{User arrays}
%==================================
Modules containing user arrays accessible from all user subroutines may
be defined in the \texttt{user\_modules.f90} file. This file is
compiled before any other Fortran user file, to ensure modules
may be accessed in other user subroutines using the \texttt{use <module>}
construct. It may contain any routines or variables the user needs,
and contains no predefined routines or variables (i.e. the only
specificity of this file is that a file with this name is compiled before
all others).

%==================================
\subsubsubsection{Parallelism and periodicity}
%==================================

Parallelism is based on domain partitioning: each processor is assigned
a part of the domain, and data for cells on parallel boundaries
is duplicated on neighbouring processors in corresponding ``ghost'',
or ``halo'' cells (both terms are used interchangeably). Values in
these cells may be accessed just the same as values in regular cells.
Communication is only required when cell values are modified
using values from neighbouring cells, as the values in the ``halo'' can
not be computed correctly (since the halo does not have access to all
its neighbours), so halo values must be updated by copying values from
the corresponding cells on the neighbouring processor.

Compared to other tools using a similar system, a specificity of
\CS is the separation of the halo in two parts: a standard part,
containing cells shared through faces on parallel boundaries, and an
extended part, containing cells shared through vertices, which is
used mainly for least squares gradient reconstruction using an
extended neighbourhood. Most updates need only to operate on the standard
halo, requiring less data communication than those on the extended halos.

\begin{figure}[!h]
\centerline{
\includegraphics*[width=14cm]{halo}}
\caption{Parallel domain partitioning: halos}\label{fig:haluile}
\end{figure}

Periodicity is handled using the same halo structures as parallelism,
with an additional treatment for vector and coordinate values: updating
coordinates requires applying the periodic transformation to the copied
values, and in the case of rotation, updating vector and tensor values
also requires applying the rotation transformation.
Ghost cells may be parallel, periodic, or both. The example of a pump
combining parallelism and periodicity is given in \figurename~\ref{fig:parperio_pump}.
In this example, all periodic boundaries match with boundaries on
the same domain, so halos are either parallel or periodic.

\begin{figure}[!h]
\centerline{
\includegraphics*[width=5.5cm]{rota_perio_parall}}
\caption{Combined parallelism and periodicity}\label{fig:parperio_pump}
\end{figure}

\label{prg_paralperio}
{\bf Activation}

Parallelism is activated by means of the GUI or of the launch scripts
in the standard cases:
\begin{list}{$\bullet$}{}

\item On clusters with batch systems, the launching of a parallel run
      requires to complete the batch cards located in the
      beginning of \texttt{runcase} or \texttt{runcase\_batch}
      script, and set the number of MPI processes, or the numbers
      of physical nodes and processors per node (\texttt{ppn}) wanted.
      This can be done through the Graphical Interface or by editing
      the \texttt{runcase}  or \texttt{runcase\_batch} file directly.
      The number of processors defined here will override the number
      defined through the GUI in a non-batch environment
      (so that studies defined on one environment may be migrated
      to larger compute resources easily), but it may be overridden
      by the \texttt{define\_case\_parameters} function from
      the \texttt{cs\_user\_scripts.py} file, or by setting the
      \texttt{n\_procs\_weight}, \texttt{n\_procs\_min}, and
      \texttt{n\_procs\_max} parameters for the different domains
      defined in \texttt{runcase\_coupling}.

\item On clusters with unsupported batch systems,
      \texttt{runcase} file may have to be modified manually.
      Please do not hesitate to contact the \CS support
      (saturne-support@edf.fr) so that these modifications can be added to
      the standard launch script to make it more general.

\item A parallel calculation may be stopped in the same manner as a
      sequential one using the file \texttt{control\_file} (see paragraph
      \ref{sec:prg_control_file}).

\item The standard elements of information displayed in the listing
      (marked out with \texttt{'v  '} for the min/max values of the
      variables), \texttt{'c  '} for the data concerning the convergence
      and \texttt{'a  '} for the values before clipping) are global
      values for the whole domain and not related to each processor.

\end{list}

\vspace{0.5cm}
{\bf User subroutines}

The user can check in a subroutine
\begin{list}{-}{}
\item that the presence of periodicity is tested with the variable
      \texttt{iperio} (=1 if periodicity is activated);
\item that the presence of rotation periodicities is tested with the variable
      \texttt{iperot} (number of rotation periodicities);
\item that running of a calculation in parallel is tested for with the
      variable \texttt{irangp} (\texttt{irangp} is worth -1 in the case of a
      non-parallel calculation and $p-1$ in the case of a parallel calculation,
      $p$ being the number of the current processor)
\end{list}
Attention must be paid to the coding of the user subroutines. If
conventional subroutines like \texttt{cs\_user\_parameters.f90} or \texttt{cs\_user\_boundary\_conditions}
usually do not cause any problem, some kind of developments are more
complicated. The most usual cases are dealt with below. \\ Examples are
given for the subroutine \texttt{cs\_user\_extra\_operations}.
\begin{list}{$\bullet$}{}
\item {\bf Access to information related to neighbouring cells in
      parallel and periodic cases}.\\
When periodicity or parallelism are brought into use, some cells of the
      mesh become physically distant from their neighbours. Concerning
      parallelism, the calculation domain is split and distributed
      between the processors: a cell located at the ``boundary'' of a
      given processor may have neighbours on different processors. \\
In the same way, in case of periodicity, the neighbouring cells of cells
      adjacent to a periodic face are generally distant. \\
When data concerning neighbouring cells are required for the
      calculation, they must first be searched on the other processors
      or on the other edge of periodic frontiers. In order to ease the
      manipulation of these data, they are stored temporarily in virtual
      cells called ``halo'' cells, as can be seen in \figurename~\ref{fig:haluile}.
It is in particular the case when the following operations are made on a
      variable $A$:
\begin{list}{-}{}
\item calculation of the gradient of $A$ (use of the subroutine \texttt{grdcel});
\item calculation of an internal face value from the values of $A$  in
      the neighbouring cells (use of \texttt{ifacel}).
\end{list}
The variable $A$ must be exchanged before these operations can be
      made: to allow it, the subroutine \texttt{synsca} may be called.

\item {\bf Global operations in parallel mode}.\\
In parallel mode, the user must pay attention when performing
      global operations. The following list is not exhaustive:
        \begin{list}{-}{}
\item calculation of extreme values on the domain (for instance, minimum
      and maximum of some calculation values);
\item test of the existence of a certain value (for instance, do faces
      of a certain color exist?);
\item verification of a condition on the domain (for instance, is a
      given flow value reached somewhere?);
\item counting out of entities (for instance, how many cells have
      pressure drops?);
\item global sum (for instance, calculation of a mass flow or the total
      mass of a pollutant).
      \end{list}
The user may refer to the different examples present in the directory \texttt{EXAMPLES} in the\\
\texttt{cs\_user\_extra\_operations-parallel\_operations.f90} file.
Care should be taken with the fact that the boundaries between
      subdomains consist of {\bf internal} faces shared between
      two processors (these are indeed internal faces, even if they are
      located at a ``processor boundary''). They should not be counted twice
      (once per processor) during global operations using internal faces
      (for instance, counting the internal faces per processor and
      summing all the obtained numbers drives into over-evaluating the
      number of internal faces of the initial mesh).

\item {\bf Writing operations that should be made on one
      processor only in parallel mode}.\\
In parallel mode, the user must pay attention during the writing of
      pieces of information. Writing to the ``listing'' can be done
      simply by using the \texttt{nfecra} logical unit (each processor will write
      to its own ``listing'' file): use
      \texttt{write(nfecra, ...}. \\
If the user wants an operation to be done by only one processor (for
      example, open or write a file), the associated instructions must
      be included inside a test on the value of \texttt{irangp} (generally it is
      the processor 0 which realises these actions, and we want the
      subroutine to work in non-parallel mode, too: \texttt{if
      (irangp.le.0) then ...}).
\end{list}

{\bf Some notes about periodicity}

Note that periodic faces are not part of the domain boundary:
periodicity is interpreted as a ``geometric'' condition
rather than a classical boundary condition.

Some particular points should be reminded:
\begin{list}{-}{}
\item Periodicity can also work when the periodic boundaries are meshed
      differently (periodicity of non-conforming faces), {\it except} for
      the case of a 180 degree rotation periodicity with faces coupled
      on the rotation axis.
\item rotation periodicity is incompatible with
  \begin{list}{-}{}
  \item semi-transparent radiation,
  \item reinforced velocity-pressure coupling (\texttt{ipucou=1)}.
  \end{list}
\item although it has not been the case so far, potential problems might be met
      in the case of rotation periodicity with the $R_{ij}-\varepsilon$ (LRR)
      model. They would come from the way of taking into account the
      orthotropic viscosity (however, this term usually has a low influence).
\end{list}

%==================================
\subsubsubsection{Geometry and particle arrays
      related to Lagrangian modelling}
%==================================

In this section is given a non-exhaustive list of the main variables
which may be seen by the user in the Lagrangian module. Most of them
should not be modified by the user. They are calculated automatically
from the data. However it may be useful to know their meaning.

\noindent
These variables are listed in the alphabetical index in the end of this
document.

\noindent
The type of each variable is given: integer [i], real number [r],
integer array [ia], real array [ra].

\minititre{Size of the Lagrangian arrays}

\variab{lndnod}{lndnod}{i}{Size of the array \texttt{icocel} concerning the
cells $\rightarrow$ faces connectivity (the faces $\rightarrow$ nodes connectivity
must be given to allow the construction of this connectivity. See note 3
of section \ref{sec:prg_dimensions})}

\variab{nbpmax}{nbpmax}{i}{Maximum number of particles
simultaneously allowed in the calculation domain}

\variab{nvp}{nvp}{i}{Number of variables describing the particles for
which a stochastic differential equation (SDE) is solved}

\variab{nvls}{nvls}{i}{Number of variables describing the supplementary
user particles for which a SDE is solved}

\variab{nvep}{nvep}{i}{Number of real state variables describing the particles}

\variab{nivep}{nivep}{i}{Number of integer state variables describing the particles}

\variab{ntersl}{ntersl}{i}{Number of source terms representing the backward
coupling of the dispersed phase on the continuous phase}

\variab{nvlsta}{nvlsta}{i}{Number of volumetric statistical variables }

\variab{nvlsts}{nvlsts}{i}{Number of supplementary user volumetric
statistical variables}

\variab{nvisbr}{nvisbr}{i}{Number of boundary statistical variables}

\variab{nusbor}{nusbor}{i}{Number of supplementary user boundary statistical
variables}

\variab{nvgaus}{nvgaus}{i}{Number of Gaussian random variables}

\minititre{Lagrangian arrays}

\variab{icocel}{icocel(lndnod)}{ia}{Cells $rightarrow$ internal/boundary faces
connectivity. The numbers of the boundary faces are marked out in \texttt{icocel}
with a negative sign}

\variab{itycel}{itycel(ncelet+1)}{ia}{Array containing the position of
the first face surrounding every cell in the array \texttt{icocel} (see subroutine
\texttt{lagdeb} for more details)}

\variablist{eptp}{eptp(nvp,*)}{ra}
{Variables forming the state vector related to the particles: either at
the current stage if the Lagrangian scheme is a second-order, or at the
current time step if the scheme is a first-order. These variables are
marked out by ``pointers'' whose value can vary between 1 and \texttt{nvp}:
\begin{list}{$\rightarrow$}{}
\item \texttt{jmp}: particle mass
\item \texttt{jdp}: particle diameter
\item \texttt{jxp}, \texttt{jyp}, \texttt{jzp}: particle coordinates
\item \texttt{jup}, \texttt{jvp}, \texttt{jwp}: particle velocity components
\item \texttt{juf}, \texttt{jvf}, \texttt{jwf}: locally undisturbed fluid flow
      velocity components
\item \texttt{jtp}, \texttt{jtf}: particle and locally undisturbed fluid flow
      temperature (\degresC)
\item \texttt{jcp}: particle specific heat
\item \texttt{jhp}: coal particle temperature (\degresC)
\item \texttt{jmch}: mass of reactive coal of the coal particle
\item \texttt{jmck}: mass of coke of the coal particle
\item \texttt{jvls(ii): ii\textit{th}} supplementary user variable
\end{list}
}

\variab{eptpa}{eptpa(nvp,nbpmax)}{ra}{Variables forming the state vector
related to the particles: either at the previous stage if the Lagrangian
scheme is a second-order, or at the previous time step if the Lagrangian
scheme is a first-order}

\variablist{ipepa}{ipepa(nivep,nbpmax)}{ia}{Integer variables
related to the particles. They are marked out by the following ``pointers'':
\begin{list}{$\rightarrow$}{}
\item \texttt{jisor}: Number of the current cell containing the particle; this
      number is re-actualised during the trajectography step
\item \texttt{jinch}: Number of the coal particle
\end{list}
}

\variablist{pepa}{pepa(nvep,nbpmax)}{ra}{Real variables
related to the particles. They are marked out by the following ``pointers'':
\begin{list}{$\rightarrow$}{}
\item \texttt{jrtsp}: particle residence time
\item \texttt{jrpoi}: particle statistic weight
\item \texttt{jrdck}: coal particle shrinking core diameter
\item \texttt{jrd0p}: coal particle initial diameter
\item \texttt{jrr0p}: coal particle initial density
\end{list}
}

\variab{indep}{indep(nbpmax)}{ia}{Storage of the cell number of every
particle at the beginning of a Lagrangian iteration; this data is not
modified during the iteration}

\variab{vitpar}{vitpar(nbpmax,3)}{ra}{At the beginning of the
trajectography, \texttt{vitpar} contains the particle velocity vector components;
the modifications of the particle velocity following every
particle/boundary interaction are saved in this array; after the
trajectography and backward coupling steps, \texttt{eptp} is updated with
\texttt{vitpar}}

\variab{vitflu}{vitflu(nbpmax,3)}{ra}{At the beginning of the
trajectography, \texttt{vitflu} contains the locally undisturbed fluid flow
velocity vector components;
the modifications of the locally undisturbed fluid flow velocity
following every
particle/boundary interaction are saved in this array; after the
trajectography and backward coupling steps, \texttt{eptp} is updated with
\texttt{vitflu}}

\variab{gradpr}{gradpr(ncelet,3)}{ra}{Pressure gradient of the
continuous phase}

\variab{gradvf}{gradvf(ncelet,9)}{ra}{Gradient of the continuous phase
fluid velocity (useful if the complete model is activated: see \texttt{modcpl})}

\variab{cpgd1}{cpgd1(nbpmax)}{ra}{First de-volatilisation term (light
volatile matters) of the coal particles (useful in the case of backward
coupling on the continuous phase)}

\variab{cpgd2}{cpgd2(nbpmax)}{ra}{Second devolatilisation term (heavy
volatile matters) of the coal particles (useful in the case of backward
coupling on the continuous phase)}

\variab{cpght}{cpght(nbpmax)}{ra}{Heterogeneous combustion term of the
coal particles (useful in the case of backward coupling on the
continuous phase)}

\variablist{statis}{statis(ncelet,nvlsta)}{ra}{Averages of the volumetric
variables related to the dispersed phase They can be accessed by the following ``pointers'':
\begin{list}{$\rightarrow$}{}
\item \texttt{ilvx,ilvy,ilvz}: mean dispersed phase velocity
\item \texttt{ilfv}: dispersed phase volumetric concentration
\item \texttt{ilpd}: sum of the statistical weights
\item \texttt{iltp}: dispersed phase temperature (\degresC)
\item \texttt{ildp}: dispersed phase mean diameter
\item \texttt{ilmp}: dispersed phase mean mass
\item \texttt{ilhp}: temperature of the coal particle cloud (\degresC)
\item \texttt{ilmch}: mass of reactive coal of the coal particle cloud
\item \texttt{ilmck}: mass of coke of the coal particle cloud
\item \texttt{ilmdk}: shrinking core diameter of the coal particle cloud
\item \texttt{ilvu(ii): ii\textit{th}} supplementary user volumetric statistics
\end{list}
}

\variablist{statis}{stativ(ncelet,nvlsta)}{ra}{Variances of the volumetric variables
related to the dispersed phase. they can be accessed by using the same ``pointers'' as
the ones used for the \texttt{stativ} array.}

\variablist{parbor}{parbor(nfabor,nvisbr)}{ra}{Boundary statistics
related the dispersed phase; after every particle/boundary
interaction it is possible to save some data and to calculate averages;
the boundary statistics are marked out by the following ``pointers'':
\begin{list}{$\rightarrow$}{}
\item \texttt{inbr}: number of particle/boundary interactions
\item \texttt{iflm}: particle mass flow at the boundary faces
\item \texttt{iang}: mean interaction angle with the boundary faces
\item \texttt{ivit}: mean interaction velocity with the boundary faces
\item \texttt{ienc}: mass of coal deposit at the walls
\item \texttt{iusb(ii): ii\textit{th}} supplementary user boundary statistics
\end{list}}

\variablist{cs\_glob\_lagr\_source\_terms}{cs\_glob\_lagr\_source\_terms(ncelet,ntersl)}{ra}{Source terms
corresponding to the backward coupling of the dispersed phase on the
continuous phase. These source terms are marked out by the following
``pointers'':
\begin{list}{$\rightarrow$}{}
\item \texttt{itsvx}, \texttt{itsvy}, \texttt{itsvz}:
      explicit source terms for the continuous phase velocity
\item \texttt{itsli}: implicit source term for the continuous phase velocity and
      for the turbulent energy if the $k-\varepsilon$ model is used
\item \texttt{itske}: explicit source term for the turbulent dissipation and the
      turbulent energy if the $k-\varepsilon$ turbulence model is used for the
      continuous phase
\item \texttt{itsr11}, ... \texttt{itsr33}: source terms for the Reynolds stress
      and the turbulent dissipation if the $R_{ij}-\varepsilon$ turbulence model
      is used for the continuous phase
\item \texttt{itsmas}: mass source term
\item \texttt{itste}, \texttt{itsti}: explicit and implicit thermal source terms
      for the thermal scalar of the continuous phase
\item \texttt{itsmv1(icha)}, \texttt{itsmv2(icha)}: source terms respectively
      for the light and heavy volatile matters
\item \texttt{itsco}: source term for the carbon released during
      heterogeneous combustion
\end{list}}

\variab{croule}{croule(ncelet)}{ra}{Importance function for the
technique of variance reduction (cloning/fusion of particles)}

\variab{vagaus}{vagaus(nbpmax,nvgaus)}{ra}{Vectors of Gaussian random
variables}

\variab{auxl}{auxl(nbpmax,3)}{ra}{Auxiliary work array}


%==================================
\subsubsubsection{Variables saved to allow calculation restarts}
%==================================

The directory \texttt{checkpoint} contains:
\begin{list}{-}{}
\item \texttt{main}: main restart file,
\item \texttt{auxiliary}: auxiliary restart file (see \texttt{ileaux\index{ileaux}},
                       \texttt{iecaux\index{iecaux}}),
\item \texttt{radiative\_transfer}: restart file for the radiation module,
\item \texttt{lagrangian}: main restart file for the Lagrangian module,
\item \texttt{lagrangian\_stats}: auxiliary restart file for the Lagrangian module (mainly for the statistics),
\item \texttt{1dwall\_module}: restart file for the 1D wall thermal module,
\item \texttt{vortex}: restart file for the vortex method (see
                       \texttt{ivrtex\index{ivrtex}}).
\end{list}

The main restart file contains the values in every cell of the mesh for
pressure, velocity, turbulence variables and all the scalars (user scalars et specific physics scalars.
Its content is sufficient for a calculation restart, but the complete continuity of the solution at
restart is not ensured\footnote{In other words, a restart calculation of n time
steps following a calculation of m time steps will not yield strictly the same
results as a direct calculation on m+n time steps, whereas it is the case when
the auxiliary file is used}.

The auxiliary restart file completes the main restart file to ensure
solution continuity in the case of a calculation restart.
If the code cannot find one or several pieces of data required for the
calculation restart in the auxiliary restart file, default values are
then used. This allows in particular to run calculation restarts even if
the number of faces has been modified (for instance in case of
modification of the mesh merging or of periodicity
conditions\footnote{Imposing a periodicity changes boundary faces into
internal faces}). More precisely, the auxiliary restart file contains
the following data:

\begin{list}{-}{}
\item type and value of the time step, turbulence model,
\item density value at the cells and boundary faces, if it is variable,
\item values at the cells of the other variable physical properties,
when they are extrapolated in time (molecular dynamic viscosity, turbulent or
sub-grid scale viscosity, specific heat, scalar diffusivity). The specific heat is stored automatically for the Joule
effect (in case the user should need
it at restart to calculate the temperature from the enthalpy before the new
specific heat has been estimated),
\item time step value at the cells, if it is variable,
\item mass flow value at the internal and boundary faces (at the last
time step, and also at the previous time step if required by the time scheme),
\item boundary conditions,
\item values at the cells of the source terms when they are extrapolated in time,
\item number of time-averages, and values at the cells of the associated
cumulated values,
\item for each cell, distance to the wall when it is required (and
index-number of the nearest boundary face, depending on \texttt{icdpar}%
\index{\texttt{icdpar}}),
\item values at the cells of the external forces in balance with a part
of the pressure (hydrostatic, in general),
\item for the D3P gas combustion model: massic enthalpies and temperatures at entry,
type of boundary zones and entry indicators,
\item for the EBU gas combustion model: temperature of the fresh gas, constant
mixing rate (for the models without mixing rate transport), types of boundary
zones, entry indicators, temperatures and mixing rates at entry,
\item for the LWC gas combustion model: the boundaries of the probability
density functions for enthalpy and mixing rate, types of boundary
zones, entry indicators, temperatures and mixing rates at entry,
\item for the pulverised coal combustion: coal density,  types of boundary
zones, variables \texttt{ientat}, \texttt{ientcp}, \texttt{inmoxy}, \texttt{timpat}, \texttt{x20}
(in case of coupling with the Lagrangian module, \texttt{iencp} and \texttt{x20}
are not saved),
\item for the pulverised fuel combustion: types of boundary
zones, variables \texttt{ientat}, \texttt{ientfl}, \texttt{inmoxy}, \texttt{timpat}, \texttt{qimpat}
, \texttt{qimpfl},
\item for the electric module: the tuned potential difference \texttt{dpot}%
\index{\texttt{dpot}}
and, for the electric arcs module, the tuning coefficient \texttt{coejou}%
\index{\texttt{coejou}}
(when the boundary conditions are tuned), the Joule source term for the enthalpy
(when the Joule effect is activated) and the Laplace forces (with the electric
arc module).
\end{list}

It should be noted that, if the auxiliary restart file is read, it is
possible to run calculation restarts with relaxation of the
density\footnote{Such a relaxation only makes sense for a steady
calculation}(when it is variable), because this variable is stored in the
restart file. On the other hand, it is generally not possible to do the
same with the other physical properties (they are stored in the restart
file only when they are extrapolated in time, or with the Joule effect for the
specific heat).

Apart from \texttt{vortex} which has a different structure and is
always in text format, all the restart files are binary
files. Nonetheless, they may be dumped or compared using
the \texttt{cs\_io\_dump} tool.

In the case of parallel calculations, it should be noted that all the processors
will write their restart data in the same files. Hence, for instance, there will
always be one and only one \texttt{main} file, whatever the number of
processors used. The data in the file are written according to the initial full
domain ids for the cells, faces and nodes. This allows in particular
to restart using {\it p} processors a calculation begun with {\it n} processors,
or to make the restart files independent of any mesh renumbering that may
be carried out in each domain.

{\em WARNING: if the mesh is composed of several files, the order
in which they appear in the launch script or in the Graphical Interface must not
be modified in case of a calculation restart\footnote{When uncertain, the user
can check the saved copy of the launch script in the \texttt{RESU} directory, or
the head of the \texttt{preprocessor*.log} files, which repeat the
command lines passed to the Preprocessor module}.}

{\em NOTE: when joining of faces or periodicity is used, two nodes closer
than a certain (small) tolerance will be merged. Hence, due to numerical
round-up errors, two different machines may yield different results.
This might change the number of faces in the global
domain\footnote{The number of cells will not be modified, it is always the sum of the
number of cells of the different meshes} and make restart files
incompatible. Should that problem arise when making a calculation restart on a
different architecture, the solution is to ignore the \texttt{auxiliary}
file and use only the \texttt{main} file, by setting \texttt{ileaux = 0}
in \texttt{cs\_user\_parameters.f90}}

%==================================
\subsubsection{Using selection criteria in user subroutines}
%==================================
\label{sec:fvm_selector}

In order to use selection criteria (cf. \S\ref{sec:selection_criteria}) in Fortran
user subroutines, a collection of utility subroutines is provided. The aim is to
define a subset of the mesh, for example:

\begin{list}{-}{}
\item boundary regions (cf. \texttt{cs\_user\_boundary\_conditions}, \texttt{usalcl},
\texttt{cs\_user\_radiative\_transfer\_bcs.f90}, \texttt{uslag2}, ...),
\item volume initialization (cf. \texttt{cs\_user\_initialization}, ...),
\item head-loss region (cf. \texttt{cs\_user\_head\_losses.f90}),
\item source terms region (cf. \texttt{cs\_user\_source\_terms}),
\item advanced post-processing (cf. \texttt{cs\_user\_postprocess.c},
      \texttt{cs\_user\_extra\_operations}, ...),
\end{list}

This section explains how to define surface or volume sections,
in the form of lists \texttt{lstelt} of \texttt{nlelt} elements
(internal faces, boundary faces or cells).
For each type of element, the user calls the appropriate Fortran
subroutine: \texttt{getfbr}
for boundary faces, \texttt{getfac} for internal faces
and \texttt{getcel} for cells. All of these take
the three following arguments:
\begin{list}{-}{}
\item the character string which contains the selection
      criterion (see some examples below),
\item the returned number of elements \texttt{nlelt},
\item the returned list of elements \texttt{lstelt}.
\end{list}

Several examples of possible selections are given here:
\begin{list}{-}{}
\item \verb+call getfbr('Face_1, Face_2', nlelt, lstelt)+ to select
  boundary faces in groups Face\_1 or Face\_2,
\item \verb+call getfac('4', nlelt, lstelt)+ to select internal
  faces of color 4,
\item \verb+call getfac('not(4)', nlelt, lstelt)+ to select internal
  faces which have a different color than 4,
\item \verb+call getfac('4 to 8', nlelt, lstelt)+ to internal faces
with color between 4 and 8 internal faces,
\item \verb+call getcel('1 or 2', nlelt, lstelt)+ to select cells
  with colors 1 or 2,
\item \verb+call getfbr('1 and y > 0', nlelt, lstelt)+ to select boundary
  faces of color 1 which have the coordinate $Y > 0$,
\item \verb+call getfac('normal[1, 0, 0, 0.0001]', nlelt, lstelt)+ to select
internal faces which have a normal direction to the vector (1,0,0),
\item \verb+call getcel('all[]', nlelt, lstelt)+ to select all cells.
\end{list}

The user may then use a loop on the selected elements.\\
For instance, in the subroutine \texttt{cs\_user\_boundary\_conditions} used to impose
boundary  conditions, let us consider the boundary faces of color
number 2 and which have the coordinate $X <= 0.01$
(so that \verb+call getfbr('2 and x <= 0.01', nlelt,lstelt)+);
we can do a loop (\verb+do ilelt = 1, nlelt+) and
obtain \verb+ifac = lstelt(ilelt)+.

\minititre{Note: legacy method using explicit families and properties}

The selection method for user subroutines by prior versions of \CS
is still available, though it may be removed in future versions.
This method was better adapted to working with colors than with groups,
and is explained here:

From \CS 's point of view, all the references to mesh entities (boundary faces
and volume elements) correspond to a number (color number or negative
of group number) associated with the entity. An entity may have several
references (for instance, one entity may have one color and belong to
several groups). In \CS, these references may be designated as
``properties''. \\
The mesh entities are gathered in equivalence classes on the base of
their properties. These equivalence classes are called ``families''. All
the entities of one family have the same properties. In order to know
the properties (in particular the color) of an entity (a boundary face
for example), the user must first determine the family to which it
belongs. \\
For instance, let's consider a mesh whose boundary faces have all been
given one color (for example using SIMAIL). The family of the boundary
face \texttt{ifac} is \texttt{ifml=ifmfbr(ifac)}. The first (and only)
property of this family is the color \texttt{icoul}, obtained for the face
\texttt{ifac} with \texttt{icoul=iprfml(ifml,1)}. In order to know the
property number corresponding to a group, the utility function
\texttt{numgrp(nomgrp, lngnom)} (with a name
\texttt{nomgrp} of the type \texttt{character*} and its length
\texttt{lngnom} of the type \texttt{integer}) can be used.

%==================================
\subsection{Face and cell mesh-defined properties and selection}
%==================================
\label{sec:selection_criteria}

The mesh entities may be referenced by the user during the mesh
creation. These references may then be used to mark out some mesh entities
according to the need (specification of boundary conditions, pressure
drop zones, ...). The references are generally of one of the two
following types:
\begin{list}{$\bullet$}{}
\item color.
A color is an integer possibly associated with boundary faces and
volume elements by the mesh generator. Depending on the tool,
this concept may have different names, which \CS interprets
as colors. Most tools allow only one color per face or element.
      \begin{list}{-}{}
      \item I-deas uses a color number with a default of
            7 (green) for elements, be they volume elements or boundary
            ``surface coating'' elements. Color 11 (red) is used for
            for vertices, but vertex properties are ignored by \CS.
      \item SIMAIL uses the equivalent notions of ``reference''
            for element faces, and ``subdomain'' for volume elements.
            By default, element faces are assigned no reference (0),
            and volume elements domain 1.
      \item Gmsh uses ``physical property'' numbers.
      \item EnSight has no similar notion, but if several parts
            are present in an EnSight 6 file, or several parts
            are present \emph{and} vertex ids are given in an
            Ensight Gold file, the part number is interpreted as
            a color number by the Preprocessor.
      \item The MED 2.3 model allowed integer ``attributes'', though
            many tools working with this format ignored those
            and only handle groups.
      \end{list}
\item groups.
Named ``groups'' of mesh entities may also be used with many
mesh generators or formats. In some cases, a given cell or face may belong
to multiple groups (as some tools allow new groups to be defined
by boolean operations on existing groups).
In \CS, every group is assigned a group number (base on alphabetical
ordering of groups).
      \begin{list}{-}{}
      \item I-deas assigns a group number with each
            group, but by default, this number is just a counter.
            Only the group name is considered by \CS (so that elements
            belonging to two groups with identical names and different
            numbers are considered as belonging to the same group).
      \item CGNS allows both for named boundary conditions and mesh
            sections. If present, boundary condition names are
            interpreted as group names, and groups may also be defined
            based on element section or zone names using additional
            Preprocessor options (\texttt{-grp-cel} or
            \texttt{-grp-fac} followed by \texttt{section} or
            \texttt{zone}).
      \item Using the MED format, it is preferable to use ``groups''
            than colors, as many tools ignore the latter.
      \end{list}
\end{list}

Selection criteria may be defined in a similar fashion whether
using the GUI or in user subroutines.
Typically, a selection criteria is simply a string containing
the required color numbers or group names, possibly combined
using boolean expressions. Simple geometric criteria are also
possible.

A few examples are given below:

\verb+ENTRY+\\
\verb+1 or 7+\\
\verb+all[]+\\
\verb+3.1 >= z >= -2 or not (15 or entry)+\\
\verb+range[04, 13, attribute]+\\
\verb+sphere[0, 0, 0, 2] and (not no_group[])+

Strings such as group names containing white-space
or having names similar to reserved operators may be protected
using ``escape characters''.\footnote{Note that for defining a
string in Fortran, double quotes are easier to use, as they do not
conflict with Fortran's single quotes delimiting a string.
In C, the converse is true. Also, in C, to define a string
such as \texttt{{$\backslash$}plane}, the string
\texttt{{$\backslash$}{$\backslash$}plane} must be
used, as the first $\backslash$ character is used by the
compiler itself. Using the GUI, either notation is easy.}
More complex examples of strings with protected strings are given here:

\verb+"First entry" or Wall\ or\ sym+\\
\verb+entry or \plane or "noone's output"+

The following operators and syntaxes are allowed (fully capitalized
versions of keywords are also allowed, but mixed upper-case/lower-case
versions are not):

\begin{tabular}[top]{p{6cm} l}
\multicolumn{2}{l}{\bf escape characters }\\
protect next character only: & \texttt{$\backslash$} \\
protect string:              & \texttt{{'}$string${'}} \quad \texttt{"$string$"}\\
\end{tabular}

\begin{tabular}[top]{p{6cm} l}
\multicolumn{2}{l}{\bf basic operators }\\
priority: & \texttt{(} \quad \texttt{)} \\
not:      &  \texttt{not} \quad \texttt{!} \quad \texttt{!=} \\
and:      &  \texttt{and} \quad \texttt{\&} \quad \texttt{\&\&} \\
or:       &  \texttt{or} \quad \texttt{|} \quad \texttt{||} \quad \texttt{,} \quad \texttt{;} \\
xor:      &  \texttt{xor} \quad \texttt{\^} \\
\end{tabular}

\begin{tabular}[top]{p{6cm} l}
\multicolumn{2}{l}{\bf general functions }\\
select all:                        &  \texttt{all[]}\\
entities having no group or color: &  \texttt{no\_group[]} \\
select a range of groups or colors: &  \texttt{range[$first$, $last$]} \\
                                    &  \texttt{range[$first$, $last$, group]} \\
                                    &  \texttt{range[$first$, $last$, attribute]} \\
\end{tabular}

For the range operator, $first$ and $last$ values are inclusive.
For attribute (color) numbers, natural integer value ordering is used,
while for group names, alphabetical ordering is used. Note also that in
the bizarre (not recommended) case in which a mesh would contain for
example both a color number 15 and a group named ``15'', using
\texttt{range[15, 15, group]} or \texttt{range[15, 15, attribute]}
could be used to distinguish the two.

Geometric functions are also available. The coordinates considered are
those of the cell or face centres. Normals are of course
usable only for face selections, not cell selections.

\begin{tabular}[top]{p{6cm} l}
\multicolumn{2}{l}{\bf geometric functions }\\
face normals:   &  \texttt{normal[$x$, $y$, $z$, $epsilon$]} \\
                &  \texttt{normal[$x$, $y$, $z$, epsilon = $epsilon$]} \\
plane, $ax + by + cz + d = 0$ form: &  \texttt{plane[$a$, $b$, $c$, $d$, $epsilon$]} \\
                &  \texttt{plane[$a$, $b$, $c$, $d$, epsilon = $epsilon$]} \\
                &  \texttt{plane[$a$, $b$, $c$, $d$, inside]} \\
                &  \texttt{plane[$a$, $b$, $c$, $d$, outside]} \\
plane, normal + point in plane form: &  \texttt{plane[$n_x$, $n_y$, $n_z$, $x$, $y$, $z$, $epsilon$]} \\
                &  \texttt{plane[$n_x$, $n_y$, $n_z$, $x$, $y$, $z$, epsilon = $epsilon$]} \\
                &  \texttt{plane[$n_x$, $n_y$, $n_z$, $x$, $y$, $z$, inside]} \\
                &  \texttt{plane[$n_x$, $n_y$, $n_z$, $x$, $y$, $z$, outside]} \\
box, extents form: &  \texttt{box[$x_{min}$, $y_{min}$, $z_{min}$,
                                  $x_{max}$, $y_{max}$, $z_{max}$]} \\
box, origin + axes form: &  \texttt{box[$x_0$, $y_0$, $z_0$,}\\
                         &  \texttt{\qquad $dx_1$, $dy_1$, $dz_1$,
                                        $dx_2$, $dy_2$, $dz_2$,
                                        $dx_3$, $dy_3$, $dz_3$]} \\
cylinder:       &  \texttt{cylinder[$x_0$, $y_0$, $z_0$, $x_1$, $y_1$, $z_1$, $radius$]} \\
sphere:         &  \texttt{sphere[$x_c$, $y_c$, $z_c$, $radius$]} \\
inequalities:   &  \texttt{>}, \texttt{<}, \texttt{>=}, \texttt{<=} associated
                   with \texttt{x}, \texttt{y}, \texttt{z}  or
                   \texttt{X}, \texttt{Y}, \texttt{Z} keywords\\
                &   and coordinate value; \\
                &  \texttt{$x_{min}$ <= x < $x_{max}$} type syntax is allowed. \\
\end{tabular}

In the current version of \CS, all selection criteria used
are maintained in a list, so that re-interpreting a criterion already
encountered (such as at the previous time step) is avoided.
Lists of entities corresponding to a criteria containing no geometric
functions are also saved in a compact manner, so re-using a previously
used selection should be very fast. For criteria containing geometric
functions, the full list of corresponding entities is not maintained,
so each entity must be compared to the criterion at each time step.
Heavy use of many selection criteria containing geometric functions
may thus lead to reduced performance.

%==================================
\section{Importing and preprocessing meshes}
%==================================

Importing meshes is done by the \pcs module, while and preprocessing
is done mainly by the code Kernel (except for element orientation
checking, which is done by the \pcs).

The \pcs module of \CS reads the
mesh file(s) (under any supported format) and translates the necessary
information into a Kernel input file.

When multiple meshes are used, the \pcs is called once per mesh,
and each resulting output is added in a \texttt{mesh\_input}
directory (instead of a single \texttt{mesh\_input} file).

The executable of the \pcs module is \texttt{cs\_preprocess}, and
is normally called through the run script, so it is not in standard paths
(it is at \texttt{<prefix>/libexec/code\_saturne/cs\_preprocess}).
Its most useful options and sub-options are described briefly here.
To obtain a complete and up-to-date list of options and environment
variables, use the following command:
\texttt{cs\_preprocess~-h} or \texttt{cs\_preprocess~--help}.
Many options, such as this one, accept a short and a long version.

Nonetheless, it may be useful to call the \pcs manually
in certain situations, especially for frequent verification when
building a mesh, so its use is described here. Verification
may also be done using the GUI or the mesh quality check mode
of the general run script.

The \pcs is controlled using command-line arguments.
A few environment variables allow advanced users to modify
some behaviours or to obtain a trace of memory management.

%==================================
\subsection{\pcs options\label{sec:optpcs}}
%==================================

Main choices are done using command-line options. For example:

\texttt{cs\_preprocess --num 2 fluid.med}

\noindent means that we read the second mesh defined in the
\texttt{fluid.med} file, while:

\texttt{cs\_preprocess --no-write --post-volume med fluid.msh}

\noindent means that we read file \texttt{fluid.msh}, and
do not produce a \texttt{mesh\_input}  file, but do output
a \texttt{fluid.med} file (effectively converting a Gmsh file to
a MED file).

\subsubsection{Mesh selection\label{sec:optpcs:mesh}}

Any use of the preprocessor requires one mesh file (except for
\texttt{cs\_preprocess} and \texttt{cs\_preprocess~-h} which respectively
print the version number and list of options).
This file is selected as the last argument to \texttt{cs\_preprocess},
and its format is usually automatically determined based on its
extension (c.f. \ref{sec:formats_in} page~\pageref{sec:formats_in})
but a \texttt{--format} option allows forcing the format choice of
the selected file.

For formats allowing multiple meshes in a single file, the
\texttt{--num} option followed by a strictly positive integer allows
selection of a specific mesh; by default, the first mesh is selected.

For meshes in CGNS format, we may in addition use the \texttt{--grp-cel}
or \texttt{--grp-fac} options, followed by the \texttt{section}
or \texttt{zone} keywords, to define additional groups of cell or faces
based on the organization of the mesh in sections or zones. The sub-options
have no effect on meshes of other formats.

\subsubsection{Post-processing output\label{sec:optpcs:post}}

By default, the \pcs does not generate any post-processor output.
By adding \texttt{--post-volume [format]},
with the optional \texttt{format} argument being one of \texttt{ensight},
\texttt{med}, or \texttt{cgns} to the command-line arguments,
the output of the volume mesh to the default or indicated format
is provoked.

In case of errors, output of error visualization output is always
produced, and by adding \texttt{--post-error [format]},
the format of that output may be selected (from one of \texttt{ensight},
\texttt{med}, or \texttt{cgns}, assuming MED and CGNS are
available),

\subsubsection{Element orientation correction\label{sec:optpcs:orient}}

Correction of element orientation is possible and can be activated using the
\texttt{--reorient} option.

Note that we cannot guarantee correction (or even detection) of a bad
orientation in all cases.
Not all local numbering possibilities of elements are tested,
as we focus on ``usual'' numbering permutations. Moreover,
the algorithms used may produce false positives or fail to find
a correct renumbering in the case of highly non convex elements.
In this case, nothing may be done short of modifying the mesh, as
without a convexity hypothesis, it is not always possible to choose
between two possible definitions starting from a point set.

With a post-processing option such as \texttt{--post-error}
or, \texttt{--post-volume},
visualizable meshes of corrected elements as well as remaining
badly oriented elements are generated.

\subsection{Environment variables\label{sec:envvpcs}}

Setting a few environment variables specific to the \pcs allows modifying
its default behaviour. In general, if a given behaviour is modifiable
through an environment variable rather than by a command-line option,
it has little interest for a non-developer, or its modification is
potentially hazardous. The environment variables used by the \pcs
are described here:

\envvar{CS\_RENUMBER}

Deactivating renumbering is possible by setting \code{CS\_RENUMBER=off}.

\envvar{CS\_PREPROCESS\_MEM\_LOG}

Allows defining a file name in which memory allocation, reallocation,
and freeing is logged.

\envvar{CS\_PREPROCESS\_MIN\_EDGE\_LEN}

Under the indicated length ($10^{-15}$ by default), an edge is considered
to be degenerate and its vertices will be merged after the transformation
to descending connectivity. Degenerate edges and faces will thus be
removed. Hence, the post-processed element does not change, but the
Kernel may handle a prism where the preprocessor input contained a
hexahedron with two identical vertex couples (and thus a face of zero
surface). If the \pcs does not print any information relative to this
type of correction, it means that it has not been necessary. To completely
deactivate this automatic correction, a negative value may be assigned
to this environment variable.

\envvar{CS\_PREPROCESS\_IGNORE\_IDEAS\_COO\_SYS}

If this variable is defined and is a strictly positive integer, coordinate
systems in \ideas universal format files will be ignored. The behaviour
of the \pcs will thus be the same as that of versions 1.0 and 1.1.
Note that in any case, non Cartesian coordinate systems are not handled yet.

\subsubsection{System environment variables\label{sec:envpcs:sys}}

Some system environment variables may also modify the behaviour of
the \pcs. For example, if the \pcs was compiled with \med support
on an architecture allowing shared (dynamic) libraries, the
\texttt {LD\_PRELOAD} environment variable may be used to define a
``prioritary'' path to load \med or HDF5 libraries, thus allowing the user to experiment
with other versions of these libraries without recompiling the \pcs.
To determine which shared libraries are used by an executable file, use
the following command: \texttt{ldd~\{executable\_path\}}.

\subsection{Optional functionality\label{sec:pcs:lib_opt}}

Some functions of the \pcs are based on external libraries,
which may not always be available. It is thus possible to configure
and compile the \pcs so as not to use these libraries.
When running the \pcs, the supported options are printed.
The following optional libraries may be used:

\begin{itemize}

\item CGNS library. In its absence, \hyperref[sec:fmtdesc_cgns]{CGNS}
      format support is deactivated.

\item \med-file library. In its absence, \hyperref[sec:fmtdesc_med]{\med}
      format is simply deactivated.

\item libCCMIO library. In its absence, \hyperref[fmtdesc:ccm]{CCM}
      format is simply deactivated.

\item Read compressed files using Zlib. With this option, it is
      possible to directly read mesh files compressed with a
      \emph{gzip} type algorithm and bearing a \emph{.gz} extension.
      This is limited to formats not already based on an external
      library (i.e. it is not usable with CGNS, \med, or CCM files),
      and has memory and CPU time overhead, but may be practical.
      Without this library, files must be uncompressed before use.

\end{itemize}

\subsection{General remarks}

Note that the \pcs is in general capable of reading all ``classical''
element types present in mesh files (triangles, quadrangles, tetrahedra,
pyramids, prisms, and hexahedra).
Quadratic or cubic elements are converted upon reading into their
linear counterparts. Vertices referenced by no element (isolated vertices
or centres of higher-degree elements) are discarded. Meshes are read
in the order defined by the user and are appended, vertex and element
indices being incremented appropriately.
\footnote{Possible entity labels are not maintained, as they would
probably not be unique when appending multiple meshes.}

At this stage, volume elements are sorted by type, and the fluid domain
post-processing output is generated if required.

In general, groups assigned to vertices are ignored.
selections are thus based on faces or cells. with tools such
as \simail, faces of volume elements may be referenced directly, while
with \ideas or SALOME, a layer of surface elements bearing the required
colors and groups must be added. Internally, the \pcs always considers
that a layer of surface elements is added (i.e. when reading a \simail
mesh, additional faces are generated to bear cell face colors.
When building the $faces \rightarrow cells$ connectivity, all faces with the
same topology are merged: the initial presence of two layers of identical
surface elements belonging to different groups would thus lead to
a calculation mesh with faces belonging to two groups).

\subsection{Files passed to the Kernel\label{sec:pcs:mode_comm}}

Data passed to the Kernel by the \pcs is transmitted using a
binary file, using ``big endian'' data representation, named
\texttt{mesh\_input} (or contained in a directory of that name).

When using the \pcs for mesh verification, data for the Kernel
is not always needed. In this case, the \texttt{--no-write} option may
avoid creating a \pcs output file.

%==================================
\subsection{Mesh preprocessing%
\label{sec:prepro}}
%==================================

\subsubsection{Joining of non-conforming meshes}\label{sec:optpcs:join}

Conforming joining of possibly non-conforming meshes may be done by the
solver, and defined either using the Graphical User Interface (GUI) or the
\texttt{cs\_user\_join} user function. In the GUI, the user must
add entries in the ``Face joining'' section of the ``Meshes'' tab in the item
``Calculation environment $\rightarrow$ Meshes selection''.
The user may specify faces to be joined, and can also modify basic joining
parameters, see \figurename\ref{fig:joining}.
%
\begin{figure}[!h]
\begin{center}
\includegraphics[width=\textwidth]{gui_mesh_join}
\caption{Conformal or non-conformal joining}
\label{fig:joining}
\end{center}
\end{figure}
%
For a simple mesh, it is rarely useful to specify strict face selection
criteria, as joining is sufficiently automated to detect which faces
may actually be joined. For a more complex mesh, or a mesh with thin
walls which we want to avoid transforming into interior faces, it is
recommended to filter boundary faces that may be joined by using
face selection criteria. This has the
additional advantage of reducing the number of faces to test for
in the intersection/overlap search, and thus reduced to the time
required by the joining algorithm.

One may also modify tolerance criteria using 2 options:

\noindent
\begin{tabular}[top]{p{3.0cm}%
                     >{\PreserveBackslash\raggedright\hspace{0pt}}p{12.0cm}}
\texttt{fraction} $r$  &
assigns value $r$ (where $0 < r < 0.49$) to the maximum
intersection distance multiplier ($0.1$ by default). The maximum
intersection distance for a given vertex is based on the length of
the shortest incident edge, multiplied by $r$. The maximum intersection
at a given point along an edge is interpolated from that at its
vertices, as shown on the left of \figurename~\ref{fig:join_tolerance}; \\
\texttt{plane} $c$ &
assigns the maximum angle between normals for two faces to be
considered coplanar ($25^{\circ}$ by default);
this parameter is used in the second stage of the algorithm, to
reconstruct conforming faces, as shown on the right of figure
\ref{fig:join_tolerance}.\\
\end{tabular}

\begin{figure}[!hp]
\centerline{
\includegraphics*[width=14cm]{join_tolerance}}
\caption{Maximum intersection tolerance and faces normal angle
\label{fig:join_tolerance}}
\end{figure}

In practice, we are sometimes led to increase the maximum intersection
distance multiplier to $0.2$ or even $0.3$ when joining curved surfaces,
so that all intersection are detected. As this influences merging
of vertices and thus simplification of reconstructed faces, but also
deformation of ``lateral'' faces, it is recommended only to modify it
if necessary. As for the \texttt{plane} parameter, its use has
only been necessary on a few meshes up to now, and always in the
sense of reducing the tolerance so that face reconstruction does not
try to generate faces from initial faces on different surfaces.

\subsubsection{Periodicity\label{sec:optpcs:period}}

Handling of periodicity is based on an extension of conforming joining,
as shown on \figurename~\ref{fig:join_periodic}. It is thus not necessary
for the periodic faces to be conforming (though it usually leads to better
mesh quality). All options relative to conforming joining of
non-conforming faces also apply to periodicity. Note also that
once pre-processed, 2 periodic faces have the same orientation
(possibly adjusted by periodicity of rotation).

This operation can also be performed by the solver and specified
either using the GUI or the \\\texttt{cs\_user\_periodicity} function.

\begin{figure}[!hp]
\centerline{
\includegraphics*[height=8cm]{join_periodic}}
\caption{Matching of periodic faces
\label{fig:join_periodic}}
\end{figure}

As with joining, it is recommended to filter boundary faces to process
using a selection criterion. As many periodicities may be built as desired,
as long as boundary faces are present. Once a periodicity is handled,
faces having periodic matches do not appear as boundary faces, but as
interior faces, and are thus not available anymore for other
periodicities.

%==================================
\subsubsection{Parameters for conforming or non-conforming mesh joinings}
%==================================

The setting of these parameters is done in the user subroutine \texttt{cs\_user\_join} (called once). The user can specify the parameters used for the joining of different meshes. Below is given the list of the standard parameters which can me modified:
\begin{list}{-}{}
\item \texttt{fract}: the initial tolerance radius associated to each vertex is equal to the length of the shortest incident edge, multiplied by this fraction,
\item \texttt{plane}: when subdividing faces, 2 faces are considered coplanar and may be joined if the angle between their unit normals (in degrees) does not exceed this parameter,
\item \texttt{iwarnj}: the associated verbosity level (debug level if over 3).
\end{list}
In the call of the function \texttt{cs\_join\_add}, a selection criteria for
mesh faces to be joined is specified.

The call to the function '\texttt{cs\_join\_set\_advanced\_param}' allows defining parameters not available through the GUI.

The list of advanced modifiable parameters is given below:
\begin{list}{-}{}
\item \texttt{mtf}: a merge tolerance factor, used to locally modify the tolerance associated to each vertex before the merge step. Depending on its value, four
scenarios are possible:
\begin{list}{$\rightarrow$}{}
\item if $mtf=0$, no vertex merge
\item if $mtf<1$, the vertex merge is more strict. It may increase the number of tolerance reduction and therefore define smaller subset of vertices to merge together but it can drive to loose intersections.
\item if $mtf=1$, no change occurs
\item if $mtf>1$, the vertex merge is less strict. The subset of vertices able to merge is greater.
\end{list}
\item \texttt{pmf}: a pre-merge factor. This parameter is used to define a limit under which two vertices are merged before the merge step,
\item \texttt{tcm}: a tolerance computation mode. If its value is:
\begin{list}{$\rightarrow$}{}
\item 1 (default), the tolerance is the minimal edge length related to a vertex, multiplied by a fraction.
\item 2, the tolerance is computed like for 1 with, in addition, the multiplication by a coefficient equal to the maximum between $sin(e1)$ and $sin(e2)$; where $e1$ and $e2$ are two edges sharing the same vertex V for which we want to compute the tolerance.
\item 11, it is the same as 1 but taking into account only the selected faces.
\item 12, it is the same as 2 but taking into account only the selected faces.
\end{list}
\item \texttt{icm}: the intersection computation mode. If its value is:
\begin{list}{$\rightarrow$}{}
\item 1 (default), the original algorithm is used. Care should be taken to clip the intersection on an extremity.
\item 2, a new intersection algorithm is used. Caution should be used to avoid clipping the intersection on an extremity.
\end{list}
\item \texttt{maxbrk}: defines the maximum number of equivalence breaks which is enabled during the merge step,
\item \texttt{maxsf}: defines the maximum number of sub-faces used when splitting a selected face
\end{list}
%
The following are advanced parameters used in the search algorithm for face intersections between selected faces (octree structure). They are useful in case of memory limitation:
\begin{list}{-}{}
\item \texttt{tml}: the tree maximum level is the deepest level reachable during the tree building,
\item \texttt{tmb}: the tree maximum boxes is the maximum number of bounding boxes (BB) which can be linked to a leaf of the tree (not necessary true for the deepest level),
\item \texttt{tmr}: the tree maximum ratio. The building of the tree structure stops when the number of bounding boxes is superior to the product of \texttt{tmr} with the number of faces to locate. This is an efficient parameter to reduce memory consumption.
\end{list}

%==================================
\subsubsection{Parameters for periodicity}
%==================================

Periodicities can be set directly in the Graphical User Interface (GUI) or using the user function \texttt{cs\_user\_periodicity} (called once during the calculation initialisation). In both cases, the user can choose between 3 types of periodicities: translation, rotation, or mixed
(see \figurename\ref{fig:periodicities}).
Then specific parameters must be set.

\begin{figure}[!h]
\begin{center}
\includegraphics[width=14cm]{gui_mesh_periodicity}
\caption{Periodicity}
\label{fig:periodicities}
\end{center}
\end{figure}

As periodicity is an extension of mesh joining, all parameters (whether basic or advanced) available for mesh joining are also applicable for periodicity, in addition to the parameters defining the periodicity transformation.

%==================================
\subsubsection{Modification of the mesh geometry}
%==================================
\noindent
\textit{Functions called only once during the calculation initialisation.}

The user function \texttt{cs\_user\_mesh\_input} allows a detailed
selection of imported meshes read, reading files multiple times,
applying geometric transformations, and renaming groups.

The user function \texttt{cs\_user\_mesh\_modify} may be used
for advanced modification of the main \texttt{cs\_mesh\_t}} structure.

{\em WARNING: Caution must be exercised when using this function
along with periodicity. Indeed, the periodicity parameters are not
updated accordingly, meaning that the periodicity may not be valid
after mesh vertex coordinates have changed. It is particularly
true when one rescales the mesh. Rescaling should thus be done
in a separate run, before defining periodicity.}

The user function \texttt{cs\_user\_mesh\_thinwall} allows
insertion of thin walls in the calculation mesh. Currently, this
function simply transforms the selected internal faces into boundary
faces, on which boundary conditions can (and must) be applied.

Faces on each side of a thin wall will share the same vertices,
so post-processing of the main volume mesh may not show the
inserted walls, though they will appear in the main boundary
output mesh.

%==================================
\subsection{Mesh smoothing utilities}
%==================================
\noindent
\textit{Function called once only during the calculation initialisation.}

The smoothing utilities may be useful when the calculation mesh has local
defects.
The principle of smoothers is to mitigate the local defects by averaging
the mesh quality. This procedure can help for calculation robustness or/and
results quality.

The user function \texttt{cs\_user\_mesh\_smoothe} allows to use different
smoothing functions detailed below.

{\em WARNING 1: Caution must be exercised when using this function
along with periodicity. Indeed, the periodicity parameters are not
currently updated accordingly, meaning that the periodicity may be
unadapted after one changes the mesh vertex coordinates. It is particularly
true when one rescales the mesh. Rescaling should thus be done
in a separate run, before defining periodicity.}

{\em WARNING 2: Caution must be exercised when using smoothing utilities
because the geometry may be modified. In order to preserve geometry,
the function} \texttt{cs\_mesh\_smoother\_fix\_by\_feature} {\em allows to
fix by a feature angle criterion the mobility of boundary vertices.}

\subsubsection{Fix by feature}
The vertex normals are defined by the average of the normals of the
faces sharing the vertex.
The feature angle between a vertex and one of its adjacent faces is defined
by the angle between the vertex normal and the face normal.

This function sets a vertex if one of its feature angles is less than
$cos(\theta)$ where $\theta$ is the maximum feature angle (in degrees)
defined by the user.
In fact, if $\theta = 0^{\circ}$ all boundary vertices will be fixed, and
if $\theta = 90^{\circ}$ all boundary vertices will be free.

Fixing all boundary vertices ensures the geometry is preserved, but reduces
the smoothing algorithm's effectiveness.

\subsubsection{Warped faces smoother}

The function \texttt{cs\_mesh\_smoother\_unwarp} allows reducing face warping
in the calculation mesh.

Be aware that, in some cases, this algorithm may degrade other mesh quality
criteria.

%==================================
\section{Partitioning for parallel runs\label{sec:parall:part}}
%==================================

Graph partitioning (using one of the optional \metis or
\scotch libraries) is done by the Kernel. Unless explicitly
deactivated, this stage produces one or several ``cell $\rightarrow$ domain''
distribution files, named {\tt domain\_number\_\it{p}} for a partitioning on
$p$ sub-domains, which may be read when starting a subsequent computation so as
to avoid re-running that stage. These files are placed in a directory named
{\tt partition\_output}.

The Kernel redistributes data read in {\tt mesh\_input} based on the associated
(re-read or computed) partitioning, so there is no need to run any
prior script when running on a different number of processors, although
a previous partitioning may optionally be re-used.

Without a graph-based partitioning library, or based on the user's choice,
the Kernel will use a built-in partitioning using a space-filling curve
(Z or Hilbert curve) technique. This usually leads to partitionings
of lower quality than with graph partitioning, but parallel
performance remains reasonable.

Partitioning options may be defined using the GUI or by calling
the appropriate functions in the
\texttt{cs\_user\_partition\_options} user function.

\subsection{Partitioning stages\label{sec:parall:part:stages}}

Partitioning is always done just after reading the mesh, unless a
partitioning input file is available, in which case the partitioning
replaces this stage.

When a mesh modification implying a change of cell connectivity graph
is expected, the mesh may be repartitioned after the pre-processing
stage, prior to calculation. By default, re-partitioning is only done
if the partitioning algorithm chosen for that stage is expected to
produce different results due to the connectivity change. This is
the case for graph-based algorithms such as those of \metis or \scotch,
when mesh joining is defined, or additional periodic matching is defined
(and the algorithm is not configured to ignore periodicity information).

There are thus two possible partitioning stages:

\begin{itemize}
\item \texttt{CS\_PARTITION\_FOR\_PREPROCESS}, which is optional, and
      occurs just  after reading the mesh.
\item \texttt{CS\_PARTITION\_MAIN}, which occurs just after reading the
      mesh if it is the only stage, or after mesh preprocessing (and
      before computation), if the partitioning for preprocessing stage
      is activated.
\end{itemize}

The number of partitioning stages is determined automatically based on
information provided through \texttt{cs\_partition\_set\_preprocess\_hints()},
but repartitioning may also be forced or inhibited using the
\texttt{cs\_partition\_set\_preprocess()} user function.

\subsection{Partitioner choice\label{sec:parall:part:partlib}}

If the Kernel has been configured with both \ptscotch or \scotch and
\parmetis or \metis libraries, \ptscotch will be used by default\footnote{
Though \parmetis will be chosen before serial \scotch in a parallel
run}, but the user may force the selection of another partitioning
type using either the GUI or user routines.

In addition to graph-based partitionings, a space-filling curve based
algorithm is available, using either a Morton (Z) or Peano-Hilbert
curve, in the computation domain's bounding box or bounding curve.

When partitioning for preprocessing, a space-filling curve is used,
unless forced by calling\\ \texttt{cs\_partition\_set\_algorithm()}
with the appropriate algorithm choice for the\\
\texttt{CS\_PARTITION\_FOR\_PREPROCESS} stage.

\subsection{Effect of periodicity\label{sec:parall:part:noperiod}}

By default, face periodicity relations are taken into account when building
the ``cell $\rightarrow$ cell'' connectivity graph used for partitioning.
This allows better partitioning optimization, but increases the probability
of having groups of cells at opposite sides of the domain in a same
sub-domain. This is not an issue for standard calculations, but may
degrade performance of search algorithms based on bounding boxes.
It is thus possible to ignore periodicity when partitioning a mesh.

Also, partitioning using a space-filling curve ignores periodicity.

Note that nothing guarantees that a graph partitioner will not place
disjoint cells in the same sub-domain independently of this option,
but this behaviour is rare.

%==================================
\section{Basic modelling setup}
%==================================

%==================================
\subsection{Initialisation of the main parameters}
%==================================

This operation is done in the Graphical User Interface (GUI) or by using the user subroutines in \\ \texttt{cs\_user\_parameters.f90}.
In the GUI, the initialisation is performed by filling the parameters displayed in \figurename\ref{fig:gui_calculation_features}
to \ref{fig:gui_output_profiles}. If the option 'Mobile mesh' is activated in \figurename~\ref{fig:gui_mobile_mesh},
please see Section \ref{sec:ALE} for more details. The headings filled for the initialisation of the main parameters
are the followings:
\begin{list}{-}{}
\item Thermophysical model options: Steady or unsteady algorithm, specific physics, ALE mobile mesh,
      turbulence model, thermal model and species transport (definition of the scalars and their variances),
see \figurename~\ref{fig:gui_calculation_features} to \figurename~\ref{fig:gui_species}. If a thermal scalar, temperature or enthalpy, is selected,
two other headings on conjugate heat transfer and radiative transfers can be filled in (see \figurename~\ref{fig:gui_thermal_scalar}).
\item Physical properties: reference pressure, velocity and length, fluid properties (density, viscosity, thermal conductivity,
specific heat and scalar diffusivity), gravity, see \figurename~\ref{fig:gui_reference_values} to \figurename~\ref{fig:gui_gravity}.
If non-constant values are used for the fluid properties, and if the GUI is not used, the \\ \texttt{cs\_user\_physical\_properties}
file must be used, see \S~\ref{sec:usphyv}.
\item Volume conditions: definition of volume regions (for initialisation, head losses and source terms,
see \S~\ref{sec:prg_usersourceterms} and \S~\ref{sec:prg_headlosses}), initialisation of the variables (including scalars),
Coriolis source term, see \figurename~\ref{fig:gui_initialisation} and \figurename~\ref{fig:gui_coriolis}.
\item Boundary conditions: definition and parametrisation of boundary conditions for all variables (including scalars).
If the GUI is not used, the \texttt{cs\_user\_boundary\_conditions} file must be used, see \S~\ref{sec:prg_boundaryconditions}.
\item Numerical parameters: number and type of time steps, and advanced parameters for the numerical solution of the equations,
see \figurename~\ref{fig:gui_global_parameters} to \figurename~\ref{fig:gui_time_step}.
\item Calculation control: parameters related to the time averages, the
      locations of the probes where some variables will be monitored over time
      (if the GUI is not used, this information is specified in \S~\ref{sec:cs_user_initialization}),
       the definition of the frequency of the outputs in the calculation
      listing, the post-processing output writer frequency and format options, and
      the post-processing output meshes and variables selection, see
      \figurename~\ref{fig:gui_time_averages}, \figurename~\ref{fig:gui_output_log}, \figurename~\ref{fig:gui_output_writers},
      and \figurename~\ref{fig:gui_output_meshes}. The item ``Profiles'' allows to save, with a
      frequency defined by the user, 1D profiles on a parametric curve define by its equation, see \figurename\ref{fig:gui_output_profiles}.
\end{list}

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=\textwidth]{gui_calculation_features}
\caption{Calculation features options}
\label{fig:gui_calculation_features}
\end{center}
\end{figure}

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=\textwidth]{gui_mobile_mesh}
\caption{Mobile mesh option}
\label{fig:gui_mobile_mesh}
\end{center}
\end{figure}

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=\textwidth]{gui_turbulence_models}
\caption{Turbulence model selection}
\label{fig:gui_turbulence_models}
\end{center}
\end{figure}

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=\textwidth]{gui_thermal_scalar}
\caption{Thermal scalar selection}
\label{fig:gui_thermal_scalar}
\end{center}
\end{figure}

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=\textwidth]{gui_user_scal_def_init}
\caption{Definition of the transported species/scalars}
\label{fig:gui_species}
\end{center}
\end{figure}

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=13cm]{gui_phys_prop_reference_values}
\caption{Setting of the reference values for pressure, velocity and length}
\label{fig:gui_reference_values}
\end{center}
\end{figure}

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=\textwidth]{gui_fluid_props}
\caption{Fluid properties}
\label{fig:gui_fluid_props}
\end{center}
\end{figure}

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=\textwidth]{gui_gravity}
\caption{Setting of the gravity}
\label{fig:gui_gravity}
\end{center}
\end{figure}

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=\textwidth]{gui_initialisation}
\caption{Initialisation of variables}
\label{fig:gui_initialisation}
\end{center}
\end{figure}


\begin{figure}[!ht]
\begin{center}
\includegraphics[width=\textwidth]{gui_coriolis}
\caption{Setting of the Coriolis source term}
\label{fig:gui_coriolis}
\end{center}
\end{figure}

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=\textwidth]{gui_global_res_parameters}
\caption{Global resolution parameters}
\label{fig:gui_global_parameters}
\end{center}
\end{figure}

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=\textwidth]{gui_numerical_parameters}
\caption{Numerical parameters for the main variables resolution}
\label{fig:gui_numerical_parameters}
\end{center}
\end{figure}

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=\textwidth]{gui_time_step}
\caption{Time step settings}
\label{fig:gui_time_step}
\end{center}
\end{figure}

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=\textwidth]{gui_time_averages}
\caption{Management of time averaged variables}
\label{fig:gui_time_averages}
\end{center}
\end{figure}

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=\textwidth]{gui_output_log}
\caption{Parameters of chronological logging options}
\label{fig:gui_output_log}
\end{center}
\end{figure}

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=\textwidth]{gui_output_writers}
\caption{Management of postprocessing writers}
\label{fig:gui_output_writers}
\end{center}
\end{figure}

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=\textwidth]{gui_output_meshes}
\caption{Management of postprocessing meshes}
\label{fig:gui_output_meshes}
\end{center}
\end{figure}

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=\textwidth]{gui_output_profiles}
\caption{Management of 1D profiles of the solution}
\label{fig:gui_output_profiles}
\end{center}
\end{figure}

With the GUI, the subroutine \texttt{cs\_user\_parameters.f90} is only
used to modify high-level parameters which can not be managed by the
interface. Without the GUI, this
subroutine is compulsory and some of the headings must be completed (see \S\ref{sec:prg_stepbystepcalculation}).
\texttt{cs\_user\_parameters.f90}
is used to indicate the value of different calculation
basic parameters: constant and uniform physical values, parameters of
numerical schemes, input-output management...\\
It is called only during the calculation initialisation.

For more details about the different parameters, please refer to the key
word list (\S\ref{sec:prg_motscles}).

\texttt{cs\_user\_parameters.f90} is in fact constituted of 4 separate subroutines:  \texttt{usipph}, \texttt{usppmo},
\texttt{usipsu} and \texttt{usipes}. Each one controls various
 specific parameters. The keywords which are not featured in the supplied example
can be provided by the user in \texttt{SRC/REFERENCE/base}; in this case,
understanding of the comments is required to add the keywords in the appropriate
subroutine, it will ensure that the value
has been well defined. The modifiable parameters in each of the subroutines of
\texttt{cs\_user\_parameters.f90} are:

\begin{list}{$\bullet$}{}
\item \texttt{usipph}: \texttt{iturb}, \texttt{itherm} and \texttt{icavit} (don't modify these
      parameters anywhere else)
\item \texttt{usppmo}: activation of specific physical models.
\item \texttt{usipsu}: physical parameters of the calculation (thermal scalar, physical
      properties, ...), numerical parameters (time steps, number of iterations, ...),
      definition of the time averages.
\item \texttt{usipes}: post-processing output parameters (periodicity, variable names,
      probe positions, ...)
\end{list}

For more details on the different parameters, see the list of keywords
(\S~\ref{sec:prg_motscles}).
 The names of the keywords can also be seen in the help sections of the interface.

\noindent
$\bullet\ $ When using the interface, only the
additional parameters (which can not be defined in the interface)
should appear in \texttt{cs\_user\_parameters.f90}. The user
needs then only to activate examples which are useful for his
case (replacing \texttt{if (.false.)} with \texttt{if (.true.)},
or removing such tests).

\subsection{Selection of mesh inputs: \textmd{\texttt{cs\_user\_mesh\_input}}}
%==================================

\noindent
\textit{Subroutine called only during the calculation initialisation.}

This C function may be used to select which mesh input files
are read, and apply optional coordinate transformations or group renumberings
to them. By default, the input read is a file or directory named
\texttt{mesh\_input}, but if this function is used, any file can be selected,
and the same file can be read multiple times (applying a different
coordinate transformation each time).
All inputs read through this function are automatically concatenated, and
may be later joined using the mesh joining options.

Geometric transformations are defined using a homogeneous coordinates
transformation matrix. Such a matrix has 3 lines and 4 columns, with the
first 3 columns describing a rotation/scaling factor, and the last column
describing a translation. A 4th line is implicit, containing zeroes
off-diagonal, and 1 on the diagonal. The advantages of this representation
is that any rotation/translation/scaling combination may be expressed
by matrix multiplication, while simple rotations or translations
may still be defined easily.

%==================================
%==================================
\subsection{Non-default variables initialisation} \label{sec:cs_user_initialization}
%==================================

The non-default variables initialisation is performed in the subroutine \texttt{cs\_user\_initialization} (called only during the calculation initialisation).\\ At the calculation beginning, the variables are initialised
automatically by the code. Velocities and scalars are set to 0 (or \texttt{scamax} or \texttt{scamin} if 0 is outside the acceptable
scalar variation range), and the turbulent variables are estimated from
\texttt{uref} and \texttt{almax}. \\
For $k$ (of variable index \texttt{ik}) in the $k-\varepsilon$,
$R_{ij}-\varepsilon$, v2f or $k-\omega$ models:
\begin{equation*}
k = 1.5 \left(0.02 \textrm{ \texttt{uref}}\right)^2
\end{equation*}
and in $R_{ij}-\varepsilon$:
\begin{equation*}
R_{ij}=\frac{2}{3}k\delta_{ij}
\end{equation*}
For $\varepsilon$ (of variable index \texttt{iep}) in the $k-\varepsilon$,
$R_{ij}-\varepsilon$ or v2f models:
\begin{equation*}
\varepsilon = k^{1.5} \frac{C_\mu}{\textrm{\texttt{almax}}}
\end{equation*}
For $\omega$ (of variable index \texttt{iomg}) in the $k-\omega$ model:
\begin{equation*}
\omega = k^{0.5} \frac{1}{\textrm{\texttt{almax}}}
\end{equation*}
For $\varphi$ and $\overline{f}$ (of variable indices \texttt{iphi} and
\texttt{ifb}) in the v2f models:
\begin{equation*}
\left\{\begin{array}{ll}
\varphi = & \frac{2}{3} \\
\overline{f} = & 0
\end{array}\right.
\end{equation*}
For $\alpha$ (of variable index \texttt{ial}) in the EBRSM and BL-v2/k models:
\begin{equation*}
\alpha = 1
\end{equation*}
For $\tilde{\nu}_t$ in the Spalart-Allmaras model:
\begin{equation*}
\tilde{\nu}_t = 0.02 \sqrt{\frac{3}{2}} (\textrm{\texttt{uref}}) (\textrm{\texttt{almax}})
\end{equation*}

The subroutine \texttt{cs\_user\_initialization} allows if necessary to
initialise certain variables to values closer to their estimated final values,
in order to obtain a faster convergence.

This subroutine allows also the user to make a non-standard initialisation of
physical parameters (density, viscosity, ...), to impose a local
value of the time step, or to modify some parameters (time step,
variable specific heat, ...) in the case of a calculation restart.

\minititre{Note: value of the time step}
\begin{list}{-}{}
\item For calculations with constant and uniform time step
      (\texttt{idtvar}=0), the value of the time step is \texttt{dtref},
      given in the parametric file of the interface or in
      \texttt{cs\_user\_parameters.f90}.
\item For calculations with a non-constant time step
      (\texttt{idtvar}=1 or 2), which is not a calculation restart,
      the value of \texttt{dtref} given in the parametric file of the interface
      or in \texttt{cs\_user\_parameters.f90} is used to initialise the time
      step.
\item For calculations with a non-constant time step
      (\texttt{idtvar}=1 or 2) which is a restart of a
      calculation whose time step type was different (for instance, restart
      using a variable time step of a calculation run using a constant time
      step), the value of \texttt{dtref}, given in the parametric file of the
      interface or in \texttt{cs\_user\_parameters.f90}, is used to initialise
      the time step.
\item For calculations with non-constant time step
      (\texttt{idtvar}=1 or 2) which is a restart of a
      calculation whose time step type was the same (for instance, restart with
      \texttt{idtvar}=1 of a calculation run with \texttt{idtvar}=1), the time
      step is read from the restart file and the value of \texttt{dtref} given
      in the parametric file of the interface, or in
      \texttt{cs\_user\_parameters.f90}, is not used.
\end{list}
It follows, that for a calculation with a non-constant time step
(\texttt{idtvar}=1 or 2) which is a restart of a calculation in which
\texttt{idtvar} had the same value, \texttt{dtref} does not allow to modify the
time step. The user subroutine \texttt{cs\_user\_initialization} allows
modifying the array \texttt{dt} which contains the value of the time step read
from the restart file (array whose size is \texttt{ncelet}, defined at the cell
centres whatever the chosen time step type is).

%==================================
\subsection{Manage boundary conditions}
%==================================
\label{sec:prg_boundaryconditions}
The boundary conditions can be specified in the Graphical User Interface (GUI) under the heading ``Boundary conditions''
or in the user subroutine \texttt{cs\_user\_boundary\_conditions} called every time step.
With the GUI, each region and the type of boundary condition associated to it are defined in
\figurename~\ref{fig:gui_bc_regions}. Then, the parameters of the boundary condition are specified
 in \figurename~\ref{fig:gui_bc_parameters}. The colors of the boundary faces may be read directly from
a ``listing'' file created by the Preprocessor. This file can be generated directly by the interface
under the heading ``Definition of boundary regions $\rightarrow$ Add from Preprocessor listing $\rightarrow$ import groups and references from Preprocessor listing'', see \figurename~\ref{fig:gui_bc_regions}.
%
\begin{figure}[!ht]
\begin{center}
\includegraphics[width=13cm]{gui_bc_regions}
\caption{Definition of the boundary conditions}
\label{fig:gui_bc_regions}
\end{center}
\end{figure}
%
\begin{figure}[!ht]
\begin{center}
\includegraphics[width=13cm]{gui_bc_parameters}
\caption{Parameters of the boundary conditions}
\label{fig:gui_bc_parameters}
\end{center}
\end{figure}

\texttt{cs\_user\_boundary\_conditions} is the second compulsory subroutine for every calculation launched
without interface (except in the case of specific physics where the
corresponding boundary condition user subroutine must be used).

When using the interface, only complex boundary conditions (input profiles, conditions varying in time, ...)
need to be defined with \texttt{cs\_user\_boundary\_conditions}.
In the case of a calculation launched without the
interface, all the boundary conditions must appear in \texttt{cs\_user\_boundary\_conditions}.

\texttt{cs\_user\_boundary\_conditions} is essentially constituted of loops on boundary
face subsets. Several sequences
of \verb+call getfbr+ \verb+('criterion', nlelt, lstelt)+ (cf.
\S\ref{sec:fvm_selector}) allow selecting
the boundary faces with respect to their group(s), their
color(s) or geometric criteria. If needed, geometric and
physical variables are also available to the user. These allow him
to select the boundary faces using other criteria.

For more details about the treatment of boundary conditions, the user
may refer to the theoretical and computer documentation \cite{theory} of
the subroutine \texttt{condli} (for wall conditions, see
\texttt{clptur}) (to access this document on a workstation, use
\mbox{\texttt{code\_saturne~info --guide theory}}).

From the user point of view, the boundary conditions are fully
defined by three arrays\footnote{Except with Lagrangian boundary condition}:
\texttt{itypfb(nfabor)}\index{\texttt{itypfb}},
\texttt{icodcl(nfabor,nvar)}\index{\texttt{icodcl}} and
\texttt{rcodcl(nfabor,nvar,3)}\index{\texttt{rcodcl}}.
\begin{list}{-}{}
\item \texttt{itypfb(ifac)} defines the type of the face \texttt{ifac}
      (input, wall, ...).
\item \texttt{icodcl(ifac,ivar)} defines the type of boundary
      condition for the variable \texttt{ivar} on the face \texttt{ifac}
      (Dirichlet, flux ...).
\item \texttt{rcodcl(ifac,ivar,.)} gives the numerical values associated with the
      type of boundary condition (value of the Dirichlet condition, of the flux ...).
\end{list}

In the case of standard boundary conditions (see
\S\ref{sec:prg_clstandard}), it is sufficient to complete \texttt{itypfb(ifac)} and
parts of the array \texttt{rcodcl}; the array \texttt{icodcl} and most of \texttt{rcodcl} are filled automatically. For non-standard boundary
conditions (see \S\ref{sec:prg_clnonstandard}), the arrays \texttt{icodcl} and
\texttt{rcodcl} must be fully completed.

%==================================
\subsubsection{Coding of standard boundary conditions}
%==================================
\label{sec:prg_clstandard}%
The standard keywords used by the indicator \texttt{itypfb} are:
\texttt{ientre\index{ientre}}, \texttt{iparoi\index{iparoi}},
\texttt{iparug\index{iparug}}, \texttt{isymet\index{isymet}},
\texttt{isolib\index{isolib}}, \texttt{ifrent\index{ifrent}}, \texttt{ifresf\index{ifresf}},
\texttt{i\_convective\_inlet\index{i\_convective\_inlet}} and \texttt{iindef\index{iindef}}.

\begin{list}{$\bullet$}{}
\item If \texttt{itypfb=ientre}: inlet face.

\begin{list}{$\rightarrow$}{}
\item Zero-flux condition for pressure and Dirichlet condition for all
      other variables. The value of the Dirichlet condition must be given in
      \texttt{rcodcl(ifac,ivar,1)} for every value of \texttt{ivar}, except for
      \texttt{ivar=ipr}. The other values of \texttt{rcodcl} and
      \texttt{icodcl} are filled automatically.
\end{list}

\item If \texttt{itypfb=iparoi}: smooth solid wall face, impermeable and with friction.

\begin{list}{$\rightarrow$}{}
\item the eventual sliding wall velocity of the face is
      found in \texttt{rcodcl(ifac,ivar,1)} (\texttt{ivar} being
      \texttt{iu}, \texttt{iv} or \texttt{iw}). The initial
      values of \texttt{rcodcl(ifac,ivar,1)} are zero for
      the three velocity components (and therefore are to be specified
      only if the velocity is not equal to zero). \\
{\em WARNING: the wall sliding velocity must belong to the boundary face
      plane. For safety, the code only uses the projection of this
      velocity on the face. As a consequence, if the velocity specified
      by the user does not belong to the face plane, the wall sliding velocity really
      taken into account will be different.}

\item For scalars, two kinds of boundary conditions can be
      defined:
\begin{list}{$\rightsquigarrow$}{}
\item Imposed value at the wall. The user must write\\
\hspace*{1cm}\texttt{icodcl(ifac,ivar)}=5\\
\hspace*{1cm}\texttt{rcodcl(ifac,ivar,1)}=imposed value\\
\item Imposed flux at the wall. The user must write\\
\hspace*{1cm}\texttt{icodcl(ifac,ivar)}=3\\
\hspace*{1cm}\texttt{rcodcl(ifac,ivar,3)}=imposed flux value (depending on the
variable, the user may refer to the case \texttt{icodcl}=3 of \S~\ref{sec:prg_clnonstandard} for the flux definition).
\item If the user does not fill these arrays, the default condition
      is zero flux.
\end{list}
\end{list}

\item If \texttt{itypfb=iparug}: rough solid wall face, impermeable and with friction.

\begin{list}{$\rightarrow$}{}
\item the eventual moving velocity of the wall tangent to the face is
      given by \texttt{rcodcl(ifac,ivar,1)} (\texttt{ivar} being
      \texttt{iu}, \texttt{iv} or \texttt{iw}). The initial
      value of \texttt{rcodcl(ifac,ivar,1)} is zero for
      the three velocity components (and therefore must be specified
      only in the case of the existence of a slipping velocity). \\
{\em WARNING: the wall moving velocity must be in the boundary face
      plane. By security, the code uses only the projection of this
      velocity on the face. As a consequence, if the velocity specified
      by the user is not in the face plane, the wall moving velocity really
      taken into account will be different.}
\item The dynamic roughness must be specified in \texttt{rcodcl(ifac,iu,3)}.
      The values of \texttt{rcodcl(ifac,iv,3)} stores the thermal and scalar roughness.
      The values of \texttt{rcodcl(ifac,iw,3)} is not used.
\item For scalars, two kinds of boundary conditions can be defined:
\begin{list}{$\rightsquigarrow$}{}
\item Imposed value at the wall. The user must write\\
\hspace*{1cm}\texttt{icodcl(ifac,ivar)}=6\\
\hspace*{1cm}\texttt{rcodcl(ifac,ivar,1)}=imposed value\\
\item Imposed flux at the wall. The user must write\\
\hspace*{1cm}\texttt{icodcl(ifac,ivar)}=3\\
\hspace*{1cm}\texttt{rcodcl(ifac,ivar,3)}= imposed flux value (definition
      of the flux condition according to the variable, the user can refer to the
      case \texttt{icodcl}=3 of the paragraph \ref{sec:prg_clnonstandard}).
\item If the user does not complete these arrays, the default condition
      is zero flux.
\end{list}
\end{list}
\item If \texttt{itypfb=isymet}: symmetry face (or wall without friction).
\begin{list}{$\rightarrow$}{}
\item Nothing to be writen in \texttt{icodcl} and  \texttt{rcodcl}.
\end{list}

\item If \texttt{itypfb=isolib}: free outlet face (or more precisely free
      inlet/outlet with forced pressure)
\begin{list}{$\rightarrow$}{}
\item The pressure is always treated with a Dirichlet condition, calculated
      with the constraint $\displaystyle \frac{\partial }{\partial n}\left(\frac{ \partial P}{\partial \tau}\right)=0$.
      The pressure is set to $P_0$ at the first \texttt{isolib} face met.
      The pressure calibration is always done on a single face, even if there are
      several outlets.
\item If the mass flow is coming in, the velocity is set to zero
      and a Dirichlet condition for the scalars and the turbulent quantities is used
      (or zero-flux condition if no Dirichlet value has been specified).
\item If the mass flow is going out, zero-flux condition are set for the velocity,
      the scalars and the turbulent quantities.
\item Nothing is written in \texttt{icodcl} or \texttt{rcodcl} for the pressure or
      the velocity. An optional Dirichlet condition can be specified for the scalars
      and turbulent quantities.
\end{list}

\item If \texttt{itypfb=ifrent}: free outlet, free inlet (based on Bernoulli relationship) face.

\item If \texttt{itypfb=ifresf}: free-surface boundary condition.

\begin{list}{$\rightarrow$}{}
\item if outlet, the equivalent to standard outlet.
      In case of ingoing flux, the Benoulli relationship which links pressure and velocity is used (see the thory guide for more information). An additional head loss modelling what is going on outward of the domain can be added by the user.
\end{list}

\item If \texttt{itypfb=i\_convective\_inlet}: inlet with zero diffusive flux for all transported variables (species and velocity). This allows to exactly impose the ingoing flux.

\item If \texttt{itypfb=iindef}: undefined type face (non-standard case).
\begin{list}{$\rightarrow$}{}
\item Coding is done in a non-standard way by filling both arrays \texttt{rcodcl} and
      \texttt{icodcl} (see \S~\ref{sec:prg_clnonstandard}).
\end{list}
\end{list}

\minititre{Notes}
$\bullet\ $ Whatever is the value of the indicator \texttt{itypfb(ifac)}, if
the array \texttt{icodcl(ifac,ivar)} is modified by the user ({\em i.e.} filled
with a non-zero value), the code will not use the default
conditions for the variable \texttt{ivar} at the face \texttt{ifac}. It will
take into account only the values of \texttt{icodcl} and \texttt{rcodcl} provided by the
user (these arrays must then be fully completed, like in the non-standard case). \\
For instance, for a normal symmetry face where scalar 1 is associated with a
Dirichlet condition equal to 23.8 (with an infinite exchange
coefficient):\\
\hspace*{2cm}\texttt{itypfb(ifac)=isymet}\\
\hspace*{2cm}\texttt{icodcl(ifac,isca(1))=1}\\
\hspace*{2cm}\texttt{rcodcl(ifac,isca(1),1)=23.8}\\
(\texttt{rcodcl(ifac,isca(1),2)=rinfin} is the default value, therefore it is
not necessary to specify a value)\\
The boundary conditions for the other variables are automatically
defined.

\noindent
$\bullet\ $The user can define new types of boundary faces. He only must
choose a value $N$ and to fully specify the boundary conditions (see
\S\ref{sec:prg_clnonstandard}). He must specify
\texttt{itypfb(ifac)}=$N$ where $N$ range is 1 to
\texttt{ntypmx\index{ntypmx}} (maximum number of boundary face types), and of
course different from the values \texttt{ientre}, \texttt{iparoi},
\texttt{iparug}, \texttt{isymet}, \texttt{isolib} and \texttt{iindef} (the values
of these variables are given in the \texttt{paramx} module). This allows to
easily isolate some boundary faces, in order for instance to calculate balances.

%==================================
\subsubsection{Coding of non-standard boundary conditions}
%==================================
\label{sec:prg_clnonstandard}%
Ifa face does not correspond to a standard type, the user
must completely fill the arrays \texttt{itypfb}, \texttt{icodcl} and
\texttt{rcodcl}. \texttt{itypfb(ifac)} is then equal to \texttt{iindef}
or another value defined by the user (see note at the end of
\S~\ref{sec:prg_clstandard}). The arrays \texttt{icodcl} and \texttt{rcodcl}
must be filled as follows:

\begin{list}{$\bullet$}{}
\item If \texttt{icodcl(ifac,ivar)}=1: Dirichlet condition at the face
      \texttt{ifac} for the variable \texttt{ivar}.

\begin{list}{$\rightarrow$}{}
\item \texttt{rcodcl(ifac,ivar,1)} is the value of the variable \texttt{ivar}
      at the face \texttt{ifac}.

\item \texttt{rcodcl(ifac,ivar,2)} is the value of the exchange coefficient
      between the outside and the fluid for the variable \texttt{ivar}. An
      infinite value (\texttt{rcodcl(ifac,ivar,2)=rinfin}) indicates an
      ideal transfer between the outside and the fluid (default case).

\item \texttt{rcodcl(ifac,ivar,3)} is not used.

\item \texttt{rcodcl(ifac,ivar,1)} has the units of the variable
      \texttt{ivar}, {\em i.e.}:
\begin{list}{$\rightsquigarrow$}{}
\item $m/s$ for the velocity

\item $m^2/s^2$ for the Reynolds stress

\item $m^2/s^3$ for the dissipation

\item $Pa$ for the pressure

\item \degresC\ for the temperature

\item $J.kg^{-1}$ for the enthalpy

\item \degresC$^2$ for temperature fluctuations

\item $J^2.kg^{-2}$ for enthalpy fluctuations
\end{list}

\item \texttt{rcodcl(ifac,ivar,2)} has the following units (defined in such way
      that when multiplying the exchange coefficient by the variable, the
      given flux has the same units as the flux defined below when
      \texttt{icodcl=3}):

\begin{list}{$\rightsquigarrow$}{}
\item $kg.m^{-2}.s^{-1}$ for the velocity

\item $kg.m^{-2}.s^{-1}$ for the Reynolds stress

\item $s.m^{-1}$ for the pressure

\item $W.m^{-2}.\mbox{\degresC}^{-1}$ for the temperature

\item $kg.m^{-2}.s^{-1}$ for the enthalpy
\end{list}

\end{list}

\item If \texttt{icodcl(ifac,ivar)=2}: radiative outlet at the face \texttt{ifac}
      for the variable \texttt{ivar}. It reads $ \dfrac{\partial \varia }{\partial t} + C \dfrac{\partial \varia}{\partial n} = 0 $, where $C$ is a to be defined celerity of radiation.

\begin{list}{$\rightarrow$}{}
\item \texttt{rcodcl(ifac,ivar,3)} is not used.

\item \texttt{rcodcl(ifac,ivar,1)} is the flux value of \texttt{ivar} at the cell center $\centip$,
      projection of the center of the adjacent cell on the straight line
      perpendicular to the boundary face and crossing its center,
      at the previous time step.
      It corresponds to:
\item \texttt{rcodcl(ifac,ivar,2)} is CFL number based on the parameter $C$,
      the distance to the boundary $\centip \centf$ and the time step:
      $CFL = \dfrac{C dt }{\centip \centf}$,

\end{list}

\item If \texttt{icodcl(ifac,ivar)=3}: flux condition at the face \texttt{ifac}
      for the variable \texttt{ivar}.

\begin{list}{$\rightarrow$}{}
\item \texttt{rcodcl(ifac,ivar,1)} and \texttt{rcodcl(ifac,ivar,2)} are not used.

\item \texttt{rcodcl(ifac,ivar,3)} is the flux value of \texttt{ivar} at the
      wall. This flux is negative if it is a source for the fluid. It corresponds to:
\begin{list}{$\rightsquigarrow$}{}
\item
$\displaystyle -(\lambda_T+C_p\frac{\mu_t}{\sigma_T})\grad T\cdot\vect{n}$ for a temperature (in $W/m^2$)

$\displaystyle -(\frac{\lambda_T}{C_p}+\frac{\mu_t}{\sigma_h})\grad h\cdot\vect{n}$
     for an enthalpy (in $W/m^2$).

$\displaystyle -(\lambda_\varphi+\frac{\mu_t}{\sigma_\varphi})\grad\varphi\cdot\vect{n}$ in the case of another scalar $\varphi$ (in $kg.m^{-2}.s^{-1}.[\varphi]$, where $[\varphi]$ are the units of $\varphi$).

\item $-\Delta t\ \grad P\cdot\vect{n}$ for the pressure (in $kg.m^{-2}.s^{-1}$).

\item $-(\mu+\mu_t)\grad U_i\cdot\vect{n}$ for a velocity component (in $kg.m^{-1}.s^{-2}$).

\item $-\mu\grad R_{ij}\cdot\vect{n}$ for a $R_{ij}$ tensor component (in $W/m^2$).
\end{list}

\end{list}

\item If \texttt{icodcl(ifac,ivar)}=4: symmetry condition, for the symmetry
      faces or wall faces without friction. This condition can only be
      used for velocity components ($\vect{U}\cdot\vect{n}=0$) and
      the $R_{ij}$ tensor components (for other variables, a zero-flux
      condition type is usually used).\\

\item If \texttt{icodcl(ifac,ivar)}=5: friction condition, for wall faces
      with friction. This condition can not be applied to the pressure.
\begin{list}{$\rightsquigarrow$}{}
\item For the velocity and (if necessary) the turbulent variables, the
      values at the wall are calculated from theoretical profiles. In
      the case of a sliding wall, the three components of the sliding
      velocity are given by (\texttt{rcodcl(ifac,iu,1)},
      \texttt{rcodcl(ifac,iv,1)}, and \\\texttt{rcodcl(ifac,iw,1)}).\\
{\em WARNING: the wall sliding velocity must belong to the boundary face
      plane. For safety, the code uses only the projection of this
      velocity on the face. Therefore, if the velocity vector specified
      by the user does not belong to the face plane, the wall sliding velocity really
      taken into account will be different.}

\item For other scalars, the condition \texttt{icodcl}=5 is similar to
      \texttt{icodcl=1}, but with a wall exchange coefficient calculated from a
      theoretical law. Therefore, the values of \\\texttt{rcodcl(ifac,ivar,1)} and
      \texttt{rcodcl(ifac,ivar,2)} must be specified: see \cite{theory}.
\end{list}

\item If \texttt{icodcl(ifac,ivar)}=6: friction condition, for the rough-wall faces
      with friction. This condition can not be used with the pressure.
\begin{list}{$\rightsquigarrow$}{}
\item For the velocity and (if necessary) the turbulent variables, the
      values at the wall are calculated from theoretical profiles. In
      the case of a sliding wall, the three components of the sliding
      velocity are given by (\texttt{rcodcl(ifac,iu,1)},
      \texttt{rcodcl(ifac,iv,1)}, and \\\texttt{rcodcl(ifac,iw,1)}).\\
      {\em WARNING: the wall sliding velocity must belong to the boundary face
      plane. For safety, the code uses only the projection of this
      velocity on the face. Therefore, if the velocity vector specified
      by the user does not belong to the face plane, the wall sliding velocity really
      taken into account will be different.}\\
      The dynamic roughness height is given by \texttt{rcodcl(ifac,iu,3)} only.

\item For the other scalars, the condition \texttt{icodcl}=6 is similar to
      \texttt{icodcl}=1, but with a wall exchange coefficient calculated from a
      theoretical law. The values of \texttt{rcodcl(ifac,ivar,1)} and
      \texttt{rcodcl(ifac,ivar,2)} must therefore be specified: see \cite{theory}.
      The thermal roughness height is then given by \texttt{rcodcl(ifac,ivar,3)}.
\end{list}

\item If \texttt{icodcl(ifac,ivar)}=9: free outlet condition for the
      velocity. This condition is only applicable to velocity
      components.\\
      If the mass flow at the face is negative, this condition is equivalent
      to a zero-flux condition.\\
      If the mass flow at the face is positive, the velocity at the face is set to zero (but not the mass flow).\\
\texttt{rcodcl} is not used.

\item If \texttt{icodcl(ifac,ivar)}=14: generalized symmetry boundary condition for vectors (Marangoni
      effect for the velocity for instance).
      This condition is only applicable to vectors and set a Dirichlet boundary condition on the normal
      component and a Neumann condition on the tangential components.\\
      If the three components are  \texttt{ivar1, ivar2, ivar3}, the required values are:

\begin{list}{$\rightarrow$}{}
      \item \texttt{rcodcl(ifac,ivar1,1)}: Dirichlet value in the $x$ direction.
      \item \texttt{rcodcl(ifac,ivar2,1)}: Dirichlet value in the $y$ direction.
      \item \texttt{rcodcl(ifac,ivar3,1)}: Dirichlet value in the $z$ direction.
      \item \texttt{rcodcl(ifac,ivar1,3)}: flux value for the $x$ direction.
      \item \texttt{rcodcl(ifac,ivar2,3)}: flux value for the $y$ direction.
      \item \texttt{rcodcl(ifac,ivar3,3)}: flux value for the $z$ direction.
\end{list}
      Therefore, the code automatically computes the boundary condition to impose to the normal and to
      the tangential components.

\end{list}

\minititre{Note}
$\bullet\ $A standard \texttt{isolib} outlet face amounts to a Dirichlet
condition (\texttt{icodcl}=1) for the pressure, a free outlet condition
(\texttt{icodcl}=9) for the velocity and a Dirichlet condition
(\texttt{icodcl}=1) if the user has specified a Dirichlet value or a zero-flux
condition (\texttt{icodcl}=3) for the other variables.\\

%==================================
\subsubsection{Checking of the boundary conditions}
%==================================

The code checks the main compatibilities between the boundary
conditions. In particular, the following rules must be respected: \\
$\bullet\ $On each face, the boundary conditions of the three velocity components must belong to the same type. The same is true for the components of the $R_{ij}$ tensor.\\
$\bullet\ $If the boundary conditions for the velocity belong to the
``sliding'' type (\texttt{icodcl}=4), the conditions for $R_{ij}$ must belong to
the ``symmetry'' type (\texttt{icodcl}=4), and vice versa.\\
$\bullet\ $If the boundary conditions for the velocity belong to the
``friction'' type (\texttt{icodcl}=5 or 6), the boundary conditions for the turbulent variables
must belong to the ``friction'' type, too.\\
$\bullet\ $If the boundary condition of a scalar belongs to the
``friction'' type, the boundary condition of the velocity must belong to
the ``friction'' type, too.

In case of mistakes, if the post-processing output is activated (which is the default setting),
a special error output, similar to the mesh format, is produced in order to help
correcting boundary condition definitions.

%==================================
\subsubsection{Sorting of the boundary faces}
%==================================

In the code, it may be necessary to have access to all the boundary
faces of a given type. To ease this kind of search, an array made of
sorted faces is automatically filled (and updated at each time step):
\texttt{itrifb(nfabor)\index{itrifb}}.\\
\texttt{ifac=itrifb(i)} is the number of the i$^{\text{th}}$  face of type
1.\\
\texttt{ifac=itrifb(i+n)} is the number of the i$^{\text{th}}$ face of type
2, if there are $n$ faces of type 1.\\
... etc.

Two auxiliary arrays of size \texttt{ntypmx} are also defined.\\
\texttt{idebty(ityp)\index{idebty}} is the index
corresponding to the first
face of type \texttt{ityp} in the array \texttt{itrifb}.\\
\texttt{ifinty(ityp)\index{ifinty}} is the index
corresponding to the last
face of type \texttt{ityp} in the array \texttt{itrifb}.

Therefore, a value \texttt{ifac0} found between \texttt{idebty(ityp)} and
\texttt{ifinty(ityp)} is associated to each face \texttt{ifac} of type
\texttt{ityp=itypfb(ifac)}, so that \texttt{ifac=itrifb(ifac0)}.

If there is no face of type \texttt{ityp}, the code set \\
\texttt{ifinty(ityp)=idebty(ityp)-1},\\
which enables to bypass, for all the missing \texttt{ityp}, the loops such as \\
\texttt{do ii=idebty(ityp),ifinty(ityp)}.

The values of all these indicators are displayed at the beginning of the
code execution listing.

%=============================================================
\subsubsection[Boundary conditions with LES]
{Boundary conditions with LES}
%===============================================================
\subsubsubsection{Vortex method}
\label{sec:prg_usvort}%
The subroutine \texttt{usvort} allows generating the unsteady inlet boundary
conditions for the LES by the vortex method. The method is based on
 the generation of vortices in the 2D inlet plane with help from
the pre-defined functions. The fluctuation normal to the inlet plane
is generated by a Langevin equation. It is in the subroutine \texttt{usvort}
 where the parameters of this method are given.

\noindent
\textit{Subroutine called at each time step}

To allow the application of the vortex method, an indicator must be informed of
the method in the user subroutine \texttt{cs\_user\_parameters.f90} (ivrtex=1)

The subroutine \texttt{usvort} contains 3 separate parts:

\begin{list}{-}{}
\item The 1st part defines the number of inlets concerned with the vortex
method (\texttt{nnentt\index{nnent}}) and the number of vortex for each inlet
(\texttt{nvort\index{nvort}}), where \texttt{ient} represents the number of inlets.
\item The 2nd part (\texttt{iappel=1}) defines the boundary faces at which the
      vortex method is applicable. The \texttt{irepvo\index{irepvo}} array is informed
      by \texttt{ient} which defines the number of inlets concerned with the vortex
(essentially, the vortex method can be applied with many independent inlets).
\item The 3rd section defines the main parameters of the method at each inlet.
      With the complexity of any given geometry, 4 cases are distinguished
      (the first 3 use the data file \texttt{ficvor} and in the final case only 1
      initial velocity and energy are imposed.):

\begin{list}{*}{}
\item \texttt{icas}=1, For the outlet of a rectangular pipe; 1 boundary condition is defined
for each side of the rectangle taking into account their interaction
with the vortex.
\item \texttt{icas}=2, For the outlet of a circular pipe; the entry face is considered as a
 wall (as far as interaction with the vortex is concerned)
\item \texttt{icas}=3, For inlets of any geometry; no boundary conditions are defined at the
 inlet face (i.e no specific treatment on the interaction between the
 vortex and the boundary)
\item \texttt{icas}=4, similar to \texttt{icas}=3 except the data file is not
 used (\texttt{ficvor}); the outflow
 parameters are estimated by the code from the global data (initial
 velocity, level of turbulence and dissipation), information which is
 supplied by the user.
\end{list}

When the geometry allows, cases 1 and 2 are used. Case 4 is only used
 if it is not possible to use the other three.

In the first 3 cases, the 2 base vectors in the plane of each inlet
must be defined (vectors \texttt{dir1} and \texttt{dir2}). The 3rd vector is
automatically calculated by the code, defined as a product of \texttt{dir1} and
\texttt{dir2}. \texttt{dir1} and \texttt{dir2} must be chosen imperatively to
give (\texttt{cen}, \texttt{dir1}, \texttt{dir2}) an orthogonal reference of the
inlet plane and so \texttt{dir3} is oriented in the entry domain. If
\texttt{icas}=2, the \texttt{cen} position must be the center of gravity of the
rectangle or disc.

The reference points (\texttt{cen}, \texttt{dir1}, \texttt{dir2}, \texttt{dir3})
 define the values of the variable in the \texttt{ficvor} file.\\
In the case where \texttt{icas}=4, the vectors \texttt{dir1} and \texttt{dir2}
are generated by the code.

If \texttt{icas}=1, the boundary conditions at the rectangle's edges must be
defined. They are defined in the array
\texttt{iclvor\index{iclvor}}. \texttt{iclvor(ii,ient)} represents the standard boundary
conditions at the edge II (1$\leqslant$II$\leqslant$4) of the inlet
\texttt{ient}. The code for the boundary conditions is as follows:
\begin{list}{*}{}
\item \texttt{iclvor}=1 for a wall
\item \texttt{iclvor}=2 for symmetry
\item \texttt{iclvor}=3 for periodicity of translation (the face corresponding
      to periodicity will automatically be taken as 3)
\end{list}
The 4 edges are numbered relative to the directions \texttt{dir1} and
\texttt{dir2} as shown in \figurename~\ref{fig:vortex}:

\begin{figure}[!ht]
\centerline{
\includegraphics*[width=8cm]{vortex}}
\caption{Numbering of the edges of a rectangular inlet(\texttt{icas}=1)
 treated by the vortex method}\label{fig:vortex}
\end{figure}

If \texttt{icas}=1, the user must define \texttt{llx} and \texttt{lly} which give
the lengths of the rectangular pipe in the directions \texttt{dir1} and \texttt{dir2}.\\
If \texttt{icas}=2, \texttt{lld} represents the diameter of the circular pipe.
If \texttt{icas}=4, \texttt{udebit}, \texttt{kdebit} and \texttt{edebit} are
defined for each inlet, these give respectively,
initial speed, turbulent energy level and the dissipation level. These can be used to
 obtain their magnitude using the correlations in the user routine \texttt{cs\_user\_boundary\_conditions} for
 fully developed flow in a pipe.

 The independent parameters are defined as follows:
\begin{list}{*}{}
\item \texttt{itmpl} represents the indicator of the advancement in time of the
  vortex. If \texttt{itmpli}=1, the vortex will be regenerated after a fixed
  time of
  \texttt{tmplim} second (defined as \texttt{itmpli}=1).
  If \texttt{itmpli}=2, following the data indicated in \texttt{ficvor} file,
  the vortex will have a variable life span equal to
  $5 \displaystyle C_\mu \displaystyle \frac{k^{\frac{3}{2}}}{\varepsilon U}$ ,
  where $C_\mu=0.09$ and $k$, $\varepsilon$ and $U$  represent respectively, turbulent energy,
  turbulent dissipation and the convective velocity in the direction normal to the inlet plane.

\item \texttt{xsgmvo} represents the support functions used in the vortex
  method. These are representative of the eddy sizes entered in the vortex
  method.
  \texttt{isgmvo} is used to define their size: if \texttt{isgmvo}=1,
  \texttt{xsgmvo} will be constant across the inlet face and is defined in
  \texttt{usvort}, if \texttt{isgmvo}=2, \texttt{xsgmvo} will be variable and
  equal to the mixing length of the standard $k-\varepsilon$ model
  ($\displaystyle {C_\mu}^{\frac{3}{4}} \displaystyle \frac{k^{\frac{3}{2}}}{\varepsilon}$), if
  \texttt{isgmvo}=3, \texttt{xsgmvo} will be equal to the maximum of $L_t$ et
  $L_K$ where $L_t$ and $L_K$ are the $\displaystyle \frac{\partial U}{\partial y}$
  $\displaystyle \frac{\partial U}{\partial y}$
  Taylor and Kolmogrov coefficients
  ($\displaystyle L_T=(5 \nu \frac{k}{\displaystyle \varepsilon})^{\frac{1}{2}}$,
  $\displaystyle L_K= 200 (\frac{\nu^3}{\varepsilon})^{\frac{1}{4}}$).

\item \texttt{idepvo} gives the vortex displacement method in the 2D inlet plane
  (the vortex method is a Lagrangian method in which the eddy centres are
  replaced by a set velocity). If \texttt{idepvo}=1, the velocity displacement
  referred to by \texttt{ud} which is the vortex following a random sampling
  (a sample number r, is taken for each vortex, at each time step and for each direction and
  the center of the vortex is replaced by the 2 principal directions,
  $r \mbox{ud} \Delta t$ where $\Delta t$ is the time step of the calculation).
  If \texttt{idepvo}=2, the vortex will be convected by itself (with the speed
  given by the time step before the vortex method)
\end{list}

A data file, \texttt{ficvor}, must be defined in the cases of
\texttt{icas}=1,2,3, for each inlet. The data file must contain the following
data in order ($x$, $y$, $U$, $\displaystyle \frac{\partial U}
{\partial y}$, $k$, $\varepsilon$). The number of lines of the file is given by
the integer \texttt{ndat}. $x$ and $y$ are the co-ordinates in the inlet plane
defined by the vectors \texttt{dir1} and \texttt{dir2}. $U$, $k$ and
$\varepsilon$ are respectively, the average speed normal to the inlet,
the turbulent energy and the turbulent dissipation.
$\displaystyle \frac{\partial U}{\partial y}
$ is the derivative in the direction normal to the
 inlet boundary in the cases, \texttt{icas}=1, \texttt{icas}=2.
 Where \texttt{icas}=3 and \texttt{icas}=4 this variable is not applied
 (it is given the value 0) so the Langevin equations, used to generate
 fluctuations normal to the inlet plane, is de-activated
 (the fluctuations normal to the inlet is 0 on both these cases).
 Note that the application of
 many different test of the Langevin equation doesn't have a notable influence
 on the results and that, by contrast it simply increases the computing time per
 iteration and so it decreases the random sampling which slows down the pressure
 solver. The interpolation used in the vortex method is defined by the function
 \texttt{phidat}. An example is given at the end of the subroutine \texttt{usvort} where the
 user can define the interpolation required. In the \texttt{phidat} function,
 \texttt{xx} and \texttt{yy} are the co-ordinates by which the value of
 \texttt{phidat} is calculated. \texttt{xdat} and \texttt{ydat} are the
 co-ordinates in the \texttt{ficvor} file. \texttt{vardat} is the
 value of the \texttt{phidat} function with the co-ordinates \texttt{xdat}
 and \texttt{ydat} (given in the \texttt{ficvor} file). Note that using an
 indicator \texttt{iii} accelerates the calculations (the user need not modify or delete).
 The user must also define the parameter \texttt{isuivo} which indicates if the
 vortex was started at 0 or if the file must be re-read (\texttt{ficmvo}).

\end{list}

{\bf \underline{WARNING}}
\begin{list}{$\bullet$}{}
\item Be sure that the \texttt{ficvor} file and  the interpolation in the user
  function \texttt{phidat} are compatible (in particular that all the entry
  region is covered by \texttt{ficvor})
\item If the user wants to use a 1D profile in the \texttt{dir2} direction,
 set $x$ =0 in the \texttt{ficvor} file and define the interpolation in
 \texttt{phidat}.
\end{list}

\subsubsubsection{Synthetic Eddy Method}
The user file \texttt{cs\_user\_les\_inflow.f90} allows to generate the
unsteady boundary conditions for the LES by the Synthetic Eddy
Method.
The basic principle of this method is illustrated in figure
\ref{sem_principle}: the turbulent fluctuations at the inlet are generated
by a set of synthetic eddies advected across the inlet
boundaries. The eddies evolve in a virtual ``box'' surrounding the inlet
boudaries and each of them contributes to the normalized velocity
fluctuations, depending on its relative position with the inlet faces
and on a form function characterizing the shape of the eddies. By this
way, the Synthetic Eddy Method provides a coherent flow with a target
mean velocity and target Reynolds stresses at LES inlet.

\begin{figure}
\centering
\includegraphics[width=0.6\linewidth]{sem_principle}
\put(-95,147){$\mathbf{U}_c$}
\put(-35,40){$v'^{SEM}$}
\put(-185,48){$\mathcal{S}$}
\put(-232,83){$\mathcal{B}$}
\caption{\label{sem_principle}Illustration of the principle of the
  Synthetic Eddy Method, with $\mathcal{S}$ the inlet boundary,
  $\mathcal{B}$ the virtual box and $\mathbf{U}_c$ the advection
  velocity of the eddies}
\end{figure}

{\bf \underline{WARNING}}: As for laminar or RANS inlets, the type of
boundary for LES inlets is \texttt{ientre}. It has to be specified in
the GUI or in the \texttt{cs\_user\_boundary\_conditions}
surboutine. On the contrary, if Dirichlet values are given for these
faces in the GUI or in the \texttt{cs\_user\_boundary\_conditions}
subroutine (\texttt{rcodcl(ifac,ivar,1)} array), they are erased by
those provided by the Synthetic Eddy Method.

In the current version of \CS, the Synthetic Eddy Method is not
available through the GUI but only through the
\texttt{cs\_user\_les\_inflow.f90} user file. The user file contains 3
subroutines:

\begin{itemize}
\item \texttt{cs\_user\_les\_inflow\_init} (mandatory): global definition
  of synthetic turbulence inlets
\item \texttt{cs\_user\_les\_inflow\_define} (mandatory): specific
  definition of each synthetic turbulence inlet
\item \texttt{cs\_user\_les\_inflow\_advanced} (not mandatory): advanced
  definition of each synthetic turbulence inlet
\end{itemize}

\texttt{cs\_user\_les\_inflow\_init}: this subroutine defines
some global parameters shared by all LES inlets. These parameters are:

\begin{itemize}
\item \texttt{nent}: number of LES inlet boundaries
\item \texttt{isuisy}: in case of a restart calculation, it indicates if
  the synthetic turbulence is re-initialize (0) or read from the
  previous calculation (1). In that case, the checkpoint folder must
  contain the \texttt{les\_inflow} restart file. This file is
  generated during a computation with synthetic turbulence, at the
  same physical times as the main and auxiliary restart files.
\end{itemize}

\texttt{cs\_user\_les\_inflow\_define}: this subroutine defines
the specific parameters of each LES inlet. These parameters are:

\begin{itemize}
\item \texttt{typent}: type of LES inflow method. The Synthetic Eddy
  Method corresponds to \texttt{typent=3}. For the sake of
  comparision, other methods can be selected through this user file
  (see remark 2).

\item \texttt{nelent}: number of synthetic eddies in the ``box''. This
  parameter might be adjusted, depending on the case (in particular
  the size of the inlet plane and the level of turbulence). As a
  general rule, the greater is the better since an insufficient number
  can lead to an intermittent signal while some numerical tests have
  shown that this parameter does not have a great influence beyond a
  threshold value. Given the inlet of size $h^2$ of a shear flow at a
  given Reynolds number $Re=u_\tau h/\nu$, an appropriate number of
  eddies can be evaluated by $(Re/50)^3$ ($Re$ and 50 approximates
  respectively the size, in wall unit, of the largest and the smallest
  synthetic eddy. Note the latter can depend on the grid size, see
  remark 1).

\item \texttt{iverbo}: level of verbosity in the
  listing. \texttt{iverbo=1} provides mainly informations about the
  size of the eddies and the size of the ``box'' surrounding the inlet
  boundary.

\item \texttt{nfbent} and \texttt{lfbent}: number and list of boundary
  faces composing the LES inlet boundary.

\item \texttt{vitent}: reference mean velocity at inlet. This
  parameter imposes the target mean velocity at inlet. A finer (non
  homogeneous) definition of the mean velocity can be done in the
  \texttt{cs\_user\_les\_inflow\_advanced} subroutine (see below).

\item \texttt{enrent}: reference turbulence kinetic energy $k$ at
  inlet. This parameter imposes the target Reynolds stresses $R_{ij}$
  at inlet, computed by $R_{ij}=\frac{2}{3}k\delta_{ij}$ (isotropy). A
  finer (non isotropic and/or non homogeneous) definition of the
  Reynolds stresses can be done in the
  \texttt{cs\_user\_les\_inflow\_advanced} subroutine (see below).

\item \texttt{dspent}: reference dissipation rate $\varepsilon$ at
  inlet. This parameter is used to compute the size of the synthetic
  eddies (see remark 1). A finer (non homogeneous) definition of
  the dissipation rate can be done in the
  \texttt{cs\_user\_les\_inflow\_advanced} subroutine (see below).
\end{itemize}

\texttt{cs\_user\_les\_inflow\_advanced}: this optional subroutine
enables to give an accurate (non homogeneous) specification of inflow
statistics: mean velocity (\texttt{uvwent} array), Reynolds stresses
(\texttt{rijent} array) and dissipation rate (\texttt{epsent}
array). In that case, this accurate specification replaces the
one given in \texttt{cs\_user\_les\_inflow\_define} subroutine
(\texttt{vitent}, \texttt{enrent} and \texttt{dspent} variables).

{\bf \underline{REMARK 1}}: The specification of the dissipation rate
$\varepsilon$ at inlet is used to compute the size $\sigma_i$ of the
synthetic eddies in the $i$ cartesian direction. One has:
$$\sigma_i=\max\left\{C\frac{\big(\frac{3}{2}R_{ii}\big)^{3/2}}{\varepsilon},\Delta\right\},\qquad
C=0.5.$$
$\Delta$ is a reference size of the grid, in order to assume that all
synthetic eddies are discretized. In the implementation of \CS, it is
computed at each inlet boundary face $F$ as:
$$\Delta=2\max_{i\le3,V\in\mathcal{V}}\Big\{\big|x_i^V-x_i^C\big|\Big\}$$
with $\mathcal{V}$ the subset of the vertices of the boundary face $F$
and $C$ the cell adjacent to $F$.

{\bf \underline{REMARK 2}}: For the sake of comparison, others LES inflow methods are
available through the
\texttt{cs\_user\_les\_inflow.f90} user file, in addition to the
Synthetic Eddy Method:

\begin{itemize}
\item The Batten method corresponds to \texttt{typent=2} in
  \texttt{cs\_user\_les\_inflow\_define} subroutine. With this method,
  the inflow velocity signal is the superposition of several Fourier
  modes. The number of modes is indicated through the
  \texttt{nelent} keyword. As   for Synthetic Eddy Method, the mean
  velocity, the turbulent kinetic energy and the dissipation rate have
  to be specified at inlet: either giving their reference values
  (\texttt{vitent}, \texttt{enrent} and \texttt{dspent}) in the
  \texttt{cs\_user\_les\_inflow\_define} subroutine, either providing
  an accurate local description in the
  \texttt{cs\_user\_les\_inflow\_advanced} subroutine.

\item \texttt{typent=1}: turbulent fluctuations are given by a Gaussian
  noise. The mean velocity and Reynolds stresses have to be specified
  (in \texttt{cs\_user\_les\_inflow\_define} or in
  \texttt{cs\_user\_les\_inflow\_advanced}). The other parameters of
  the user subroutines are useless. The turbulent fluctuations
  provided by this method are much less realistic than those provided
  by the Synthetic Eddy Method or the Batten method. Especially for
  low Reynolds number flows, this could lead to the rapid dissipation
  of this fluctuations and the laminarization of the flow.

\item \texttt{typent=0}: No fluctuation. This method does not require
  any parameter. It should be reserved to regions where the flow is
  laminar.
\end{itemize}

%==================================
\subsection{Manage the variable physical properties}
%==================================
%==================================
\subsubsection{Basic variable physical properties}\label{sec:usphyv}
%==================================
When the fluid properties are not constant, the user is offered the choice to define the variation laws in the Graphical User Interface (GUI) or in the subroutine \texttt{cs\_user\_physical\_properties} which is called at each time step. In the GUI, in the item ``Fluid properties'' under the heading ``Physical properties'', the variation laws are defined for the fluid density, viscosity, specific heat, thermal conductivity and scalar diffusivity through the use of a formula editor, see \figurename~\ref{fig:gui_fluid_props2} and \figurename~\ref{fig:UL1_GUI}.
%
\begin{figure}[!ht]
\begin{center}
\includegraphics[width=\textwidth]{gui_fluid_props}
\caption{Physical properties - Fluid properties}
\label{fig:gui_fluid_props2}
\end{center}
\end{figure}
%
\begin{figure}[!ht]
\begin{center}
\includegraphics[width=10cm]{gui_density_law}
\caption{Definition of a user law for the density}
\label{fig:UL1_GUI}
\end{center}
\end{figure}

If necessary, all the variation laws related to the fluid physical properties are written in the subroutine \texttt{cs\_user\_physical\_properties}.

The validity of the variation laws must be checked, particularly when
non-linear laws are defined (for instance, a third-degree polynomial law
may produce negative density values).

{\bf \underline{WARNING}}\label{sec:prg_propvar}%
\begin{list}{$\bullet$}{}
\item If the user wishes to impose a variable density or variable viscosity in
      \texttt{usphyv}, it must be flagged either in the interface or in
      \texttt{cs\_user\_parameters.f90}(\texttt{irovar}=1, \texttt{ivivar}=1).
\item In order to impose a physical property ($\rho$, $\mu$,
      $\lambda$, $C_p$)\footnote{Except for some specific physics}, a reference
      value should be provided in the interface or in \texttt{cs\_user\_parameters.f90} (in
      particular for $\rho$, the pressure will be function of $\rho_0 gz$)
\item By default, the $C_p$ coefficient and the
      diffusivity for the scalars \texttt{iscal} ($\lambda_T$ for the
      temperature) are considered as constant in time and uniform in
      space, with the values \texttt{cp0} and \texttt{visls0(iscal)}
      specified in the interface or in \texttt{cs\_user\_parameters.f90}.\\
To assign a variable value to $C_p$, the user \textbf{must} specify it in the
      interface (with a user law) or assign the value 1 to \texttt{icp} in \texttt{cs\_user\_parameters.f90},
      and fill for each cell \texttt{iel} the array
      \texttt{propce(iel,ipccp)} in \texttt{cs\_user\_physical\_properties}. NB: completing the array
      \\\texttt{propce(iel,ipccp)} while \texttt{icp=0} induces array
      overwriting problems and produces wrong results.

\item In the same way, to have variable diffusivities for the scalars
      \texttt{iscal}, the user \textbf{must} specify it in the interface (with a user law) or calling \texttt{field\_set\_key\_int(ivarfl(isca(iscal)), kivisl, 0)}
      in \texttt{cs\_user\_parameters.f90} (in \texttt{usipsu}), and complete for each cell
      \texttt{iel} the values array of the field id \texttt{ifcvsl} returned by calling
      \texttt{field\_get\_key\_id(ivarfl(isca(iscal)), kivisl, ifcvsl)}
      in \texttt{cs\_user\_physical\_properties}.\\

{\em Note}: The scalar diffusivity id must not be defined for
      user scalars representing the average of the square of the
      fluctuations of another scalar, because the diffusivity of a user
      scalar \texttt{jj} representing the average of the square of the
      fluctuations of a user scalar \texttt{kk} comes directly from the
      diffusivity of this last scalar. In particular, the diffusivity
      of the scalar \texttt{jj} is variable if the diffusivity of \texttt{kk}
      is variable.
\end{list}


%==================================
\subsubsection{Modification of the turbulent viscosity}
%==================================

The subroutine \texttt{usvist} is used to modify the calculation of the turbulent
viscosity, {\em i.e.} $\mu_t$ in $kg.m^{-1}.s^{-1}$
(this piece of information, at the mesh cell centres, is conveyed by the
variable \texttt{propce(iel,ipcvst)}, with
\texttt{ipcvst =  ipproc(ivisct)}). The
subroutine is called at the beginning of every time step, after the
calculation of the physical parameters of the flow and of the
``conventional'' value of $\mu_t$ corresponding to the chosen turbulence
model (indicator \texttt{iturb}).\\
{\em WARNING: The calculation of the turbulent viscosity being a
particularly sensible stage, a wrong use of {\em\texttt{usvist}} may
seriously distort the results.}

%==================================
\subsubsection{Modification of the variable $C$ of the dynamic LES model}
%==================================

\noindent
\textit{Subroutine called every time step in the case of LES with the
dynamic model.}

The subroutine \texttt{ussmag} is used to modify the calculation of the variable $C$ of
the LES sub-grid scale dynamic model.

It worth to recalling that the LES approach introduces the notion of
filtering between large eddies and small motions. The solved variables
are said to be filtered in an ``implicit'' way. Sub-grid scale models
(``dynamic'' models) introduce in addition an explicit filtering.

The notations used for the definition of the variable $C$ used in the
dynamic models of \CS are specified below. These notations are the ones
assumed in the document \cite{benhamadouche01}, to which the user may
refer to for more details.

The value of $a$ filtered by the explicit filter (of width
$\widetilde{\overline{\Delta}}$) is called $\widetilde{a}$ and the value
of $a$ filtered by the implicit filter (of width $\overline{\Delta}$) is
called $\overline{a}$.
We define:
\begin{equation}
\begin{array}{ll}
\overline{S}_{ij}=\frac{1}{2}(\frac{\partial \overline{u}_i}{\partial x_j}
                  +\frac{\partial \overline{u}_j}{\partial x_i})  &
||\overline{S}||=\sqrt{2 \overline{S}_{ij}\overline{S}_{ij}}\\
\alpha_{ij}=-2\widetilde{\overline{\Delta}}^2
             ||\widetilde{\overline{S}}||
             \widetilde{\overline{S}}_{ij}&
\beta_{ij}=-2\overline{\Delta}^2
             ||\overline{S}||
               \overline{S}_{ij}\\
L_{ij}=\widetilde{\overline{u}_i\overline{u}_j}-
 \widetilde{\overline{u}}_i\widetilde{\overline{u}}_j&
M_{ij}=\alpha_{ij}-\widetilde{\beta}_{ij}\\
\end{array}
\end{equation}


In the framework of LES, the total viscosity (molecular + sub-grid) in
$kg.m^{-1}.s^{-1}$ may be written in \CS:
\begin{equation}
\begin{array}{llll}
\mu_{\text{total}}&=&\mu+\mu_{\text{sub-grid}} &
    \text{\ \ if\ \ }\mu_{\text{sub-grid}}>0\\
                   &=&\mu                          &
    \text{\ \ otherwise }\\
\text{with\ }\mu_{\text{sub-grid}}&=&\rho C \overline{\Delta}^2 ||\overline{S}||
\end{array}
\end{equation}

$\overline{\Delta}$ is the width of the implicit filter, defined at the
cell $\Omega_i$ by \\
$\overline{\Delta}=XLESFL*(ALES*|\Omega_i|)^{BLES}$
\index{xlesfl}\index{ales}\index{bles}.

In the case of the Smagorinsky model (\texttt{iturb=40}), $C$ is a
constant which is worth $C_s^2$. $C_s^2$ is the so-called Smagorinsky
constant and is stored in the variable \texttt{$csmago$\index{csmago}}.

In the case of the dynamic model (\texttt{iturb=41}), $C$ is variable in
time and in space. It is determined by
$\displaystyle C=\frac{M_{ij}L{ij}}{M_{kl}M_{kl}}$.

In practice, in order to increase the stability, the code does not use the
value of $C$ obtained in each cell, but an average with the values
obtained in the neighbouring cells (this average uses the extended
neighbourhood and corresponds to the explicit filter). By default, the
value calculated by the code is
\begin{displaymath}
C=\frac{\widetilde{M_{ij}L{ij}}}{\widetilde{M_{kl}M_{kl}}}
\end{displaymath}

The subroutine \texttt{ussmag} allows to modify this value. It is for
example possible to calculate the local average after having calculated the
ratio
\begin{displaymath}
C=\widetilde{\left[\frac{M_{ij}L{ij}}{M_{kl}M_{kl}}\right]}
\end{displaymath}

{\em WARNING: The subroutine {\em\texttt{ussmag}} can be activated only when
the dynamic model is used.}

%==================================
\subsection{User source terms}
%==================================
\label{sec:prg_usersourceterms}

Assume, for example, that the user source terms modify the equation of a
variable $\varphi$ in the following way:
\begin{displaymath}
\rho\frac{\partial \varphi}{\partial t}+\ldots = \ldots + S_{impl}\times\varphi+S_{expl}
\end{displaymath}
The example is valid for a velocity component, for a turbulent variable ($k$, $\varepsilon$, $R_{ij}$, $\omega$,
$\varphi$ or $\overline{f}$) and for a scalar (or for the average of the
square of the fluctuations of a scalar), because the syntax of all the
subroutines \texttt{ustsnv}, \texttt{cs\_user\_turbulence\_source\_terms}
 and \texttt{ustssc} in the \texttt{cs\_user\_source\_terms} file is similar.

In the finite volume formulation, the solved system is then modified as
follows:
\begin{displaymath}
\left(\frac{\rho_i\Omega_i}{\Delta t_i}-\Omega_iS_{impl,i}\right)
\left(\varphi_i^{(n+1)}-\varphi_i^{(n)}\right)
+\ldots = \ldots + \Omega_iS_{impl,i}\varphi_i^{(n)} + \Omega_iS_{expl,i}
\end{displaymath}
The user needs therefore to provide the following values:\\
$\text{\texttt{crvimp}}_i=\Omega_iS_{impl,i}$\\
$\text{\texttt{crvexp}}_i=\Omega_iS_{expl,i}$

In practice, it is essential for the term
$\displaystyle \left(\frac{\rho_i\Omega_i}{\Delta
t_i}-\Omega_iS_{impl,i}\right)$ to be positive. To ensure this property,
the equation really taken into account by the code is the following:
\begin{displaymath}
\left(\frac{\rho_i\Omega_i}{\Delta t_i}-
\text{Min}(\Omega_iS_{impl,i};0)\right)
\left(\varphi_i^{(n+1)}-\varphi_i^{(n)}\right)
+\ldots = \ldots + \Omega_iS_{impl,i}\varphi_i^{(n)} + \Omega_iS_{expl,i}
\end{displaymath}
To make the ``implicitation'' effective, the source term decomposition
between the implicit and explicit parts will be done by the user who must
ensure that $\text{\texttt{crvimp}}_i=\Omega_iS_{impl,i}$ is always negative
(otherwise the solved equation remains right, but there will not be
``implicitation'').

{\em WARNING: When the second-order in time is used along with the extrapolation of the
source terms\footnote{indicator \texttt{isno2t} for the velocity,
\texttt{isto2t} for the turbulence and \texttt{isso2t} for the scalars}, it is no longer possible to test the sign of $S_{impl,i}$,
because of coherence reasons (for more details, the user may refer to
the theoretical and computer documentation \cite{theory} of the
subroutine \texttt{preduv}). The user must therefore make sure it is
always positive (or take the risk to affect the calculation stability).}

\minititre{Particular case of a linearised source term}

In some cases, the added source term is not linear, but the user may
want to linearise it using a first-order Taylor development, in order to
make it partially implicit.\\
Consider an equation of the type:
\begin{displaymath}
\rho\frac{\partial\varphi}{\partial t}=F(\varphi)
\end{displaymath}
To make it implicit using the following method:
\begin{eqnarray*}
\frac{\rho_i\Omega_i}{\Delta t}\left(\varphi_i^{(n+1)}-\varphi_i^{(n)}\right) & = &
\Omega_i\left[F(\varphi_i^{(n)})+\left(\varphi_i^{(n+1)}-\varphi_i^{(n)}\right)
\frac{dF}{d\varphi}(\varphi_i^{(n)})\right]\\
& = & \Omega_i\frac{dF}{d\varphi}(\varphi_i^{(n)})\times\varphi_i^{(n+1)}
+\Omega_i\left[F(\varphi_i^{(n)})-\frac{dF}{d\varphi}(\varphi_i^{(n)})
\times\varphi_i^{(n)}\right]
\end{eqnarray*}

The user must therefore specify:\\
$\displaystyle\text{\texttt{crvimp}}_i=\Omega_i\frac{dF}{d\varphi}(\varphi_i^{(n)})$\\
$\displaystyle\text{\texttt{crvexp}}_i=
\Omega_i\left[F(\varphi_i^{(n)})-\frac{dF}{d\varphi}(\varphi_i^{(n)})\times\varphi_i^{(n)}\right]$

\underline{\em Example}:\\
If the equation is
$\displaystyle \rho\frac{\partial\varphi}{\partial t}=-K\varphi^2$,
the user must set:\\
$\text{\texttt{crvimp}}_i=-2K\Omega_i\varphi_i^{(n)}$\\
$\text{\texttt{crvexp}}_i=K\Omega_i[\varphi_i^{(n)}]^2$

%==================================
\subsubsection{In Navier-Stokes}
%==================================

The source term in Navier-Stokes can be filled in thanks to the GUI or the
\texttt{cs\_user\_source\_terms} user file.
Without the GUI, the subroutine \texttt{ustsnv} is used to
add user source terms to the Navier-Stokes equations (at each time step).

\texttt{ustsnv} is called only once per time step; for each cell \texttt{iel},
the vector \texttt{crvexp(.,iel)} (explicit part) and the matrix \texttt{crvimp(.,.,iel)}
(implicit part) must be filled in for the whole velocity vector.


%==================================
\subsubsection{For $k$ and $\varepsilon$}
%==================================

\noindent
\textit{Subroutine called every time step, for the $k-\varepsilon$ and
the v2f models.}

The subroutine \texttt{cs\_user\_turbulence\_source\_terms} is used to add source terms to the transport equations
related to the turbulent kinetics energy $k$ and to the turbulent
dissipation $\varepsilon$.
This subroutine is called every time step (the
treatment of the two variables $k$ and $\varepsilon$ is made
simultaneously). The user is expected to provide the arrays \texttt{crkimp} and
\texttt{crkexp} for $k$, and \texttt{creimp} and \texttt{creexp} for
$\varepsilon$. These arrays are similar to the arrays \texttt{crvimp} and
\texttt{crvexp} given for the velocity in the user subroutine \texttt{ustsnv}.
The way of making implicit the resulting source terms is the same as the one
presented in \texttt{ustsnv}. For $\varphi$ and $\bar{f}$
in the v2f model, see \texttt{cs\_user\_turbulence\_source\_terms}, \S\ref{sec:prg_ustsv2}.

%==================================
\subsubsection{For $R_{ij}$ and $\varepsilon$}
%==================================

\noindent
\textit{Subroutine called every time step, for the $R_{ij}-\varepsilon$ models.}

The subroutine \texttt{cs\_user\_turbulence\_source\_terms} is used to add source terms to the transport equations
related to the Reynolds stress variables $R_{ij}$ and to the turbulent
dissipation $\varepsilon$.
This subroutine is called 7 times every time step
(once for each Reynolds stress component and once for the
dissipation). The user must provide the arrays \texttt{crvimp} and
\texttt{crvexp} for the field variable of index \texttt{f\_id} (referring successively to
\texttt{ir11}, \texttt{ir22}, \texttt{ir33},
\texttt{ir12}, \texttt{ir13}, \texttt{ir23} and
\texttt{iep}). These arrays are similar to the arrays \texttt{crvimp}
and \texttt{crvexp} given for the velocity in the user subroutine
\texttt{ustsnv}. The method for impliciting the resulting source terms is the
same as that presented in \texttt{ustsnv}.

%==================================
\subsubsection{For $\varphi$ and $\overline{f}$}
%==================================
\label{sec:prg_ustsv2}

\noindent
\textit{Subroutine called every time step, for the v2f models.}

The subroutine \texttt{cs\_user\_turbulence\_source\_terms} is used to add source terms to the transport equations
related to the variables $\varphi$ and $\overline{f}$ of the v2f
$\varphi$-model. This subroutine is called twice
every time step (once for $\varphi$ and once for $\overline{f}$).
The user is expected to provide the arrays \texttt{crvimp} and
\texttt{crvexp} for \texttt{ivar} referring successively to \texttt{iphi}
and \texttt{ifb}. Concerning $\varphi$, these arrays are similar to the arrays
\texttt{crvimp} and \texttt{crvexp} given for the velocity in the user subroutine
\texttt{ustsnv}. Concerning $\overline{f}$, the equation is slightly
different:
\begin{displaymath}
L^2 div(\grad(\overline{f})) = \overline{f} + \ldots + S_{impl}\times\overline{f}+S_{expl}
\end{displaymath}
In the finite volume formulation, the solved system is written as:
\begin{displaymath}
\int_{\partial\Omega_i}\grad(\overline{f})^{(n+1)}dS=\frac{1}{L_i^2}\left(
\Omega_i\overline{f}^{(n+1)}_i + \ldots +  \Omega_iS_{impl,i}\overline{f}_i^{(n+1)} +
\Omega_iS_{expl,i} \right)
\end{displaymath}
The user must then specify:\\
$\text{\texttt{crvimp}}_i=\Omega_iS_{impl,i}$\\
$\text{\texttt{crvexp}}_i=\Omega_iS_{expl,i}$

The way of making implicit the resulting source terms is the same as the
one presented in \texttt{ustsnv}.

%==================================
\subsubsection{For $k$ and $\omega$}
%==================================

\noindent
\textit{Subroutine called every time step, for the $k-\omega$ SST model.}

The subroutine \texttt{cs\_user\_turbulence\_source\_terms} is used to add source terms to the transport equations
related to the turbulent kinetics energy $k$ and to the specific
dissipation rate $\omega$. This subroutine is
called every time step (the treatment of the two
variables $k$ and $\omega$ is made simultaneously). The user is expected
to provide the arrays \texttt{crkimp} and \texttt{crkexp} for the variable
$k$, and the arrays \texttt{crwimp} and \texttt{crwexp} for the variable $\omega$.
These arrays are similar to the arrays \texttt{crvimp} and \texttt{crvexp}
given for the velocity in the user subroutine \texttt{ustsnv}. The way of
making implicit the resulting source terms is the same as the one presented in
\texttt{ustsnv}.

%==================================
\subsubsection{For $\tilde{\nu}_t$}
%==================================

\noindent
\textit{Subroutine called every time step, or the Spalart-Allmaras model.}

The subroutine \texttt{cs\_user\_turbulence\_source\_terms} is used to add source terms to the transport equations
related to the turbulent viscosity $\nu_t$ for the Spalart-Allmaras model.
This subroutine is called every time step. The user is expected
to provide the arrays \texttt{crkimp} and \texttt{crkexp} for the variable
$\tilde{\nu}_t$. These arrays are similar to the arrays \texttt{crvimp} and \texttt{crvexp}
given for the velocity in the user subroutine \texttt{ustsnv}. The way of
making implicit the resulting source terms is the same as the one presented in
\texttt{ustsnv}.

%==================================
\subsubsection{For user scalars}
%==================================

\noindent
\textit{Subroutine called every time step.}

The source terms in the transport equations related to the user scalars
(passive or not, average of the square of the fluctuations of a scalar, ...)
can be filled in thanks to the GUI or the \texttt{cs\_user\_source\_terms} user file.
Without the GUI, the subroutine \texttt{ustssc} is used to add source terms to the
transport equations related to the user scalars. In the same way as
\texttt{ustsnv}, this subroutine is called every time step, once for
each user scalar. The user must provide the arrays \texttt{crvimp}
and \texttt{crvexp} related to each scalar. \texttt{cvimp} and \texttt{crvexp}
must be set to 0 for the scalars on which it is not wished for the user source
term to be applied (the arrays are initially set to 0 at each inlet in the subroutine).

%==================================
\subsection{Pressure drops (head losses) and porosity}
%==================================
\label{sec:prg_headlosses}

%==================================
\subsubsection{Head losses}
%==================================

Pressure drops can be defined in the Graphical User Interface (GUI) or in the subroutine \texttt{cs\_user\_head\_losses.f90} (called three times every time step). In the GUI, under the heading ``Volume conditions'', the item ``Volume regions definition'' allows to define areas where pressure drops occur, see an example in fig \ref{fig:hl1}. The item ``Head losses'' allows to specify the head loss coefficients, see \figurename~\ref{fig:hl2}. The tensor representing the pressure drops is supposed to be symmetric
and positive.

\begin{figure}[!ht]
\begin{center}
%\begin{tabular}{c}
%\includegraphics[width=14cm]{gui_volume_regions} \\
\includegraphics[width=\textwidth]{gui_head_loss_regions}
%\end{tabular}
\caption{Creation of head losses region}
\label{fig:hl1}
\end{center}
\end{figure}
%
\begin{figure}[!ht]
\begin{center}
\includegraphics[width=\textwidth]{gui_head_loss_coeffs}
\caption{Head losses coefficients}
\label{fig:hl2}
\end{center}
\end{figure}

If necessary, the pressure drops are written in the
subroutine \texttt{cs\_user\_head\_losses.f90}.

\begin{list}{$\bullet$}{}
\item During the first call, all the cells are checked to know the
      number of cells in which a pressure drop is present.
      This number is called \texttt{ncepdp\index{ncepdp}} in
      \texttt{cs\_user\_head\_losses.f90} (and corresponds to
      \texttt{ncepdc\index{ncepdc}}). It is used to lay out the arrays
      related to the pressure drops. If there is no pressure drop,
      \texttt{ncepdp} must be equal to zero (it is the default value, and the
      rest of the subroutine is then useless).

\item During the second call, all the cells are checked again to
      complete the array \texttt{icepdp\index{icepdp}} whose size is
      \texttt{ncepdp}. \mbox{\texttt{icepdc(ielpdc)}} is the number of the
      \texttt{ielpdc}\raisebox{1ex}{\small th} cell containing pressure drops.

\item During the third call, all the cells containing pressure drops
      are checked in order to complete the array
      containing the components of the tensor of pressure drops
      \mbox{\texttt{ckupdc(ncepdp,6)}\index{ckupdc}}. This array is so that
      the equation related to the velocity may be written:
\begin{displaymath}
\rho\frac{\partial}{\partial t}\vect{u}=\ldots -\rho\tens{K}\ind{pdc}\cdot\vect{u}
\end{displaymath}
The tensor components are given in the following order (in the general
      reference frame): \texttt{k11}, \texttt{k22}, \texttt{k33}, \texttt{k12},
      \texttt{k23}, \texttt{k13} with \texttt{k12}, \texttt{k23} and \texttt{k13}
      being zero if the tensor is diagonal.

\end{list}


The three calls are made at every time step, so that variable pressure drop
zones or values may be treated.

%==================================
\subsubsection{Porosity}
%==================================

The management of the porosity is not yet available in the GUI. To define the porosity, the user
must fill in the subroutine \texttt{usporo} and set the keyword \texttt{iporos} to $\texttt{1}$ in the
\texttt{cs\_user\_parameters} file. This subroutine is called every time step.\\
When using the subroutine \texttt{usporo}, the user is expected to fill in the array \texttt{porosi}
for each cell in order to give the porosity.

%==================================
\subsection{Management of the mass sources}
%==================================

The subroutine \texttt{cs\_user\_mass\_source\_terms} is used to add a density source term in some cells of
the domain (called at each time step). The mass conservation equation is then modified as follows:
\begin{displaymath}
\frac{\partial \rho}{\partial t} + div(\rho\vect{u})=\Gamma
\end{displaymath}

$\Gamma$ is the mass source term expressed in $kg.m^{-3}.s^{-1}$.

The presence of a mass source term modifies the evolution equation of
the other variables, too. Let $\varphi$ be any solved variable apart
from the pressure (velocity component, turbulent energy, dissipation,
scalar, ...). Its evolution equation becomes:
\begin{displaymath}
\rho\frac{\partial \varphi}{\partial t} + \ldots = \ldots + \Gamma(\varphi_i-\varphi)
\end{displaymath}

$\varphi_i$ is the value of $\varphi$ associated with the mass entering
or leaving the domain. After discretisation, the equation may be written:
\begin{displaymath}
\rho\frac{\varphi^{(n+1)}-\varphi^{(n)}}{\Delta t} + \ldots
= \ldots + \Gamma(\varphi_i-\varphi^{(n+1)})
\end{displaymath}

For each variable $\varphi$, there are two possibilities:
\begin{list}{$\bullet$}{}
\item We can consider that the mass is added (or removed) with the
      ambient value of $\varphi$. In this case
      $\varphi_i=\varphi^{(n+1)}$ and the equation of $\varphi$ is not
      modified.
\item Or we can consider that the mass is added with an
      imposed value $\varphi_i$ (this solution is physically correct
      only when the mass is effectively added, $\Gamma>0$).
\end{list}

\bigskip

This subroutine is called three times every time step.

\begin{list}{$\bullet$}{}
\item During the first call, all the cells are checked to know the
      number of cells containing a mass source term.
      This number is called \texttt{ncesmp\index{ncesmp}} in
      \texttt{cs\_user\_mass\_source\_terms} (and corresponds to
      \texttt{ncetsm\index{ncetsm}}). It is used to lay out the arrays
      related to the mass sources. If there is no mass source,
      \texttt{ncesmp} must be equal to zero (it is the default value, and the
      rest of the subroutine is then useless).

\item During the second call, all the cells are checked again to
      complete the array \texttt{icetsm\index{icetsm}} whose dimension is
      \texttt{ncesmp}. \mbox{\texttt{icetsm(ieltsm)}} is the number of the
      \texttt{ieltsm}\raisebox{1ex}{\small th} cell containing a mass source.

\item During the third call, all the cells containing mass sources are
      checked in order to complete the arrays
      \mbox{\texttt{itypsm(ncesmp,nvar)}\index{itypsm}} and
      \mbox{\texttt{smacel(ncesmp,nvar)}\index{smacel}}:\\
- \texttt{itypsm(ieltsm,ivar)} is the flow type associated with the variable
      \texttt{ivar} in the \texttt{ielstm}\raisebox{1ex}{\small th} cell
      containing a mass source.\\
\hspace*{1cm}\texttt{itypsm}=0: $\varphi_i=\varphi^{(n+1)}$ condition\\
\hspace*{1cm}\texttt{itypsm}=1: imposed $\varphi_i$ condition\\
\hspace*{1cm}\texttt{itypsm} is not used for \texttt{ivar=ipr}\\
- \texttt{smacel(ieltsm,ipr)} is the value of the mass source term $\Gamma$, in
$kg.m^{-3}.s^{-1}$.\\
- \texttt{smacel(ieltsm,ivar)}, for \texttt{ivar} different from
\texttt{ipr}, is the value
of $\varphi_i$ for the variable \texttt{ivar} in the
\texttt{ielstm}\raisebox{1ex}{\small th} cell containing a mass source.\\

\minititre{Notes}
$\bullet$ If \texttt{itypsm(ieltsm,ivar)=0}, \texttt{smacel(ieltsm,ivar)}
      is not used.\\
$\bullet$ If $\Gamma$=\texttt{smacel(ieltsm,ipr)}$<$0, mass is removed from
      the system, and \CS considers automatically a
      $\varphi_i=\varphi^{(n+1)}$ condition, whatever the values given
      to \texttt{itypsm(ieltsm,ivar)} and \texttt{smacel(ieltsm,ivar)}
      (the extraction of a variable is done at ambient value).
\end{list}

The three calls are made every time step, so that variable mass source
zones or values may be treated.\\

For the variance, do not take into account the scalar $\varphi_i$ in the environment
where $\varphi\ne\varphi_i$ generates a variance source.

%==================================
\subsection{User law editor of the GUI}
%==================================

A formula interpreter is embedded in \CS, which can be used through the GUI.
In order to call the formula editor of the GUI, click on the button:

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=1cm]{gui_formula_button}
\label{fig:mei_button}
\end{center}
\end{figure}

The formula editor is a window with three tabs:
\begin{list}{$\bullet$}{}
\item User expression

This tab is the formula editor. At the opening of the
window only the required symbols are displayed.
The syntax colorization shows to the user symbols which are
required symbols, functions, or user variables.
Each expression must be closed by a semicolon (``;''). The
required symbols must be present in the final user law. A
syntax checker is used when the user clicks on the OK button.

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=10cm]{gui_density_law}
\caption{Example of the user law editor}
\label{fig:mei_editor}
\end{center}
\end{figure}

\item Predefined symbols

There are three types of symbols \\

\underline{Useful functions:}

\texttt{cos}: cosine

\texttt{sin}: sine

\texttt{tan}: tangent

\texttt{exp}: exponential

\texttt{sqrt}: square root

\texttt{log}: Napierian logarithm

\texttt{acos}: arc cosine

\texttt{asin}: arc sine

\texttt{atan(x)}: arc tangent (arc tangent of x in radians; the return value is in the range [-pi/2, pi/2])

\texttt{atan2(y,x)}: arc tangent (arc tangent of y/x in radians; the return value is in the range [-pi, pi])

\texttt{cosh}: hyperbolic cosine

\texttt{sinh}: hyperbolic sine

\texttt{tanh}: hyperbolic tangent

\texttt{abs}: absolute value

\texttt{mod}: modulo

\texttt{int}: floor

\texttt{min}: minimum

\texttt{max}: maximum

\texttt{interp1d}: 1D linear interpolation\\


\underline{Useful constants:}

pi = 3.14159265358979323846

e = 2.718281828459045235\\


\underline{Operators and statements:}

$+$ \qquad$-$\qquad $*$\qquad $/$ \qquad$\wedge$

! \qquad $<$ \qquad $>$ \qquad $<=$ \qquad $>=$ \qquad $==$ \qquad $!=$ \qquad $\&\&$ \qquad $\mid\mid$

\texttt{while} \texttt{if} \texttt{else} \texttt{print}


\item Examples

This tab displays examples of formula, which could be copy and paste.

\end{list}

\underline{External data}

Through the predefined function \texttt{interp1d} it is possible to use
external data inside a user law. This function allows to make linear interpolation
from a data file (\texttt{.dat} or \texttt{.csv}) by selecting two columns in this file
for the abscissa and ordinate of a function. In a ``\texttt{.dat}'' data file, tabular
values are stored in plain-text form and are separated by a blank. In a comma-separated values
(``\texttt{.csv}'') file, values are separated by a comma ``\texttt{,}''.
For both formats, lines beginning with ``\texttt{\#}'' are considered as comments.

To use this function:

\texttt{interp1d(file.csv, i, j, x)}

with:
\begin{itemize}
\item file.csv, the name of the file
\item i, the column number for the abscissa
\item j, the column number for the ordinate
\item x, the number for which the linear interpolation is looked
\end{itemize}
The function return a linear interpolation for the x value
if it is in the range of the abscissa definition, or a
linear extrapolation if it is out of this range.

NB: the abscissa value must be in the growing order.

%==================================
\subsection{Modification of the variables at the end of a time step}
%==================================

The subroutine \texttt{cs\_user\_extra\_operations} is called at the end
of every time step. It is used to print of modify any variable at the end
of every time step.

Several examples are given in the directory \texttt{EXAMPLES}:
\begin{list}{-}{}
\item Calculation of a thermal balance at  the boundaries and in the
      domain (including the mass source terms)

\item Modification of the temperature in a given area starting from a
      given time

\item Extraction of a 1D profile (which is also possible with the GUI,
see \figurename~\ref{fig:gui_output_profiles})

\item Printing of a moment

\item Usage of utility
      subroutines in the case of a parallel calculation
      (calculation of a sum on the processors, of a maximum, ...)
\end{list}

{\em WARNING: As all the variables (solved variables, physical
properties, geometric parameters) can be modified in this subroutine, a
wrong use may distort totally the calculation.}

The thermal balance example is particularly interesting.
\begin{list}{-}{}
\item It can be easily adapted to another scalar (only three simple
      modifications to do, as indicated in the subroutine).
\item It shows how to make a sum on all the sub-domains in the framework
      of a parallel calculation (see the calls to the subroutines
      \texttt{par*}).
\item It shows the precautions to take before doing some operations in
      the framework of periodic or parallel calculations (in particular
      when we want to calculate the gradient of a variable or to have
      access to values at the neighbouring cells of a face).
\item Finally it must not be forgotten that the resolution with
      temperature (and not enthalpy) as a solved variable is questionable when the specific
      heat is not constant.
\end{list}

%==================================
\subsection{Non-standard management of the chronological record files}
%==================================
\label{sec:prg_ushist}

Both the interface and the subroutine \texttt{cs\_user\_parameters.f90} allow autonomous management of the
``automatic'' chronological record files:
position of the probes, printing frequency and related variables. The
results are written in a different file for each variable. These files
are written in text (readable by {\em xmgrace} or {\em gnuplot}}) or CSV format and contain the profiles corresponding to
every probe. This type of output format may not be well adapted if, for
instance, the number of probes is too high. The subroutine
\texttt{ushist}, called at each time step, allows then to personalise the output format of the
chronological record files. The example in the
directory works as follows:

\begin{list}{-}{}
\item Positioning of the probes (only at the first passage): the index
      \texttt{ii} varies between 1 and the number of probes. The coordinates
      \texttt{xx}, \texttt{yy} and \texttt{zz} of each probe are given.
      The subroutine \texttt{findpt}
      gives then the number \texttt{icapt(ii)\index{icapt}} of the cell center
      which is the closest to the defined probe.

\item Opening of the output files (only at the first pass): in the
      example, the program opens a different file for
      all the \texttt{nvar} variables. \texttt{ficush(j)} contains the name of the
      J\raisebox{1ex}{\small th} file and \texttt{impush(j)} its unit number
      (\texttt{impush} is initialised by default so that the user has at his
      disposal specific unit numbers and does not run the risk to overwrite an
      already open file).

\item Writing to the files: in the version given as example, the program
      writes the time step number, the physical time step (based on the
      standard time step in the case of a variable time step) and the
      value of the selected variable at the different probes.

\item Closing of the files (only at the last time step).

\end{list}

{\em WARNING: The use of {\em\texttt{ushist}} neither erases nor replaces the
parameters given in the interface or in {\em\texttt{cs\_user\_parameters.f90}}. Therefore, in
the case of the use of {\em\texttt{ushist}}, and to avoid the creation
of useless files, the user should set {\em \texttt{ncapt}=0} in the interface or
in {\em \texttt{cs\_user\_parameters.f90}} to deactivate the automatic production of
chronological records}.\\
In addition, {\em \texttt{ushist}} generates supplementary result
files.


%==================================
\section{Advanced modelling setup}
%==================================

%==================================
\subsection{Use of a specific physics}
%==================================
\label{sec:prg_usppmo}%
Specific physics such as dispersed phase, atmospheric flows, gas combustion,
pulverised fuel combustion, electrical model and compressible model can be
added by the user from the interface, or by using the subroutine \texttt{usppmo} of
the \texttt{cs\_user\_parameters} file (called only during the calculation initialisation).
With the interface, when a specific physics is activated in \figurename~\ref{fig:5_GUI},
additional items or headings may appear (see for instance Sections \ref{sec:Ini-lag}
and \ref{sec:Ini-coal}).

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=\textwidth]{gui_calculation_features}
\caption{Specific physics models selection}
\label{fig:5_GUI}
\end{center}
\end{figure}

When the interface is not used, \texttt{usppmo} is one of the three subroutines
which must be obligatory completed by the user in order to use a specific physics module
(only heavy fuel combustion is not available with the GUI).
At the moment, \CS allows to use two ``pulverised coal'' modules
(with Lagrangian coupling or not) and one ``pulverised heavy fuel'' module,
two ``gas combustion'' modules, two ``electrical'' modules,
a ``compressible'' module and an ``atmospheric'' module. To activate one of
these modules, the user must complete one (and only one) of the
indicators \texttt{ippmod(i.....)\index{ippmod}} in the subroutine
\texttt{usppmo}. By default, all the indicators \texttt{ippmod(i.....)} are
initialised at -1, which means that no specific physics is activated.

\begin{list}{$\bullet$}{}
       \item Diffusion flame in the framework of ``3 points'' rapid complete
             chemistry: indicator {\bf \tt ippmod(icod3p\index{icod3p})}
        \begin{list}{$\rightarrow$}{}
               \item \texttt{ippmod(icod3p)} = 0 adiabatic conditions
               \item \texttt{ippmod(icod3p)} = 1 permeatic conditions (enthalpy
                     transport)
               \item \texttt{ippmod(icod3p)} =-1 module not activated
         \end{list}
        \item Eddy Break Up pre-mixed flame: indicator {\bf \tt
             ippmod(icoebu\index{icoebu})}
         \begin{list}{$\rightarrow$}{}
                \item \texttt{ippmod(icoebu\index{icoebu})} = 0 adiabatic
                      conditions at constant richness
                \item \texttt{ippmod(icoebu)} = 1 permeatic conditions at
                      constant richness
                \item \texttt{ippmod(icoebu)} = 2 adiabatic conditions at
                      variable richness
                \item \texttt{ippmod(icoebu)} = 3 permeatic conditions at
                      variable richness
                \item \texttt{ippmod(icoebu)} =-1 module not activated
         \end{list}
        \item Libby-Williams pre-mixed flame: indicator {\bf \tt ippmod(icolwc\index{icolwc})}
         \begin{list}{$\rightarrow$}{}
               \item \texttt{ippmod(icolwc)}=0 two peak model with adiabiatic conditions.
               \item \texttt{ippmod(icolwc)}=1 two peak model with permeatic conditions.
               \item \texttt{ippmod(icolwc)}=2 three peak model with adiabiatic conditions.
               \item \texttt{ippmod(icolwc)}=3 three peak model with permeatic conditions.
               \item \texttt{ippmod(icolwc)}=4 four peak model with adiabiatic conditions.
               \item \texttt{ippmod(icolwc)}=5 four peak model with permeatic conditions.
               \item \texttt{ippmod(icolwc)}=-1 module not activated.
          \end{list}
        \item Multi-coals and multi-classes pulverised coal combustion:
              indicator {\bf \tt ippmod(iccoal\index{iccoal})}
              The number of different coals must be less than or equal to
              \texttt{ncharm\index{ncharm}} = 3. The number of particle size
             classes \texttt{nclpch\index{nclpch}(icha)} for the coal
             \texttt{icha}, must
             be less than or equal to \texttt{ncpcmx\index{ncpcmx}} = 10.
         \begin{list}{$\rightarrow$}{}
                \item \texttt{ippmod(iccoal)} = 0 imbalance between the
                      temperature of the continuous and the solid phases
                \item \texttt{ippmod(iccoal)} = 1 otherwise
                \item \texttt{ippmod(iccoal)} =-1 module not activated
         \end{list}

        \item Multi-classes pulverised heavy fuel combustion:
              indicator {\bf \tt ippmod(icfuel\index{icfuel})}
         \begin{list}{$\rightarrow$}{}
                \item \texttt{ippmod(icfuel)} = 0 module activated
                \item \texttt{ippmod(icfuel)} =-1 module not activated
         \end{list}

        \item Lagrangian modelling of multi-coals and
             multi-classes pulverised coal combustion:
                 indicator {\bf \tt ippmod(icpl3c\index{icpl3c})}
              The number of different coals must be less than or equal to
              \texttt{ncharm\index{ncharm}} = 3. The number of particle size
             classes \texttt{nclpch\index{nclpch}(icha)} for the coal
             \texttt{icha}, must be less than or equal to
             \texttt{ncpcmx\index{ncpcmx}} = 10.
         \begin{list}{$\rightarrow$}{}
                \item \texttt{ippmod(icpl3c)} = 1 coupling with the Lagrangian
                      module, with transport of $H_2$
                \item \texttt{ippmod(icpl3c)} =-1 module not activated
         \end{list}
       \item Electric arcs module (Joule effect and Laplace forces):
             indicator {\bf \tt ippmod(ielarc\index{ielarc})}
        \begin{list}{$\rightarrow$}{}
               \item \texttt{ippmod(ielarc)} = 1 determination of the magnetic field by
                     means of the Ampere's theorem (not available)
               \item \texttt{ippmod(ielarc)} = 2 determination of the magnetic
                     field by means of the vector potential
               \item \texttt{ippmod(ielarc)} =-1 module not activated
         \end{list}
       \item Joule effect module (Laplace forces not taken into account):
             indicator {\bf \tt ippmod(ieljou\index{ieljou})}
        \begin{list}{$\rightarrow$}{}
               \item \texttt{ippmod(ieljou)} = 1 use of a real potential
               \item \texttt{ippmod(ieljou)} = 2 use of a complex potential
               \item \texttt{ippmod(ieljou)} = 3 use of real potential and specific boundary conditions for transformers.
               \item \texttt{ippmod(ieljou)} = 4 use of complex potential and specific boundary conditions for transformers.
               \item \texttt{ippmod(ieljou)} =-1 module not activated
         \end{list}
       \item Compressible module: indicator {\bf \tt
             ippmod(icompf\index{icompf})}
        \begin{list}{$\rightarrow$}{}
               \item \texttt{ippmod(icompf)} = 0 module activated
               \item \texttt{ippmod(icompf)} =-1 module not activated
         \end{list}
       \item Atmospheric flow module: indicator {\bf \tt
             ippmod(iatmos\index{icompf})}
        \begin{list}{$\rightarrow$}{}
               \item \texttt{ippmod(iatmos)} =-1 module not activated
               \item \texttt{ippmod(iatmos)} = 0 standard modelling
               \item \texttt{ippmod(iatmos)} = 1 dry atmosphere
               \item \texttt{ippmod(iatmos)} = 2 humid atmosphere
         \end{list}
%       \item cooling towers module: indicator {\bf \tt
%             ippmod(iaeros\index{icompf})}
%        \begin{list}{$\rightarrow$}{}
%               \item \texttt{ippmod(iaeros\index)} =-1 module not activated
%               \item \texttt{ippmod(iaeros\index)} = 0 no model (NOT functional)
%               \item \texttt{ippmod(iaeros\index)} = 1 Poppe's model
%               \item \texttt{ippmod(iaeros\index)} = 2 Merkel's model
%         \end{list}
\end{list}

{\em WARNING: Only one specific physics module can be activated at the
same time.}

In the framework of the gas combustion modelling, the user may impose
his own enthalpy-temperature tabulation (conversion law). He needs then
to give the
value zero to the indicator \texttt{indjon\index{indjon}} (the default value
being 1). For more details, the user may refer to the following note
(thermochemical files).

\minititre{Note: the thermo-chemical files}
The user must not forget to place in the directory DATA the
thermochemical file \texttt{dp\_FCP.xml}, \texttt{dp\_C3P}, \texttt{dp\_C3PSJ} or
\texttt{dp\_ELE} (depending on the specific physics module he activated)
Some example files are placed in the directory \texttt{DATA/REFERENCE} at the creation of the
study case. Their content is described below.

\begin{list}{$\bullet$}{}
       \item Example of file for the gas combustion:
        \begin{list}{$\rightarrow$}{}
               \item if the enthalpy-temperature conversion data base
                     JANAF is used: \texttt{dp\_C3P} (see
                     array \ref{tab:dpC3P}).

\begin{table}[htbp]
\begin{center}
\small{
\begin{tabular}{|c|c|c|c|} \hline
 Lines  &Examples of values &        Variables             & Observations                                     \\ \hline
  1     &         5         &  \texttt{ngaze\index{ngaze}} & Number of current species                        \\ \hline
  2     &        10         &   \texttt{npo\index{npo}}    & Number of points for the                         \\
        &                   &                              & enthalpy-temperature table                       \\ \hline
  3     &       300.        &  \texttt{tmin\index{tmin}}   & Lower temperature limit                          \\
        &                   &                              & for the table                                    \\ \hline
  4     &      3000.        &  \texttt{tmax\index{tmax}}   & Upper temperature limi   t                       \\
        &                   &                              & for the tabulation                               \\ \hline
  5     &                   &                              & Empty line                                       \\ \hline
  6     & CH4 O2 CO2 H2O N2 &  \texttt{nomcoe\index{nomcoe}}(\texttt{ngaze}) & List of the current species                      \\ \hline
  7     &.35 .35 .35 .35 .35&  \texttt{kabse\index{kabse}}(\texttt{ngaze})   & Absorption coefficient                           \\
        &                   &                              & of  the current species                          \\ \hline
  8     &         4         &  \texttt{nato\index{nato}}   & Number of elemental species                      \\ \hline
  9     &.012  1  0  1  0  0& \texttt{wmolat\index{wmolat}}(\texttt{nato}),  & Molar mass of the elemental                      \\
 10     &.001  4  0  0  2  0&                              & species (first column)                           \\
 11     &.016  0  2  2  1  0&\texttt{atgaze\index{atgaze}}(\texttt{ngaze},\texttt{nato})& Composition of the current species             \\
 12     &.014  0  0  0  0  2&                              & as a function of the elemental species           \\
        &                   &                              & (\texttt{ngaze} following columns)                        \\ \hline
 13     &         3         &  \texttt{ngazg\index{ngazg}} & Number of global species                         \\
        &                   &                              & Here, \texttt{ngazg} = 3 (Fuel, Oxidiser and Products)    \\ \hline
 14     &  1. 0. 0. 0. 0.   &                              & Composition of the global species as a           \\
 15     &  0. 1. 0. 0. 3.76 &\texttt{compog\index{compog}}(\texttt{ngaze},\texttt{ngazg})& function of the current species of line 6 \\
 16     &  0. 0. 1. 2. 7.52 &                              & In the order: Fuel (line 15),                    \\
        &                   &                              & Oxidiser (line 16) and Product (line 17)         \\ \hline
 17     &         1         &  \texttt{nrgaz\index{nrgaz}} & Number of global reactions                       \\
        &                   &                              & Here \texttt{nrgaz} = 1 (always equal to 1                \\
        &                   &                              & in this version)                                 \\ \hline
 18     &                   & \texttt{igfuel\index{igfuel}}(\texttt{nrgaz}), & Numbers of the global species concerned by       \\
        & 1 2 -1 -9.52 10.52&  \texttt{igoxy\index{igoxy}}(\texttt{nrgaz}),  & the stoichiometric ratio                         \\
        &                   &                              & (first 2 integers)                               \\
        &                   &\texttt{stoeg\index{stoeg}}(\texttt{ngazg},\texttt{nrgaz})& Stoichiometry in global species reaction.       \\
        &                   &                               & Negative for the reactants (here                \\
        &                   &                               & ``Fuel'' and ``Oxidiser'') and positive for      \\
        &                   &                               & the products (here ``Products'')                \\ \hline
\end{tabular}
}
\caption{Example of file for the gas combustion when JANAF is used: \texttt{dp\_C3P}}\label{tab:dpC3P}
\end{center}
\end{table}

               \item if the user provides his own enthalpy-temperature tabulation
                     (there must be three chemical species and only
                     one reaction): \texttt{dp\_C3PSJ} (see
                     array \ref{tab:dpC3PSJ}). This file replaces \texttt{dp\_C3P}.

\begin{table}[htbp]
\begin{center}
\small{
\begin{tabular}{|c|c|c|c|} \hline
 Lines  &            Examples of values     &        Variables            & Observations                                \\ \hline
   1    &                    6              &   \texttt{npo}              & Number of tabulation points                 \\ \hline
   2    &  50. -0.32E+07 -0.22E+06 -0.13E+08&                             &                                             \\
   3    & 250. -0.68E+06 -0.44E+05 -0.13E+08&\texttt{th\index{th}}(\texttt{npo}),           & Temperature(first column),                  \\
   4    & 450.  0.21E+07  0.14E+06 -0.13E+08& \texttt{ehgazg\index{ehgazg}}(1,\texttt{npo}),& mass enthalpies of fuel, oxidiser             \\
   5    & 650.  0.50E+07  0.33E+06 -0.12E+08& \texttt{ehgazg}(2,\texttt{npo}),              & and products (columns 2,3 and 4)            \\
   6    & 850.  0.80E+07  0.54E+06 -0.12E+08& \texttt{ehgazg}(3,\texttt{npo})               & from line 2 to line \texttt{npo}+1                   \\
   7    &1050.  0.11E+08  0.76E+06 -0.11E+08&                             &                                             \\ \hline
   8    & .00219       .1387        .159    &\texttt{wmolg(1)\index{wmolg}},       & Molar masses of fuel,                         \\
        &                                   &                    \texttt{wmolg(2)},& oxidiser                                    \\
        &                                   &                    \texttt{wmolg(3)} & and products                                \\ \hline
   9    &                .11111             &          \texttt{fs(1)\index{fs(1)}} & Mixing rate at the stoichiometry            \\
        &                                   &                             & (relating to Fuel and Oxidiser)             \\ \hline
  10    &    0.4      0.5       0.87        &\texttt{ckabsg\index{ckabsg}(1)},     & Absorption coefficients of the fuel,             \\
        &                                   &                  \texttt{ckabsg(2)}, & oxidiser                                    \\
        &                                   &                  \texttt{ckabsg(3)}  & and products                                \\ \hline
  11    &    1.       2.                    & \texttt{xco2\index{xco2}},   \texttt{xh2o\index{xh2o}}& Molar coefficients of $CO_2$         \\
        &                                   &                             & and $H_2O$ in the products                  \\
        &                                   &                             & (using Modak radiation)                     \\ \hline
\end{tabular}
}
\caption{Example of file for the gas combustion when the user provides
 his own enthalpy-temperature table
                     (there must be three species and only one
                     reaction): \texttt{dp\_C3PSJ} (this file replaces
 \texttt{dp\_C3P})}\label{tab:dpC3PSJ}
\end{center}
\end{table}
        \end{list}

       \item Example of file for the pulverised coal combustion:
             \texttt{dp\_FCP.xml} (see the example in the directory \texttt{DATA/REFERENCE},
             this file can be filled in thanks to the GUI).

%\begin{table}[htbp]
%\begin{center}
%\tiny{
%\begin{tabular}{|c|c|c|c|} \hline
% Lines  &      Examples of values        &        Variables              & Observations                      %                   \\ \hline
%   1    &          THERMOCHEMISTRY       &                               & Comment line                      %                   \\ \hline
%   2    &               8                & \texttt{ncoel\index{ncoel}}   & Number of current species         %                   \\ \hline
%   3    &               8                &  \texttt{npo\index{npo}}      & Number of points for the          %                   \\
%        &                                &                               & enthalpy-temperature tabulation   %                   \\ \hline
%   4    &        CURRENT SPECIES         &                               & Comment line                      %                   \\ \hline
%   5    & CH4 C2H4 CO O2 CO2 H2O N2 C(S) & \texttt{nomcoel\index{nomcoel}}(\texttt{ncoel}) & List of the     %                                     \\
%        &                                &                               & current species                   %                   \\ \hline
%   6    &               300.             &  \texttt{tmin\index{tmin}}    & Temperature inferior limit (Kelvin)                  \\
%        &                                &                               & for the enthalpy-temperature tabulation              \\ \hline
%   7    &              2400.             &  \texttt{tmax\index{tmax}}    & Temperature superior limit (Kelvin)                  \\
%        &                                &                               & for the enthalpy-temperature tabulation              \\ \hline
%   8    &               4                &  \texttt{nato\index{nato}}    & Number of elementary species                          \\ \hline
%   9    &  .012  1  2  1  0  1  0  0  1  &                               & Molar mass of the elemental species                  \\
%  10    &  .001  4  4  0  0  0  2  0  0  &\texttt{wmolat\index{wmolat}}(\texttt{nato}),    & (first column)                                       \\
%  11    &  .016  0  0  1  2  2  1  0  0  &\texttt{atcoel\index{atcoel}}(\texttt{ncoel,nato})& and composition of the current species              \\
%  12    &  .014  0  0  0  0  0  0  2  0  &                               & as a function of the elemental species               \\ \hline
%  13    &          RADIATION             &                               & Comment line                                         \\ \hline
%  14    &               0.1              & \texttt{ckabs1\index{ckabs1}}  & Constant absorption coefficient                      \\
%        &                                &                               & for the gas mixture                                  \\ \hline
%  15    &   COAL CHARACTERISTICS         &                               & Comment line                                         \\ \hline
%  16    &               2                & \texttt{ncharb\index{ncharb}} & Number of coal types                                 \\ \hline
%  17    &         1            1         & \texttt{nclpch\index{nclpch}}(\texttt{ncharb}) & Number of classes for each coal                      \\
%        &                                &                               & (each column corresponding to                        \\
%        &                                &                               & one coal type )                                      \\ \hline
%  18    &    50.E-6        50.E-6        & \texttt{diam20\index{diam20}}(\texttt{nclacp})  & Initial diameter of each class (m)                   \\
%        &                                &                               & \texttt{nclacp\index{nclacp}} is the total number of classes. \\
%        &                                &                               & All the diameters are written on the same line       \\
%        &                                &                               & (sucessively for each coal, we give the              \\
%        &                                &                               & diameter corresponding to each class)                \\ \hline
%  19    &    74.8          60.5          & \texttt{cch\index{cch}}(\texttt{ncharb})& Composition in C (mass.-\%, dry) of each coal        \\ \hline
%  20    &     5.1           4.14         & \texttt{hch\index{hch}}(\texttt{ncharb})& Composition in H (mass.-\%, dry) of each coal        \\ \hline
%  21    &    12.01          5.55         & \texttt{och\index{och}}(\texttt{ncharb})& Composition in O (mass.-\%, dry) of each coal        \\ \hline
%  22    & 0  31524000.    0  31524000.   & \texttt{ipci\index{ipci}}(\texttt{ncharb})      & Value of the PCI ($Jkg^{-1}$) for each coal,         \\
%        &                                & \texttt{pcich\index{pcich}}(\texttt{ncharb})    & the first integer indicating if this value refers    \\
%        &                                &                               & to pure (0) or dry coal (1)                          \\ \hline
%  23    &   1800.      1800.             & \texttt{cp2ch\index{cp2ch}}(\texttt{ncharb})& Heat-storage capacity at constant pressure        \\
%        &                                &                               & ($Jkg^{-1}K^{-1}$) for each coal                     \\ \hline
%  24    &   1200.      1200.             & \texttt{rho0ch\index{rho0ch}}(\texttt{ncharb})   & Initial density ($kgm^{-3}$) of each                 \\ \hline
%  25    &          Coke                  &                               & Comment line                                         \\ \hline
%  26    &      0.         0.             & \texttt{cck\index{cck}}(\texttt{ncharb}) & Composition in C (mass.-\%, dry) of the coke         \\
%        &                                &                               & for each coal                                        \\ \hline
%  27    &      0.         0.             & \texttt{hck\index{hck}}(\texttt{ncharb}) & Composition in H (mass.-\%, dry) of the coke         \\
%        &                                &                               & for each coal                                        \\ \hline
%  28    &      0.         0.             & \texttt{ock\index{ock}}(\texttt{ncharb}) & Composition in O (mass.\%, dry) of the coke          \\
%        &                                &                               & for each coal                                        \\ \hline
%  29    &      0.         0.             & \texttt{pcick\index{pcick}}(\texttt{ncharb})& PCI of the dry coke ($Jkg^{-1}$) for each coal       \\ \hline
%  30    &          Ashes               &                               & Comment line                                         \\ \hline
%  31    &      6.3        6.3            & \texttt{xashch\index{xashch}}(\texttt{ncharb})& Ash mass fraction (mass.-\%, dry) in each coal       \\ \hline
%  32    &      0.         0.             & \texttt{h0ashc\index{h0ashc}}(\texttt{ncharb})  & Ash formation enthalpy ($Jkg^{-1}$)                  \\
%        &                                &                               & for each coal                                        \\ \hline
%  33    &      0.         0.             & \texttt{cpashc\index{cpashc}}(\texttt{ncharb})   & CP of the ashes ($Jkg^{-1}K^{-1}$) for each coal     \\ \hline
%  34    &      0.         0.             & \texttt{xwatch\index{cpashc}}(\texttt{ncharb})   & humidity rate of the ashes (mass.-\%) for each coal     \\ \hline
%  35    &  Devolatilisation (Kobayashi)  &                             & Comment line                                         \\ \hline
%  36    &  1  0.37      0  0.37          & \texttt{iy1ch\index{iy1ch}}(\texttt{ncharb}),   & For each coal, pairs (\texttt{iy1ch}, \texttt{y1ch}).                  \\
%        &                                & \texttt{y1ch\index{y1ch}}(\texttt{ncharb})     & The real \texttt{y1ch} is the dimensionless stoich. coefficient\\
%        &                                &                               & If the integer \texttt{iy1ch} is worth 1,                     \\
%        &                                &                               & the provided value of \texttt{y1ch} is adopted and            \\
%        &                                &                               & the composition of the light volatile matters        \\
%        &                                &                               & is calculated automatically.                         \\
%        &                                &                               & If the integer \texttt{iy1ch} is worth 0,                     \\
%        &                                &                               & the provided value of \texttt{y1ch} is ignored:               \\
%        &                                &                               & \texttt{y1ch} is calculated automatically (the light          \\
%        &                                &                               & volatiles are then composed of {$CH_{4}$}, {$CO$}).  \\ \hline
%  37    &  1  0.74      1  0.74          & \texttt{iy2ch\index{iy2ch}}(\texttt{ncharb}),   & For each coal, pairs (\texttt{iy2ch}, \texttt{y2ch}).                  \\
%        &                                & \texttt{y2ch\index{y2ch}}(\texttt{ncharb})     & The real \texttt{y2ch} is the adimensional stoich. coefficient\\
%        &                                &                               & If the integer \texttt{iy2ch} is worth 1,                     \\
%        &                                &                               & the provided value of \texttt{y2ch} is adopted and            \\
%        &                                &                               & the composition of the heavy volatile matters        \\
%        &                                &                               & is calculated automatically.                         \\
%        &                                &                               & If the integer \texttt{iy2ch} is worth 0,                     \\
%        &                                &                               & the provided value of \texttt{y2ch} is ignored:               \\
%        &                                &                               & \texttt{y2ch} is calculated automatically (the heavy          \\
%        &                                &                               & volatiles are then composed of {$C_{2}H_{4}$}, {$CO$}).\\ \hline
%  38    &  370000.      410000.          & \texttt{a1ch\index{a1ch}}(\texttt{ncharb})& Devolatilisation pre-exponential factor A1 ($s^{-1}$)\\
%        &                                &                               & for each coal (light volatile matters)               \\ \hline
%  39    &  1.3E13       1.52E13          & \texttt{a2ch\index{a2ch}}(\texttt{ncharb})& Devolatilisation pre-exponential factor A2 ($s^{-1}$)\\
%        &                                &                               & for each coal (heavy volatile matters)               \\ \hline
%  40    &   74000.       80000.          & \texttt{e1ch\index{e1ch}}(\texttt{ncharb})& Devolatilisation activation energy E1 ($Jmol^{-1}$)  \\
%        &                                &                               & for each coal (light volatile matters)               \\ \hline
%  41    &  250000.      310000.          & \texttt{e2ch\index{e2ch}}(\texttt{ncharb})& Activation energy E2 ($Jmol^{-1}$) of devolatilisation\\
%        &                                &                               & for each coal (heavy volatile matters)               \\ \hline
%  42    &  heterogeneous combustion $O_2$   &                               & Comment lign                                 \\ \hline
%  43    &      17.88        17.88        & \texttt{ahetch\index{ahetch}}(\texttt{ncharb})   & Char burnout pre-exponential constant                \\
%        &                                &                               & ($kgm^{-2}s^{-1}atm^{-1}$) for each coal             \\ \hline
%  44    &      16.55        16.55        & \texttt{ehetch\index{ehetch}}(\texttt{ncharb})   & Char burnout activation energy ($kcalmol^{-1}$)       \\
%        &                                &                               & for each coal                                        \\ \hline
%  45    &       1            1           & \texttt{iochet\index{iochet}}(\texttt{ncharb})   & Char burnout reaction order for each coal            \\
%        &                                &                               & 0.5 if \texttt{iochet} = 0 and 1 if \texttt{iochet} = 1                \\ \hline
%  46    &  heterogeneous combustion $CO_2$  &                               & Comment lign                               \\ \hline
%  47    &      1.788        1.788        & \texttt{ahetch\index{ahetch}}(\texttt{ncharb})   & Char burnout pre-exponential constant                \\
%        &                                &                               & ($kgm^{-2}s^{-1}atm^{-1}$) for each coal             \\ \hline
%  48    &      1.655        1.655        & \texttt{ehetch\index{ehetch}}(\texttt{ncharb})   & Char burnout activation energy ($kcalmol^{-1}$)       \\
%        &                                &                               & for each coal                                        \\ \hline
%  49    &       1            1           & \texttt{iochet\index{iochet}}(\texttt{ncharb})   & Char burnout reaction order for each coal            \\
%        &                                &                               & 0.5 if \texttt{iochet} = 0 and 1 if \texttt{iochet} = 1                \\ \hline
%  50    &    OXYDIZERS CHARACTERISTICS   &                               & Comment lign                               \\ \hline
%  51    &              3                 & \texttt{noxyd\index{noxyd}}    & Number of oxydizers  \\
%        &                                &                               & (mixtures of $O_2,N_2,H_2O,CO_2$) \\ \hline
%  52    &    1.        0.        1.      & \texttt{oxyo2\index{oxyo2}}(\texttt{noxyd})   & Composition in $O_2$ of each oxydizer ($moles$) \\
%  53    &    0.        0.        1.      & \texttt{oxyn2\index{oxyn2}}(\texttt{noxyd})   & Composition in $N_2$ of each oxydizer ($moles$) \\
%  54    &    0.        0.        1.      & \texttt{oxyh2o\index{oxyh2o}}(\texttt{noxyd}) & Composition in $H_2O$ of each oxydizer ($moles$) \\
%  55    &  2.39        1.        1.      & \texttt{oxyco2\index{oxyco2}}(\texttt{noxyd}) & Composition in $CO_2$ of each oxydizer ($moles$) \\ \hline
%\end{tabular}
%}
%\caption{Example of file for the pulverised coal combustion:
% \texttt{dp\_FCP}}\label{tab_dpFCP}
%\end{center}
%\end{table}

       \item Example of file for the heavy fuel combustion: \texttt{DP\_FUE\_new} (see
             the example in \texttt{DATA/REFERENCE}).

       \item Example of file for the electric arcs: \texttt{dp\_ELE} (see
             array \ref{tab:dpELE}).

\begin{table}[htbp]
\begin{center}
\small{
\begin{tabular}{|c|l|c|c|} \hline
 Lines  &        Examples of values        & Variables & Observations                                       \\ \hline
  1     &\# Free format ASCII file ...     &           & Free comment                                       \\ \hline
  2     &\# Comment lines ...              &           & Free comment                                       \\ \hline
  3     &\#                            ... &           & Free comment                                       \\ \hline
  4     &\# Argon propoerties ...          &           & Free comment                                       \\ \hline
  5     &\#                            ... &           & Free comment                                       \\ \hline
  6     &\# No of NGAZG and No   ... &           & Free comment                                       \\ \hline
  7     &\# NGAZG NPO                  ... &           & Free comment                                       \\ \hline
  8     &    1   238         &    \texttt{ngazg\index{ngazg}}   & Number of species                         \\
        &                    &    \texttt{npo\index{npo}}       & Number of given temperature points for    \\
        &                    &                         & the tabulated physical properties                  \\
        &                    &                         & (\texttt{npo} $\leqslant$ \texttt{npot} set in \texttt{ppthch})             \\
        &                    &                         & So there will be \texttt{ngazg} blocks of \texttt{npo} lines each    \\ \hline
  9     &\#                            ... &           & Free comment                                       \\ \hline
 14     &        0           & \texttt{ixkabe\index{ixkabe}} & Radiation options for \texttt{xkabe\index{xkabe}}   \\ \hline
 15     &\#                            ... &           & Free comment                                       \\ \hline
 16     &\#  Propreties                ... &           & Free comment                                       \\ \hline
 17     &\#  ~~~~T~~~~~~~~~~~H         ... &           & Free comment                                       \\ \hline
 18     &\#  Temperature  Enthalpy    ... &           & Free comment                                       \\ \hline
 19     &\#                            ... &           & Free comment                                       \\ \hline
 20     &\#  ~~~~~K~~~~~~~~~J/kg       ... &           & Free comment                                       \\ \hline
 21     &\#                            ... &           & Free comment                                       \\ \hline
 22     &    ~~~300.~~~~~~14000.       ... &           & In line tabulation of the physical properties      \\
        &                                  &           & as a function of the temperature in Kelvin         \\
        &                                  &           & for each of the \texttt{ngazg} species             \\
        &                    &    \texttt{h}                    & Enthalpy in J/kg                                   \\
        &                    &    \texttt{roel}                 & Density in kg/m3                                   \\
        &                    &    \texttt{cpel}                 & Specific heat in J/(kg K)                          \\
        &                    &    \texttt{sigel}                & Electric conductivity in Ohm/m                     \\
        &                    &    \texttt{visel}                & Dynamic viscosity in kg/(m s)                      \\
        &                    &    \texttt{xlabel}               & Thermal conductivity in W/(m K)                    \\
        &                    &    \texttt{xkabel\index{xkabel}} & Absorption coefficient (radiation)                 \\   \hline
\end{tabular}
}
\caption{Example of file for the electric arcs module:
 \texttt{dp\_ELE}}\label{tab:dpELE}
\end{center}
\end{table}

\end{list}

\clearpage


%==================================
\subsection[Pulverised coal and gas combustion module]
{Pulverised coal and gas combustion module (needs update)}
%==================================
%==================================
\subsubsubsection{Initialisation of the variables}\label{sec:Ini-coal}
%==================================
For coal combustion, it is possible to initialise the specific variables in the Graphical User Interface (GUI) or in the subroutine
\texttt{cs\_user\_initialization}. In the GUI, when a coal combustion physics is selected in the item ``Calculation features'' under the heading
``Thermophysical models'', an additional item appears: ``Pulverized coal combustion''. In this item the user can define coal types, their composition, the oxidant and reactions parameters, see \figurename~\ref{fig:Ini-coal1} to \figurename~\ref{fig:Ini-coal5}.

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=12cm]{gui_coal_classes}
\caption{Thermophysical models - Pulverized coal combustion, coal classes}
\label{fig:Ini-coal1}
\end{center}
\end{figure}

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=11cm]{gui_coal_coke}
\caption{Pulverized coal combustion, coke}
\label{fig:Ini-coal2}
\end{center}
\end{figure}

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=11cm]{gui_coal_composition}
\caption{Pulverized coal combustion, coal composition}
\label{fig:Ini-coal3}
\end{center}
\end{figure}

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=11cm]{gui_coal_reaction}
\caption{Pulverized coal combustion, reaction parameters}
\label{fig:Ini-coal4}
\end{center}
\end{figure}

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=11cm]{gui_coal_oxydant}
\caption{Pulverized coal combustion, oxydant}
\label{fig:Ini-coal5}
\end{center}
\end{figure}

If the user deals with gas combustion or if he (or she) does not want to use the
GUI for coal combustion, the subroutine \texttt{cs\_user\_initialization} must be used (only during the calculation initialisation).\\
In this section, ``specific physics'' will refer to gas combustion or
to pulverised coal combustion.

These subroutines allow the user to initialise some variables specific
to the specific physics activated {\em via} \texttt{usppmo}. As usual,
the user may have access to several geometric variables to discriminate
between different initialisation zones if needed.

It should be recalled again that the user can access the array of values of the
variables as described in the \doxygenfile{field.html}{the \texttt{doxygen}
documentation dedicated to the fields management}. In the following description,
only variables indices \texttt{ivar} are given, but field indices can be
retrieved easily by using \texttt{ivarfl(ivar)}.

{\em WARNING: in the case of a specific physics modelling, all the
variables will be initialised here, even the potential user scalars: {\em
\texttt{cs\_user\_initialization}} is no longer used.}


\begin{list}{$\bullet$}{}
       \item in the case of the EBU pre-mixed flame module, the user can
             initialise in every cell \texttt{iel}: the mixing rate
             \texttt{isca(ifm)} in variable richness, the
             fresh gas mass fraction \\
             \texttt{isca(iygfm\index{iygfm})}
             and the mixture enthalpy \texttt{isca(iscalt)} in
             permeatic conditions

        \item in the case of the rapid complete chemistry diffusion flame
             module, the user can initialise in every cell \texttt{iel}: the
             mixing rate \texttt{isca(ifm\index{ifm})}, its variance
             \texttt{isca(ifp2m\index{ifp2m})} and the mixture mass
             enthalpy \texttt{isca(iscalt)} in permeatic conditions

        \item in the case of the pulverised coal combustion module, the
             user can initialise in every cell \texttt{iel}:
              \begin{list}{$\rightarrow$}{}
                     \item the transport variables related to the solid phase
                           \begin{list}{}{}
                                  \item \texttt{isca(ixch\index{ixch}(icla))}
                                  the reactive coal mass fraction related to the
                                  class \texttt{icla} (\texttt{icla} from 1 to
                                  \texttt{nclacp} which is the total number of
                                  classes, {\em i.e.} for all the coal type)
                                  \item \texttt{isca(ixck(\index{ixck}icla))}
                                  the coke mass fraction related to the class
                                  \texttt{icla}
                                  \item \texttt{isca(inp\index{inp}(icla))} the
                                  number of particles related to class
                                  \texttt{icla} per kg of air-coal mixture
                                  \item \texttt{isca(ih2\index{ih2}(icla))} the
                                   mass enthalpy related to the class
                                   \texttt{icla} in permeatic conditions
                           \end{list}
                     \item \texttt{isca(iscalt)} the mixture enthalpy
                     \item the transport variables related to the gas phase
                           \begin{list}{}{}
                                  \item
                                       \texttt{isca(if1m\index{if1m}(icha))} the
                                       mean value of the tracer 1 representing
                                       the light volatile matters released by
                                       the coal \texttt{icha}
                                  \item
                                       \texttt{isca(if2m\index{if2m}(icha))} the
                                        mean value of the tracer 2 representing
                                        the heavy volatile matters released by
                                        the coal \texttt{icha}
                                  \item \texttt{isca(if3m\index{if3m})}
                                        the mean value of the tracer 3
                                        representing the carbon released
                                        as CO during coke burnout
                                  \item \texttt{isca(if4p2m\index{if4p2m})} the
                                  variance associated with the tracer 4
                                  representing the air (the mean value of this
                                  tracer is not transported, it can be deduced
                                  directly from the three others)
                                  \item \texttt{isca(ifp3m\index{ifp3m})} the
                                  variance associated with the tracer 3
                           \end{list}
              \end{list}
\end{list}

%==================================
\subsubsection{Boundary conditions}\label{sec:coal-cl}
%==================================
In this section, ``specific physics'' refers to gas combustion or
to pulverised coal combustion.\\
For coal combustion, it is possible to manage the boundary conditions in the Graphical User Interface (GUI). When the coal combustion physics is selected in the heading ``Thermophysical models'', specific boundary conditions are activated for inlets, see \figurename~\ref{fig:cond_lim-coal}. The user fills for each type of coal previously defined (see \S~\ref{sec:Ini-coal}) the initial temperature and initial composition of the inlet flow, as well as the mass flow rate.

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=13cm]{gui_coal_bc}
\caption{Boundary conditions for the combustion of coal}
\label{fig:cond_lim-coal}
\end{center}
\end{figure}

For gas combustion or if the GUI is not used for coal combustion, the use of
\texttt{cs\_user\_boundary\_conditions} (called at every time step) is as
mandatory as \texttt{cs\_user\_parameters.f90} and \texttt{usppmo} to run a calculation involving specific physics. The way of using them is the same as using
 in the framework of standard calculations, that is, run several loops on the boundary faces lists (cf. \S\ref{sec:fvm_selector})
marked out by their colors, groups, or  geometrical criterion, where
the type of face, the type of boundary condition for each variable and
eventually the value of each variable are defined.

{\em WARNING: In the case of a specific physics modelling, all the
boundary conditions for every variable must be defined here, even for
the eventual user scalars: {\em \texttt{cs\_user\_boundary\_conditions}} is not used at all.}\\

In the case of a specific physics modelling, a zone number \texttt{izone}
\footnote{\texttt{izone} must be less than the maximum number of boundary
zone allowable by the code, \texttt{nozppm}. This is fixed at 2000 in
 \texttt{pppvar};not to be modified} (for
instance the color \texttt{icoul}) is associated with every boundary face, in
order to gather together all the boundary faces of the same type. In
comparison to \texttt{cs\_user\_boundary\_conditions}, the main change from the user point of
view concerns the faces whose boundary conditions belong to the type
\texttt{itypfb=ientre\index{ientre}}:

\begin{list}{$\bullet$}{}
       \item for the EBU pre-mixed flame module:
             \begin{list}{$\rightarrow$}{}
                    \item the user can choose between the ``burned gas
                          inlet'' type (marked out by the burned gas indicator
                          \texttt{ientgb\index{ientgb}(izone\index{izone})}=1) and the
                          ``fresh gas inlet'' type (marked out by
                          the fresh gas indicator
                          \texttt{ientgf\index{ientgf}(izone)}=1)
                    \item for each inlet type (fresh or burned
                          gas), a mass flow or a velocity must be imposed:

                          \begin{list}{-}{}
                                 \item to impose the mass flow,
                                     \begin{list}{-}{}
                                       \item the user gives to
                                             the indicator
                                             \texttt{iqimp\index{iqimp}(izone)}
                                             the value 1,
                                       \item  the
                                             mass flow value is set in
                                             \texttt{qimp\index{qimp}(izone)}
                                             (positive value, in $kgs^{-1}$)
                                       \item finally he imposes the
                                             velocity vector direction
                                             by giving the components of
                                             a direction vector in
                                             \texttt{rcodcl\index{rcodcl}(ifac,iu\index{iu})}, \texttt{rcodcl(ifac,iv\index{iv})} and \texttt{rcodcl(ifac,iw\index{iw})}
                                     \end{list}

{\em WARNING:
\begin{list}{-}{}
\item the variable \texttt{qimp(izone)} refers to the mass flow across the whole
      zone \texttt{izone} and not across a boundary face (specifically for the axi-symmetric calculations, the inlet surface of the mesh must be broken up)
\item the variable \texttt{qimp(izone)} deals with the inflow across the area \texttt{izoz} and only across this zone; it is recommended to pay attention to the boundary conditions.
\item the velocity direction vector is neither necessarily normed, nor
      necessarily incoming.
\end{list}}

                                 \item to impose a velocity, the user
                                       must give to the indicator
                                       \texttt{iqimp(izone)} the value 0 and set
                                       the three velocity components (in
                                       $m.s^{-1}$) in
                                       \texttt{rcodcl(ifac,iu)},
                                       \texttt{rcodcl(ifac,iv)} and
                                       \texttt{rcodcl(ifac,iw)}
                          \end{list}
                    \item finally he specifies for each gas inlet type
                          the mixing rate \texttt{fment\index{fment}(izone)} and
                          the temperature \texttt{tkent\index{tkent}(izone)} in Kelvin
             \end{list}

       \item for the ``3 points'' diffusion flame module:
             \begin{list}{$\rightarrow$}{}
                    \item the user can choose between the ``oxidiser
                          inlet'' type marked out by
                          \texttt{ientox\index{ientox}(izone)}=1 and the ``fuel
                          inlet'' type marked out by
                          \texttt{ientfu\index{ientfu}(izone)}=1
                    \item concerning the input mass flow or the input
                          velocity, the method is the same as for the
                          EBU pre-mixed flame module
                    \item finally, the user sets the temperatures
                          \texttt{tinoxy\index{tinoxy}} for each oxidiser inlet
                          and \texttt{tinfue\index{tinfue}}, for each fuel inlet

{\em Note: In the standard version, only the cases with only one
                          oxidising inlet type and one fuel inlet type
                          can be treated. In particular, there must be
                          only one input temperature for the oxidiser
                          (\texttt{tinoxy}) and one input temperature for the
                          fuel (\texttt{tinfuel}).}
             \end{list}

       \item for the pulverised coal module:
             \begin{list}{$\rightarrow$}{}
                    \item the inlet faces can belong to the ``primary
                          air and pulverised coal inlet'' type, marked
                          out by \texttt{ientcp\index{ientcp}(izone)}=1, or to
                          the ``secondary or tertiary air inlet'' type,
                          marked out by \texttt{ientat\index{ientat}(izone)}=1
                    \item in a way which is similar to the process
                          described in the framework of the EBU module,
                          the user chooses for every inlet face to
                          impose the mass flow or not (\texttt{iqimp(izone)}=1 or
                          0). If the mass flow is imposed, the user
                          must set the air mass flow value
                          \texttt{qimpat\index{qimpat}(izone)}, its direction in
                          \texttt{rcodcl(ifac,iu)}, \texttt{rcodcl(ifac,iv)}
                          and \\ \texttt{rcodcl(ifac,iw)} and if
                    \item incoming air temperature \texttt{timpat\index{timpat}(izone)} in
                          Kelvin. If the velocity is imposed, he must
                          set  \texttt{rcodcl(ifac,iu)}, \\
                          \texttt{rcodcl(ifac,iv)} and \texttt{rcodcl(ifac,iw)}.

                    \item if the inlet belongs to the ``primary air and
                          pluverised coal'' \texttt{type (ientcp(izone) = 1)} the
                          user must also define for each coal type \texttt{icha}:
                          the mass flow
                          \texttt{qimpcp\index{qimpcp}(izone,icha)}, the
                          granulometric distribution
                          \texttt{distch\index{distch}(izone,icha,iclapc)}
                          related to each class \texttt{iclacp}, and the
                          injection temperature
                          \texttt{timpcp\index{timpcp}(izone,icha)}

             \end{list}
\end{list}

%==================================
\subsubsection{Initialisation of the options of the variables}
%==================================
In the case of coal combustion, time averages, chronological records and listings follow-ups can be set in the Graphical User Interface (GUI) or in the subroutines \texttt{cs\_user\_combustion}. In the GUI, under the heading ``Calculation control'', additional variables appear in the list in the items ``Time averages'' and ``Profiles'', as well as in the item Volume solution control'', see \figurename~\ref{fig:t_average-coal} and \figurename~\ref{fig:V_control-coal}.

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=12cm]{gui_coal_time_average}
\caption{Calculation control - Time averages}
\label{fig:t_average-coal}
\end{center}
\end{figure}

\begin{figure}[!ht]
\begin{center}
\includegraphics[width=13cm]{gui_coal_solution_control}
\caption{Calculation control - Volume solution control}
\label{fig:V_control-coal}
\end{center}
\end{figure}

In this section, ``specific physics'' refers to gas combustion or
pulverised coal combustion.

For gas combustion or if the GUI is not used for coal combustion, the 3
subroutines \texttt{cs\_user\_combustion} can be used to complete
\texttt{cs\_user\_parameters.f90} for the considered specific physics. These
subroutines are called at the calculation start.
They allow to:
\begin{list}{$\bullet$}{}
\item activate, for the variables which are specific to the activated specific
      physics module, chronological records at the probes defined in
      \texttt{cs\_user\_parameters.f90} (indicators \texttt{ihisvr(ipp)}).\\
      Concerning the main variables (velocity, pressure, etc ...) the user
      must still complete \texttt{cs\_user\_parameters.f90} if he wants to get
      chronological records, printings in the listing or chronological
      outputs.
      The variables which can be activated by the user for each specific
      physics are listed below. The solved variables (of variable indices
      \texttt{ivar}) and the properties of indices \texttt{iprop} (defined at
      the cell \texttt{iel} by \texttt{propce(iel,ipproc(iprop))}) are listed
      below:
      \begin{list}{$\rightarrow$}{}
       \item EBU pre-mixed flame modelling:
       \begin{list}{-}{}
        \item Solved variables
              \begin{list}{\texttt{ivar} = }{}
               \item \texttt{isca(iygfm\index{iygfm})} fresh gas mass fraction
               \item \texttt{isca(ifm\index{ifm})} mixing rate
               \item \texttt{isca(ihm\index{ihm})} enthalpy, if transported
              \end{list}
        \item Properties \texttt{propce(iel,ipproc(iprop))}
              \begin{list}{\texttt{iprop} = }{}
               \item \texttt{itemp\index{itemp}} temperature
               \item \texttt{iym(1)\index{iym(1)}} fuel mass fraction
               \item \texttt{iym(2)\index{iym(2)}} oxidiser mass fraction
               \item \texttt{iym(3)\index{iym(3)}} product mass fraction
               \item \texttt{ickabs\index{ickabs}} absorption
                     coefficient, when the radiation modelling is
                     activated
               \item \texttt{it3m\index{it3m}} and \texttt{it4m\index{it4m}}
                     ``$T^3$'' and ``$T^4$'' terms, when the radiation
                     modelling is activated
              \end{list}
       \end{list}
       \item rapid complete chemistry diffusion flame modelling:
             \begin{list}{}{}
              \item  everything is identical to the ``EBU'' case, except
                     the fresh gas mass fraction which is replaced by the
                     variance of the mixing rate
                     \texttt{ivar=isca(ifp2m\index{ifp2m})}
             \end{list}
       \item pulverised coal modelling with 3 combustibles:
             \begin{list}{}{}
              \item {\em variables shared by the two phases}:
                    \begin{list}{-}{}
                     \item Solved variables
                           \begin{list}{\texttt{ivar} = }{}
                            \item \texttt{isca(ihm\index{ihm})}: gas-coal
                                  mixture enthalpy
                            \item \texttt{isca(immel\index{immel})}: molar mass
                                  of the gas mixture
                           \end{list}
                    \end{list}
              \item {\em variables specific to the dispersed phase}:
              \begin{list}{-}{}
               \item Solved variables
                     \begin{list}{\texttt{ivar} = }{}
                      \item \texttt{isca(ixck\index{ixck}(icla))}: coke mass
                            fraction related to the class \texttt{icla}
                      \item \texttt{isca(ixch\index{ixch}(icla))}: reactive coal
                            mass fraction related to the class \texttt{icla}
                      \item \texttt{isca(inp\index{inp}(icla))}: number of
                            particles of the class \texttt{icla} per kg of
                            air-coal mixture
                      \item \texttt{isca(ih2\index{ih2}(icla))}: mass enthalpy
                            of the coal of class \texttt{icla}, if we are in
                            permeatic conditions
                     \end{list}
               \item Properties \texttt{propce(iel,ipproc(iprop))}
                     \begin{list}{\texttt{iprop} = }{}
                      \item \texttt{immel\index{immel}}: molar mass of the gas
                            mixture
                      \item \texttt{itemp2\index{itemp2}(icla)}: temperature of
                            the particles of the class \texttt{icla}
                      \item \texttt{irom2\index{irom2}(icla)}: density of
                            the particles of the class \texttt{icla}
                      \item \texttt{idiam2\index{idiam2}(icla)}: diameter of the
                            particles of the class \texttt{icla}
                      \item \texttt{igmdch\index{igmdch}(icla)}: disappearance
                            rate of the reactive coal of the class \texttt{icla}
                      \item \texttt{igmdv1\index{igmdv1}(icla)}: mass transfer
                            caused by the release of light volatiles
                            from the class \texttt{icla}
                      \item \texttt{igmdv2\index{igmdv2}(icla)}: mass transfer
                            caused by the release of heavy volatiles
                            from the class \texttt{icla}
                      \item \texttt{igmhet\index{igmhet}(icla)}: coke
                            disappearance rate during the coke burnout
                            of the class \texttt{icla}
                      \item \texttt{ix2\index{ix2}(icla)}: solid mass fraction
                            of the class \texttt{icla}
                     \end{list}
              \end{list}
              \item {\em variables specific to the continuous phase}:
              \begin{list}{-}{}
               \item Solved variables
                     \begin{list}{\texttt{ivar} = }{}
                      \item \texttt{isca(if1m\index{if1m}(icha))}: mean value of
                            the tracer 1 representing the light
                            volatiles released by the coal \texttt{icha}
                      \item \texttt{isca(if2m\index{if2m}(icha))}: mean value of
                            the tracer 2 representing the heavy
                            volatiles released by the coal \texttt{icha}
                      \item \texttt{isca(if3m)\index{if3m}}: mean value of the
                            tracer 3 representing the carbon released as
                            CO during coke burnout
                      \item \texttt{isca(if4pm\index{if4pm})}: variance of the
                            tracer 4 representing the air
                      \item \texttt{isca(if3p2m\index{if3p2m})}: variance of the
                            tracer 3
                     \end{list}
               \item Properties \texttt{propce(iel,ipproc(iprop))}
                     \begin{list}{\texttt{iprop} = }{}
                      \item \texttt{itemp1\index{itemp1}}: temperature of the
                            gas mixture
                      \item \texttt{iym1(1)\index{iym1(1)}}: mass fraction of
                            $CH_{X1m}$ (light volatiles) in the gas
                            mixture
                      \item \texttt{iym1(2)\index{iym1(2)}}: mass fraction of
                            $CH_{X2m}$ (heavy volatiles) in the gas
                            mixture
                      \item \texttt{iym1(3)\index{iym1(3)}}: mass fraction of
                            CO in the gas mixture
                      \item \texttt{iym1(4)\index{iym1(4)}}: mass fraction of
                            $O_2$ in the gas mixture
                      \item \texttt{iym1(5)\index{iym1(5)}}: mass fraction of
                            $CO_2$ in the gas mixture
                      \item \texttt{iym1(6)\index{iym1(6)}}: mass fraction of
                            $H_2O$ in the gas mixture
                      \item \texttt{iym1(7)\index{iym1(7)}}: mass fraction of
                            $N_2$ in the gas mixture
                     \end{list}
              \end{list}
             \end{list}
      \end{list}

 \item set the relaxation coefficient of the density \texttt{srrom}, with \\
$\rho^{n+1}=\texttt{srrom}*\rho^{n}+(1-\texttt{srrom})\rho^{n+1}$\\
(the default value is \texttt{srrom\index{srrom}} = 0.8. At the
      beginning of a calculation, a sub-relaxation of 0.95 may reduce
      the numerical ``shocks'').

 \item set the dynamic viscosity \texttt{diftl0}. By default
      \texttt{diftl0\index{diftl0}}= 4.25 $kgm^{-1}s^{-1}$
(the dynamic diffusivity being the ratio between the thermal
      conductivity $\lambda$ and the mixture specific heat $C_p$ in the
      equation of enthalpy).

 \item set the value of the constant \texttt{cebu\index{cebu}} of the Eddy Break
      Up model (only in \texttt{cs\_user\_combustion}. By default \texttt{cebu}=2.5)
\end{list}

%==================================
\subsection{Heavy fuel oil combustion module}
%==================================
%==================================
\subsubsection{Initialisation of transported variables}
%==================================
To initialise or modify (in case of a continuation) values of transported
variables and of the time step, the standard subroutine \texttt{cs\_user\_initialization} is used.

Physical properties are stored in \texttt{propce} (cell center). For instance, \texttt{propce(iel, ipproc(irom  ))} is \texttt{rom(iel)}, the mean density (in $kg.m^{-3}$).\\
Physical properties (\texttt{rom, viscl, cp, ...}) are computed in \texttt{ppphyv} and are not to be modified here.

The \texttt{cs\_user\_initialization-fuel.f90} example illustrates how the user
may initialise quantities related to gaseous species and droplets compositions
in addition to the chosen turbulent model.

%==================================
\subsubsection{Boundary conditions}
%==================================
Boundary conditions are defined as usual on a per-face basis in
\texttt{cs\_user\_boundary\_conditions}. They may be assigned in two ways:
\begin{list}{.}{}
\item for ``standard'' boundary conditions (inlet, free outlet, wall, symmetry): a code is defined in the array \texttt{itypfb} (of dimensions equal to the number of boundary faces). This code will then be used by a non-user subroutine to assign the conditions.
\item for ``non-standard'' conditions: see details given in
  \texttt{cs\_user\_boundary\_conditions-fuel.f90} example.
\end{list}

%==================================
\subsection{Radiative thermal transfers in semi-transparent gray media}
%==================================
%==================================
\subsubsection{Initialisation of the radiation main parameters}
%==================================

The main radiation parameters can be initialise in the Graphical User Interface (GUI) or in the user subroutine \texttt{cs\_user\_radiative\_transfer\_param}. In the GUI, under the heading ``Thermophysical models'', when one of the two thermal radiative transfers models is selected, see \figurename~\ref{fig:0_ray}, additional items appear. The user is asked to choose the number of directions for angular discretisation, to define the absorption coefficient and select if the radiative calculation are restarted or not,
see \figurename~\ref{fig:1_ray} and \figurename~\ref{fig:3_ray}. When ``Advanced options'' is selected for both models \figurename~\ref{fig:2_ray} or \figurename~\ref{fig:4_ray} appear, the user must fill the resolution frequency and verbosity levels. In addition, the activation of the radiative transfer leads to the creation of an item ``Surface solution control'' under the heading ``Calculation control'',
see \figurename~\ref{fig:5_ray}, where radiative transfer variables can be selected to appear in the output listing.

\begin{figure}[ht]
\begin{center}
\includegraphics[width=13cm]{gui_rad_transf_models}
\caption{Radiative transfers models}
\label{fig:0_ray}
\end{center}
\end{figure}

\begin{figure}[ht]
\begin{center}
\includegraphics[width=10cm]{gui_rad_transf_do_params}
\caption{Radiative transfers - parameters of the DO method}
\label{fig:1_ray}
\end{center}
\end{figure}

\begin{figure}[ht]
\begin{center}
\includegraphics[width=7cm]{gui_rad_transf_do_advanced}
\caption{Radiative transfers - advanced parameters of the DO method}
\label{fig:2_ray}
\end{center}
\end{figure}

\begin{figure}[ht]
\begin{center}
\includegraphics[width=10cm]{gui_rad_transf_p1_params}
\caption{Radiative transfers - parameters of the P-1 model}
\label{fig:3_ray}
\end{center}
\end{figure}

\begin{figure}[ht]
\begin{center}
\includegraphics[width=7cm]{gui_rad_transf_p1_advanced}
\caption{Radiative transfers - advanced parameters of the P-1 model}
\label{fig:4_ray}
\end{center}
\end{figure}

\begin{figure}[ht]
\begin{center}
\includegraphics[width=6cm]{gui_rad_transf_post_output}
\caption{Calculation control - Radiative transfers post-processing output}
\label{fig:5_ray}
\end{center}
\end{figure}

If the GUI is not used, \texttt{cs\_user\_radiative\_transfer\_param} is one of the two subroutine which must be completed by the user for all
calculations including radiative thermal transfers. It is called only during the calculation initialisation. It is composed of three headings. The first one is dedicated to the activation
of the radiation module, only in the case of classic physics. \\
{\em WARNING: when a calculation is ran using a specific physics module,
this first heading must not be completed. The radiation module is then
activated or not, according to the parameter file related to the considered
specific physics.} \\

\noindent
In the second heading the basic parameters of the radiation module are indicated.\\
Finally, the third heading deals with the selection of the
post-processing graphic outputs. The variables to treat are splitted
into two categories: the volumetric variables and those related to the
boundary faces.\\

\noindent
For more details about the different parameters, the user may refer to the
keyword list (\S~\ref{sec:prg_motscles}).


%==================================
\subsubsection{Radiative transfers boundary conditions}
%==================================
These informations can be filled by the user through the Graphical User Interface
(GUI) or by using the subroutine \texttt{cs\_user\_radiative\_transfer\_bcs.f90} (called every time step). If the
interface is used, when one of the ``Radiative transfers'' options is selected in
\figurename~\ref{fig:0_ray}, it activates specific boundary conditions each time
a ``Wall'' is defined, see \figurename~\ref{fig:6_ray}. The user can then choose
between 3 cases. The parameters the user must specify are displayed for one of
them in \figurename~\ref{fig:7_ray}.

\begin{figure}[ht]
\begin{center}
\includegraphics[width=11cm]{gui_rad_transf_wall_model}
\caption{Boundary conditions - choice of wall thermal radiative transfers}
\label{fig:6_ray}
\end{center}
\end{figure}

\begin{figure}[ht]
\begin{center}
\includegraphics[width=11cm]{gui_rad_transf_wall_params}
\caption{Boundary conditions - example of wall thermal radiative transfer}
\label{fig:7_ray}
\end{center}
\end{figure}

When the GUI is not used, \texttt{cs\_user\_radiative\_transfer\_bcs.f90} is the second subroutine necessary for
every calculation which includes radiative thermal transfers. It is used to give all the
necessary parameters concerning, in the one case, the wall temperature
calculation, and in the other, the coupling between the thermal
scalar (temperature or enthalpy), and the radiation module at the
calculation domain boundaries. It must be noted that the boundary conditions
concerning the thermal scalar which may have been defined in the
subroutine \texttt{cs\_user\_boundary\_conditions} will be modified by the radiation module
according to the data given in \texttt{cs\_user\_radiative\_transfer\_bcs.f90} (cf. \S\ref{sec:fvm_selector}).\\
A zone number must be given to each boundary face \footnote{This must be less
 than the maximum allowable by the code, \texttt{nozrdm}. This is fixed at 2000
 in \texttt{radiat} and cannot be modified.} and, specifically for
the walls, a boundary condition type and an initialisation temperature
(in Kelvin). The initialisation temperature is only used to make the
solving implicit at the first time step. The zone number allows assigning
an arbitrary integer to a set of boundary faces having the same
radiation boundary condition type. This gathering is used by the
calculation, and in the listing to print some physical values (mean
temperature, net radiative flux ...). An independent graphic output in
{\em EnSight} format is associated with each zone and allows the display on
the boundary faces of the variables selected in the third heading of the
subroutine \texttt{cs\_user\_radiative\_transfer\_param}.\\
A boundary condition type stored in the array ISOTHP is associated with
each boundary face. There are five different types:

\begin{list}{$\bullet$}{}

\item \texttt{\textbf{itpimp}}: wall face with imposed temperature,

\item \texttt{\textbf{ipgrno}}: for a grey or black wall face, calculation of the
      temperature by means of a flux balance,

\item \texttt{\textbf{iprefl}}: for a reflecting wall face, calculation of the
      temperature by means of a flux balance.
 This is fixed at 2000 in \texttt{radiat} and cannot be modified.

\item \texttt{\textbf{ifgrno}}: grey or black wall face to which a conduction
      flux is imposed,

\item \texttt{\textbf{ifrefl}}: reflecting wall face to which a conduction
      flux is imposed, which is equivalent to impose this flux directly
      to the fluid.

\end{list}

\noindent
Depending on the selected boundary condition type at every wall face,
the code needs to be given some additional information:

\begin{list}{$\bullet$}{}

\item \texttt{\textbf{itpimp}}: the array \texttt{tintp} must be completed
      with the imposed temperature value and the array \texttt{epsp} must
      be completed with the emissivity value (strictly positive).

\item \texttt{\textbf{ipgrno}}: must be given: an initialisation temperature in
      the array \texttt{tintp}, the wall emissivity (strictly positive, in
      \texttt{epsp}), thickness (in \texttt{epap}), thermal conductivity
      (in \texttt{xlamp}) and an external temperature (in \texttt{textp})
      in order to calculate a conduction flux across the wall.

\item \texttt{\textbf{iprefl}}: must be given: an initialisation temperature (in
      \texttt{tintp}), the wall thickness (in \texttt{epap}) and thermal conductivity (in
      \texttt{xlamp}) and an external temperature (in \texttt{textp}).

\item \texttt{\textbf{ifgrno}}: must be given: an initialisation temperature (in
      \texttt{tintp}), the wall emissivity (in \texttt{epsp}) and the conduction
      flux (in $W/m^2$ whatever the thermal scalar, enthalpy or temperature) in
      the array \texttt{rcodcl}. The value of \texttt{rcodcl} is positive when the
      conduction flux is directed from the inside of the fluid domain to the
      outside (for instance, when the fluid heats the walls). If the
      conduction flux is null, the wall is adiabatic.

\item \texttt{\textbf{ifrefl}}: must be given: an initialisation temperature (in
      \texttt{tintp}) and the conduction flux (in $W/m^2$ whatever the thermal
      scalar) in the array \texttt{rcodcl}. The value of \texttt{rcodcl} is
      positive when the conduction flux is directed from the inside of the
      fluid domain to the outside (for instance, when the fluid heats the
      walls). If the conduction flux is null, the wall is adiabatic. The flux
      received by \texttt{rcodcl} is directly imposed as boundary condition for
      the fluid.

\end{list}

\noindent
{\em WARNING: it is mandatory to set a zone number to every boundary
face, even those which are not wall faces. These zones will be used during the
printing in the listing. It is recommended to gather together the
boundary faces of the same type, in order to ease the reading of the
listing.}\\

%==================================
\subsubsection{Absorption coefficient of the medium, boundary conditions
   for the luminance and calculation of the net radiative flux}
%==================================
When the absorption coefficient is not constant, the subroutine \texttt{usray3} is called instead at each time step. It is composed of three parts. In the first one, the user
must provide the absorption coefficient of the medium in the array CK,
for each cell of the fluid mesh. By default, the absorption coefficient
of the medium is 0, which corresponds to a transparent medium.\\

{\em WARNING: when a specific physics is activated, it is forbidden to
give a value to the absorption coefficient in this subroutine. In this
case, the coefficient is either calculated automatically, or provided by the user {\em via} a
thermo-chemical parameter file (dp\_C3P or dp\_C3PSJ for gas combustion,
and dp\_FCP for pulverised coal combustion).}\\

\noindent
The two following parts of this subroutine concern a more advanced use
of the radiation module. It is about imposing boundary conditions to the
equation of radiative transfer and net radiative flux calculation, in
coherence with the luminance at the boundary faces, when the user wants
to give it a particular value. In most cases, the given examples do not
need to be modified.

%==================================
\subsubsection{Input of radiative transfer parameters}
%==================================

\noindent
\textit{The routine \texttt{usray5} is called twice. The first time is for boundary
conditions. The second time is for the net radiation flux computation}

In this subroutine, during the first call (\texttt{iappel=1}), the boundary conditions
 are filled:
\begin{list}{-}{}
\item the radiative intensity must be set in the array \texttt{cofrua} when the discrete
 ordinates model is used; an example is given in \texttt{usray5} for an isotropic radiation
 field on a grey wall. Proposed boundary conditions for the intensity in \texttt{usray5} are:
symmetry, inlet/outlet, and wall boundary,
\item the entering intensity for free boundaries is set to zero in \texttt{cofrua} (if the
user has more information, he can improve it),
\item arrays \texttt{cofrua} and \texttt{cofrub} must be filled when the P-1 model is
 used. The boundary conditions proposed are the same as with the discrete ordinates model.
\end{list}
During the second call (\texttt{iappel=2}), the density of the net radiation flux must be
 calculated consistently with the boundary conditions of the intensity considering that the
density of net flux is the balance between the radiative emitting part of a boundary face
(and not the reflecting one) and the radiative absorbing part. The provided example is
consistent with the example of the intensity boundary conditions given when the discrete
 ordinates model is used.


%==================================
\subsection{Conjugate heat transfer}
%==================================
%========================================
\subsubsection{Thermal module in a 1D wall}
%========================================

\noindent
\textit{subroutine called at every time step}