Initial output for object set to disappear at T_BEGIN still shows the presence of the object. #13763
Description
Not sure if this is a bug or a feature. For the case:
&HEAD CHID='test_init', TITLE='Initial output write demonstration case.' /
&MESH IJK=36,24,24, XB=0.0,3.6,0.0,2.4,0.0,2.4 /
&TIME T_END=1.0, DT=0.1 /
&SURF ID='HOT', TMP_FRONT=100, COLOR='RED' /
&DEVC ID='clock_1', XYZ=0,0,0, QUANTITY='TIME', SETPOINT=0.75, INITIAL_STATE=F /
&OBST XB=1.0,1.4,1.0,1.4,0.5,1.5, SURF_ID='HOT', DEVC_ID='clock_1' /
&VENT MB='XMIN',SURF_ID='OPEN'/
&VENT MB='XMAX',SURF_ID='OPEN'/
&VENT MB='YMIN',SURF_ID='OPEN'/
&VENT MB='YMAX',SURF_ID='OPEN'/
&VENT MB='ZMAX',SURF_ID='OPEN'/
&BNDF QUANTITY='WALL TEMPERATURE' /
&SLCF PBY=1.2, QUANTITY='CELL PHASE', CELL_CENTERED=.TRUE. /
&DEVC XYZ=1.25,1.25,1.25, QUANTITY='CELL PHASE', ID='CellPhaseInObst', TEMPORAL_STATISTIC='INSTANT VALUE'/
&DUMP DT_SLCF=0.5, DT_DEVC=0.5 /
&TAIL /
I set an obst originally to be not present and use a timer device to make it appear in the domain later at t=0.75.
The slice for CELL PHASE still shows the obstacle at t=0. Same with the DEVC 'CellPhaseInObst', it shows a 1 (i.e. SOLID) at t=0 in the devc.csv file. Interestingly a BNDF does not show the obst surface temperature at t=0, probably a SMV trick.
I came across this as we are trying to streamline the connected zones, obst creation and removal and poisson matrix generation at setup time. Turns out the issue we are seeing today (geom_stretched_grid.fds) refers to DUMP_MESH_OUTPUTS being called before MESH_ZONE entries get generated (in GLOBAL_MATRIX_REASSIGN).
An option Ill try next is to move mesh and global dumps after the first call to GLOBAL_MATRIX_REASSIGN.