Kevin McGrattan

Dave, are you still having this problem?

Glenn -- I would suggest that we combine the BNDF menu item for the conventional OBSTs and the new "geometry" boundaries. That might eliminate these kinds of discrepancies.

An alternative is to see if the area under the curve has a physical meaning that we can check against.

DSC integrated over time comes to 1867 J/g. I can get close to this by summing the contributions of reaction energy and specific heat. But, yes, it's hard to pin...

I'm going to punt on this. Adding FED to every grid cell in a 2D or 3D slice would be too memory intensive. I suggest you remove it from Smokeview...

I did some work on this case. The problem has to do with the fact that we have a really thin solid just before it burns away and the heat...

What is the `TMP_FACTOR`?

We would need a few min/max cushions, but I get the idea. Maybe a problem is that the `SUBSTEP_POWER` is capped at 2. I'll play with it some more.

That could work. Let's get the bundle out and then test this idea on more cases.

I'm going to test this ``` IF (NWP>1) THEN DELTA_TMP(1) = (DT_BC/ONE_D%RHO_C_S(1))*& (RDX_S(1)*(ONE_D%K_S(1)*RDXN_S(1)*(ONE_D%TMP(2)-ONE_D%TMP(1))+Q_NET_F) + Q_S(1)) DELTA_TMP(NWP) = (DT_BC/ONE_D%RHO_C_S(NWP))*& (RDX_S(NWP)*(Q_NET_B-ONE_D%K_S(NWP-1)*RDXN_S(NWP-1)*(ONE_D%TMP(NWP)-ONE_D%TMP(NWP-1))) + Q_S(NWP)) ELSE DELTA_TMP(1) = (DT_BC/ONE_D%RHO_C_S(1))*(RDX_S(1)*(Q_NET_B+Q_NET_F) + Q_S(1)) ENDIF TMP_RATIO =...