Surface Wall Flux Validation

[sprankleray]

Hello,
I’m hoping you can tell me what I’m missing here. I’m trying to run some of the validation cases found on github for V7.7.5; currently looking at wall surface flux. Figure 10.1 indicates that the NIST/NRC Long wall and Short wall test CFAST runs are relatively accurate when compared to the test data. In fact, the Long Wall Heat Flux curve for NIST/NRC Test 8 shown on page 262 of the V&V Guide and reproduced in the attached file shows maximum flux values of about 3 kW/m2 and 4 kW/m2 for Targets 1 and 2 respectively.

However, when I run NIST_NRC_T8.in and plot incident flux curves for Target 1 and Target 2 from the devices.csv output file, I get different (lower) values; see attached.

The shape of the curves is about the same but the maximum flux values are lower at about 2 kW/m2 for Target 1 and slightly over 3 kW/m2 for Target 2. It seems the most likely reason I’m getting different results is that I’m not looking at the correct data. I want to look at wall surface flux and only the curves above from page 262 of the V&V Guide connect Target 1 and Target 2 flux to wall flux. There’s nothing in the input file itself other than that these targets are indeed located on the surface of the long wall. And there’s nothing I can find in the output files; that is, I can’t find a *.csv output file that has a column labeled “Wall Surface Flux.”

I note that the descriptive text provided in Section 10.1 (page 127) of the V&V Guide, for the NIST/NRC tests indicates that the actual measured flux values are net flux rather than gauge flux. But that would seem to show a need to alter the empirical test data in the curves presented above and on page 262, and I assume that has been done. It would not alter a comparison of the results I obtained using v 7.7.5 and those presented on page 262, also using v 7.7.5. Any light you can shed on this would be appreciated.

Thanks,
Ray

Wall Flux comparison of V&V Guide pg 262 to new case-1.pdf

[mcgratta]

My first thought is that this might be an issue with "net", or "total", or "incident" heat flux. Depending on the device, sometimes we measure a "gauge" heat flux. As I recall, some of the devices used in these experiments measured the heat flux to the hot wall, as opposed to a cold water-cooled heat flux gauge. Take a look at all output columns with heat flux and see if any match with what is in the guide.

[sprankleray]

Kevin, Thanks for the quick response. I noted the aspect you mention here as it appears in Section 10.1 (pg 127) of the V&V Guide. However, wouldn't that only affect the model output v the empirical test results? I'm looking at running v7.7.5 for NIST/NRC_T8.in and getting the same results that are presented, in chart form, for the same input file and the same version of CFAST, v7.7.5. Unless you altered the *.csv files, my output of v7.7.5 should be identical to yours, as long as I'm reading the right columns. I couldn't find any other columns or any other *.csv that gave flux results to match those curves in the V&V Guide, even when in validation output mode. Thanks Ray From: Kevin McGrattan ***@***.***> Sent: Thursday, February 5, 2026 4:45 PM To: firemodels/cfast ***@***.***> Cc: sprankleray ***@***.***>; Author ***@***.***> Subject: Re: [firemodels/cfast] Surface Wall Flux Validation (Discussion #2230) My first thought is that this might be an issue with "net", or "total", or "incident" heat flux. Depending on the device, sometimes we measure a "gauge" heat flux. As I recall, some of the devices used in these experiments measured the heat flux to the hot wall, as opposed to a cold water-cooled heat flux gauge. Take a look at all output columns with heat flux and see if any match with what is in the guide. - Reply to this email directly, view it on GitHub<#2230 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AFZGHI5RN5URQOFI7EG6YA34KO2WPAVCNFSM6AAAAACUEWIPUCVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTKNZRGIYDENY>. You are receiving this because you authored the thread.Message ID: ***@***.******@***.***>>

[mcgratta]

I plotted all the heat flux outputs for Target 1 for the most recent version of this case in the GitHub repository. There appears to be one curve that peaks at 3 kW/m2, but I'm too old to distinguish all these pastel colors. I'll take a closer look tomorrow, or take another look at your output file and the one in the git repo.