Validation of FDS and smoke concentration #13386
Replies: 2 comments 4 replies
-
The results are based on two different sets experiments with different fuels. There are some challenges here. We should probably revisit this section of the Validation Guide to see if other datasets that we already use elsewhere in the guide might give more data to make a judgment on. Is FDS extremely conservative? Depends. What is the source of your soot yield input? If you are assuming a soot yield or using a soot yield for a small scale flame and applying it to a large fire, then your input is going to be a large driver of predictive error that might be hard to quantify. If you are looking at detection, which typically happens early on at low soot densities, the details of the validation cases suggest that if you have a reasonable soot yield, you are probably doing OK in terms of detection time. Also helping out is for a growing t^2 type fire, soot levels are rapidly increasing and even a factor of 2 error in soot concentration may not be a large change in detection time. If you are looking at long term smoke levels for egress, then the FDS reported visibility may be too high (not only due to the assumed extinction coefficient but also deposition as removal term). |
Beta Was this translation helpful? Give feedback.
-
Hi drfloyd. Thanks for your answer. The reason why i'm even asking about the smoke koncentration in the first place, is because i use FDS for fire safety engineering. And almost every single case i have, smoke is the main issue, and often a building is completely filled with smoke even though it has a pretty big fireventilationssystem. Im not saying that the FDS predictiong is wrong, im just curious why the smoke almost always is such an issue. Thats when i read the 2,76 bias in the validation guide. Ultimately this means that FDS overpredicts a lot right? For information im using a*t^2 fire with a constant soot_yield of 0,06 g/g. Constant HRRPUA on 1000 kW/m2 and Hc = 20.000 kJ/kg. |
Beta Was this translation helpful? Give feedback.
-
I've read through the validation guide for FDS and in the summary where all the different quantities are shown in a tabel with the calculated bias, i cant help but notice the bias regarding 1) Smoke concentration = 2,76 and 2) Smoke obscuration = 1,05.
First of all. Shouldnt these two values follow eachother in bias. I mean if smoke is overpredicted, then the smoke obscuration should also be overpredicted right?
Second. Does the bias for smoke concentration mean that a fire in FDS will produce 2,76 times the amount of an actual fire? And therefor is FDS extremely conservative.
Hopefully someone can clear things up for me. Thanks.
Beta Was this translation helpful? Give feedback.
All reactions