Fire in high rack

[Łukasz]

Dear all,

how to properly model a fire in a high-bay rack (fire in a high-rack warehouse). From the analyzes that I had the opportunity to review, I noticed that many people model the fire as a single plane on the floor (t2 fire on a plane with a specific area and HRRPUA). Wouldn't it be a better way, closer to reality, to model two or three surfaces one above the other, which would be ignited successively giving a certain  total heat release rate in total (sketch attached). Alternatively, wouldn't it be better to model a cube whose individual surfaces would ignite over time, giving a specific total heat release rate.
Is there any literature/articles on this topic? Do any of you have examples for such a fire?

Best Regards,

[michele caserta]

the modeling depends on the scenario you assumed which depends on the material you have, the assumptions and the boundary conditions. If you think it's realistic that three blocks start burning at once then go for it.

[michele caserta]

the modeling depends on the scenario you assumed which depends on the material you have, the assumptions and the boundary conditions. If you think it's realistic that three blocks start burning at once then go for it.

On Saturday, May 20, 2023 at 7:25:49 PM UTC+2 Łukasz wrote:

[dr_jfloyd]

See the UL/NFRPF Sprinkler, Vent, and Draft Curtain Study in the Validation Guide along with the referenced NIST report (available online). The plastic commodity cases have fire spread over a stack of Group A commodity. 

[o...@aquacoustics.biz]

Modelling fire spread is still in the realms of research, but FDS is getting ever-closer.  Here is a rack fire (FDS source) that was development for a commercial project some years back.  I cannot share the rationalization of the model, fire load, or growth rate (commercial in confidence) - but the project involved many models looking at comparative performance of automatic sprinklers between defined and undefined storage configurations.  The analysis showed that, within the justified limitations of the model under numerous ignition scenarios, the undefined storage configuration performed 'better' than permitted storage configurations.  I'd include the SmokeView files but these are somewhat large.

[o...@aquacoustics.biz]

On reflection I can redact the related fire report and I can post the SmokeView files in a DropBox or similar if anyone is interested (although it will take a day or two).  The analysis was accepted by the BCA (AHJ), the client was delighted, and (through significant conservatism) I am confident that the design is compliant.  I must add this to my list of projects to write up.

[Łukasz]

Thank You

[Łukasz]

Thank you for the information. I read that report once, I need to remember

[Łukasz]

Please share the files if possible. This will expand knowledge in this area.

Best Regards,