Issue with CFAST Model

[JohnIsaac147]

WS-01_RESULTS.csv
Hi Everyone,

I have a CFAST model that is acting a little bit weirdly, the model consist of two workshop and service bit compartments, there are roller shutters that are represented as opening in different location in two compartments, there are hanging canopies above the openings and each canopy each side above the openings are provided with compartments that are split by each end of each compartment.

The results are acting alittle bit wierdly and the smoke volume and smoke layer height in the non-fire compartment (service bit) are showing very similiar results to the results of the compartment where the fire zone is (workshop), the smoke temperature in the non-fire compartment does not exceed 29 degrees.

We have looked into overlapping of the compartments, we have also looked into the smoke decesnding below the compartment openings heights and the smoke filling the other compartment, also the canopy right - service pit shows a sudeen drop.

I have attached the CFAST file and the results of the of each of the smoke layer height, smoke volume and smoke temperature.

If any of the community have a suggestion on reasoning of the CFAST malfunction, please let me know.

[rmcdermo]

@rpeacoc @preneke @gforney Please have a look. Thanks

[drjfloyd]

@JohnIsaac147, you have only attached the results file. Please attach your CFAST input. You will need to change the extension to .txt to attach it to the issue,

[JohnIsaac147]

WS-03.txt
WS-01.txt

Hi Floyd,

Thanks for the reply, please find the txt files attached to the issue.

[gforney]

could you restate the problem you are seeing in terms of what you loaded in smokeview, what you saw, what you expected to see

[JohnIsaac147]

The CFAST model indicates that smoke is present in the non-fire compartment. The smoke volume is significant, and the smoke layer descends to the floor, although the smoke temperature remains relatively low (not exceeding 30°C)!

[preneke]

I am not sure what your problem is. Smoke from fires does flow into rooms without a fire and can be cool because of entrainment of cooler air and radiation and conduction to cooler surfaces. If the upper layer in the fire room is below the top of a vent into another room smoke will accumulate in that room. In any floorplan in CFAST it is generally the case that with a fire that is big enough or runs long enough smoke will accumulate in all rooms that have an open path to the fire room. Can you share the file you are running and be a little more specific in why you believe there shouldn't be any smoke in the room you are talking about. From: JohnIsaac147 ***@***.***> Sent: Monday, December 8, 2025 8:56 PM To: firemodels/cfast ***@***.***> Cc: Reneke, Paul A. (Fed) ***@***.***>; Mention ***@***.***> Subject: [EXTERNAL] Re: [firemodels/cfast] Issue with CFAST Model (Discussion #2223) The CFAST model indicates that smoke is present in the non-fire compartment. The smoke volume is significant, and the smoke layer descends to the floor, although the smoke temperature remains relatively low (not exceeding 30°C)! - Reply to this email directly, view it on GitHub<#2223 (reply in thread)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ADDV4MAC7IFZOR53EKRPQGD4AYT3HAVCNFSM6AAAAACNX3UHOOVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTKMRQGM3TGMQ>. You are receiving this because you were mentioned.Message ID: ***@***.******@***.***>>

[JohnIsaac147]

The issue is that the two compartments are not open to each other. There are two additional compartments at both ends of the fire and non-fire compartments, functioning as canopies that are open to the outside. Smoke should fill the fire compartment and then vent externally. However, the model treats the two canopy compartments as a channel, allowing smoke to spread into the non-fire compartment.

Please find the attached files.
WS-01.txt
WS-03.txt

Kind regards,

[drjfloyd]

CFAST is not a computational fluid dyanmics model. It is a zone model with the significant assumptions that flows through openings can be characterized with an orifce equation and that compartments have two layers that each have perfectly uniform conditions. The further your simulation strays from these assumptions, the worse you should expect CFAST to perform. An open canopy is not going to to have a uniform layer and the openings for that canopy are not an orifice.