Maximum density warning

[Ioanna Koromila]

Hello Community,

I am trying to simulate a fire on a ship car deck where fire initiates from a heavy truck that is able to release 140 MW. Supply and exhaust vents are also taken into account.

​As I understood this problem resembles tunnel fire scenarios, so I used the pressure considerations according to section 9.3 of the FDS user guide

&PRES TUNNEL_PRECONDITIONER=.TRUE.,  VELOCITY_TOLERANCE =0.001, MAX_PRESSURE_ITERATIONS=200/

The simulation stops at approximately 1000s (simulation time)  with the following warning: "Maximum density , 12.080 kg/m3, clipped in Mesh XX". In the .out file, I noticed that the pressure error exceeds 10^5. 

Is the heat release really high for simulation? Should I increase the pressure iterations? Do you have any advice on this issue?

Thank you in advance, 

Ioanna

[Kevin McGrattan]

Are there any regions in the domain that are completely enclosed? Are there regions where gases are trapped? Search for areas in the domain where the gas density is very high. 

[Ioanna Koromila]

Yes, the entire geometry is enclosed (length 98m, width 18.8, height 5.2). The only "openings" are the areas of supply and exhaust vents. I' ll check out also for gas densities. In case gas densities are very high, should I consider some openings even if it does not represent my real geometry? My case study is to investigate fire propagation when there are no openings in the geometry.  

Thank you

Στις Δευ, 1 Νοε 2021 στις 5:43 μ.μ., ο/η Kevin McGrattan <mcgr...@gmail.com> έγραψε:

Are there any regions in the domain that are completely enclosed? Are there regions where gases are trapped? Search for areas in the domain where the gas density is very high. 

--
You received this message because you are subscribed to a topic in the Google Groups "FDS and Smokeview Discussions" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/fds-smv/zYBBisU7VlM/unsubscribe.
To unsubscribe from this group and all its topics, send an email to fds-smv+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/CAAJimDE6XaQC33U4w%3DXPXw-tCKzCu%2BTsh%2BJd-peZz8LE2WrDwQ%40mail.gmail.com.

[fde]

Is there any chance both supply and extract have prescriptive flow rate? Maybe it is overconstrained. Try setting either of them as OPEN and watch the pressure throughout domain. 

[Kevin McGrattan]

140 MW and no openings? I suspect the enclosure will vent itself when the walls blow off. Track slice files of density, temperature and pressure and decide if this is physically correct.

[Ioanna Koromila]

I have indeed set a prescriptive flow rate for vents. I will convert some vents into  "OPEN". Thank you. 

It seems that the problem appears when I apply supply & exhaust vents. I simulated a case without them and the fire extinguishes due to limited oxygen (dashed HRR in the attached picture; the continuous line corresponds to HRR of the case with ventilation). 

 

I ll check out also the scice files. Thank you very much.

Ioanna 

Στις Δευ, 1 Νοε 2021 στις 7:27 μ.μ., ο/η Kevin McGrattan <mcgr...@gmail.com> έγραψε:

140 MW and no openings? I suspect the enclosure will vent itself when the walls blow off. Track slice files of density, temperature and pressure and decide if this is physically correct.

--

You received this message because you are subscribed to a topic in the Google Groups "FDS and Smokeview Discussions" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/fds-smv/zYBBisU7VlM/unsubscribe.
To unsubscribe from this group and all its topics, send an email to fds-smv+u...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/CAAJimDG1W%3DUn3BRW%2BCqeGDXxhVfsYCJX2nsR13o9akejA9L%3DuQ%40mail.gmail.com.

[Kevin McGrattan]

I suggest you run a case with no fire. Just turn on the HVAC system and let it run. Make sure that you have no unnatural pressure build-up. 

[Ioanna Koromila]

Dear Kevin, this is a good idea to understand if vents cause pressure issues. I’ll proceed with this run. Thank you very much for your support.

On Tue, 2 Nov 2021 at 4:42 PM, Kevin McGrattan <mcgr...@gmail.com> wrote:

I suggest you run a case with no fire. Just turn on the HVAC system and let it run. Make sure that you have no unnatural pressure build-up. 

--

You received this message because you are subscribed to a topic in the Google Groups "FDS and Smokeview Discussions" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/fds-smv/zYBBisU7VlM/unsubscribe.
To unsubscribe from this group and all its topics, send an email to fds-smv+u...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/CAAJimDHEx2NXqT0MaGj2T8yrFzJw5sbp88kr7cHZPmPhXW%3D3%2Bw%40mail.gmail.com.

--

Ioanna A. Koromila

Research Associate | NCSR Demokritos 

Institute of Nuclear & Radiological Sciences & Technology, Energy and Safety

Systems Reliability and Industrial Safety Laboratory 

Patr. Gregoriou E & 27 Neapoleos St. | Agia Paraskevi, 15341  |Greece

Contact

email: ioanna....@gmail.com | LinkedIn: Ioanna Koromila

[dr_jfloyd]

The thing to realize when fully specifying all inlets and outlets is that if you specify the volume flow, then you will only have the same mass flow if the densities at the inlet and outlet are the same.  With no fire, this will only happen if the inlets and outlets are at the same elevation as density decreases with height. A supply vent at the bottom and an exahust vent at the top will add mass over time. A supply vent at the top and an exhaust vent at the bottom will loss mass over time.

With a fire, if the exhaust vent sees higher temperature (lower density) gas, then you will be adding mass over time.

[Ioanna Koromila]

Both vents have the same flow and are at the same height. The supply is at the front of the geometry and the exhaust at the back, both are at the bottom. The area they cover is about 5m3. Do you think this could be a problem? Should I reduce the vent surface area?

I will do all the tests you have suggested, but it will take me a few days to get the results and discuss them again if necessary.

--

You received this message because you are subscribed to a topic in the Google Groups "FDS and Smokeview Discussions" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/fds-smv/zYBBisU7VlM/unsubscribe.
To unsubscribe from this group and all its topics, send an email to fds-smv+u...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/19b17e1d-e9d0-4ed2-b9a6-4a74033e858en%40googlegroups.com.

[Ioanna Koromila]

Dear developers, 

I examined the "no fire case". Attached are some graphs resulting from the _hrr.csv file (hrr_file_no_fire_case.xlsx). As far as I understand no significant pressure issues resulted (maybe in pressure zone4 ?). 

However, observing the .out file (part of this is attached in "out_file_no_fire_case.txt"),  I wonder if there was a problem with defining the reaction or/and the fire. 

In brief, the procedure I pursue is the following:
1.  I assume a fuel package capable of burning with a pre-defined HRR curve that I have numerically calculated by assuming a weighted average of the materials found in a truck/car.

2. The reaction characteristics (yield, heat of combustion, etc) result from the weighted average of the materials participating in the fire (truck/car). 

A part of the .fds file is attached. The missing details concern the HRR, devices, ventilation, and obstacles details. 

I would appreciate your opinion on this and your guidance in case my assumptions and the fds file setup have significant errors or misunderstandings. 

Thank you very much for your kind and prompt response so far. 

Ioanna Koromila

[Kevin McGrattan]

Do you need to have 22 regions in the domain that are sealed off from the rest of the domain? Sometimes these regions can have pressure build-up to due to heating. I suggest you eliminate the sealed regions and make sure that there is either adequate leakage or openings to the outside.

[Ioanna Koromila]

Dear Kevin, 

The geometry for the case study I examine has many enclosed areas. If I consider any additional open space the case study changes. I will consider some extra leakage.  

Thanks for the suggestions and the valuable assistance. 

Kind regards,

Ioanna

Στις Δευ, 8 Νοε 2021 στις 7:16 μ.μ., ο/η Kevin McGrattan <mcgr...@gmail.com> έγραψε:

Do you need to have 22 regions in the domain that are sealed off from the rest of the domain? Sometimes these regions can have pressure build-up to due to heating. I suggest you eliminate the sealed regions and make sure that there is either adequate leakage or openings to the outside.

--

You received this message because you are subscribed to a topic in the Google Groups "FDS and Smokeview Discussions" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/fds-smv/zYBBisU7VlM/unsubscribe.
To unsubscribe from this group and all its topics, send an email to fds-smv+u...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/CAAJimDHQ8w6N1bH_4KgHxuJVExttqqp%2B4_-XiBDTHrhVSxo6mQ%40mail.gmail.com.

[dr_jfloyd]

You have a high molecular weight fuel for which you have not defined a fuel specific heat. The User's Guide says to avoid this as it can cause problems with the extinction model. You can normalize your fuel to a lower molecular weight (make it C=1 and keep the ratios of H,O,N the same), turn off the extinction model, or used &SPEC to define a molecular weight.

[Ioanna Koromila]

Dear Floyd, 

I didn't have in mind this issue. I'll take a closer look at this part of the User's Guide.

Ioanna

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/8173d291-f72d-4ca8-a818-6bbb8a7db12en%40googlegroups.com.