Travelling Fire

[carlop...@gmail.com]

Good morning,

I should simulate fire propagation between several cars in FDS.
I created a small model to verify that the propagation was happening correctly (attached). 

I created a single mesh with cell size of 0.20 m.

A first car is ignited and should propagate the fire to the adjacent car and the car above.

According to your experience, I kindly ask if the FDS file is set up correctly.

Also, I would like to ask, is there any way in FDS to figure out which cars ignited as a result of the ignition?  

Thank you all. Have a good day

[dr_jfloyd]

Does the case run?

Do you get the behavior you expect?

If the answer to either is no, then you should edit the fds file. 

You could use control functions to evaluate each car. For example, one possible approach is to have a DEVC integrating the BURNING RATE over each car with a SETPOINT slightly above zero.  The ctrl_devc_log file will contain a record of anytime a DEVC with a SETPOINT or a CTRL function changes state.

[carlop...@gmail.com]

Thank you dr_jfloyd you are absolutely right. I first need to check if the model matches the literature studies being employed.

The use of DEVC Burnig Rate sounds like a very good idea to me. I will try to use it.

Besides the approach I used, is it possible to model fire spreading in other ways with FDS? 

[dr_jfloyd]

You basically have three options for fire spread:

1. Fully prescribe the spread in the input file (i.e. define explicitly what in the input file what surfaces ignite at what time and with what HRR). 

2. Define criteria where wall cells or a VENT starts to follow a predefined HRR curve (this is what you are doing now with IGNITION_TEMPERAUTRE starting a wall cell but you could also activate a VENT with an HRRPUA when some condition is reached).

3. Define MATL reactions and fully predict fire growth and spread. 

[carlop...@gmail.com]

Thank you for your reply  dr_jfloyd , I just take the opportunity to ask you for one last clarification on this.

The only thing I don't understand is the following:
as I modeled once one of the 5 faces reaches the ignition temperature (350 °C in my case) it ignites and starts to burn. The fire should then spread to the remaining faces.

I, on the other hand, basically want that once any of the 5 faces reaches its ignition temperature, the whole block starts to burn releasing HRR as I set it. is this possible?

Thank you

[carlop...@gmail.com]

I currently have FDS set up like this:

Car size 4.60 m x 2.00 m x 1.60 m (h). HRRmax = 8300 kW

If I divide HRRmax (8300 kw) by 5 faces ( 30.32 m2) I get 273, 75 kW/sq m. So when the whole block burns the heat release is correct.

If only one face burns the modeling is wrong--I will have less heat release than I expect

I don't understand how I can solve

[dr_jfloyd]

You could try something like this: Define the target cars just with no ignition temperature but keeping the MATL properties so they heat up. Then you can monitor each car to see when any cell reaches its ignition tempertarure.  At that point you could add VENTs to the car OBST where the VENTs have the desired HRR profile. 

[carlop...@gmail.com]

Do you mean as in the attached example?

1) I create a first car that functions as a trigger;
2) The target car has MATL ID (rubber) surface properties;
3) On the target car I place a Device AST set to 350°C;
4) When the temperature is reached, it activates, through a control, the VENT burner on the target car.

Do I understand this correctly?

If so I had thought about it. Not continued in this way because I have about 50 cars, I should put 50 controls + 300 Device AST controlling the ignition (1 device on each side of the car). 

[dr_jfloyd]

You may want to use XB and SPATIAL_STATISTIC='MAX' so any cell on the face can trigger the VENT but other than that it looks like you have the general idea correct with your inputs.

You can alway use Excel, Python, or other tool to take the list of OBST for the target cars and generate all the AST and CTRL functions needed.  

[carlop...@gmail.com]

Thank you for the suggestion.

With XB and SPATIAL_STATISTIC='MAX' I monitor the time within which my obstructions reach 350 °C and then with another simulation with a time control I ignite the obstructions?

Alternatively, may I ask how to set up a control with XB and SPATIAL_STATISTIC='MAX'.

I entered this code on FDS

&DEVC ID='Adiabatic Surface Temperature01_MAX', QUANTITY='ADIABATIC SURFACE TEMPERATURE', SPATIAL_STATISTIC='MAX', XB=1.4,3.4,2.8,7.4,0.0,1.6, SETPOINT=350.0/

But it doesn't seem to work.

[dr_jfloyd]

In your example case from earlier today your target car was:

&OBST ID='Ignition Car', XB=4.8,6.8,2.8,7.4,-2.775558E-16,1.6, THICKEN=.TRUE., SURF_ID='RubberCar'/

The XB for the DEVC should match the XB for the OBST. 

[carlop...@gmail.com]

Yes you are right. I had changed the model. This is the one from this morning.
I am attaching the corrected FDS file.
However, I do not understand how I connect the ignition with XB and SPATIAL_STATISTIC='MAX'.

If I understand you correctly, you intend to use this instrument only to monitor the temperatures on the various targets and then through a time control make them burn out. 

It would seem that the ignition can only be controlled with DEVC.   

[dr_jfloyd]

Get rid of IGNITION_TEMPERATURE. When you activate the VENT you just want it to immeadiately start following the HRR curve.

[carlop...@gmail.com]

Right. But how do I activate the VENT?
 
The DEVC XB and SPATIAL_STATISTIC='MAX' do not allow me to command the activation of the VENT. I don't understand how I can use this device to my advantage. Sorry...

I still have to put AST DEVC and then add a control, am I wrong?  

thanks for the answers. You are very kind.

[dr_jfloyd]

VENT has inputs of DEVC_ID and CTRL_ID. Replace the CTRL_ID with DEVC_ID, and give the ID of the DEVC measuring max AST. Add SETPOINT to the DEVC.

Instead of:

&VENT ID='Vent04', SURF_ID='Grounded car', XB=4.8,6.8,2.8,7.4,1.6,1.6, CTRL_ID='Control01'/

&DEVC ID='Adiabatic Surface Temperature_MAX', QUANTITY='ADIABATIC SURFACE TEMPERATURE', STATISTICS='MAX', XB=4.8,6.8,2.8,7.4,0.0,1.6/

Do:

&VENT ID='Vent04', SURF_ID='Grounded car', XB=4.8,6.8,2.8,7.4,1.6,1.6, DEVC_ID='Adiabatic Surface Temperature_MAX'/

&DEVC ID='Adiabatic Surface Temperature_MAX', QUANTITY='ADIABATIC SURFACE TEMPERATURE', STATISTICS='MAX', XB=4.8,6.8,2.8,7.4,0.0,1.6,SETPOINT=whatever temperature you want to ignite at/

When the DEVC hits the SETPOINT its logical state will change from FALSE to TRUE and a VENT with a control will be active when its controling item is TRUE.

[carlop...@gmail.com]

Thank you again.
I had thought about writing the code in this as well and in fact I did, but it didn't work, and do you know why?
Because I was opening the FDS file with Pyrosim which I believe does not support such control....

I will have to go to technical support.

While waiting I will continue to compile the project FDS file without using PyroSim.

Thanks again for the great advice, have a nice day and good work.