vertical spread of fire + spread to other objects

[Bojan]

is it possible in fds to simulate vertical spread of fire across for example, bigger shaped curtain? i would like my fire to begin at curtain, and spread to other furniture (for example, some couch nearby). so far, i discovered in user guide how to make simple radial spread of fire, but something more complicated would be nice as well.

i came across some video on youtube where i saw spread of fire, and that furniture disappears during the process: https://www.youtube.com/watch?v=fH80vrB2KEs

[dr_jfloyd]

You can either script fire spread (explicitly define using RAMPs and VENTs how the fire spreads across the room) or you could define pyrolysis models for the objects in the room (See Section 8.5). 

[ada darmon]

Hi Bojan .Before enabling the combustion model u should state why are u doing the simulation. Pretty pictures? Engineers like to simplify things...

On 11 Apr 2016 22:18, "Bojan" <bojan...@gmail.com> wrote:

is it possible in fds to simulate vertical spread of fire across for example, bigger shaped curtain? i would like my fire to begin at curtain, and spread to other furniture (for example, some couch nearby). so far, i discovered in user guide how to make simple radial spread of fire, but something more complicated would be nice as well.

i came across some video on youtube where i saw spread of fire, and that furniture disappears during the process: https://www.youtube.com/watch?v=fH80vrB2KEs

--
You received this message because you are subscribed to the Google Groups "FDS and Smokeview Discussions" group.
To unsubscribe from this group and stop receiving emails from it, send an email to fds-smv+u...@googlegroups.com.
To post to this group, send email to fds...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/b7797935-40c4-4f27-82cc-020689733424%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Bojan]

it is clear to me that fire spread is quite difficult to model, since it is mainly based on information about material properties, which are hard to define. my first idea was to create a simple burner and ramp the curve, which i could develop using the lab data for each of the furniture which could burn. it is not the best method, but it certainly better than the fire spread mechanism in fds, due to the limitations i mentioned before.

problem is that, when the simulation is showed to people who are not fire safety engineers, it is hard to explain why we use a simple burned with ramp hrr curve, instead of having realistic fire, which spread throughout the building. because the simulation which i do is going to be showed not only to experts, i want to make it more realistic for auditorium, but on the other hand, follow the curve which i develop using the data from real experiments.

[hanna]

You can also specify a spread rate and define the maximum area for fire. The fire is still predicted ( follows a t squred ramp) but it may look more realistic.

[Bojan]

Thanks Hana!

[ada darmon]

Ok. If you are only concerned how to fool the laypersons...that is easy. You can work out your fire following the ramp and assume how much it will extend horizontally. Then, you set an initial red square called "fire" and assign the FIRE SURF property. Then simulate  progressive fire spread by assigning the fire property to adjacent surfaces (aslo defined by you) On those, you set a timer with the moment when these will become active.

For example at time=0 the fire is 1 m^2, at time=40 cm it is double (new surfaces become active) and so on. 

FDS will still compute the fire prescribed by RAMP.

Try...

Regards,

Ada

On Fri, Apr 15, 2016 at 12:40 PM, Bojan <bojan...@gmail.com> wrote:

it is clear to me that fire spread is quite difficult to model, since it is mainly based on information about material properties, which are hard to define. my first idea was to create a simple burner and ramp the curve, which i could develop using the lab data for each of the furniture which could burn. it is not the best method, but it certainly better than the fire spread mechanism in fds, due to the limitations i mentioned before.

problem is that, when the simulation is showed to people who are not fire safety engineers, it is hard to explain why we use a simple burned with ramp hrr curve, instead of having realistic fire, which spread throughout the building. because the simulation which i do is going to be showed not only to experts, i want to make it more realistic for auditorium, but on the other hand, follow the curve which i develop using the data from real experiments.

--

You received this message because you are subscribed to the Google Groups "FDS and Smokeview Discussions" group.
To unsubscribe from this group and stop receiving emails from it, send an email to fds-smv+u...@googlegroups.com.
To post to this group, send email to fds...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/629e0fda-2b9d-4a7a-b651-d5b65739abac%40googlegroups.com.

[Bojan]

Ada, thank you for your suggestion. Some other idea also came to my mind. For example, I have initial fire from sofa. The fire follows defined HRR curve (based on real experiment in lab which is ramped in FDS). Two meters away is the next part of furniture which starts to burn and also follows its own HRR curve (for instance, armchair which will start to burn when its surface reaches say 350 degrees or gets 20 kW/m2 from the flames). Four meters away is the next furniture which catches fire (and maybe the first part of furniture has already burned out) and so on. Is its possible to simulate this in FDS? In this way, each furniture would follow own HRR curve, but I will manage to predict process of ignition and burning on basis of critical temperature/radiation (which is now very hard for me to predict).

[dr_jfloyd]

There are examples in the User's Guide on how to start a fire based on an ignition temperature or any other quantity that can be measured by a DEVC.

If your goal is just a pure demonstration, then engineering a pretty simulation for people not familiar with modeling is OK in my opinion. If the purpose is to prove a design, then I think as modelers we need to take care that all our modeling inputs are defendable based upon our best current knowledge. We should strive not to re-enforce the belief amongst some AHJs that modeling is all smoke and mirrors and that we just make the model say we what want it to.

[Bojan]

I completely agree with you. However, I am trying to investigate one fire incident and so far, I have images from the scene and statements from witnesses. I managed to develop some kind of scenario e.g. what could burn in which order, and my idea is to develop main HRR curve based on separate HRR curve for each part of furniture, obtained in lab experiment. Of course, it is not the perfect prediction since fire experiment in lab is quite different to real fire, but that is everything I have right now. On the other hand, I would like to stick to the main HRR curve, and make more realistic fire spread (instead of using the ordinary burner) since, as I mentioned before, it is quite hard to explain to people who are not fire safety experts, why we use burned with ramp, instead of simulating the whole fire.

[Craig Weinschenk]

Perhaps an alternate course of action here may be to stick to with the ordinary burner and ramp and invest time on explaining to the "non-experts" why burners with ramps are often used in fire models. Could be very useful for all parties involved.

[Bojan]

That is also an option, probably the most reasonable one. However, in my last post, I wanted to check if I am going to the right direction, by combining HRR curves from each furniture in order to build one main curve, and ramp it in FDS. Because, right now, with all resources I have, I did not find better solution, so just wanted to check.

[dr_jfloyd]

Using test data from individual furniture items coupled to some form of ignition criteria is definitely one approach to illustrate one possible fire scenario. However, it will almost certainly not be the fire that actually occurred (differences in timing of ignition for the various pieces of furniture, burning rate of the objects in the room vs. the test, were even the identical pieces of furniture tested, etc.). There would be a limit to the conclusions you could draw to a degree of probability.

[Bojan]

From the beginning I knew that this is just prediction of possible fire, what could occur, but is certainly not the exact one that happened. However, to the some point, using simplifications and predictions based on various resources among literature, it is possible to re-create at least some reasonable fire. 

Do you have maybe some other idea, taking in account that I have only pictures after the fire, and few statements of witnesses?

[Bojan]

Is it possible to somehow incorporate different ramps into one scenario? For example, I have one chair which burns using one ramp, the next chair has its own ramp and will burn when the temperature of its surface reaches lets say 300 degrees (caused by the fire from first chair).

On Monday, April 18, 2016 at 9:28:39 PM UTC+2, hanna wrote:

[dr_jfloyd]

Yes, you can assign different RAMPs to different SURF definitions.

[Bojan]

I read the Chapter 8.4.2 in the User Guide but it is still not completely clear to me, how I should prescribe my fire using the RAMP, but still manually control the spread to other objects (in terms of surface under fire).

So, for example, I already defined my fire using RAMP over the time. But I have three surfaces. I would like to order FDS to just make the fire surface bigger, spreading fire to other objects. So, my fire lasts 20 minutes, the first object starts burning at 0 minutes, the next object at 5 minutes and the last one at 10 minutes. The surface of the fire becomes larger, but the fire is still ramped. It would be nice though to be able to say, at 15 minutes the first surface will be taken out (the whole fire surface will be smaller, because the first object already burned out). Is there some example I could follow?

Thanks!

[dr_jfloyd]

If the objects are not identical in their heat release rate (other than the time shift), then assign a different SURF and RAMP to each object.

&SURF ID='1',HRRPUA=xxxx,RAMP_ID='1'/

&SURF ID='2',HRRPUA=xxxx,RAMP_ID='2'/

&SURF ID='3',HRRPUA=xxxx,RAMP_ID='3'/

If the objects are identical in their heat release rate (other than the time shity), then you can assign the same RAMP and just use TIME DEVCs to turn on the VENTs for the fires at the desired time.  The time value in a RAMP starts upon creation of the OBST or VENT using the RAMP (i.e. the time used for input for the ramp is simulation time - creation time).

[Bojan]

Sorry, but I did not understand that completely. I am not sure if the area of fire surface is important, or I have only to prescribe a ramp. What I want to do is the following:

At the beginning, I have a fire surface of 1 m2 which follows the ramp. After 5 minutes I have another surface, which is 2 m2, and also follows the ramp. In this case I have actually fire surface of 3 m2. After 10 minutes, there is again one more surface of 1 m2. So, the surface becomes even bigger (4 m2) but is should again follow the prescribed ramp. In other words, I want to change the area of fire surface (as fire advances through the building) but I want to generate the same amount of kW, that I prescribed with the ramp.

[dr_jfloyd]

The Q = AREA * HRRPUA * RAMP over all the burning surfaces you have defined in your input file. 

Q1 = AREA1 * HRRPUA1 * RAMP1

Q2 = AREA2 * HRRPUA2 * RAMP2

If Q1=Q2 and RAMP1=RAMP2, then AREA1/AREA2 = HRRPUA2/HRRPUA1.