search

firemodels / fds

Fire Dynamics Simulator
https://pages.nist.gov/fds-smv/
Other
650 stars 618 forks source link

HRR dropping sudenly #760

Closed gforney closed 9 years ago

gforney commented 9 years ago 
Please complete the following lines...

Application Version:5.2.3
SVN Revision Number:
Compile Date:23.06.2009
Operating System:MS XP sp2

Describe details of the issue below:

I am trying to simulate 0.3 dia Diesel oil open pool fire.I am getting some
non physical output.
HRR and Mass burning rate dropping to zero within five second inspite of
having enough fuel in pan.Fuel layer thickness is 0.141m.

I dont know whats going wrong. pls help

Looking forward for your support

Regards..
Saumil

Original issue reported on code.google.com by saumil2525 on 2009-06-23 05:29:02

gforney commented 9 years ago 
Diesel fuel has a relatively high flash point, meaning that it does burn as easily 
as other liquid hydrocarbons. I am not sure that a pool of diesel fuel will self-
ignite with just a pilot. In FDS, the liquid fuel evaporates and combines with 
oxygen and burns regardless of the temperature. It is as if there is a pilot light

or spark in the calculation. However, in order to sustain the reaction, liquid fuel

must be heated up and continue evaporating. This is not happening in your case, but

I am not sure if the problem is FDS or if you had a pool of diesel fuel it would 
burn with no external heating. Do you have test data to support the selection of 
input parameters and the validation of the burning rate?

Original issue reported on code.google.com by mcgratta on 2009-06-23 13:56:29

gforney commented 9 years ago 
yes sir
 we have conducted some experiments on diesel pool. We have measured mass loss
rate,radiative flux at various height&locations and now planning to measure
centerline temperature.

we have fixed parameters like numerical grid size and solid angle after conducting
sensitivity analysis.Fluid properties are taken from material safety data sheet.
only suspicious input parameters are absorption coefficient,co yield ,h2 yield,soot
yield.
To start with i have taken 10% soot yield for the diesel(for diesel its od the order
of  0.1-0.16).Also I tested above scenario at absorption coefficient of
40,100,250,500 but with no luck .

As per our experimental data mass burning rate should be around 0.01289 kg/m^2-sec.ie
HRR~552 kW/m^2

Regards..
Saumil

Original issue reported on code.google.com by saumil2525 on 2009-06-25 10:38:08

gforney commented 9 years ago 
Assuming all the material properties are reasonable, it may still be difficult to 
match the experiment. There may be a limitation in the physical model. I'm going to

pass the issue to Simo Hostikka, who might have some suggestions.

Original issue reported on code.google.com by mcgratta on 2009-06-25 22:08:46

gforney commented 9 years ago 
FDS's liquid model does not work very well for diesel because the model tries to
ignite the pool by quessing an initial value for the evaporation rate -- afterwards
the evaporation rate is ajusted to keep the gas partial pressures in equilibrium.
This is the initial peak you see in the HRR. 

During this initial peak, the pool surface temperature increases only to 80...90
degC. For many liquids, this would be high enough to start continuous burning, but
for diesel it is not.

How do you ignite the pool in reality? How long time you must apply a heat source?

Suggestion: To fix the problem, I will implement a new feature
AUTO_IGNITE = .FALSE. on the SURF line. Setting that, there will be no initial
quessed burning rate. Instead, you need to explicitly heat up the surface to make it
burn.

Original issue reported on code.google.com by shostikk on 2009-06-26 06:54:15

gforney commented 9 years ago 
In reality we are pouring few 5ml of Heptane on diesel pool and we lit up heptane
with match.

Instead of putting heat source can we put a small layer of heptane on diesel pool or
can you suggest any other method of applying instantaneous heat source which doesn't
intrude diesel pool.

Regards..
Saumil

Original issue reported on code.google.com by saumil2525 on 2009-06-26 11:56:59

gforney commented 9 years ago 
I'm not sure if the layered liquid fuel would work or not. Please give that a try and
thell what happens.

Original issue reported on code.google.com by shostikk on 2009-06-26 12:01:15

gforney commented 9 years ago 
I have tried 1 mm layer of heptane over diesel pool & It worked. still not sure whether
heptane has got consumed or not(forgot to put thickness measurements device), 
may be by tomorrow I will get some results.Also I tried to lit up diesel with some

constant temperature heat source of 2000 degree Celsius which was kept on pool for
3 
sec but FDS prompted numerical instability .Reason might be sharp temperature 
gradient.

Regards..
Saumil

Original issue reported on code.google.com by saumil2525 on 2009-06-27 16:48:23

gforney commented 9 years ago 
Dear sirs

I ran Diesel pool fire simulation with 0.5 mm thick layer (~37ml) of heptane on 
diesel pool.I have got some results which are attached here with however simulation

was ran with very course grid and results are not matching with our experimental data

but the primary concern of fuel self ignition is sorted out.

I am very thankful to both Dr kevin and Dr Hostikka for giving valuable inputs.

I just want to know that 
1)I have defined 2 reaction simulteneously.Is it valid? 
2)we are not giving any initial guess for the evaporation rate then how initial guess

is taken for different fuel& pool dia? Is there any roll of equn 3.94 of FDS tech 
guide for calculation of initial guess?

Regards...
Saumil

Original issue reported on code.google.com by saumil2525 on 2009-06-28 13:06:11

gforney commented 9 years ago 
What do you mean "2 reaction simultaneously"? Do you mean two REAC lines? If so, 
then the answer is no, you cannot use two reactions.

There are no parameters that you can control to influence the evaporation algorithm.

Original issue reported on code.google.com by mcgratta on 2009-06-30 17:19:11

gforney commented 9 years ago 
I added a new feature: INITIAL_VAPOR_FLUX on MATL line, which can be used to control
the initial 'guess' value of evaporation. Default is 5E-4 m3/m2. 

I also ran your diesel pool case, and noticed that we are getting HRR of about
1500...2000 kW/m2. Your expriment showed 550 kW/m2.

Based on the ethanol pool tests, I have found that this liquid model is currently
extremely difficult to get converged results. The reason is that when we approximate
the 'on-the-surface' value of vapor partial pressure, we use weighted sum of first
grid cell and the 'wall' values (Y_W). These values are often order of magnitude
different, and the results depends strongly on weighting.

Original issue reported on code.google.com by shostikk on 2009-07-01 13:19:03

gforney commented 9 years ago 
What does m3/m2 mean? Should it be m3/s/m2?

Original issue reported on code.google.com by mcgratta on 2009-07-01 13:39:00

gforney commented 9 years ago 
Oops. I must fix the guides too...

Original issue reported on code.google.com by shostikk on 2009-07-01 13:47:07

gforney commented 9 years ago 
I tried to enter this feature but end up with some error.Can you pls send the .fds 
file.As I have been using Pyrosim as preprocessor I am facing difficulty with fds code
syntax.

Thanks 
Regards...
Saumil

Original issue reported on code.google.com by saumil2525 on 2009-07-02 17:28:14

gforney commented 9 years ago 
This version of FDS has not been released. You must wait for 5.4.0.

Original issue reported on code.google.com by mcgratta on 2009-07-02 17:33:36

gforney commented 9 years ago 
In above case I have taken the radiative loss fraction as 0.35 and trying to match
burnrate changing the absorption co efficient of the liquid fuel & probably will be
in position to match simulated burn rate with the exp one.

But when I took radiative fraction as 0 , burn rate fell dramatically to 4e-4 kg/sec.
while simulation with same input parameter except radiative loss fraction (when taken
as 0.35)gives  2.7e-3 kg/sec as burn rate.so i changed absorption coefficient of
liquid to 200 but the burn rate was almost same.(insensitive to liquid absorption
coefficient)

Is it because of grid size ? (.fds file is attacheched here with)

In fds tech guide only one statement is given on liquid aborption coefficient.Heat
balance given in book by Dougal Drysdale pp162 eqn 5.8 uses absorption co efficient
of flame rather liquid.can you please throw some light on the issue?

Original issue reported on code.google.com by saumil2525 on 2009-07-15 04:46:28

gforney commented 9 years ago 
Saumil,
Have you been able to run the diesel fires using the latest release?

Simo

Original issue reported on code.google.com by shostikk on 2009-12-15 09:50:56

gforney commented 9 years ago 
hello sir
I have done gasoline with new release V5.4. I am facing some problem but i haven't

tried mass burning rate prediction. 

I request you to follow:

http://code.google.com/p/fds-smv/issues/detail?id=909&sort=-id  

since i was not in position to upload the slice files (which are 22mb each) our 
discussion halted.It will be great if you can give some input on issue 909.

Regards.
Saumil

Original issue reported on code.google.com by saumil2525 on 2009-12-26 15:11:12

gforney commented 9 years ago 
I'm changing this to OnHold, since the active work is going on in issue 909.

Original issue reported on code.google.com by shostikk on 2010-03-02 11:08:19

gforney commented 9 years ago 
I checked that FDS 6 close-to-beta (svn 13550) runs this case, and gives HRR between
40 and 120 kW during the first 100 s. Further evaluation needs to be done using the
best estimate for diesel absorption coefficient.

I close this issue.

Original issue reported on code.google.com by shostikk on 2012-10-31 11:46:48