search

firemodels / fds

Fire Dynamics Simulator
https://pages.nist.gov/fds-smv/
Other
636 stars 613 forks source link

HRRPUA error #4789

Closed RakeshRanga closed 7 years ago

RakeshRanga commented 7 years ago

Please find the code where I am trying to place the fire on top of the PMMA slab. The code is running fine but the HRRPUA is not being realised. HRRPUA ERROR.pdf

drjfloyd commented 7 years ago

Please attached your input as a .txt file so we can work with it. Copying and pasting from a pdf is not a reliable method.

RakeshRanga commented 7 years ago

Respected Sir, Please find attached the txt file for the same.

Thank you. Regards, Rocky

On Fri, Mar 3, 2017 at 11:53 PM, Jason Floyd notifications@github.com wrote:

Please attached your input as a .txt file so we can work with it. Copying and pasting from a pdf is not a reliable method.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/firemodels/fds/issues/4789#issuecomment-284030886, or mute the thread https://github.com/notifications/unsubscribe-auth/AY6MeuTWqk4wJ72ZyjUFwXY8tb3q4wgWks5riFqSgaJpZM4MSk6c .

############### 1-3-2017 ###########

&HEAD CHID='PMMA VERTICAL',TITLE='VERTICAL BURNING STUDY' /

&MESH IJK=52,3,30,XB=0.014,0.066,0.005,0.0065,0,0.150 / &MESH IJK=26,3,20,XB=0.014,0.066,0.005,0.0065,0.150,0.250/ &MESH IJK=26,5,50,XB=0.014,0.066,0,0.005,0,0.250 / &MESH IJK=26,5,50,XB=0.014,0.066,0.0065,0.0115,0,0.250 / &MESH IJK=7,3,50,XB=0,0.014,0.005,0.0065,0,0.250 / &MESH IJK=7,3,50,XB=0.066,0.080,0.005,0.0065,0,0.250 / &MESH IJK=7,5,50,XB=0,0.014,0,0.005,0,0.250 / &MESH IJK=7,5,50,XB=0.066,0.080,0,0.005,0,0.250 / &MESH IJK=7,5,50,XB=0,0.014,0.0065,0.0115,0,0.250 / &MESH IJK=7,5,50,XB=0.066,0.080,0.0065,0.0115,0,0.250 /

&MISC RESTART=.FALSE. / &TIME T_END = 10.0 /

&SPEC ID ='PMMA FUEL',FORMULA='C5.0H8.0O2.0'/ &SPEC ID ='OXYGEN',MASS_FRACTION_0=0.232/ &SPEC ID ='NITROGEN',MASS_FRACTION_0=0.768/ &SPEC ID ='CARBON DIOXIDE'/ &SPEC ID ='WATER VAPOR'/

&REAC FUEL='PMMA FUEL' HEAT_OF_COMBUSTION=28750./

&MATL ID ='PMMA' CONDUCTIVITY = 0.20 SPECIFIC_HEAT = 2.2 DENSITY = 1100.0 EMISSIVITY = 0.95 N_REACTIONS = 1.0 NU_SPEC = 1.0 SPEC_ID = 'PMMA FUEL' A = 8.5E12 E = 1.8E5 HEAT_OF_REACTION = 870./

&MATL ID = 'ALUMINIUM' EMISSIVITY = 0.09 DENSITY = 7850. CONDUCTIVITY = 237 SPECIFIC_HEAT = 0.91 /

&SURF ID = 'VERTICAL_PMMA' COLOR = 'BURNT UMBER' BURN_AWAY = .TRUE. MATL_ID = 'PMMA' THICKNESS = 0.0015 /

&SURF ID ='Al' COLOR ='SILVER' BURN_AWAY =.FALSE. MATL_ID ='ALUMINIUM' THICKNESS =0.001 /

&SURF ID='FIRE',RAMP_Q='BURN', HRRPUA=1000./ &RAMP ID='BURN',T=0,F=0 / &RAMP ID='BURN',T=0.2,F=1/

Creating the PMMA block

&OBST XB=0.015,0.065,0.005,0.0065,0,0.150,SURF_IDS='FIRE','VERTICAL_PMMA','VERTICAL_PMMA' /

Creating the aluminium side coverings

&OBST XB=0.014,0.020,0.004,0.005,0,0.150,SURF_ID='Al' / &OBST XB=0.014,0.015,0.005,0.0065,0,0.150,SURF_ID='Al' / &OBST XB=0.014,0.020,0.0065,0.0075,0,0.150,SURF_ID='Al' / &OBST XB=0.060,0.066,0.004,0.005,0,0.150,SURF_ID='Al' / &OBST XB=0.065,0.066,0.005,0.0065,0,0.150,SURF_ID='Al' / &OBST XB=0.060,0.066,0.0065,0.0075,0,0.150,SURF_ID='Al' /

&VENT MB='XMIN',SURF_ID='OPEN'/ &VENT MB='XMAX',SURF_ID='OPEN'/ &VENT MB='YMIN',SURF_ID='OPEN'/ &VENT MB='YMAX',SURF_ID='OPEN'/

&SLCF PBX=0.03,QUANTITY='TEMPERATURE'/ &SLCF PBX=0.04,QUANTITY='TEMPERATURE'/ &SLCF PBX=0.05,QUANTITY='TEMPERATURE'/ &SLCF PBY=0.0055,QUANTITY='TEMPERATURE'/

&SLCF XB=0,0.080,0,0.0115,0,0.250,QUANTITY='TEMPERATURE' / &SLCF XB=0,0.080,0,0.0115,0,0.250,QUANTITY='TEMPERATURE',VECTOR=.TRUE. /

&BNDF QUANTITY='INCIDENT HEAT FLUX'/

&DUMP DT_RESTART=2. / &TAIL /

drjfloyd commented 7 years ago

If you are using the simple chemistry model, you should not specify the species for OXYGEN and NITROGEN. Simple chemistry uses the default species of AIR and by giving initial mass fractions for OXYGEN AND NITROGEN you now have no AIR present. If you want to specify your own air mixture, then you cannot use the simple chemistry model and need to fully define the reaction with NU_SPEC and NU.

You haven't specified a specific heat for PMMA so it is being computed using the default assumption that gamma = 1.4. This is a bad assumption for high molecular weight species as in your case it winds up giving a specific heat of 0.29 kJ/kg. It is more likely to be closer to 1.5 kJ/kg.

Don't make the block of PMMA it's own mesh. I also would recommend not having only a handful of cells for one of the IJK mesh directions. This is not how the multiple mesh approach was designed to work.

Avoid high grid aspect ratios. We do not recommend that you exceed 2 to 3.

RakeshRanga commented 7 years ago

Respected Sir,

I have specified the specific heat. It is 2.2 kJ/kg. Also my domain dimension along J th direction is only 11.5mm. How can I avoid having handful of cells? If I increase the cells to a large number i.e. make the grid size very fine, the computational time will increase tremendously.

Thank you. Regards, RakeshRanga

On Sat, Mar 4, 2017 at 1:25 AM, Jason Floyd notifications@github.com wrote:

If you are using the simple chemistry model, you should not specify the species for OXYGEN and NITROGEN. Simple chemistry uses the default species of AIR and by giving initial mass fractions for OXYGEN AND NITROGEN you now have no AIR present. If you want to specify your own air mixture, then you cannot use the simple chemistry model and need to fully define the reaction with NU_SPEC and NU.

You haven't specified a specific heat for PMMA so it is being computed using the default assumption that gamma = 1.4. This is a bad assumption for high molecular weight species as in your case it winds up giving a specific heat of 0.29 kJ/kg. It is more likely to be closer to 1.5 kJ/kg.

Don't make the block of PMMA it's own mesh. I also would recommend not having only a handful of cells for one of the IJK mesh directions. This is not how the multiple mesh approach was designed to work.

Avoid high grid aspect ratios. We do not recommend that you exceed 2 to 3.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/firemodels/fds/issues/4789#issuecomment-284054190, or mute the thread https://github.com/notifications/unsubscribe-auth/AY6MegyZps3igJD4LwbaEHGvV4x2NSIkks5riHAegaJpZM4MSk6c .

sbenkorichi commented 7 years ago

RakeshRanga,

I recommend you check section 8.4 & 8.5 of the User Guide. Jason is right, since you have specified HRRPUA along with the RAMP, you don't need to use complex pyrolysis model, in fact it's wrong to do so. I believe Kevin once said he will add a logic to prevent the user from using both models at the same time. I'm not sure if it's implemented in the latest version or not yet. If you leave only the simple pyrolysis model it would burn fine. However, this goes back to you what you are intending to achieve from this work. Have you conducted already experiments? If so, you might just specify your HRR evolution with RAMP, rather than working on complex model.

As for the Mesh, you might check section 6.3 from the user guide. Generally, you don't want your heat flux escape your domain. You simply can replace all these meshes with only one coarse mesh, if you are not have any reason. &MESH IJK=40,15,50,XB=0,0.080,0,0.0115,0,0.250 / , since you are not trying to specifying the processor, or threads. Actually, this would take less time than what you have above. A rule of thumb is to check Kris tool, however you still need to perform sensitivity analysis and verify the grid resolution. http://www.koverholt.com/fds-mesh-size-calc/ Note: What Jason meant by uploading files with txt, is just to change the extension of the file instead of case.fds to case.fds.txt . Not pasting it the whole text file here.

Hope this helps. Kind regards, Salah