Temperature inside the solid wall?

[RNstu08]

Hello Everyone,

I would like to know how the temperature is transmitted through a wall and how we measure the temperature inside the wall.

We have a case where two rooms are separated by a wall (COMBUSTIBLE) and have only one common door without any windows or mechanical ventilation. The fire started in room 1, and we would like to understand its effects on the other part of the room. We are interested in measuring the temperature inside the wall and the surface temperature as well.

I had used different measuring techniques that are available on FDS, such as 
1. Wall temperature
2. Inside wall temperature
3. Adiabatic surface temperature
4. Thermocouples

Only the adiabatic surface temperature shows some deflection for IOR = 2 on the graph, while the other devices didn't work at all.

I would be happy if someone shed light on these topics. If required, I will share the FDS input file.

Thank you everyone in advance.

[fde]

See user guide:  21.2.3 Quantities at Certain Depth  

Also see example fds file in Heat transfer/convective cooling

[dr_jfloyd]

Adiabatic surface temperature is really a measurement of the heat flux to the wall. There is really no way to know what is happening without an input file (as simple an input as possible that still shows the problem). If wall temperature or inside wall temperature is not showing a temperature rise, then you may want to check for the following:

1. Did you use the correct XYZ for the DEVC?

2. Did you correctly apply a SURF that has thermophysical properties? If you have applied a SURF containing TMP_FRONT, then that will set the wall temperature to be fixed.

3. FDS does not transfer heat between different OBST that are touching (unless you are using HT3D). If you are trying to measure temperature in one OBST that is attached to a second OBST that is seeing the heat from the fire, then this will not be successful.

4. If you did correctly apply a SURF with thickness, conductivity, density, and specific heat, do those inputs have the correct units? For example, the FDS input for specific heat should be in kJ/kg/K and if you input J/kg/K then your surface will heat 1000 times more slowly. 

[RNstu08]

I have added the input file. This input file is working for the IWT device as I reduced the wall thickness to the size of the grid. However, I am wondering if my model has other geometries with even smaller dimensions than the wall thickness, then I can't maintain the wall thickness equal to the grid size. It would drastically increase the mesh size or If I keep the grid equal to the wall thickness, then I will not capture the important phenomena happening at the boundary layer of the smaller dimension objects. For example, in the input file, the wall thickness is 0.1 m, which is equal to the grid size. However, the object crib has a dimension of 0.05 m.

I found your 3rd point interesting. Would you care to elaborate?

[dr_jfloyd]

If you want to see heat transferred through an obstruction, it needs to be one grid cell thick or less.  However, THICKNESS on SURF is independent of the grid size. If you have a 5 cm grid, you can still assign the OBST a 10 cm THICKNESS on the SURF.

11.3.3 Back Side Boundary Conditions

The layers of a solid boundary are listed in order from the surface. By default, if the obstruction is less than
or equal to one cell thick, then the innermost layer will be exposed to the air temperature on the back side.
If the obstruction is on the boundary of the domain or is more than one cell thick, then it is assumed to back
up to an air gap at ambient temperature. For example, a thin steel plate (i.e. thickness less than or equal to
the grid) would use the FDS predicted temperatures on either side of the plate for predicting heat transfer.
There are other back side boundary conditions that can be applied. One is to assume that the wall
backs up to an insulated material in which case no heat is lost to the backing material. The expression
BACKING=’INSULATED’ on the SURF line prevents any heat loss from the back side of the material. Use
of this condition means that you do not have to specify properties of the inner insulating material because it
is assumed to be perfectly insulated.

If the wall is assumed to back up to the room on the other side of the wall and you want FDS to calculate
the heat transfer through the wall into the space behind the wall, the attribute BACKING=’EXPOSED’ should
be listed on the SURF line. This feature only works if the wall OBST is less than or equal to one mesh
cell thick, and if there is a non-zero volume of computational domain on the other side of the wall OBST.
Obviously, if the wall is an external boundary of the domain, the heat is lost to an ambient temperature void.
The same happens if the back side gas cell cannot be found (i.e. the wall OBST is greater than one cell thick).

[RNstu08]

I understand that the thickness of the surf is independent of the grid.

However, in the input case, if I keep the grid size at 0.1m with the obstruction wood_wall at 0.1m, the IWT device for the temperature measurement inside the obstruction works. If I increase the grid size to 0.05m, then the IWT does not work at all.

[dr_jfloyd]

THERMOCOUPLE is gas phase only. If you put this inside of a wall, you will just see TMPA.

Your DEVC have y-coordinates of 2.24 and 2.22.  With a 10 cm grid your grid boundaries are at 2.2 and 2.3 With a 5 cm grid your boundaries are 2.2 and 2.25. With 10 cm both DEVC are closer to the front face of the wall. This is not the case with a 5 cm grid as you now have two cells for your wall thickness.  Try making the y-coordinate of your DEVC 2.2.

[RNstu08]

I tried your recommendation but IWT device still does not work. It only works when grid size is equal to wall thickness which would be 10cm. As I mentioned before, If I make the domain with 10cm grid size, the important phenomena happening at the objects smaller than 5cm would not capture well. other thing, is when grid size = wall thickness, then both the IWT devices works even though they are closer to each other.

[dr_jfloyd]

Per the User's Guide Section discussion for the use of INSIDE WALL TEMPERATURE, did you assign a DEPTH for your IWT device in your latest input file?

[RNstu08]

Yeah, it certainly works after mentioning the depth. Thank you for your input.

I have additional questions about the same input file.

I decided to check the burning rate of the wall. Therefore, I have added density, burning rate, and wall thickness. However, nothing works as of now.

Am I using the correct devices to find it or are there any other ways to determine it? I would also like to know how the charring rate and char depth can be calculated.

[Kevin]

"Therefore, I have added density, burning rate, and wall thickness. However, nothing works as of now."

If you want help, you have to be more specific than this.

[RNstu08]

I mean to the same input file I have added the device to measure the density, burning rate, and wall thickness along the wall over time.

[dr_jfloyd]

Your issue before was you did not follow the instructions in the User's Guide. I assume that in this case that you have also incorrectly specified the inputs, but how can we possibly know what you did wrong without knowing exactly what you did? 

[RNstu08]

I have added the following devices in addition to the previous ones (attached input file below) to find the burning rate, density, and wall thickness. I used two different techniques. However, none of them has worked so far. My question is, Am I using the correct devices to find it or are there any other ways to determine it? I would also like to know how the charring rate and char depth can be calculated.

&DEVC ID='DENSITY_0', QUANTITY='DENSITY', XYZ=1.5,2.2,1.175/
&DEVC ID='DENSITY_1', QUANTITY='DENSITY', XYZ=1.5,2.218538,1.175/
&DEVC ID='DENSITY_2', QUANTITY='DENSITY', XYZ=1.5,2.245326,1.175/

&DEVC ID='WALL_THICKNESS_1', QUANTITY='WALL THICKNESS', XYZ=1.5,2.218538,1.175, IOR=-3/
&DEVC ID='WALL_THICKNESS_2', QUANTITY='WALL THICKNESS', XYZ=1.5,2.245326,1.175, IOR=-3/

&DEVC ID='NET_HEAT_FLUX', QUANTITY='NET HEAT FLUX', XYZ=1.5,2.218538,1.375, IOR=-2/
&DEVC ID='HRR', QUANTITY='HRR', XB=0.0,3.0,-0.6,3.7,0.0,2.5/
&DEVC ID='FLOW', QUANTITY='HEAT FLOW', XB=2.0,2.7,0.0,0.0,0.0,2.0/

&DEVC ID='Wall Thickness_MEAN', QUANTITY='WALL THICKNESS', SPATIAL_STATISTIC='MEAN', XB=0.8,2.2,2.2,2.3,0.0,2.0/
&DEVC ID='Density_VOLUME INTEGRAL', QUANTITY='DENSITY', SPATIAL_STATISTIC='VOLUME INTEGRAL', XB=0.8,2.2,2.201975,2.3,0.0,2.0/
&DEVC ID='Burning Rate_SURFACE INTEGRAL', QUANTITY='BURNING RATE', SPATIAL_STATISTIC='SURFACE INTEGRAL', XB=0.8,2.2,2.2,2.3,0.0,2.0/

[Kevin McGrattan]

Read about the PROFile output quantity in the User's Guide.

[RNstu08]

Yeah, but using PROF can only be used to find three quantities, such as density, temperature, and the density of the material components. What about the other things that I mentioned earlier? And what is the difference between the DEVC and PROF? Since both density and temperature quantities can be determined using these two types.

[Kevin McGrattan]

Run this case

https://github.com/firemodels/fds/blob/master/Verification/WUI/ground_vegetation_load.fds

Then study its inputs and outputs. Get back to me with questions if this does not explain things.

[RNstu08]

I have seen the case file and also ran the case. That answers some of my questions. However, I am still wondering how to determine the char depth (mm) and charring rate (mm/min).

[dr_jfloyd]

Your solid phase reaction is not producing char; therefore, there is no char depth to measure.

[RNstu08]

  Yes, this input file does not have a solid phase reaction for the char production. I tried to include the char in this case. However, when I include it, I get the material density of char using PROF output quantity. My question is, how to find the char depth (mm) and charring rate (mm/min)?  

[Kevin McGrattan]

These are not direct outputs. You can infer them from the PROF file.

[RNstu08]

Would you elaborate on that?