Conservation of mass

[lganeshs]

Hi all,

I am a new FDS/Pyrosim user. As a simplification of my problem, I have simulated a closed box with a single inlet. The inlet is modeled as a vent located on an internal wall of the box which pumps air (green line in plots below)at 0.2m/s. The simulation is allowed to run for 100s. I have extracted the 3-D density data at every 5s interval to compute the total mass of gas inside the box. I expect the mass to increase in a linear fashion but the mass increases non-linearly. I repeated the simulation with propane (blue line in plots below)pumped into the box in place of air and the mass increases linearly. Kindly refer to the plots below. The figure on top tracks the mass of fuel in the box and the figure below shows the total mass inside the box. I have repeated the simulations, by specifying equivalent volume flow rate in place of velocity and the same behavior is repeated.

How do I resolve this problem? Do let me know if any other details are required. Thank you in advance

[marcosvanella]

Can you post the input file?

[drjfloyd]

VOLUME_FLOW and VEL both specify a fixed velocity. They do not specify the density. The density is given by the surface temperature and zone pressure at the elevation of the vent. As you add mass, the pressure goes up so over time the density of the air being injected goes up, If you want to have a constant flow of mass then use MASS_FLUX.

[lganeshs]

I shall try to explain the problem using a simpler configuration.A closed box with just one fuel inlet. I have run 2 simulations: one with Propane as fuel and other with wood. The mass flux is set at 0.0245kg/m2s. So, I expect the mass increase in the box from t=0s to t=300s should be the same. I have set up a stat device measuring volume integral of density i.e. total mass (labelled mass in devc.csv) and mass integral of mixture fraction(labelled mixture frcation in devc.csv). The plot comparing the total mass from 2 the two simulations is given below. I have also attached the two fds files. My question boils down to why is mass integral of mixture fraction different at 300s when the inlet mass flow rates match.

![image](https://github.com/user-attachments/assets/c4d60615-bc46-4cdc-b
Archive 2.zip
7f8-c3b4455fff2f)

[drjfloyd]

You were specifically requested to provide the exact input files you are running. You have not done that. This is what happens when I try to run the files you just attached:

C:\Projects\FDS_test>fds simple_propane.fds

 Starting FDS ...

 MPI Process      0 started on L-PW078R66

 Reading FDS input file ...


ERROR(605): SURF_ID fuel_inlet does not exist. (MPI Process: 0, CHID: simple_propane)

ERROR: FDS was improperly set-up - FDS stopped (CHID: simple_propane)

C:\Projects\FDS_test>

[lganeshs]

Hi,

I am really sorry. I uploaded the wrong files. Please
Archive 3.zip
find attached the correct file.

[drjfloyd]

Propane has an MW of 44 g/mol and your wood gas has an MW of 26 g/mol. Say you turned off combustion by setting Y_O2_INFTY to a very low value or setting AUTO_IGNITION_TEMPERATURE to a high value. Eventually you would fill the box up with pure propane or pure wood gas. Temperature is going to be the same in both cases and pressure will be the same since the domain is OPEN. Will you have the same mass of fuel inside the box for both cases?

[lganeshs]

No I wouldn't expect it. I was getting confused earlier. Thank you so much for the clarification.

[lganeshs]

I ran the simple case using mass flux and it works. Now, I would like to simulate a 1D heat conduction on the walls of the box with a small opening on the ceiling . So, i have extended the mesh boundaries beyond the walls and changed the external boundary conditions to open vent. In this case, when I try to set air inlet using mass flux, i get an error that background species cannot be used.

[mcgratta]

Is there an error code? Have you looked at the User's Guide for that error code?

[lganeshs]

I shall check for the error code

[lganeshs]

Hi Mcgratta,
I was unable to get the error code. On defining the location of the vent against an obstruction, I stopped getting the error.
I am not sure if the two things are connected.

[mcgratta]

This is getting confusing, and it is hard to follow a thread that extends over several weeks. If you have a simple case that produces a clearly erroneous output or an error that you do not understand. Start an issue on the Issue Tracker so that one person can be assigned to it and work it to completion. I am closing this discussion.