energy balance

[ra az]

Hi all users,

I’m usually using for year’s commercial CFD software such as phoenics
and after that CFX (RANS) models. I used to check the mass & energy
balance output to control the accuracy of the model (conservation
laws)

I started using FDS since few years, but still not understanding its
balances, I made a simple example of a car park with 8MW fire and I
applied devices to measure heat flow at the exhausts plus temperature
measure volumes just one cell before the exhausts. I noticed the
following:

- using an 8MW fire I would expect a higher temperature than 570°C at
fire center
- the radiation model is turned off, model boundaries are adiabatic,
yet the HRR.csv file indicates that a 35% radiation fraction is still
active.
- The resulted mean temperature at the volume flow device is 130°C
- The resulted heat flow at the device on the exhaust objects measure
both total 3.2MW

The crazy thing now is that I’m using an 8MW source and exhaust of
50m3/s in the model.
- Why the heat outflow is measuring only 3.2 MW while I have an 8MW
fire plus the
energy that is coming with the air from outside, where is the rest of
the energy (about 5MW)?

- by measuring 130°C exhaust temperature this means we have a delta T
of 110°C, so calculating back the energy Q= 110x1004x50x0,87 = 4,8 MW,
where are the rest??

So please can somebody help me understanding how FDS is dealing with
the conservation of mass and energy, and why I get these strange
values.

Thanks,

[dr_jfloyd]

Whether or not you see flame temperatures is highly dependent on grid resolution.  If you have a coarse grid you will not get a flame temperature.  Consider the extreme of modeling a 1 m x 1 m x 1 m cube with a single grid cell and a fire that is a lit match.  Would you expect to see flame temperatures?

Please carefully read the FDS documentation (they are text searchable pdfs).  In Section 11.3 Radiation Transport:

It is possible to turn off the radiation transport solver (saving roughly 20 % in CPU time) by adding the statement RADIATION=.FALSE. to the MISC line. For isothermal calculations, the radiation is turned off automatically. If burning is taking place and radiation is turned off, then the total heat release rate is reduced by the RADIATIVE_FRACTION, which is input on the RADI line. This radiated energy completely disappears from the calculation. More on this feature can be found in Section 11.1.2.

Did you set the boundaries adiabatic by using ADIABATIC=.TRUE. or by using SURF_ID='INERT'.  The latter is a cold wall boundary and not an adiabatic condition. 

If your fuel source is 20 C, then unless you have set the heat transfer coefficient to zero for the burner you will have heat losses to burner surface from convection.

With 8 MW and the default 35 % radiative fraction, one would expect an energy flow of 5.2 MW. Your 4.8 MW is a 7 % error.  Without your exact inputs and what version of FDS you are running, etc.  why you are getting what your are getting is unknowable.