carbon monoxide simulation

[andrew]

Hi,everybody!

I'm modelling a gasoline fire in a closed-end tunnel and the fire is
located near the end.One of my purposes is to get better understanding
of the carbon monoxide distribution along the tunnel.And there are
three questions remained:

1 )As the fuel is 90# gasoline,which means octane(C8H18) accounts for
90% and heptane(C7H16) 10%.So,on the &REAC line ,I consider 90#
gasoline as pure C7.9H17.8,is it OK to be addressed like that?


2)Since the combustion condition is most probably under-ventilated,so
I add CO_PRODUCTION=.TRUE. to the &MISC line,can this method produce
more accurate results?


3)I set SOOT_YIELD=0.1 for this fuel,is it appropriate with this
magnitude?


I attached the fds input data(CO.fds) to the file group.


Thanks a lot for your help!

[Kevin]

>
> 1 )As the fuel is 90# gasoline,which means octane(C8H18) accounts for
> 90% and heptane(C7H16) 10%.So,on the &REAC line ,I consider 90#
> gasoline as pure C7.9H17.8,is it OK to be addressed like that?

Yes, this is how you combine the different fuels.

>
> 2)Since the combustion condition is most probably under-ventilated,so
> I add CO_PRODUCTION=.TRUE. to the  &MISC line,can this method produce
> more accurate results?

CO_PRODUCTION is going to allow for CO production that can be much
higher than a fixed CO_YIELD. However, you still need to specify
CO_YIELD as the "post-fire" value. Read more in the User's Guide.

>
> 3)I set  SOOT_YIELD=0.1 for this fuel,is it appropriate with this
> magnitude?
>

There was a long discussion on this point several weeks ago. Search
for it.

[andrew]

Thank you very much ,Kevin.

[Chris]

I don't really understand what is meant by post-flame yield of CO. You
mean the CO yield of a post-flash-over fire can not be predicted
accurately or can be much higher? That means the value of CO should be
treated carefully and compared against ventilation-conditions? And in
the User Guide it says the CO_YIELD that is expected of a well
ventilated fire has to be entered anyway. I understand this point.
What I did now was to implement the CO_PRODUCTION=.TRUE. to the
device_test.fds-TEST-CASE. Is this approach ok? The default CO yield
value of a Propane-Burner is 0.0.
So without CO_PRODUCTION=.TRUE. I have just a single-step reaction
with no CO production at all.

> > for it.- Zitierten Text ausblenden -
>
> - Zitierten Text anzeigen -

[Kevin]

Within the flame envelop (or within a flashed over compartment), you
will find CO concentrations that are well in excess of what would have
been predicted using only a fixed yield of about 1%. However, if the
compartment opens up to a well-ventilated space (like the outside),
much of the excess CO is converted to CO2 at the flame envelop. But
not all, and the amount of CO that survives would be difficult for FDS
(or any model for that matter) to predict. So the "post-flame" CO
yield tells FDS how much of the CO NOT to convert.

> > - Zitierten Text anzeigen -- Hide quoted text -
>
> - Show quoted text -

[Ankit]

HRRPUA for Gasoline and Ethanol are known. How can we define HRR for mixture of these two fuels(E90 i.e. 90% Ethanol by volume mixed with 10% gasoline ) Should we take weighted average mean of HRR of both fuels ?

[dr_jfloyd]

Yes, you should produce the same total heat release for the combined fuel as you would if you burned the same masses of fuel separately and added the heat releases together.