How to derive full size fire from 2D cylindrical FDS

[gforney]

Hi

I have made an 2D fds file and would like to derive the full size fire from these calculations
but it seems:

-HRR_FDS = A_square * HRRPUA

where:
A_square = (y_max-y_min)*r
r = radius of the fire

Shouldn't it be:
-HRR_FDS = A_circle_segment * HRRPUA

where:
A_circle_segment = (y_max-y_min)/(2*PI*r)*PI*r^2 
                 = 1/2*(y_max-y_min)*r

??

One last thing:
-How to check for cylindrical coordinates in the .out file?

Original issue reported on code.google.com by Kindler.85 on 2014-04-07 11:30:55

[gforney]

I'll take a look at it.

Original issue reported on code.google.com by mcgratta on 2014-04-07 12:51:12

[gforney]

Issue 2128 has been merged into this issue.

Original issue reported on code.google.com by mcgratta on 2014-04-07 12:54:45

[gforney]

FDS takes HRRPUA and divides it by the heat of combustion to derive the mass flux rate
of fuel per unit area:

HRRPUA (kW/m2) / HoC (kJ/kg) = mdot'' (kg/m2/s)

So the mass flux, and hence the convection/diffusion of fuel at the surface, is determined
purely from HRRPUA. Given that the equations are solved in cylindrical coordinates,
the total HRR is thus the HRRPUA times the area of the circle of radius r.

I'm going to cc some others to confirm if this is a correct interpretation.

Original issue reported on code.google.com by mcgratta on 2014-04-07 21:35:20

[gforney]

It seems that the Y-coordinates in cylindrical co-ordinates are treated as the dimension
of the mesh at X = 1, not at X=XMAX.

Therefore, 
A_fire_circle_segment = (pi r**2) * (y_max-y_min)/(2*pi), where r is the fire's radius
and inner radius is 0.

Original issue reported on code.google.com by shostikk on 2014-04-08 06:03:28

[gforney]

I added a test case to the Verification Guide called cyl_test. It's a case similar to
the one posted in this Issue, with dy=2 mm and dy=2 cm. If we just blow hot air out
of a vent, the two simulations are exactly the same. However, if there is combustion,
they are close, but not exact. I suspect this has to do with estimates of cell mass
in fire.f90, but I can't pin it down. 

Original issue reported on code.google.com by mcgratta on 2014-04-08 20:28:20

[gforney]

The result should be independent of y_min, y_max.

Each "cell" should be thought of as a ring of thickness DZ.  The volume of the cell
is

DV = DZ * \int_{0}^{2pi} \int_{r0}^{r1} r dr dtheta = DZ * pi * (r1^2 - r0^2)

I need some time to convert r0, r1 into X(I), etc., fix fire and then the HRR output.
 In the HRR file, we should integrate Q * DV and this should be independent of the
y dimension in the input file.

Original issue reported on code.google.com by randy.mcdermott on 2014-04-08 20:53:39

[gforney]

I added another case to the Verification Guide -- a simple DNS laminar diffusion flame.
It worked -- there was no dy dependence.

Original issue reported on code.google.com by mcgratta on 2014-04-08 23:23:23

[gforney]

Sorry, I lost track of this issue. Currently in FDS, when the calculation is in cylindrical
coordinates, y_max - y_min is interpreted as dtheta. The total HRR is computed

int qdot''' r dr dtheta dz

In the attached test case, the fire's radius, R, is 0.5 m. The HRRPUA is 250 kW. The
area of the entire circle is pi*R^2 and the HRR of the entire fire would be 250*pi*R^2
= 196.3 kW. However, dtheta=0.02 radians, which means that the FDS reported HRR for
this slice of the circle is 

0.02/(2*pi)*196.3 = 0.625 kW

I will add some clarification to the User's Guide.

Original issue reported on code.google.com by mcgratta on 2014-10-15 18:21:57

---

• _Attachment: cyl_hrr.fds_

[gforney]

(No text was entered with this change)

Original issue reported on code.google.com by mcgratta on 2015-02-06 18:41:47