CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   Area weighted or mass weighted average (http://www.cfd-online.com/Forums/main/8151-area-weighted-mass-weighted-average.html)

SAM October 1, 2004 05:01

Area weighted or mass weighted average
 
I wonder wat is difference b/w the area weighted and mass weighted average values of any varaible.

and which one is better to use and in what type of situations

Regards

Apurva October 1, 2004 09:33

Re: Area weighted or mass weighted average
 
Hi,

For pressure use area weighted averaged value. For rest mass averaged is fine.

Thanks

Apurva

Jon October 1, 2004 11:55

Re: Area weighted or mass weighted average
 
Won't this depend on the profile through your boundary?

If you want ave vel. you want area ave, if you want ave. temp. you will want Mf ave.

centaur_ks October 2, 2004 18:44

Re: Area weighted or mass weighted average
 
consult. "Computational Fluid Dynamics" - T J Chung. This things are better explained in that. hope this helps

Jan Rusås October 4, 2004 03:55

Re: Area weighted or mass weighted average
 
For incompressible flow it is physical correct to use mass average for the pressure. For velocities you should use area weighted. You can also do a search in the forum, it has been discussed before.

sam October 4, 2004 04:21

Re: Area weighted or mass weighted average
 
thank i will look into it

but any body else can give me more suggestion , coz my experience says that both normally gives same resutls

Junseok Kim October 4, 2004 04:33

Re: Area weighted or mass weighted average
 
Take a look at below paper.

Junseok

INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN FLUIDS Int. J. Numer. Meth. Fluids 2003; 41:551–560 (DOI: 10.1002/d.455) Interfacial transfer in Tryggvason's method Joel H. Ferziger∗; † Department of Mechanical Engineering; Stanford University; Stanford CA; U.S.A. SUMMARY Tryggvason's method computes two phase ows by smoothing the regions near the boundaries between the two uids. This requires averaging of the uid properties, which may dier by orders of magnitude and can lead to large errors. Traditional linear averaging produces a rst order error that can be extremely large. In this paper, we show that for a number of simple test cases, averaging of the inverses of the properties can lead to a signicant improvement in the quality of the results.

Jan Rusås October 4, 2004 05:31

Re: Area weighted or mass weighted average
 
Regarding the average methode for pressure, you can find that from the energy equation. It should actually be volume flow averaged, but if the density is constant you can use mass flow average. Only if the velocity is near constant, same mass flow at each position, will result in same value for mass flow and area average.

Jon October 4, 2004 06:29

Re: Area weighted or mass weighted average
 
If the flow is reasonably spatially constant w.r.t. the variable you are intersted in AND the velocity then both will be similar.

If not, see above.

sam October 4, 2004 10:33

Re: Area weighted or mass weighted average
 
in fluent how can we take the inverse (as stated in above mentioned paper )

I will grateful to u for ur help

Regards

sam October 4, 2004 10:35

Re: Area weighted or mass weighted average
 
can u send me papers or any refernce in fluent users guide .


Junseok Kim October 4, 2004 12:42

Re: Area weighted or mass weighted average
 
I don't use fluent.

sam October 5, 2004 03:30

Re: Area weighted or mass weighted average
 
OR DO U HAVE ANY REFERECNE

deepesh nanda October 14, 2004 08:00

Re: Area weighted or mass weighted average
 
Dear friend, U use area weighted avg. when ever u r using incompressible flow. The reason is that the density of the fuild is constant from inlet to outlet and hence area weigh. avg. will do. But for incompressible flow the density vary with temp and pressure and hence during that time u should used mass weight avg. All the best. with regards deepesh

SAM October 14, 2004 12:59

Re: Area weighted or mass weighted average
 
thanks for ur reply

but i have to give reason for this in my thesis , could u please give some references

i will be grateful to u

regards

deepesh October 16, 2004 08:53

Re: Area weighted or mass weighted average
 
Buddy, If u r using fluent software then u can go to the fluent help file where they clearly give the difference and the reason for the same. By chance if u do not have the same, mail me, i will mail the required document.

all the best,

deepesh nanda

SAM October 16, 2004 08:54

Re: Area weighted or mass weighted average
 
please mail me

deepesh nanda October 18, 2004 12:16

Re: Area weighted or mass weighted average
 
Correction,

I replied for the area/mass weight avg, u use mass weight avg. only when u have compressible flow becoz. density varies with pressure and temperature for such flow. i will soon mail u the document.

thanks

chauhanji January 31, 2011 02:40

area weighted or mass weighted average
 
I have tried for both area weighted and mass weighted average for my problem. And I am giving incompressible flow, still the averages values are not same which contradicts many of u all posts here above.
Can any one please counter check and tell us what is the reason and with correct explanation.

jola January 31, 2011 06:31

Area-averaging and mass-averaging will give different results and what you should use depends on what variable you are interested in. For conserved variables like total-pressure you should use mass-averaging, whereas for other variables, like static-pressure, you should use area-averaging. Check out some of the refences mentioned.

To make things even more complex another type of averaging often used is flux-averaging. With flux-averaging all gradients are mixed out and a flux-averaged value is the value you would get if you measure at the outlet of an infinently long friction-free pipe put at the boundary. In turbomachinery applications flux-averaging is common, and sometimes you do not flux-average everything, you do flux-averaging circumferentially to mix-out wakes, but you do not do flux-averaging radially because you do not want to mix-out the end-wall boundary layers.


All times are GMT -4. The time now is 10:25.