compressible flow over the front wing

[a6zz]

Hi

I am modelling compressible flow, with boundary conditions including:
mass flow inlet at the inlet and pressure outlet at the outlet of the wind tunnel. Both of these conditions ask me for inital gauge pressure? has anyone come across this or know what I should put in this?

Abz

[a6zz]

just to add to above, I am modelling with air speed of 95 m/s which is compressible flow and subsonic. And I believe it will remain compressible and subsonic after the flow has gone past the front wing.

[slimjim8201]

At 95 m/s why don't you just assume incompressible. Even in the higher velocity zones around the wing, I wouldn't expect the air speed to go up above 150-200 m/s, still valid with an assumption of incompressible.

[kilcoo316]

I am modelling compressible flow, with boundary conditions including:
mass flow inlet at the inlet and pressure outlet at the outlet of the wind tunnel. Both of these conditions ask me for inital gauge pressure? has anyone come across this or know what I should put in this?

Sounds like CFX?


You'd be safer using a total pressure inlet boundary condition and static pressure outlet.

[Scotracer]

As slimjim said, I'd assume it is incompressible as you are at about 0.25 Mach number.

[a6zz]

thanks everyone

Anything above 0.3Mach(~100m/s) is considered compressible...so it would reach speeds will above this around the wing, as my inlet is about 95m/s and I have been told by a Professor to use compressible flow

and Kilcoo I am using CFD(fluent 6.3.26), all it lets me to do is put in gauge pressures?

[Jersey Tom]

Why not ask your professor or a TA?

[kilcoo316]

and Kilcoo I am using CFD(fluent 6.3.26), all it lets me to do is put in gauge pressures?

Ahhh, its been too long since I've used fluent...


IIRC you set a total gauge pressure and an initial pressure?

Total gauge (Pt) comes from compressibility eqn* and initial pressure (also P in below equation) is atmospheric.

* eqn:
Pt/P = {1+ [((gamma-1)/2)*(M^2)]}^(gamma/(gamma-1))



Be aware that your results are may not be great, and if you've any separation on the 2nd flap, they'll probably be quite a bit off. Fluent is far too 'stiff'. Its a jack of all trades master of none solver.


Just also - you almost certainly know this anyway - 95 m/s is roughly 215 mph... an F1 car won't corner at anything like that. Cornering speeds are more like 180mph downward (~80 m/s).

[joselu43]

Having done a fair amount of compressible flow work I must agree with everybody that tells you to ignore compressibility in looking at flow around a car surface. An upper bound of your error is given by 1 - Square root (1-Msquare)[At least up to transonic Mach numbers, M=0.8 or so]. For M = 0.25 that is less than 3%. You true error will probably be much lower, depending wether you are looking at pressure, velocity, density etc.
Believe me, in modeling this kind of flow compressibility should be the least of your worries.

[kilcoo316]

Having done a fair amount of compressible flow work I must agree with everybody that tells you to ignore compressibility in looking at flow around a car surface.

Believe me, in modeling this kind of flow compressibility should be the least of your worries.

If the guy marking the coursework asks him to do it with compressible fluid, he should do it in compressible fluid. Saying "some guys on the internet said not to" is not much of a defence (However right you might be!).



But a6zz - for extra marks you should do both compressible and incompressible and demonstrate that using compressible flow changes things very little, while increasing the overall simulation time significantly.

That demonstrates a knowledge of fluid mechanics rather than just blindly doing what your prof tells you.

[AeroGT3]

95 m/s is compressible guys. It's Mach .28 freestream. You could easily have flow at mach .8-.9 somewhere along the geometry. The differences between static and stagnation density become significant at speeds that high.

[mikhak]

Indeed, i ran a simulation of a double element front wing and from a freestream velocity of 30m/s this accelerated to as high as 83m/s. So with a freestream of 95m/s the flow will definitely accelerate to speeds where compressibility becomes important.
Note that F1 cars would only reach speeds near 95m/s on tracks like monza right at the speed trap at the end of the straight. For example at the last 2 races in australia and malaysia the cars were maxing the speed traps at between 80 and 85m/s.
You could try both models as mentioned above and compare or make life easy for yourself and reduce the speed so that compressibility doesnt come into effect.

[joselu43]

The question wether the flow is compressible or incompressible is bogus. Airflow is compressible at any speed. Using compressible formulae will always yield more accurate results that using incompressible ones -at any speed- . The question is, as pointed out by kilco316, one of tradeoff between complexity/cost and accuracy.
And I stand corrected, if the prof wants compressible flow, give him compressible flow. (It is a lot more fun anyway)
As far as achieving transonic Mach numbers (0.8 - 0.9) at 0.28 freestream in a F1 geometry, I am very skeptical. I would take any CFD results showing that with a grain of salt. Just think of the pressure gradients required.

[a6zz]

Thanks for everyones help,

Yeah, by just not assuming flow over 0.3Mach to be treated as incompressible flow, he wants me to prove it in CFD that compressible and incompressible solutions are roughly the same.