plz help
 

hello everyone

Im working on a CFD analysis of a cone cylinder at different angles of attack. For my current analysis,….. Mach = 2.5, Re = 1.2 million, base diameter is 3.7inches.

The problem is that, Fluent can take Mach no.'s input but doesn't take Reynolds number's input. Instead, it requires operating pressure and temperature inputs. Since I don't know these values for my analysis, I have used many combinations of temperature, pressure and density values that will give me the same Reynold i.e (1.2 million) & Mach no. 2.5.

Solving in this way, I did not get the same results as wind tunnel's. My solution is grid independent. Was I doing it right? Or plz guide me, how should I solve the problem?

Regards.

Re: plz help
 

"it requires operating pressure and temperature inputs"

The operating pressure is 0 for compressible flows (here it must be compressible because high mach number). The termerature has to be set for some boundary conditions (the velocity inlet I think).

If the flow is air, you can take a standard temperature, such as 300 Kelvins.

In fact, for compressible flows, the Reynolds number has no real signifiance. This adimensional number was created for incompressible flows, and the adimensionalisation of the equations of motion is not good for compressible flows. I don't know why you have a Re number, but I guess it's not really useful. If you have Mach number, the fluid properties, the temperature of incident flow and the dimensions of your body, you can solve it easily with Fluent.

Bye

Re: plz help
 

Um, forgive me for disagreeing, but the contention that the Reynolds number has no significance for compressible flows is just plain wrong. It has as much significance for compressible flows as for incompressible flows. It is true that in wind tunnel testing one can rarely match both Mach and Reynolds number to flight conditions, and so one usually settles for the Mach number as being more important to the task at hand. But that does not imply the Reynolds number is insignificant. In fact, one area of ongoing study is the best way to extrapolate tunnel data to flight because the Reynolds number can't be matched.

To the original poster - what data are you comparing to, and what conditions are you trying to match? How are you setting up your model - are you modeling the tunnel walls as part of your system, do you have any tunnel hardware that is being left out of the model, etc. More details would help.

Re: plz help
 

"But that does not imply the Reynolds number is insignificant"

Ok, I think you are right. But I think that adimentionalizing the Navier-Stokes equations cannot be done for compressible flows, right ?

So the Re number could be a parameter, but not a way to adimensionalize the N-S equations... ?

Re: plz help
 

No, the Re number still pops out from non-dimensionalizing the NS equations. The NS equations in the most general form represent the flow of a compressible fluid. Why would you think that the NS equations can't be nondimensionalized for a compressible flow? Or are you confusing compressible with inviscid?

Re: plz help
 

"Why would you think that the NS equations can't be nondimensionalized for a compressible flow?"

Because rho is not constant any more. As the Re number is rho V L / mu and it depends on rho. You have to divide by rho, then, the adimensionalization is done by dividing by an unknown value ????

Maybe do you adimensionalize with Re= V L / nu ? At that point, think that nu is not constant, because nu = mu / rho.

Re: plz help
 

Reynolds number of course has a u (velocity) in it. But u varies even in incompressible flows. Does that mean that Reynolds number makes no sense in those flows?

For compressible flows, one uses a 'typical' value of density to nondimensionalize the equations. Inlet conditions, exit conditions, whatever makes sense in terms of measurements that might be available.

Re: plz help
 

Ok, thanx for the information