[dennis722]

Hello,
I'm simulating a radiator to apply it to a formula student car simulation. But when i simulated it the pressure is over the hole radiator 0 Pa and the velocity wents with 18m/s through the radiator (i have set the velocity inlet also to 18m/s, so no pressure drop..)

I have set my simulation settings like in the pictures. I used the surface wrapper and created regions. I did it like this, because normally I'm simulating like this with the hole car. To that i set on the "Per-Region Meshing" and for the Region of the Radiator I set the "Volume of Interest Specification" to "External" instead of the default "Largest Internal".

I set a coordinate system for the radiator like in the picture and set the porosity for x- and z-componente to 10000kg/m³s and for y-componente to 70kg/m³s.

I don't know what I'm doing wrong. Also the plot of the mass flow and pressure drop don't work.

Thank you in anticipation

[ggulgulia]

Hallo Dennis.

Can you tell me what exactly is the objective of your simulation?(PD calculation, Flow uniformity, etc? )

How did you obtain the boundary conditions and the porous values. The porous values seems completely unrealistic for a FSAE vehicle which I suppose uses around 800 CC Engine.

And lastly what is the radiator wall making interface with? In ideal conditions, we would like to have 1-D flow through radiator for best performance. If radiator walls have flow through them, then I deem the design having some serious flaws!

Will wait for your reply until I can give you some directions.

Thank you

[dennis722]

The object is a ashlar it should only represents the form of the radiator. The porous values i take from the help book of star ccm+, but what are realistic values for a 800 CC Engine?
What do you mean with obtain the boundary conditions, the prism layers and mesh size or what? I set the base size to 0.1m, number of prism layers 8, prism layer thickness 10mm.
The radiator wall makes an interface with the radiator wall in the other region like the radiator inlet and radiator outlet.

When the single simulation of the radiator works fine, then I want to simulate the radiator with the fsae car in 3D simulation. So that I can design the radiator on the basis of the sidepod and the volume flow rate. So that we know whether the radiator can cool down the engine. How can i set a 1D flow through the radiator then?

Thank you

[ggulgulia]

Hi Dennis,

Thanks for your response. Realistic values of porous conditions can be obtained by plotting PD vs air flow velocity, i.e by finding the pd at different velocity. Can you tell me the dimensions. I can try to tell you some porous values based on dimensions. Remember not to include the dimension of the tanks. Only the louvered fin and tube sections.

By boundary conditions I mean the flow conditions ( AFD , temperature and porous values). What is the dimension of the radiator? I suppose you are studying external flow through the radiator. In this case, 18 m/s is way too high for a radiator

[dennis722]

The radiator is 273mmx510mm and 43mm thick. We have a radiator that we designed together with a company (AKG), so there should have a data sheet with the pressure drop or where I get the pressure drop coeff?

Originally Posted by ggulgulia

I suppose you are studying external flow through the radiator. In this case, 18 m/s is way too high for a radiator.

Yes, but why is 18m/s too high? We also drive 27m/s for short times.

That's the first time I simulate a radiator and I am the first in my team who simulate a radiator with star ccm+ to get the mass flow and design the radiator with it. Sorry if i play the fool

In the picture you can see the radiator at it's position in the sidepod.

[ggulgulia]

Hi,

If you've developed the radiator with the help of a company then it might be a good idea to get the PD vs velocity data from the company itself.

[dennis722]

I dont get the simulation running, with that.
Can I do it like I am doing it with the surface wrapper without doing a "Unite" and "Subtract"?

[me3840]

If you want to do a radiator calculation like this flow should not be allowed to move around the radiator. Flow doesn't like going through porous objects, it'll just go around it and you'll get the wrong answer.

If your goal is to get the mass flow through the radiator your current set up will not give you that - you need to install the radiator in the car and do the simulation with the car to get that value. Pressure drop through the radiator in a sealed tunnel will not change the mass flow rate (mass flow is conserved). A radiator in a tunnel like yours will have a lower mass flow rate than the freestream because the flow will try to avoid the radiator when it can. In either case you won't get anything useful for designing the car...