How to create exhaust flow in the wind tunnel

[hardingfv32]

How do you create a accurate engine exhaust flow in the wind tunnel?

Brian

[Tozza Mazza]

F1 teams can't, it's simply not possible. To test how heat works, with the exhausts and engine heat for cooling packaging, they have to use CFD. I saw this is an interview with James Allison of Renault, but can't find the link.
TM.

[Richied76]

Thats not entirely true, But you are right. Mclaren use hot air to around 200c thats pumped out of the hot air ducts in the car (to gain more reliable figures) although at very low pressure. I dont think it would be possiable to pump hot gasses at such a great volume and heat unless it was actually coming from an engine.

I mean it may even matter if the exhaust gas itself has its own cemical composition that may effect the figures. It is full of spent carbon, nitrogen, ect.. as silly as it may seam

Short of taking 60% of the exhaust from the dyno thats running at the same time as the tunnel does anyone have an idea how you could replicate exhaust flow, in practical terms?

[hardingfv32]

Is it practical to use a small turbine built into the model to get the proper temp and flow? What would be the implications to the ambient temp of the wind tunnel. They already have have very substantial air conditioning systems just to maintain constant atmosphere in the tunnel without adding a new heat source.

Brian

[dansus]

Surprised they cant suck through the air intake to a heater and pump it out the exhausts. If its a good idea to simulate exhausts or not, i have no idea.

[strad]

Y'all did see where Ferrari found that their windtunnel problems were due to increasing their model from 50% to 60% and that at that size the interaction of the air and the walls of the tunnel were giving them false readings?
I guess not enough room left for the turbulence and it affects the boundary layer.

[hardingfv32]

The idea for this thread was based on a rumor that Ferrari was have trouble with exhaust flow in their tunnel.

Maybe an auxiliary generator used in commercial jets as an example could be placed in the model. Remember this all has to be contained in the model assuming a rolling road precludes any routing from below.

If you assume that the 700 HP output of a F1 represents an efficiency of 25%. We have 75% of the energy lost, of which say 30% is exhaust. I figure the equivalent of 840 HP goes out the exhaust. I would assume that can be converted to Watts to get a feel of the electrical load. We then need to scale this to say a 50% model.

Is this logic correct?

Even so this is starting to sound impractical.

Brian

[Robbobnob]

id assume Red Bull have a system that is modelling resonably well, such a pitty they are winning and dont want to share their secrets.

[nathanturner.spc]

I wonder how much effect the data has in terms of drag compared with that of, for example, the blown exhaust system introduced by Lotus Renault this season?

Nathan

[marcush.]

Isn´t the model suspended by a balance from above? it must be possible to route two tubes through there .
But still you got the problem with the hot air heating up your test set up... You would need a really serious Heat exchanger in the return line to cool down the air back to exactly what you have asked for....another hole in your pocket draining money no end...

[wunderkind]

id assume Red Bull have a system that is modelling resonably well, such a pitty they are winning and dont want to share their secrets.

How about a water tunnel used by the marine industry?

Mercedes Benz uses water tunnels to test the aerodynamics of their roadcars, especially the underbody aerodynamics. If you duck underneath most of the recent Mercedes roadcars, you sometimes see flaps and pieces used to smooth the air flows underneath the car.

The 'spray' travels ten times slower in water and the flow-viz, and any vortices are much easier to see in water. Exhaust gases can also be simulated using similar means.

I believe Red Bull might be using a water tunnel at some university in designing and developing their car. Adrian Newey is known to be keen on yacht designs.

[hardingfv32]

The teams have a hard time getting data that accurately correlates with on track data using 60% models with air as the medium. How can the data possibly be better using water or any liquid?

What sort of temps and flow are we talking about when we scale to say 60%? Can we expect this scale exhaust flow to have a good correlation the the actual exhaust flow?

Brian

[wunderkind]

The teams have a hard time getting data that accurately correlates with on track data using 60% models with air as the medium. How can the data possibly be better using water or any liquid?

What sort of temps and flow are we talking about when we scale to say 60%? Can we expect this scale exhaust flow to have a good correlation the the actual exhaust flow?

Brian

I'm no aerodynamicist, I think a water tunnel can be useful for designing upper body elements of a F1 car. For the underbody, a rolling road is required. But I think liquid may actually be a better medium than air for aero analysis. You can turn a aero model in any yaw angle you want as long as the pool in large enough and the flow channel is wide enough. Flows travel up to ten times slower in water than in air and allow for more detailed observations to be made. But a rolling road in water is not going to be practicable.

I don't think there is a perfect way of replicating/simulating gas flows from exhaust pipes in a windtunnel enviroment. The heated gases must be generated within the 60% model and that is just not feasible. But if we discard the temperature issue for one moment, what about using a pulse compressor or simply a electic motor powered jet to expel water out of the exhaust pipes to replicate exhaust gases in an aero model in a water tunnel? It's not perfect, but it gives you a reasonably accurate picture of what goes on.

[hardingfv32]

Then, why don't the teams all have big water tunnel facilities?

Water is not a better medium than air for accurate aero analysis.

Brian

[HampusA]

Does water even interact the same way as air does?

I think to get close estimates of the exhaust pulses you need to build a miniature V8 engine.
It´s constantly changing so i can´t see how water or air would replicate that.