Aerodynamic effect of fluctuating volume of air under car

[garygph]

Firstly if there is or was a thread on this my apologies but I have not been able to find it.

I am curious as to the effect of the floor tray moving up and down very rapidly whilst a car is travelling as seen on some awesome "looking back under the car" camera shots. For the life of me I have not been able to find a clip to refer to here. I am sure that one of you boffs here have one, or could find one with more sucess than I have. However I am sure you know what I am referring to.

I understand the gap between the floor and the surface of the road is very important and have read a lot about that and seen a lot of discussion about the diffuser design etc. What I am curious about is when the gap inbetween the floor and the road increases there is a large (relativel speaking) volume to be filled and then very rapidly emptied. This is obviously going on lessons learned here that the air will not be compressed or expanded as the velocity is not high enough. I have read how the air flow entering and leaving under the floor is managed(or attempted to be managed as best as possible) with the discussions as if the gap was constant but have not read anything about the flow of air in and out whilst the gap was rapidly diminishing or expanding. Almost like a pumping action. Can this be used to enhance downforce or other objectives? Various thoughts come to my mind but am keen to hear what the aero boffs here have to say.

Are rolling roads in wind tunnels set up to simulate this with varying the gap between the car and the road via maybe a varying thickness of rolling road. This is proabably not feasable practically but was wondering..I seem to do a lot of that

Thanks for reading and am very keen to either be pointed in the right direction to an article or better still your thoughts on this.

[hardingfv32]

There are rolling roads with separate belts, but I would say it would be easier to install an active suspension to vary the car's ride height.

Brian

[DaveW]

I'm not sure that is entirely relevant for boundary layer flows, & certainly not for stalling flows, but some ideas might be found by Googling "Wagner Function". This might be a good starting point.

[garygph]

I'm not sure that is entirely relevant for boundary layer flows, & certainly not for stalling flows, but some ideas might be found by Googling "Wagner Function". This might be a good starting point.

Thank you for that link. I am on my way out but took a quick look and it is something that I have not read about before so that is brilliant!! I will certainly be sitting down to a good read when I get back. The little I have read is really interesting and informative already.

Thanks for the replies.

Gary

[DaveW]

But this old report might be more useful.

[flyboy2160]

daveW, i also thank you for those references.

[garygph]

DaveW thank you very much indeed. There is a lot of info there to digest and to comprehend..awesome. Nothing is straight forward for a designer on a F1 car thats for sure! This also helps me understand better why tunnel data is not always repeated on the track.

[Kiril Varbanov]

This also helps me understand better why tunnel data is not always repeated on the track.

Indeed, track testing is invaluable. In regards to wind tunnel, CFD and some data correlation you can check my recent article, although it's not that informative in details. I've had hard times talking to the teams ...

[hardingfv32]

In layman terms what is the significance of the "Wagner Function" in this case?

Trying to make a measurement of a system with a constantly changing variable?

Brian

[gixxer_drew]

Trying to answer the original question. On cars without any serious underbody aero (rules limited) and a relatively low ride height normal articulation can move forces around HUGE. You cant take a single value and not have it fluctuate out into timbuktu if you are talking about a car that is low (say 25mm or less and pitches at 1 deg). You can set it higher at the cost of reducing the forces.

Now you take that same car and put significant underbody aerodynamics in play... things can get really ridiculous. A big part of what I do is not just to design a car that makes some cool forces on your spreadsheet but one that is stable and easy to drive across all conditions and that can get increasingly difficult with lower ride heights and wings in ground effect. Then you get aero dominating all the setup parameters even on a car with a lot less downforce than F1 or LMP.

With GT2 for example they test with the roller at static RH for 50mm clearance, so the pitch sensitivity is much reduced compared to say Japanese GT500. But there is still value in pulling out some tricks to reduce pitch sensitivity. I would say most cars in the field have decent downforce numbers, the cream of the crop have nicer L/D and are less pitch sensitive

Hope that helps.

[garygph]

To gixxer_drew,

Thank you yes it does answer my question. I was wondering if I was just not thinking right as I thought it was a nightmare area for a designer and yet there does not seem to be much discussion about it...maybe because it is not such a visible area of the car as a turning vane or barge board etc?

it makes a lot of sense to me that your objective would be to have not the max possible downforce at a specific set of ideal ride heights etc but across a range of heights and pitches. Much like getting a nice torque/power curve on an engine and not just a high peak in a very narrow rev range. But that is easy to do compared to under the floor aerodynamics I think

My mind boggles as to how big a job it must be to try and accomodate the above described situation as well as changing yaw angles, wind direction and disturbed air flow from following another car. Waaayyy above my head!

[DaveW]

Trying to answer the original question. On cars without any serious underbody aero (rules limited) and a relatively low ride height normal articulation can move forces around HUGE. You cant take a single value and not have it fluctuate out into timbuktu if you are talking about a car that is low (say 25mm or less and pitches at 1 deg). You can set it higher at the cost of reducing the forces.

You are quite right of course. However, aeroforce changes must lag displacement changes and, arguably, the Wagner function might give a clue how the forces are likely to develop with time. It occurs to me that the front ride height of F1 vehicles might sometimes be unnecessarily constrained because this is overlooked

p.s.: It is interesting to compare front suspension movement of McLaren vs Lotus, for example.

[gixxer_drew]

Trying to answer the original question. On cars without any serious underbody aero (rules limited) and a relatively low ride height normal articulation can move forces around HUGE. You cant take a single value and not have it fluctuate out into timbuktu if you are talking about a car that is low (say 25mm or less and pitches at 1 deg). You can set it higher at the cost of reducing the forces.

You are quite right of course. However, aeroforce changes must lag displacement changes and, arguably, the Wagner function might give a clue how the forces are likely to develop with time. It occurs to me that the front ride height of F1 vehicles might sometimes be unnecessarily constrained because this is overlooked

p.s.: It is interesting to compare front suspension movement of McLaren vs Lotus, for example.

This is known to really be one of the next frontiers. Actually I recently sent out a bid on a project right up this alley a couple weeks back, but its pretty much a pipe dream at this point. As far as I know this stuff has always been known to be of interest but has never been done. The logistics of combining say, a 7 post rig and a wind tunnel is a massive project and you just cant control the conditions enough in all the other ways. Perhaps some very special CFD stuff, still a big project and less accurate.

To be honest I think it is extremely complicated to try and model this mathematically. At least from everything I have seen in my short career. Going from no model at all to an even partially accurate model still can have big benfits. Its that old addage "all models are wrong, some are useful".

[gixxer_drew]

To gixxer_drew,

Thank you yes it does answer my question. I was wondering if I was just not thinking right as I thought it was a nightmare area for a designer and yet there does not seem to be much discussion about it...maybe because it is not such a visible area of the car as a turning vane or barge board etc?

it makes a lot of sense to me that your objective would be to have not the max possible downforce at a specific set of ideal ride heights etc but across a range of heights and pitches. Much like getting a nice torque/power curve on an engine and not just a high peak in a very narrow rev range. But that is easy to do compared to under the floor aerodynamics I think

My mind boggles as to how big a job it must be to try and accomodate the above described situation as well as changing yaw angles, wind direction and disturbed air flow from following another car. Waaayyy above my head!

That is correct. I typically use a combination of simulation work to get a feel for attitude changes and talking with the race engineers about setup compromises to get a range for aero. If you have nice budgets you can optimize to track specific setups. I'll often have addons I work out ahead of time for less and more sensitive setups.

Lets say you are talking about a hill climb car, thats a real interesting pickle. You need max downforce and have no rules, so you go to tunnels but you also need a lot of ride height and big changes in pitch. Everything goes way beyond a single result at no pitch, yaw or roll in a single tunnel test. You want the car to have a higher average downforce and a lower average drag across a lap. Otherwise you are just competing for wind tunnel print outs that don't mean anything. My $.02