Aero track testing

[Jaaps]

Hey everyone,

My team is for the first time starting aero effects on our car this year.

What I'm having trouble with is how to collaborate our results found using CFD with real world track results and since our University doesn't have a wind tunnel it kinda hard to gather 'perfect' data. We are running an unsprung diffuser it wouldn't be worth running linear potentiometer since we won't get any data from it.

I don't know how else we can get pressure/force results especially with a diffuser apart from running the car around a track with and without the diffuser and then seeing track times differences. In doing that I don't think that this method alone would be sufficient to impress the judges at competition.

Any help/suggestions would be greatly appreciated.

Jakob Badenhorst
QUT Motorsport
Aerodynamics Manager

[Jaaps]

Hey everyone,

My team is for the first time starting aero effects on our car this year.

What I'm having trouble with is how to collaborate our results found using CFD with real world track results and since our University doesn't have a wind tunnel it kinda hard to gather 'perfect' data. We are running an unsprung diffuser it wouldn't be worth running linear potentiometer since we won't get any data from it.

I don't know how else we can get pressure/force results especially with a diffuser apart from running the car around a track with and without the diffuser and then seeing track times differences. In doing that I don't think that this method alone would be sufficient to impress the judges at competition.

Any help/suggestions would be greatly appreciated.

Jakob Badenhorst
QUT Motorsport
Aerodynamics Manager

[Mike Cook]

How is your diffuser suspended? Does it secure straight to your hubs or go through your pushrods? In the past we have strain gaged our pushrods to measure downforce.

[mech5496]

Make a trailer-like platform with a couple of wheels, mount the undertray on it using some rods and strain gauges to measure force on each of these...you can figure the setup pretty easily. Then tow it with a car at steady speeds and log the downforce. The moving car will (of course) disturb airflow, but you should be able to have some (rough) numbers. You could avoid some of the turbulence the tow car creates by spacing the trailer by a fair amount from it (dunno if it is safe though). Just my thought on a possible "easy" solution! I would be glad to hear what you came up with at the very end...

[Jaaps]

@Mike: our diffuser is attached straight to our uprights.

@Harry: Yer that could be a quick solution. Kinda after as accurate data measuring as possible, but that is definitely worth a look.

Would there be a way in which we could make a 'pressure map' of the whole diffuser, no idea how we would go about doing it tho.

Thanks guys

[StevenWebb]

One thing you can do, is work out the "effective" downforce by driving in different radius circles with and without the diffuser. It separates all the cfd/windtunnel/BS downforce into the actual "does it make the car faster" downforce.

[mech5496]

Stevens' idea definitely worth looking into, but it requires a full working car AND space to accommodate the undertray under it...if you have both then I would probably look at his way. About "pressure mapping" do as I suggested above, but drill some small holes across the undertray (across the points you want to measure), bond small threaded aluminum inserts on each hole. This will allow you to plug the holes (with a bolt) while not measuring. Now stick some pressure sensors and measure a few points at a time (Pitot or even MAP sensors could do the trick). Time consuming, I know, but I reckon it will do the job...

[Menisk]

So this is kinda (okay, incredibly) dogey and I wouldn't talk about it in design, but could you run stupidly low tyre pressures and watch for sidewall deflection with a GoPro? Get some control data by stacking known weight onto the car and then drive it down drag strip at Willowbank to do some straight line testing. Review the video after to see how much sidewall deflection to get a rough idea of how much force.

[revolutionary]

Some old school tech would be to just run a series of coast down tests to check cd and add to that shock travel sensors to determine downforce.

[Menisk]

They're attached straight to uprights. Won't affect the shocks.

[mech5496]

Daryl, I'm quite sure that MANY of the aero teams (I know at lest one) have used the "coast down" method to determine drag, and shock sensors to determine (sprung) aero DF. But as Menisk already mentioned the DF part won't apply to unsprung or "half-sprung" aero... Jakob, if I were you I would be extremely cautious about increased unsprung mass and extra stiffness added to your suspension from the undertray (on some modes more than others). Of course this could be tackled with some innovative design. It might worth it, or maybe not, it's up to you to decide. I think some American team has run an unsprung undertray some years ago, but abandoned it in favor of a sprung one.

[Mike Cook]

Try using pancake load cells on the mounts at the uprights. Really, all the other methods here aren't that great. I wouldn't really trust wind tunnels either, unless they were really big.

[Z]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Jaaps:
My team is for the first time starting aero effects on our car this year.
...
... I don't think that this method alone would be sufficient to impress the judges... </div></BLOCKQUOTE>
Jakob,

1. Don't try to "impress the judges". You have 600+ points on offer from good aero, and only 150 from Design. Go for the 600+.

2. For testing/development, Steven above has the right idea. Keep driving in big circles until you are pulling 3G lateral. You only need a stopwatch to measure this. Put simply, with good aero your times should drop by 30%.

3. For good understanding of the underbody flows use wool-tufts and one of those small cameras (mounted downstream so as not to mess up what it is looking at). Harry's suggestion of pressure tappings is also good. A clear (eg. pvc) tube with water + food colouring in it, set up as a Pitot-Static tube, is a simple way to start. Then compare with CFD, and see how wrong it is!

4. For good downforce I suggest you don't simply copy the usual gradually rising "diffuser". Instead have approximately equal cross-sectional area of the "tunnels", front to rear, with low front and side splitters, and a curved flap or two just above/behind the rear of the tunnels. These flaps will "drive" the underbody much more effectively than a simple diffuser.

5. The more iterations you go through, the better. So start with plywood, Maxbond, tek-screws, plasticine, duct tape, fencing wire, whatever. You can make the last iteration pretty to "impress the judges"... http://fsae.com/groupee_common/emoticons/icon_smile.gif

Z

[Menisk]

Just on Z's 5th point. I have no doubt plenty of your QUT guys have seen that crazy homebrew car with kart wheels that runs at mount cotton and goes awfully fucking fast.

[Tim.Wright]

Jaaps,

is it possible to mount the undertray to the chassis just for a straight line test?

[Jaaps]

I was planning on doing the circular runs under a stopwatch, also if we can find a big enough patch of earth, set up a endurance track and do lap times.

and also to do flow visualisation, also the wool tufts with a GoPro strapped to the back, lucky we have one or two guys on the team which have one.

I was not trying to 'impress' the judges, just to back our cfd data with accurate results.

I haven't been to Mt Cotton, I'll be sure to ask the boys about it http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

@Timo,
Nah our whole design relies on the undertray being unsprung.

Thanks for all the suggestions. Really appreciate it all

[Tim.Wright]

I understand it needs to be unsprung in a dynamic track condition, but for a straight line constant velocity test, as long as it in the right location, it doesnt matter if its mounted to the chassis or upright.

The problem with skidpad and endurance laps is that you introduce soo many other variables including driver and tyre temperature. Both of these are eliminated in a constant velocity test.

If you really must go down the stopwatch route, Id forget the endurance track and do either a constant radius (at two radiuses small and large) or a constant velocity ramp steer at two speeds (maybe 40 and 80km/h). As many repititions as you can with and without the aero to make sure the tyre temps and degradation are not stuffing your results

Tim