Active rear wing test
 

So, I've been reading a lot of Can-Am books and followed a McLaren MP4-12C around the track a few months ago - so I had to try this. Not legal in almost every race class, although it is legit in Targa Newfoundland Open class ;) Still, there's better aero discussion here than just about anywhere else, so I'll post this here. Really it's just fun to play with.

The initial plan was to use a SPAL linear actuator with a cool three-position controller. I'd have a low drag setting (triggered by a momentary button on the steering wheel), normal (the default) and high drag (triggered by the brake lights). Unfortunately, the speed of the SPAL part was just too slow, 1/2" per second. I could speed that up with linkages, but it was less than ideal.

Warren at Exomotive had been playing with headlight motors and thought they'd make good actuators for an active wing. They're strong, fast, repeatable and cheap. They also have internal logic that's easy to control. 12V on one wire makes them go up, 12V on another makes them go down. The motors spin at 50 rpm, and with the standard levers that means about 6.5" per second of stroke. You can change this by changing the length of the levers.

The goal here was to build a proof of concept without cutting up the car. I took a prototype COT mount and attached the motors to the sides in a convenient place, then triggered them with a SPDT relay attached to the brake lights. There's an override switch in the cockpit that turns off the active wing. The wing takes 0.6 seconds to change position, although it's effectively faster on retraction because of the way I have the various linkages clocked.

https://www.miataturbo.net/attachmen...ine=1379000568

Test video. The fastest test was at about 75 mph, which is the best I can do without attracting undue attention. No sign of any deformation to the wing mounts. I'll be at the track on Saturday and will give it a try with a Traqmate measuring deceleration as well as lap times. I'm not as familiar with this weekend's track as my home one, so there may some noise in the lap time data due to driver inconsistency.


A timer circuit has been suggested to delay the return after the brakes have come off, to avoid a flappy wing if I get on and off the brakes a few times in rapid succession. Moving the motors would also be preferred, although I'm less concerned with drag than many other Miata drivers for some reason ;) That would require longer pushrods, which may get into flex problems.

Interesting!

So it simply goes from "normal" mode to "brake" mode at the moment.


Could you in theory make it infinitely adjustable based on speed? Or would the simple nature of the headlight motors be too cumbersome for that?

The motors don't have any sort of feedback for positioning, they simply cycle between up and down. You'd have to simply base position on a timer, and if they get slow under heavy load you might find yourself drifting away from your intended position.

The SPAL actuator does have a position sensor, so it can be set to repeatable positions and would be a better choice for that sort of control. The speed of movement would be less of a concern in that case as well.

Besides there are times when you will want all the high-speed downforce you can get.

Id rather see a pressure sensor in the brake line so it only goes up during intense braking, rather than a delay timer for release. All the little brushes of the brake pedal during around town driving (as seen towards the end of the clip) would seem annoying.

Well, obviously the video was to show the movement. I have an override switch that locks it into the flat position by simply interrupting the ground wire to the relay.

I'd rather you shut the hell up about "around town driving" on a race car showcasing a prototype active aero rear wing.








:party:

Great concept, you may want to introduce some g loading logic so the wing does not change settings in mid corner, the timer will work in the track but not sure that will be all you need at the targa

Agreed.


On the concept though...

I think its very cool and great. I feel like this may require some logic on when the spoiler brake would activate.

Would it make trail braking more difficult/unpredictable? especially in high speed corners, where the spoiler has an effect on aero balance?

If the wing is changing settings mid-corner, it's because I stabbed the brakes. Might be a good time for a little stabilizing aero, actually.

Will it make trail braking harder? Dunno. That's exactly why I built it instead of imagining it, although I don't tend to trail brake much. Here's a peek at what the rear wing on a McLaren MP4-12C does under braking and cornering (starting at 35 seconds or so). I think the brake light flickers on that car are the stability control, not the driver. The wing seems to deploy at the same times regardless of cornering load, you're basically seeing the same spots I'd be braking on that track.

The brake light flickers are probably from the torque vectoring system that uses the brakes to help the car rotate around a corner. McLaren is a big fan of brake steer, which the actual MP4-12 used in the 1997 F1 season (there was a second brake pedal and a switch for the driver to select which rear wheel it acted upon). There's really no other way to get a car to change direction like it did around the 37s mark in your video. Cool stuff.

Yeah, "stability control" is a bit of an over-simplification :) I did find it odd that it flashes the brake lights in that situation, though. Might be a federal requirement.

I realize you're joking, but please remember the PUBLIC roads this race car was specifically built to drive on. He is indeed putting around town for small portions of it.

So suck it.

What's annoying about it? It's not like it's going to bring the car to a screeching halt every time it moves.

Man, you MiataTurbo kids really like to argue about anything. Back to my old lurkur status. :inout:

Hey Keith, if youre looking for something with a bit more adjustability, I would recommend a Concentric 5:1 linear actuator. I imagine the operation is similar to the Spal, but the speed is about 2 in/s. It cant be backdriven (rated at 34lb dynamic load, 450lb static), and has potentiometer feedback for position. You can drive it with any standard DC motor controller, or get an arduino or fancier controller for more advanced functioning. I have a lot of experience with them and theyve proven to be very reliable in our application.

Pololu - Concentric LACT4P-12V-5 Linear Actuator with Feedback: 4" Stroke, 12V, 1.7"/s

If youre interested I may have a spare one sitting around you can use for testing.

Looking at the specs and the pictures, I think the SPAL is actually the 20:1 version of that actuator with different branding. So I already have a controller for it. Thanks for the link, I didn't know about that option.

The 34lb limit when moving is a bit of a concern, but that would be double if they were paired. That would make the linkages easier as well. They would be well suited to constant fine adjustment, full airbrake might not be effective.

I'd be interested in laying my hands upon one for some testing if you'd be willing to loan it. Or two :) What I REALLY need is some sort of strain gauge to figure out just how much load is on this system.

How have you addressed possible "chatter" of the wing when in the down (brakes off) position?

It looks like the rod from the actuator to the wing would be almost horizontal in the down position (judging by the rub marks on the upright). I'd think that would make it hard for the motor/rod-ends to keep the wing "locked" in position at that angle without some sort of serious stopper/bumper on the upright, and an "over-center" sort of motion of the arm (if that makes sense). I'd hate to see that nice wing chew a hole in itself bouncing off the tops of the metal uprights.

I'm hoping from the first picture that you've replaced the brittle white plastic doohickeys that attach the headlight bucket to the arm with something more substantial?

LOL WUT?

Picture is of the wing in the down position. There is no torque load on the actuator, and the rod sees only a compression force.

Exactly. The down position is shown. The rod is under pure compression. Of course, the wing is generating downforce so it's not trying to lift off at all. There is some torque load on the actuator in the up position, but I kept it to a minimum. How will the motors react to that? Well, that's why we test, isn't it?

Any "rub marks" are simply reflections, there are none anywhere. There are no parts rubbing against each other. The front pivot rides on some nylon bushings. As you can clearly see from the picture, the plastic bushings from the original headlight mechanism have been replaced with rod ends.

Tension

WTF happened to the crossed out line thing?

IMO it would be a lot better to toss a bolt in the middle of the wing wing mount and remove the front bolt. This will make it so the wing sees its main force in the center so it balances this way the actuator will see a lot less force acting on it like it is now. While in the process toss a proper hinge to make the joint stronger.

How's the grip with that steep angle during braking? You think it would make drag to slow the car down (thus saving on brakes) as well, perhaps?

I think that is it's main purpose. I assume the wing is stalled at that AOA. At best it is acting like a spoiler at that point.

Very cool Keith. I lust for it. How does it feel under braking?

I wish we could use something like this in NASA-TT. I brainstormed some type of active aero like this, at least until NASA outlawed it last year. It would be huge on the fast courses for a TT3 car.

Im glad someone is trying it, Good job Keith.

I was looking at doing a DRS system on a dual element wing. The second element only has to move 2" on the front edge.

I made these calculations on the high end of the scale 160 mph to see about how much force the actuator would have to handle. So at 160 mph if the DRS was active it would make 565 Lbs of downforce @ a 27.5 HP draw. In the max max AOA (or in normal cornering mode) on the second element it would produce 1650 Lbs of downforce @ a 192 HP draw.

Exactly, you can clearly see the problem with the current design.

Can you guys tell me what track your hitting 160mph at in your miata? I would like to visit that track. I hit almost 140mph on the front straight at Texas World Speedway. One of a few "super speedway's" in the the US. Even in my buddy's Boss 302 mustang with the absolute best we could manage out of the last corner onto the banked front straight was 150mph.

It wasn't exactly for saying a Miata could reach 160 especially with 200 hp in drag, but it was for what the maximum the wing CFD was tested. I wanted to see the max weight the actuator would have to handle, even if it could or couldn't even reach 120.

I was thinking of adding a VSS component to the trigger logic (IE, don't change states if brake is depressed under XX MPH), but that is also a good idea.

I should have some Miata headlamp actuators around here somewhere, and will have a COT wing next week for the Exocet. This might be a fun excuse to build a cool little microcontroller that accepts inputs such as brake fluid pressure, VSS, maybe accel, and allows user definable logic to rotate the wing. Or for that matter, may be able to get it incorporated in MS3 code, which may be the excuse for a CAN based 3-axis accelerometer, something I've wanted to do for the past 2 years.

Thanks to Keith and Warren for the ideas.

FTFY!

I agree with this. Proper placement of the pivot point will minimize load on the linkages involved. As a side effect slower and/or weaker and/or shorter travel mechanisms can be used to actuate the movement.

FM duct?

Zing.

I decided to do the initial proof of concept test with minimal modification of my wing. Thus no guesstimates as to where the center of balance needed to be on the wing, no extra holes in my wing, and it also allowed me to use my existing uprights as a base. As a bonus, it gives me a wider range of front mounting points to deal with things such as pushrods that weren't quite the same dimensions as described in the McMaster-Carr website :loser: It's not the perfect, ultimate implementation. But it's the best one we've seen so far.

Track test wing cam. Max speed was around 123 mph on this. The motors have no trouble lifting and holding the wing at that speed. I did set them up so they had maximum leverage for lifting the wing - they don't pull as hard on the return, but they have the load on the wing to help them.



Unfortunately, the Traqmate didn't record the session with the active wing and I have weird readings with what it did record - I was going -3358.19 mph at one point. Argh. So actual data will have to wait for a few weeks. But I can report that the car feels excellent under braking, very stable. Since the brake lights come on with even a small movement of the pedal, modulating the brakes doesn't cause the wing to spasm. The car didn't change behavior dramatically, it felt as if it had a brake upgrade and was less likely to lock up. Of course I didn't test to see what happened if I triggered the wing by braking in the middle of a high-speed corner because that's not clever anyhow.

Oh man, I am so happy to see this in action!

I warned the starter and the corner workers that I'd be testing a moving rear wing, so please don't black-flag me for a mechanical. Big thumbs up from all the workers on the cooldown lap ;)

That is just awesome Keith! One more project for me to attempt when my Arduino gets here!
Any ideas for a cheap brake pressure sensor or pressure switch that has adjustable setpoint?

You can get pressure switches from NAPA, that's how older cars used to trigger their brake lights. The Big Book Of Switches behind the counter will have pictures, thread type and (probably) trigger pressure information.

Congrats Keith
Maybe the wing opened a new dimension to help you to 3358 mph. How to see some info next time.

well it was -3358 mph, so he must have gone through a worm hole.

Could use a variable pressure switch, that could open new possibilities.

Most of the brake pressure switches I have messed with that trigger brake lights have very low pressure setpoints so that the lights will turn on with the slightest pressure from the pedal. It would be nice to have the switch only activate at higher pressures. I am not sure what would be considered moderate brake pressure on a miata. Keith, I know you did a good bit of brake system work a while ago, do you have any idea what a good setpoint or range would be for moderate brake pressure? I will have to go take a look at the Napa book. I have sent them questions before online. They might be able to suggest some options if I can give them an actuation pressure range.

As mentioned earlier it would be nice to have something that could read a proportional signal for proportional control or just a higher setpoint so that your wing would only move to the high drag position when brakes are applied with moderate to high pressure. This should eliminate concerns with the wing actuating during light braking.

I was wondering if anyone is familiar with newer traction/stability control OEM setups that might have more useful switches/sensors that provide brake pressure info to the ECU.

A 1991 has an approximate 17:1 pedal/fluid pressure ratio. Since "moderate" is such a general term and the actual pedal pressure for a given braking amount will vary depending on what you've got for calipers, pads and rotor sizes, I'll let you measure your "moderate" pedal pressure and do the multiplication.

I work 8 years on AC-130 Hydraulics and there are plenty of different switches. Just like on the boost pump activates @ 20 psi and deactivates @30 psi just to give an idea.

Switch: McMaster-Carr

However, I would rather have a brake pressure transducer hooked up to MegaSquirt and/or Arduino. The wing would be driven by a MOSFET, just because I hate relays.

Transducer: XT Racing 1/8" Pressure Sensor | MotoSport

Oh thats a great idea using brake pressure, how hot do the actuators get just from the continuous use?

I didn't check them, I was checking something else after the long test session and forgot. They certainly have lots of cooling airflow!

Not a brake pressure switch, but a pressure transducer should be used. You can get one for $150, maybe less if you search around.

What other data should be used as inputs to rotate the wing?

VSS and/or deltaVSS?
Accel?
Other?
Is a time component needed?

Further, I don't see why a position sensor can not be added, allowing for variable control of the wing element, either multi stage or continuously variable.

Ben, check this little guy out: XT Racing 1/8" Pressure Sensor | MotoSport

VSS, deltaVSS, Accel, any sort of gyro input along the vertical axis, and definitely time. For safety, I would include some logic that holds the wing up until all inputs are "down" for a second or so. The reason why active aero is banned in most racing classes is because of the danger in transient conditions. If you are pushing the car with active aero, failure of the aero to do what you expect means an off or worse. Once the actuator is told to go down, it needs to complete a full rotation to get back up (1.2 seconds). A "down-delay" would prevent a lot of problems.

The Miata headlight actuator works off position sensors and has its own little logic system and cams in it. It is not designed for intermediate positions. It's a one-and-done deal for up or down, so variable position would require something else. The point of this exercise is to use the cheap and available headlight actuators to do this, so I'm happy living in the constraints of pure up/down.

That looks like it will work perfect for the purpose.

Same train of thought I had.

How do they respond when de-energized in mid-position?
The same controller could operate both 2-position and variable position actuators. Just a minor software difference. A position reference sensor can be incorporated for fault detection with either type of system.

It may also be worth looking at other cars with flippy headlights and low cost actuators to see if there's possibly a better option out there. S13 240SX comes to mind as a possibility, but that's because I think there's a pair of them around here somewhere.

The Concentric 5:1 linear actuators that scenturion mentioned would be a good option for full control, I think. They've got position sensors in them and the Ardunio controls are already worked out. Search eBay for linear actuators and you'll find all sorts of interesting options. I never did that before his suggestion because the SPAL unit basically fell into my lap.

If you pull the power from a Miata headlight motor mid-cycle, it will complete the cycle when power is restored. And they only spin one direction, so you'd have to go all the way down and up again.

Thoughts on using a wiper motor instead of a headlight motor? It would allow for variable position, variable speed, and bi-directional control.

The headlight actuator is basically a wiper motor with built in positional controls. You could reverse the drive polarity of the headlight actuator and apply PWM with an H-Bridge to accomplish all of that open-loop. Using a wiper motor would definitely require closed loop feedback through an encoder and home switch, or if you're feeling confident of your controls, a potentiometer.

How about this guy for us cheap bastards? $36

Pressure Transducer or Sender 1000 PSI for Oil Fuel Air | eBay

The same seller has a 1600psi sensor. I asked the seller for a datasheet(attached). These sensors can withstand 200% full scale pressure so 1600psi sensor should be good to go for our aplication and even the 1000psi could work.
The 0-4.5V input would be easy to use on Arduino.

Nice find. For $36, it's worth a shot to see if it gives good data.

Wiper motors like to just spin and spin in one direction and they only have logic to return to a single position, not two (or variable). Without complicated controls, the headlight motor is much better.

My only concern with this being hooked to the brake light is triple digit trail braking at the limit where you are counting on a certain level of grip that might not be there with the wing in air brake mode.

Edit: I typed this about 2 hours before my phone sent it. Psi switch on brakes would be better to trigger.

Indeed, that reliance on a complicated system is why active aero is banned across most racing classes. Still, it's fun to play with!

I wonder if anyone has ever applied this to a diffuser, where you would have more aggressive channels on a multi-element diffuser that are opened at low speed...You could probably make the system quite stealthy and invisible for TT classes...

The only work you need to the defuser is a motorcycle engine with a giant fan on it to suck the life out of the ground and give you 400 mph downforce at 20mph. We're messing with banned mods, might as well go nuts at it.

Sucker..

https://www.miataturbo.net/attachmen...ine=1379448926