Presented by Nine Lives Racing
1 2 3
SVreX (Forum Supporter)
SVreX (Forum Supporter) MegaDork
12/21/20 9:33 a.m.

In reply to Robbie (Forum Supporter) :

Most people would say aero efforts in autoX are not really worth the effort because of the low speed.  You've proven that wrong.

Why not an air brake?

Robbie (Forum Supporter)
Robbie (Forum Supporter) GRM+ Memberand MegaDork
12/21/20 9:38 a.m.
SVreX (Forum Supporter) said:

In reply to Robbie (Forum Supporter) :

Most people would say aero efforts in autoX are not really worth the effort because of the low speed.  You've proven that wrong.

Why not an air brake?

Because you can't assume that the air brake will leave downforce alone while increasing drag. My suspicion is that increasing drag is likely to decrease downforce, perhaps significantly. And trading tire braking for air braking probably isn't a good trade. I'm sure we can figure out some rough calcs to see where that trade would be from an ideal standpoint, but Keith makes some good points about driveability that would be hard to estimate. 

ProDarwin
ProDarwin MegaDork
12/21/20 9:45 a.m.

Out of curiosity, how much time does an autox car spend under braking?

I just grabbed a random vid.  It looks like the driver only hits the brakes hard on the course twice?  Its hard to tell.  A lot of the deceleration is just caused by cornering.

 

 

Robbie (Forum Supporter)
Robbie (Forum Supporter) GRM+ Memberand MegaDork
12/21/20 9:52 a.m.

In reply to ProDarwin :

That's the other problem. In autox, let's just ballpark that you spend effectively 100% of the time cornering at the limit. You'd probably spend about 80% of the time accelerating, and 20% decelerating. (Maybe a different ratio but since almost all cars brake better than they accelerate, we know you spend more time accelerating). 

So, Improving your cornering is job #1. Next would be improving your acceleration. Followed by improving your braking. Downforce, as it happens, improves all 3, with a tradeoff from only 1. 

I'm not saying that improving braking isn't a place to find time. It is. But it is probably not the place to find the most time, unless you are pretty maxed out at cornering and accelerating. 

But this is also all ideals. Driveability should not be underestimated. 

ProDarwin
ProDarwin MegaDork
12/21/20 9:55 a.m.

So in that vid it also looks like there are effectively 2 spots on course where he is full throttle and not under max cornering.  Assuming all 4 of those areas are combined for a total of ~6 seconds and you can optimize that for a 10% reduction, thats 0.6 seconds.  Thats well worth it for a full-prep build.

Keith Tanner
Keith Tanner GRM+ Memberand MegaDork
12/21/20 10:04 a.m.

Again, remember that the rear tires have very little to do during a max braking event. So a loss of downforce on the rear isn't necessarily such a terrible thing, as long as you get more deceleration from drag than you would from the extra downforce on the rear tires you're doing okay. That's assuming that you're running your wing at max downforce all the time otherwise and taking the drag hit on the acceleration portions.

I expect time under braking is a car-dependent variable. I was at a Corvette club autox with my classic Mini years ago, and I was actually upshifting at a point where the Corvettes were starting to brake. I suspect they could have made better use of an active aero element than I could have :)

ProDarwin, the motors I used take 0.6s to go from fully retracted to fully extended. I have them triggered by the brake lights. It's been pointed out to me that hitting the brakes mid-corner may cause problems due to a potential loss of downforce - to which I counter that hitting the brakes mid-corner already means you're in a bit of trouble :) I also don't trail-brake so I didn't take into account how that might be affected. Control could easily be passed over to something more intelligent than a lightbulb but I was after a proof of concept.

Robbie (Forum Supporter)
Robbie (Forum Supporter) GRM+ Memberand MegaDork
12/21/20 10:22 a.m.

Well, put on your nerd hats folks:

Here's some calcs from sleepy in a different thread to get us started. This was for a wing that was cranked to get max downforce.

~250#s of downforce, for ~70#s of drag... at 60mph

it's ugly as all get out... and who knows how realistic it is; but... I'll take it for a "first swipe"

also, let's assume the F-Dat is 56" wide and 30" tall, with a 2ft by 2ft box on top for the driver... thus ~15sqft of area; with a guessed coefficient of drag of a miata ~= 0.4
than, the base drag of the car, at 60mph is.... ~550#s.

Ok, so, let's assume a miata-ish car with (2k lbs, frontal area slightly larger than the above estimation if the car has a windshield).

250 lbs of downforce will increase traction about 10% (after you counter for the fact that more weight does add more traction, but at a slightly less than 1:1 ratio). 70 lbs of drag is also about 10% of the drag of the vehicle. 

BUT how does this affect cornering and acceleration and braking?

with 10% more traction, you should be able to apply 10% more force to the car through the tires. If the car already corners at 1 G, you can now corner (or brake, or accelerate) at 1.1G an improvement of 0.1 G. Adding a 70 lb force in air brake, or reducing 70 lbs in drag, has only an effect of 0.035 G. So really you need triple the drag force to be as effective (duh, 70 lbs needs to triple to get to 250). And you definitely don't want to LOSE the 250lbs of downforce in trade for a few more lbs of drag, even in a braking situation. 

So, until you're working in the 1:1 lift to drag ratio of the aero, the downforce is more important.

Also, cars are pretty good at using all of their available traction for cornering and braking. Most are not as good when accelerating. Therefore you can trade downforce for less drag when accelerating because you have traction you can't use. I wouldn't trade it any other time. This is why I think of active aero as a drag reduction tool mainly. 

Keith Tanner
Keith Tanner GRM+ Memberand MegaDork
12/21/20 11:03 a.m.

Interesting. I still disagree that downforce is all that useful on rear tires under maximum braking. With all the dynamic weight transfer going on, they're pretty lightly loaded.

On my track Miata, the difference between the amount of drag at my chosen AoA and "flat" is pretty small. But the car has a very hard time slowing down, so to me a draggy air brake is useful. Jack Olsen used a very similar design to add DRS to his 911 and didn't see any improvement at Willow Springs because the drag reduction was such a small percentage. But I don't think he took the opportunity to really crank the wing up for the resting state, possibly because of what it would do to the balance of the car.

 I originally got the idea for this experiment after following an MP4C around the track, they have a little air brake but it's more akin to an airplane wing flap.

Robbie (Forum Supporter)
Robbie (Forum Supporter) GRM+ Memberand MegaDork
12/21/20 11:17 a.m.

In reply to Keith Tanner :

Balance is a big issue. And there's where the real testing kicks in (as you know and live probably more than anyone!) I need to go back and watch your video again. It is possible your air brake is also increasing rear downforce? Maybe that's where you're getting the feeling of stability?

Robbie (Forum Supporter)
Robbie (Forum Supporter) GRM+ Memberand MegaDork
12/21/20 11:27 a.m.

I do have to say that using your headlight motors for active aero is 100% the best use of a headlight motor for the challenge that I have ever seen. Especially if you have already maxed your recoup!

Even if you don't gain much dynamic time, you'll be able to gain a lot in concour points...

Keith Tanner
Keith Tanner GRM+ Memberand MegaDork
12/21/20 11:28 a.m.

The AoA is pretty extreme. It's quite likely I'm seeing more rear downforce but it might also be stalled. I think it's more likely that it's that center of pressure moving way behind the center of gravity. Like a drogue chute.

Here it is in action.

 

sleepyhead the buffalo
sleepyhead the buffalo Mod Squad
12/21/20 11:32 a.m.

it's been a hectic day... I've got thoughts on this... I just haven't made time to sit down and get it all out

SVreX (Forum Supporter)
SVreX (Forum Supporter) MegaDork
12/21/20 11:47 a.m.

In reply to Robbie (Forum Supporter) :

No disagreement on the math. I'm not sure you are considering the time...

I'm not great at a lot of things. I'm very good at brakes, and especially at adjusting them well (including Miata rears, which seem to be a challenge for a lot of people).  I've had a lot of comments from pro drivers at the Challenge about my brakes.

They don't trust most of the cars braking capacity.  Why should they?  But once they drive mine, they get a very quick sense of confidence in the brakes, and that lets them drive it differently. They dive deeper, and get out of the corners faster. It shows in the times. I have often had cars that were top finishers in the autoX. 
 

I like the air brake idea because of its potential to convert some of the TIME.  Here's what I mean...

In your example, you suggested 80% acceleration, and 20% deceleration.  Ok. Let's run with that.  That means on a 60 second autoX, 12 seconds are consumed braking.

Lets assume there are only 6 corners in a course that require hard braking, and an air brake allows you to cut 1/2 second of braking time in each of them. That takes 3 seconds from the deceleration category and moves them to acceleration.  Instead of 80/20, you've changed it to 85/15.  25% of the time you used to use decelerating can now be used for acceleration.

(No math here. It's all guesswork!)

I'm not disagreeing about the value of downforce and grip. I am wondering whether in actual application the air brakes can convert deceleration  time to acceleration time.

One more thing... the active aero also opens the possibility of a split system that assists with cornering and helps rotate the car. Picture a 2 piece wing (like in the arduino video above)... could the right side only  "brake" while making right turns?  Could extra drag on 1 side assist in rotating?  What about active side flaps, or canards?  Roof flaps?

There is very little real math and calculations for this, because active aero is illegal in most racing.  But it is still a fascinating area to explore.

tb (minimally supportive)
tb (minimally supportive) GRM+ Memberand Dork
12/21/20 11:54 a.m.

Just wanted to say that I am enjoying the intelligent discussion. Active aero is something I have thought about frequently but never toyed with. 

 

I do have a na miata challenge car that I plan to develop for several years and four unused headlight motors...

tb (minimally supportive)
tb (minimally supportive) GRM+ Memberand Dork
12/21/20 11:59 a.m.

 

I have watched this video more than once and it only adds to my list of questionable ideas 

SVreX (Forum Supporter)
SVreX (Forum Supporter) MegaDork
12/21/20 12:15 p.m.

In reply to tb (minimally supportive) :

I can't figure out what that crazy thing is doing!  It appears to be flapping all over the place, even on straights!

But yeah... questionable ideas!

Keith Tanner
Keith Tanner GRM+ Memberand MegaDork
12/21/20 12:20 p.m.

I warned the corner workers before heading out with the active aero the first time so I wouldn't get black flagged for what looked like parts about to fall off. I assume this guy would have to do the same!

NOT A TA
NOT A TA SuperDork
12/21/20 12:24 p.m.

For challenge cars you can think "out of the box" that most cars in sanctioned classes have to adhere to. I've read the challenge rules out of curiosity though I don't ever plan on attending. I don't think there's any rules against active aero that is waaay beyond merely tipping a wing into stall as a brake. You could have a 2' tall plywood wall shoot up out of the trunk lid as a brake, or a similar flat piece shoot out of one side of a car when turning.

Adding down force to the rear under braking might allow you to adjust a brake bias valve more to the rear to take advantage of the additional traction so total braking ability would increase. Shorter braking distance allows more time at the highest speeds during a run. Add it up and maybe get an extra 20-30-40 feet of the course at full speed. . No bias valve? Just run more aggressive brake pads, pads aren't a budget hit IIRC.

ProDarwin
ProDarwin MegaDork
12/21/20 12:27 p.m.

Once you get into all of that active stuff, would it be better to just play with active diffs/yaw control first?

Keith Tanner
Keith Tanner GRM+ Memberand MegaDork
12/21/20 12:31 p.m.

I guess it comes down to the complexity of the implementation. Is it easier to move a chunk of body around or to come up with a way to activate brakes on individual wheels on a car that was not designed that way from the factory?

sleepyhead the buffalo
sleepyhead the buffalo Mod Squad
12/21/20 12:37 p.m.
ProDarwin said:

Once you get into all of that active stuff, would it be better to just play with active diffs/yaw control first?

Robbie (Forum Supporter)
Robbie (Forum Supporter) GRM+ Memberand MegaDork
12/21/20 12:46 p.m.

I've daydreamed about a split wing, but again instead of using it for braking one side or the other I think about it for variable downforce, IE a 'virtual' roll bar. Crank the inside wing AoA for more downforce on the inside of the car. The bonus is that you'd also probably have more drag on the inside. But I still worry that really what you're doing is 'reducing' downforce from the outside of the car, and that may be an overall detriment. Ie, I'd probably take more grip over less roll, if those are the only two options. 

A rudder is a fun active aero thought though.

ProDarwin - I've also fantasized about using an ABS module and an arduino or similar to try and add brake based LSD to an open diff trans on a challenge budget. If you pick a car that comes from the factory with all the right mechanicals and sensors, you'd be adding LSD for very little actual budget. But how good are you at writing and developing that code?

sleepyhead the buffalo
sleepyhead the buffalo Mod Squad
12/21/20 12:55 p.m.
Robbie (Forum Supporter) said:

A rudder is a fun active aero thought though.

One of the problems with an 'active rudder' on a car, is the car can see a Beta range of 0-360deg.  Aircraft... generally... have a more restricted Beta range, outside of some landing conditions... the adverse most of which they'll tend to avoid.  Something you won't be able to do with the car.

Edit:

One of the other reasons, thanks to one of my horrible sketches...


is that the rudder will have to be fairly sizable, since the distance from the rudder to the CG is smallish, compared to a plane (especially w.r.t. the mass of a car vs. a plane).  And, if it gets big, then it will most likely stick well up above the car's CG, and this will create a roll moment away from the turn... which is not what you want.

sleepyhead the buffalo
sleepyhead the buffalo Mod Squad
12/21/20 4:37 p.m.

So, the first thing to keep in mind, with this stuff, is that aerodynamic coefficients are generally summed around the Geometric Quarter Chord (I.e. 25% the flow-wise length of the wing from the point of tangency to the chord at the front, to the point of tangency at the back).

Drag goes backwards, and generally in the case of race cars (and hopefully only temporarily for some aircraft), Lift goes down.  In the specific case of the wing chord being parallel to the wind vector being coincident with parallel to the ground, the Downforce is aligned with Lift:

[need to replace image with vector fix]


now, when we incline the wing with respect to the ground, to make more Lift and Downforce... the code/convention still calculates the Lift/Drag with respect to the chordline:

[ need to replace image]


thus, Downforce ends up being comprised of a minor component of Lift, but it's reduced by the vector... and by an upward vectorization of Drag.

However, we shouldn't let this get too carried away... because these estimation codes are generally only reliable in the "small angles" regime... +/-15deg.  30deg, tops.


you see, one of the challenges is, as you increase in angle, the stagnation point (the point where the air goes over one side vs, the other) moves back from the geometric Leading Edge, and forward from the geometric Trailing Edge.  Somewhere around 45deg, it'd really be more accurate to replot the airfoil vertically, with new Leading Edge and Trailing Edge points...but the shape would be too extreme for the code to give meaningful answers.  If you get a Cd in excess of 1.0 (for a single element) you're certainly outside the bounds of believable drag (parachutes max out around a Cd of 1.2-1.4, iirc).

Now, of course, recall that the wing is actually just a flap, and the car is actually the wing...


and the thing is, the presence of the two bodies in proximity means they compliment each other.  So, the flap can go to a higher than normal AoA, and still change the flow field... and that change in the flow field will cause the (in this case) underside of the car/wing to generate a lower pressure.

Thus, the force/moments generated by the wing, and summed up to moments compared to the wing/car CG, are relatively small ... in comparison to the increased downforce of the wing/flap or car/wing system.

so, it's not so easy to take a wing... analyze it in isolation, and assume to know how it's going to act on the car.  Because the car itself it the largest element in the flow field.  And, the shape of the car, the shape of the wing, the position of the wing w.r.t. the car, and their position relative to the ground all impact the final dynamics of the system.

sleepyhead the buffalo
sleepyhead the buffalo Mod Squad
12/21/20 4:39 p.m.

Tomorrow I'll see if I can come up with some time-based considerations for airbrakes.

1 2 3
Our Preferred Partners
zbusRcEZJ24QLWtPKWlj5A9kEKijPXE2yY44Raxiocrh7n0yk1jYuxm1yRDTb02N