Mini-Z Aero Development

[color0]

Been a long time since I've been back here, so I thought I should make this post worthwhile by telling you guys and gals about some uber-cool stuff I/GSR have been up to in the last year or so:

Aerodynamics.

People used to tell me that aero did not matter at 1/28, that a body's weight balance was more important than the downforce it produces. We've long suspected that's wrong, but some recent developments have completely blown it out of the water -- aero matters, and a lot! I'll start off from the top to cover all the aero devices we've been playing with and just how well they work.

First thing on my to-do list was the Gurney flap -- these work on 1/1 cars, sometimes on 1/10 cars, do they work at 1/28? I had no idea. So I did some calculations based on some research papers, and, assuming the entire bodyshell was to be treated as a wing, came up with a measurement of between 3 and 4mm as the optimal height for a Mini-Z-sized Gurney flap. Here it is:



And as you guys may know, a Gurney flap needs some airflow underneath it to work. So I opened up the rear of my shell and gave it some air.



Track test: the Gurney flap weighs about 0.2g -- confirmed on a scale. That's a negligible amount of weight, so the effect on handling (where weight distribution is most obvious) should have been unaffected if there's no aero effect.

However, to my pleasure this was not true. The Gurney flap adds a little bit of traction from medium speeds and up! That little bit of extra traction makes the car a lot more stable. On the other hand, the Gurney flap is rather fragile, one touch and it cracks due to the plastic I used, and also didn't provide enough downforce to handle a more powerful (43t, neo magnets) motor. So eventually I went back to a "traditional" scoop wing, especially after seeing some analysis of flap spoilers vs. raised wings (credit to ruf on MZR for finding this pic, the actual picture is on www.auto123.com):



You can see that the airflow is definitely smoother with the raised wing model vs the flap spoiler, which means the raised wing is likely more efficient and will produce the same drag for more downforce, or the same downforce for less drag. This was my main motivation for dumping the Gurney flap and switching to a scoop wing. More later.

The next aero device I tried was a LeMans-style front diffuser. Making it was not terribly hard, but also not the easiest task on earth as you need to be PRECISE if you don't want to screw up your handling.



The front diffuser's job is to smoothly route the air upwards from underneath the bumper, thus generating downforce. Here are 2 shots with the diffuser installed that should make the point clear:





Track test: The diffuser weighs in at about 1 gram. This isn't a negligible weight, however I do know that if there's no aero effect, at high speeds the added weight will actually make the car understeer. So I tested out the diffuser bearing this in mind.

Turns out to be Absolutely not the case. The front diffuser on its own added so much front grip that I could not stop traction rolling! Any full steering input with the intent to negotiate a corner would see the car hunker down, turn really fast, and then either spin out or flip on its lid. This was at almost every corner too, not just the high speed ones. The Gurney flap was definitely not enough downforce to handle the front grip, I had to put a scoop wing on my car to balance it out. And afterwards the handling became incredible. Have you ever gone from being a mid-pack racer to lapping the field twice in 5 minutes? Yeah, that's what happened as soon as I paired a rear wing with the front diffuser. I've never seen such a ridiculous improvement before, and from then on I was convinced that aero is going to play a very important role in Mini-Z racing.

Now if front diffusers work, what about rear diffusers? I hadn't been able to test it out yet before a fellow MZR member made a beautiful underfloor and rear diffuser for his Mini-Z (credits to Rune on MZR):





I had been wondering how effective this could possibly be, since our Mini-Z's are run at very low ground clearances and air might not even make it to the back of the car in a smooth manner. But I was mistaken, Rune has since reported back that the rear diffuser also works. No laptime or race data as of yet but he notes a marked improvement in rear traction, without the car feeling like it's any heavier (his underfloor + diffuser weigh 1.8g). This to me is amazing, knowing that we can adapt existing designs that others have given up long ago and through good execution, make it work! We've made underbody aero tuning just another part of the process of setting up a Mini-Z.

Now I have one thing to discuss: basic aerodynamic theory applied to 1/1 cars dictates that aerodynamics scale with speed squared, so low speed aero effects are almost negligible and at high speeds they are far dominant to suspension settings. The rear scoop wing (see Rune's Supra above) feels like it works that way, the effect is minimal at low speeds (~3mph) and rises sharply as the speeds climb (say, to 15mph). But the front diffuser does NOT seem to work like that. Its effect was noticeable at ALL speeds high and low, and only noticeably lost its effect at the slowest one or two turns on the track where I race. The front downforce feels like it's "kicked in" in the entire rest of the layout, both infield and high speed section. This is mathematically not impossible, I know that as objects get smaller, the aerodynamic effect has been known to possibly scale with speed instead of speed squared (look up Reynolds Number and Viscous Flow if you're knowledgeable and curious). But is that the effect I'm seeing with these cars? I have no idea, and I'm not about to BS that it is without confirming that it is.

Which is why we (GSR) are doing two things right now: 1) Building a wind tunnel to try and get some numbers on the aero devices installed on my car, and 2) Trying to get legitimate wind tunnel (hey, I'm a college student now!) time or simulation software to see if there is a way to visualize the airflow going through the shell. It's good stuff.

In the meantime, I've implemented one more aero mod for testing: ducting.



See the intake and the diffuser behind it? The diffuser scoops the air upwards and directs it through vents in the hood:



And you can see the pathway the air follows looking through the hood:



I myself used to think that it wouldn't work, especially if the holes were not large enough. Air forms a boundary layer around objects, so I was thinking that small intakes like the ones pictured would never be able to move enough air to make a noticeable aero difference. But if my last track results are anything to work with, I might have just proved myself wrong. The car gained a lot of front grip at high speed -- it was actually enough to almost overwhelm my rear wing. With the speed affecting front downforce more than before (i.e. purely front diffuser) I think that I could attribute the newfound front grip to the intake/vent system that I've cut into the body. More testing will tell, but it seems like it's got potential and would rock the boat by opening up MANY more areas of our bodies to aerodynamic modification.





I'd like to end this post by posting pictures of the entire car I'm currently campaigning, a Kyosho Mini-Z MR-03 with a 911 GT1 body stretched to 94mm wheelbase. (Also the only appearance of the scoop wing in this post, sorry for showing it so late! ) This car has definitely been great to work with, and the aero mods have shown so much promise that I'm almost turning my hobby into an area of academic interest. Hope you guys enjoyed the read and I hope to have some more updates for everyone a couple months down the road.

[TheB1GDude]

Awesome! Great write-up and lots of interesting technical stuff!!

If you could get to the point of being able to produced solid reliable numbers showing the increases in performance would you look to mass produce any kits?

[color0]

Solid reliable numbers should just be a matter of time, I mean the wind tunnel IS getting built and even if it doesn't do full airflow visualization it will still give us solid, (hopefully) repeatable numbers for downforce and drag.

Mass production, however, would be a little bit sketchier. Many Mini-Z bodies don't have a very good place to mount aerodynamic devices, this particular 911 GT1 is a very special case -- but it's not a very popular racing body. I'd probably have to figure out diffusers for the Ferrari F430GT, 360GTC, Mosler MT900, McLaren F1 LM, Lexus SC430 GT500, etc. etc. as those are MUCH more popular and more people would buy a diffuser for them. Cutting the actual diffuser itself would be easy, it's just a flat piece of plastic cut to a pattern, which the user then bends to shape and adheres to the shell.

*Shrug* possible, but I'd need to buy or borrow a lot of shells.

[TripleXmodder(akaXXX)]

Awsome post for your return.. I enjoyed the read, keep at it. Since im still in Afghan the military net blocks photos, so I couldn't see the pix, but Ill be home soon and can't wait to re-read your article. Im sure the dedicated racers (Which I am not) will really be into this. Awsome man, and great to have you back.

[Kaze]

Originally Posted by color0

Now I have one thing to discuss: basic aerodynamic theory applied to 1/1 cars dictates that aerodynamics scale with speed squared, so low speed aero effects are almost negligible and at high speeds they are far dominant to suspension settings. The rear scoop wing (see Rune's Supra above) feels like it works that way, the effect is minimal at low speeds (~3mph) and rises sharply as the speeds climb (say, to 15mph). But the front diffuser does NOT seem to work like that. Its effect was noticeable at ALL speeds high and low, and only noticeably lost its effect at the slowest one or two turns on the track where I race. The front downforce feels like it's "kicked in" in the entire rest of the layout, both infield and high speed section. This is mathematically not impossible, I know that as objects get smaller, the aerodynamic effect has been known to possibly scale with speed instead of speed squared (look up Reynolds Number and Viscous Flow if you're knowledgeable and curious). But is that the effect I'm seeing with these cars? I have no idea, and I'm not about to BS that it is without confirming that it is.

Which is why we (GSR) are doing two things right now: 1) Building a wind tunnel to try and get some numbers on the aero devices installed on my car, and 2) Trying to get legitimate wind tunnel (hey, I'm a college student now!) time or simulation software to see if there is a way to visualize the airflow going through the shell. It's good stuff.

Idk about your college but ANSYS was available at mine. It's a pricey Analysis/CAE software it does amazing work though. I do know that you can program the fluid's properties (remember air is different at elevations).

As for the Aerodynamics comment, It took me a couple days to think about this and even discuss it with a friend of mine and we came to an agreement, and hopefully this helps give you a better understanding.

That agreement was this: When you apply Aero/Fluid dynamic theories, it's important to remember that "scale" is NOT something as simple as say having a 1:1 Impreza and having the 1:28th and saying the Coefficient of drag at 15 mph is proportionally the same. The best way (that we figured) to apply flow theory is to treat the Shell as its own object (in other words forget that its 1/28th and just say its a small object in a control region).

So say you're traveling at a slow speed (1 mph), picture it as everything but the car is moving (so the car is static and the air is coming to it at 1 mph). Everywhere on the shell thats perpendicular to the ground (90* with the ground) the Velocity of the air at the spot will be 0 (because air is hitting it and not deflecting so it's energy being absorbed by the car, aka drag). Now any surface thats angled will deflect the air, it won't deflect perfectly though there will be some drag and some flow. Once the air reaches the top of the angled surface (look at the top of the windshield in the pic of the airflow you posted, i'll post it below) the Velocity of the air actually increased because the deflection accelerated it.



This could help explain to you why a front deflector is working at all speeds, mind you that on a shell modeled after a real car all the "grills" are flat areas which would increase the drag force, by cutting out some of the areas that create drag and putting in something to deflect the air up (which pushes the body down), you're turning the drag force into down force.

As for the spoiler, well as flow passes an object (and you already looked up the reynolds number), the flow is interrupted and can become turbulent. In the pic you posted (I reposted above) you can see the the flap spoiler captures some of the air flowing past the car and as it passes it, the flow itself is becoming turbulent. The raised spoiler will actually capture and null (Velocity = 0) the flow hitting it to create a greater downforce. It's essentially a design of efficiency vs not using the air thats there.

Hopefully this all makes sense, the main point however is that a 1mph (low speed) drag/down forces would potentially have a strong force behind it due to such a small object. Surface area is key here, and I know Aero affects these cars because i've driven my Xmod Lan Evo VIII outdoors on windy days, and when a breeze, that is enough to slow my walking speed down, hits it, i really thought it would flip, but it didn't and even crosswinds couldn't.

Good luck with your research and wind tunnel.

[color0]

It does, but does not explain the downforce vs. speed relationship that I seem to be observing. I have enough understanding of fluid dynamics to be past the very basic description you've provided, but right now it's quantifying theories and data that's going to be fun/hard.

My understanding of intermediate-scale fluid mechanics (such as 1/28 cars) is that it's right at the boundary between viscous flow (bugs, germs, etc.) and inertial flow (real cars, planes, etc.) and there's a combination of both effects that's taking place. I went and talked to a few professors about it, if you do the math then the speed, size and shape of a Mini-Z lands it squarely in the "intermediate" Reynolds-number territory. This should mean that the boundary layer around the object is quite sizable compared to the object itself, and air hitting the front of the car doesn't actually hit the car but rather hits the boundary layer some distance ahead of it (1-3cm, depending on speed?). That makes each body a lot "rounder" than it looks, but also significantly slows down any air going under the front bumper.

The Reynolds number thing might be why the front diffuser in particular works. In the local area of the front bumper the Reynolds number gets smaller, since the diffuser is smaller than the whole car and the air the diffuser is fed is significantly slowed down vs. the vehicle's actual speed. So given that 1/28 cars start out at the "intermediate" stage, one could expect that the lower-Reynolds front bumper area (smaller size still, lower air speed as well) would predominantly follow low-Reynolds number flow characteristics, where drag/force are proportional to speed^1 (linear relationship).

Meanwhile, a raised rear spoiler is outside the boundary layer and is subject to air moving at pretty much the speed of the car, so the intermediate-Reynolds flow holds and you get a not-quite-linear relationship between drag/force and speed. That's the theory that I've come up with.



Meanwhile I made another front diffuser, this time for a more "standard" racing body, the Ferrari F430 GT.





The angle and volume are more aggressive than the 911's diffuser so I'm hoping to get more downforce on the front end, period. I do think that I may have hit the limit for diffuser angle though, the angle in these pictures seems to be too much, and I got some weird handling inconsistencies when testing the car. I took off the body, straightened the diffuser a bit (reducing its angle), put it back on the car, and immediately everything got consistent again. This is something else I want to explore, the fluid flows here are just so interesting and hard to figure out I think I'm gonna need software on top of the wind tunnel, just like you've suggested Kaze.

Meanwhile I built up the rear diffuser and central underfloor too, but these sadly don't seem to make any useful downforce other than compensating for their own weight.





I'm guessing that my ride height might be too low to allow airflow under the car (boundary layer stuff again?) because the pieces don't make any appreciable downforce other than to compensate for the weight they add. Rune's rear diffuser actually adds grip, he says, mine pulls its own weight but doesn't help out on top of that.

Hobby turned academic project is a great excuse for spending time on the hobby, lol. Glad to be (somewhat) back, happy to share any discoveries with you guys since this "initial-period" research isn't going to spread by word of mouth alone.