Get instant digital access.
Subscribe Now!

Downforce Sorcery


Story by Steve Stafford

For half a century, wings and spoilers have helped race cars stick to the track. Lately, though, another piece of equipment has joined the mix: underbody rear diffusers. Originally seen only on pro cars, these diffusers quickly made their way to the amateur ranks. Stroll through the grid at a club race, time trial or track day, and you’ll see these devices peeking out from beneath back bumpers.

Is it a case of monkey see, monkey do, or do these diffusers actually help a car stick to the road? To find out, we asked a real aero engineer, Steve Stafford. He cut his motorsports teeth in autocross, but since then he has done aero for Formula 1, Indy, NASCAR and Grand-Am efforts.

What Do Diffusers Do?

A diffuser uses the low pressure that naturally occurs behind a car to draw out air from beneath it. The result: downforce that hugs the tires to the track.

The walls of the diffuser contain the suction and increase the airflow through the unit’s entrance, or throat—normally its forward-most and lowest point. This increased flow at the throat reduces pressure by way of a venturi effect.

The length and angle of the diffuser dictates how far forward the increased downforce can act on the car. The longer and shallower the diffuser, the farther the downforce will reach.

A steeper angle on the diffuser translates linearly into more downforce, provided the diffuser doesn’t stall. A diffuser stalls because the air can only expand up to a certain rate. Stalling limits the downforce that can be created.

In a diffuser that is too steep or too short, the air will not be able to follow the surface smoothly; it will break away from the surface to find the path of least resistance. This places a theoretical maximum angle of 9 degrees, but height from the ground has much greater influence than any other factor.

How Can You Maximize Your Diffuser?

Rules may limit the size and location of a diffuser—if they don’t ban them altogether—so maximizing its performance can require a great deal of development for just those last few percent. There are a few ways to accomplish this, but they each come with caveats.

Incorporating the Exhaust: Allowing the exhaust pipes to dump into the diffuser helps accelerate air flow, providing more downforce.
Downside: The diffuser loses a significant portion of downforce as the car transitions from open throttle to closed throttle. This can shift the balance to the front when the driver least wants that to happen.

Adding Strakes: Adding strakes or fences inside the diffuser can delay the stall of the airflow or allow more angle to be used.
Downside: This increases complexity and the possibility of very narrow operating ground-clearance ranges.

Adding Wickers: Wickers can make the outside faces of the diffuser more efficient by reducing the inflow from the sides and keeping the low pressure contained in the unit. They can be used at the vertical trailing edges as well.
Downside: Because wickers are typically located close to the ground, they’re susceptible to damage and wear.

Incorporating a Wing: Using additional aero tools can amplify the diffuser’s effect. A wing or other device that creates very low pressure near the diffuser exit can reduce stall tendencies or allow more diffuser angle to be run.
Downside: Now there are more parts in the puzzle to build and tune.

How Do You Build a Diffuser?

Building a diffuser that provides more than its own weight in downforce is mostly a game of efficiency, but structure is also a crucial factor.

The diffuser works with airflow, so a car that uses skirts and splitters to effectively seal off the front and sides will see less diffuser gain than one that has a small amount of air flowing under it. Remember, like a wing, a diffuser needs airflow to work.

The diffuser’s throat should be as far forward as possible and at the lowest part of the vehicle. This increases downforce without significantly affecting car balance. Plus, this location is unlikely to become sealed off unless the entire car bottoms out.

The throat also needs to have a relatively smooth flow leading into it. That may mean trying to have a flat area ahead of the diffuser to reduce any disturbances in the flow to the diffuser.

A diffuser can generate a large amount of downforce, and the effective suction on the part can be several hundred pounds. First, ensure that the material doesn’t flex; diffusers are most commonly made of carbon fiber, but fiberglass, wood and sheet metal can work, too—they’re just a little heavier. Second, make certain that the device is securely attached to the car. Spreading out the attachment loads to several locations will help to reduce the deflection of the diffuser. Plus, if a mounting point fails, there will be several backups.

If the car exhibits more than 2 inches of suspension travel, consider running sliding skirts along the sides—if they’re legal for your class—to maximize the suction. Since the distance to the ground is such a major factor in the diffuser’s effectiveness, limiting the travel or using sliding walls that attempt to close off the sides will help. Sliding skirts can be created by adding a second wall panel pivot near the front of the diffuser and limiting it before it opens a gap at the back edge.

Since most cars don’t have unobstructed rear underbodies due to their suspensions and exhausts, sometimes a squared-off diffuser profile shape can be adjusted to maximize the benefits. This may mean that the diffuser becomes shallower but wider in those areas. This isn’t ideal, but it is a way to continue the diffuser as far forward as possible.

Armchair Aero Analysis

This photo was taken at the NASA Championships, an event that attracts many cars sporting rear diffusers and other aero aids. We asked Steve Stafford, our aero engineer, to do a little armchair analysis:

The black hatchback has several features beyond the diffuser—all done nicely. The rear wing is contoured, the skirts are very close to the ground, there are dive planes on the front corners, the front wheel arches are vented, and I suspect there is a splitter as well.

The car’s attitude suggests that it doesn’t travel very much and can keep the diffuser near the ground. The diffuser is as wide as possible between the wheels. The sides have outward-facing flanges and are close to the ground. The angle appears to be near the theoretical limit. With the width of the diffuser, the strakes help to maintain operation of the wide diffuser. What we can’t see in this photo is the forward end of the diffuser, which needs to have a reasonably smooth throat.

The car can effectively use the diffuser and maximize downforce if it sports a complete aerodynamic underbody treatment. However, keep in mind that making more downforce provides a very different driving experience. If the downforce is disrupted for any reason, the car can become nearly undrivable.

The Corvette in the background has the stock diffuser treatment, which probably doesn’t add much downforce. The Corvette appears to be quite low, so the diffuser may be fairly effective—even though the trailing edge expands too quickly and, due to that, likely loses some effectiveness. Without seeing more of the car, it’s difficult to read too much into it.

Parting Words

Since the diffuser is making downforce, there is likely to be an increase in drag. Generally this effect is small at normal speeds, but it can be substantial as a car approaches terminal speed. The key, of course, is designing an effective diffuser and then properly testing and tuning it for maximum efficiency.

Join Free Join our community to easily find more articles.

Comments

View comments on the GRM forums
stafford1500
stafford1500 Reader
11/5/15 1:40 p.m.

Hey, I know that guy...
Thanks for letting me help out on a few articles.

hotchocolate
hotchocolate Reader
11/5/15 3:59 p.m.

Speaking of downforce this really astonished me. Hope it is sppropriate to prayer here. http://www.autoblog.com/2015/11/04/2016-dodge-viper-acr-13-lap-records-video/

Ed Higginbotham
Ed Higginbotham Editorial Assistant
11/6/15 8:06 a.m.

In reply to stafford1500 :

Thanks for contributing, Steve!

David S. Wallens
David S. Wallens Editorial Director
11/9/15 7:48 p.m.

Yes, working with Steve on that piece was fun.

WaterfordMan
WaterfordMan New Reader
11/10/15 9:28 a.m.

Hey! I'm pretty sure I made the canards on that civic! They've did a bit of trimming, but I think those are mine! My buddy at DHP certainly made the splitter.

olpro
olpro Reader
11/10/15 9:55 a.m.

Since it seems to come down to drag vs downforce,why isn't it more effective to properly size a wing above the car's rear?

Keith Tanner
Keith Tanner MegaDork
11/10/15 9:59 a.m.

More effective than what? Wings and diffusers usually work together.

STM317
STM317 Reader
11/10/15 11:07 a.m.
olpro wrote: Since it seems to come down to drag vs downforce,why isn't it more effective to properly size a wing above the car's rear?

Keith is right. They work really well together, and they're often found together.

I have zero science to back this claim up, but it seems like the underside of most vehicles would be much "dirtier" aerodynamically speaking than the shiny side. Perhaps by improving the worst area of the vehicle, you get a greater gain than what you'd see with a wing that has equivalent drag/downforce properties.

stafford1500
stafford1500 Reader
11/10/15 11:19 a.m.
STM317 wrote:
olpro wrote: Since it seems to come down to drag vs downforce,why isn't it more effective to properly size a wing above the car's rear?

Keith is right. They work really well together, and they're often found together.

I have zero science to back this claim up, but it seems like the underside of most vehicles would be much "dirtier" aerodynamically speaking than the shiny side. Perhaps by improving the worst area of the vehicle, you get a greater gain than what you'd see with a wing that has equivalent drag/downforce properties.

STM317: you are on the right track with cleaning up the flow under the car. Also a rear wing will help to energize the flow under the car to make the diffuser work better. Most cars have alots of little pieces hanging out in the flow under the car, which basically makes lots of little drag sources. Cleaning up the flow will remove/reduce the amount of drag from the area and allow the air to move more quickly and generate lower pressures over a greater area (more downforce). Short stubby things hanging out in the breeze have the biggest drag impact relative to size.
OldPro: the typical wing has a fixed drag to downforce ratio that will limit its performance. Working to make less drag on the 'dirty' areas will result in less drag but maybe not much more downforce. That works out to a huge increase in efficiency defined as Lift over Drag.

stafford1500
stafford1500 Reader
11/10/15 11:21 a.m.

If anyone else has questions, post 'em up. I hang out quite a bit, but don't post unless I have something helpful to say.

SVreX
SVreX MegaDork
11/10/15 11:54 a.m.

In reply to stafford1500:

Thanks Steve.

Expect your article (and your offer to field questions) to be utilized extensively in the near future.

Tyler H
Tyler H SuperDork
11/10/15 12:30 p.m.

This was a great piece, thanks! I'd love to see an in-depth article in the magazine, (or a multipart series,) including grassroots design and construction techniques.

If you could use say...a 95 Civic coupe...that would be especially helpful.

mmurray
mmurray New Reader
11/10/15 1:42 p.m.

Articles like this are why I love GRM. Having Steve in the comments is icing on the cake.

BRB going to start mocking up aero bits in cardboard and dream about having a budget.

RXBeetle
RXBeetle Reader
11/10/15 1:59 p.m.
Tyler H wrote: This was a great piece, thanks! I'd love to see an in-depth article in the magazine, (or a multipart series,) including grassroots design and construction techniques. If you could use say...a 95 Civic coupe...that would be especially helpful.

(edited, quoted the wrong post... and pics. Also thanks Steve for the contibution! just listened to the podcast.)

I designed/built an undertray and diffuser for the FSAE team I was on a few years back. It was very proof of concept with the idea design principles of: 1. has to come off easily 2. no composites, home depot level tech 3. simple to make and modify

I settled on twin-wall polycarbonate, the price has about doubled since my build... twin wall polycarbonate sheet

It's not the most rigid material as a sheet but boxing it made for some good rigid tunnels. Flat sheet could be stiffened with a bonded aluminum angle which are also good attachment points. Everything was glued up with automotive panel bonder.

All the cutting was done with a pneumatic cut off wheel and the joints were formed with a dremel router bit. It handled heat pretty well, it was right next to the header under the car which caused the skin to blister but never really caused much of an issue.

The diffuser isn't as efficient per square area as an airfoil but it has 2 huge advantages, tremendous available square area and ground effect. The drag delta with/without the diffuser is small but the downforce # can be pretty huge giving you a pretty incredible L/D compared to a conventional wing. Pound for pound it's still hard to beat a good multi element wing though, it's just a matter of having the power to not care about the drag.

SVreX
SVreX MegaDork
11/10/15 2:13 p.m.

Question:

On a full flat bottom/ undertray that spills into a diffuser, is the primary purpose cleaning up undercarriage airflow, or creating a downforce surface? It affects construction methods.

If it is just clean air, strength (particularly strength per square inch) is insignificant.

But if it becomes a downforce surface, it would not serve it's purpose unless it was strong enough to resist the negative pressure, and draw the car down with it.

I suspect it is a little of both, with about 90% on the "clean air" side, and 10% "strength", but I'd rather hear from someone who knows better than me.

edizzle89
edizzle89 HalfDork
11/10/15 2:18 p.m.

I prefer vacuum downforce

stafford1500
stafford1500 Reader
11/12/15 7:14 a.m.

SVreX: sorry for not getting back to this for a few days.
The two goals you mentioned are linked. You can not get the best downforce if the surface is not 'clean'. Generally cleaning up the underbody means creating a surface that is closer to the ground and provides a surface for the pressure to act against. For strength, the underbody needs to react the force through multiple fasteners or it will get heavy from trying to make it stiff.
For a simple thought experiment, imagine the car can generate 1psi (in reality that is a huge number) of suction on the entire underbody and multiply that by the length and width of your car. You will have a seriously big downforce number. If you try to support all of that with one, two, three, or even four bolts/rivets you will get lots of deflection which decreases the effectiveness or rips parts off the car.
The strength of a diffuser comes from the vertical walls and trying to use lightweight but thick sections. The double walled polycarbonate in the post above is a good example. Lastly, shape matters. Smoother shapes will get better results than defined angles between major flat sections.

sachilles
sachilles UltraDork
11/12/15 8:11 a.m.
olpro wrote: Since it seems to come down to drag vs downforce,why isn't it more effective to properly size a wing above the car's rear?

Another consideration is rules. Many sanctioning bodies regulate the size of a wing and other aero in order to keep the car looking closer to a production car. Items under the car are out of site and out of mind.

STM317
STM317 Reader
11/12/15 8:20 a.m.

Obviously the article is focused on diffusers and underbody aero in racing applications, but what about street driven vehicles where more ground clearance is needed? Would there be any performance benefit if the diffuser were say 6 inches off the ground?

I'm guessing the drag reduction would still be present which would probably help fuel economy, and perhaps acceleration/top speed, but there wouldn't be much downforce generated. Does that seem reasonable or am I off base?

alfadriver
alfadriver UltimaDork
11/12/15 8:56 a.m.

In reply to STM317:

Why would someone need more downforce on the street? The extra only helps when you need to increase the grip limits on the car.

Having driven cars at their physical limits, I've never come close to that on the street.

For that matter, if you raise the top speed 5mph, I don't see that as a help on the street as well.

Maybe the acceleration. Maybe.

For sure, the fuel economy.

AaronBalto
AaronBalto Reader
11/12/15 9:20 a.m.

I read an old review of one of my cars--a '92 NSX--and the writer mentioned that the weight distribution is exactly the same at 168MPH as it is standing still. Apparently, the engineers tuned the shape of the car to maintain the same dynamics at any speed. I have confirmed this on my home-built Grassroots rolling-floor wind tunnel that I made using 64 attic fans, some plastic sheeting, and a staple gun. (not)

Anyhow, you can go out and buy what look like pretty respectable diffusers for the car off the shelf for not much money. What I am wondering is whether it is really a good idea to tinker with the work of the gods? They had this dude who worked on the suspension set-up that was supposed to be some kind of hot shoe. Ayrton Senna or something? Sounds like an illegal, if you ask me, but people seem to think he knew his stuff.

But seriously, do you mess with this knowing that you are going to have to try to work out a more effective splitter or something to help the front of the car? Or am I just trying to improve something that can't really be improved?

Keith Tanner
Keith Tanner MegaDork
11/12/15 9:54 a.m.

You can do some interesting things with aero and balance. Usually, you try to set it up so that you have the F/R downforce the same as the static weight distribution - this keeps the handling balance consistent regardless of what's going on with the air. But I've got my car set up to generate a bit more rear downforce. This means that it's more prone to oversteer on tight corners to let me rotate the car, but moves towards understeer at high speeds for a bit more stability. It's easy to tune that with the wing.

Of course it's a good idea to tinker. You'll never really learn anything if you don't. When Honda raced the NSX, they altered the aero to make it work better. And I'll bet they changed the suspension as well.

The guys at Slick Auto put a lot of time into aero tweaks with their Miata. It's not fully documented, but you can see quite a few of the tidbits on their blog. Definitely attainable by the garage builder. http://slickauto.net/?q=blog&page=1

NOT A TA
NOT A TA Dork
11/12/15 10:20 a.m.
stafford1500 wrote: If anyone else has questions, post 'em up. I hang out quite a bit, but don't post unless I have something helpful to say.

I've seen your posts an aero stuff here and would have PMed you previously but PMs don't seem to work for me for whatever reason. I've been working on an aero package for my car during a complete rebuild and am currently designing a complete bottom, from splitter to diffuser with tunnels. I don't race the car in any particular class so no regulations to follow. I plan on different setups for road tracks, LSR's, drags, open road races. The full project thread is here https://grassrootsmotorsports.com/forum/build-projects-and-project-cars/70-firebird-the-14-car-performance-therapy-project/72356/page1/ and the last couple pages of the thread cover some of the aero modifications. If I start a thread on the just the aero stuff would you have time to offer advice?

As for a question I've been unable to find an answer to. Is there an advantage/disadvantage to using a squared off roof tunnel as opposed to a round roof? I understand the bottom edges being square to create a rolled vortex in the tunnel but why the top?

I gave away my drafting table when I moved to FL, and never learned Solidworks or other CAD programs so I'm designing GRM style using DIG (Draw In Garage). Old school, but it'll work.

[URL=http://s240.photobucket.com/user/NOTATA/media/001_zps91ycuhio.jpg.html][/URL]

STM317
STM317 Reader
11/12/15 11:54 a.m.
alfadriver wrote: In reply to STM317: Why would someone need more downforce on the street? The extra only helps when you need to increase the grip limits on the car. Having driven cars at their physical limits, I've never come close to that on the street. For that matter, if you raise the top speed 5mph, I don't see that as a help on the street as well. Maybe the acceleration. Maybe. For sure, the fuel economy.

I wouldn't be tickling the limits on the street. I was thinking about a dual purpose vehicle that spends time doing some track work/ autocross as well as being driven on the street on nice days. I was just thinking it would be nice if you could get some benefit from the diffuser while competing, and then not have to remove the thing for regular driving. Obviously, it wouldn't be the most effective tool for creating downforce, but any improvement in that area would be welcomed during motorsports events, and the fuel economy gains would be nice on the street. If it could have some usefulness during each driving situation that would be ideal, even if it's a compromise in one specific area.

SVreX
SVreX MegaDork
11/12/15 4:05 p.m.
AaronBalto wrote: I read an old review of one of my cars--a '92 NSX--and the writer mentioned that the weight distribution is exactly the same at 168MPH as it is standing still. Apparently, the engineers tuned the shape of the car to maintain the same dynamics at any speed. Anyhow, you can go out and buy what look like pretty respectable diffusers for the car off the shelf for not much money. What I am wondering is whether it is really a good idea to tinker with the work of the gods?

Aaron Balto:

I think it is important to note that weight distribution is not the same thing as downforce, grip, or even balance.

An F1 car can generate 5g's of downforce (that's 5X the downforce), without changing the weight distribution.

Weight distribution is generally expressed as a percentage. Downforce is expressed as a weight. A 3000 lb car with a 50/ 50 weight distribution that develops 1000 lbs of downforce would have 1500 lbs on each axle standing still, but 2000 lbs on each axle at speed if it maintained the same distribution.

The NSX has 58% on the rear axle. Your statistic only means that it also has 58% on the rear axle at speed. It does not mean it has the same downforce.

It also does not mean 58% is the optimum distribution for autocrossing, or drag racing, or hotlapping, or land speed records. Each of those would have an optimum distribution AND an optimum downforce for a given grip.

The designers were not gods, and did not design the car to do all things. Feel free to improve on anything they did for a particular purpose.

AaronBalto
AaronBalto Reader
12/5/15 9:49 a.m.

In reply to SVreX:

Um, no. The NSX was set up by some guy named Ayrton Senna. So I would take issue with your final premise. This car was, in fact, set up by a god. Screwing around with fluid dynamics in the shadow of the people who developed this car could very likely be a fool's errand.

At least until I finish the grassroots rolling floor wind tunnel in my basement.

MrJoshua
MrJoshua UltimaDork
12/5/15 12:02 p.m.

In reply to AaronBalto:

Setup by Senna to do what? Achieve the highest speed possible? Do the best on a wide open road course or a tight one? Autocross? Drive down the highway at 70 mph? Go over speed bumps? Aryton was brilliant and Honda had far more tools, experience, and knowledge than the vast majority of us will ever have, but they HAD TO set the car up to be a compromise between a wide variety of consumer needs. Thinking that no body can set up the car to do ONE thing better on a car that was designed to to MANY things well is limiting yourself.

SVreX
SVreX MegaDork
12/6/15 7:32 a.m.

In reply to AaronBalto:

I'll let Mr Joshua respond for me on this one.

The gods are fallible. Well... the gods design for different parameters than we generally do.

Smitty
Smitty New Reader
3/23/16 6:13 p.m.

A car is a rolling compromise. Ayrton Senna had some input into the setup, but in reality, probably not as much as the EPA, NTSB, the bean counters, and the product liability lawyers. Most of the people on this web site would prefer to maximize Mr. Senna's influence, and minimize that of the others. There is always room for improvement (for a particular purpose). and Not A TA Dork should try CAD (Cardboard Aided Design)

NOT A TA
NOT A TA Dork
3/24/16 9:12 p.m.

I've been a cardboard aided design fabricator for well over 40 years! ahahaha Box in the bottom pic is for my undertray.

[URL=http://s240.photobucket.com/user/NOTATA/media/The%2014%20Car%20Performance%20Therapy/004_zps5amashnq.jpg.html][/URL]

[URL=http://s240.photobucket.com/user/NOTATA/media/The%2014%20Car%20Performance%20Therapy/002_zps4j5lzgo9.jpg.html][/URL]

Grassroots Motorsports Magazine

Subscribe Today

Also get your instant access to the digital edition of Grassroots Motorsports Magazine!

Learn More
g5Ajl9QR68vjHhrQA31LYSccET0PXKdn