The Mustang Forum for Track & Racing Enthusiasts

Taking your Mustang to an open track/HPDE event for the first time? Do you race competitively? This forum is for you! Log in to remove most ads.

  • Welcome to the Ford Mustang forum built for owners of the Mustang GT350, BOSS 302, GT500, and all other S550, S197, SN95, Fox Body and older Mustangs set up for open track days, road racing, and/or autocross. Join our forum, interact with others, share your build, and help us strengthen this community!

Let's Talk Aero

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Grant 302

basic and well known psychic
For example, I'd challenge anyone to explain all the functions of the canards on the Mustang GT4 *without* CFD analysis. It's hard to understand what we can't immediately visualize.

Since we have new people checking in on this thread, how about expand this 'challenge' to those *with* CFD capabilities...or access to wind tunnel info?
 
Since we have new people checking in on this thread, how about expand this 'challenge' to those *with* CFD capabilities...or access to wind tunnel info?

" I'd challenge anyone to explain all the functions of the canards on the Mustang GT4 *without* CFD analysis. It's hard to understand what we can't immediately visualize."

If you could do this, you would be a millionaire. The flow structures generated by complex bodies cannot be visualized a priori, they need to be calculated or evaluated in a simulation environment like a wind tunnel.

I’m sure multimatic was solving a problem with those canards, could be vortex bursting, streeting, and it might have had advantages further down the body of the car than the strictly at the nose.

A brief primer, but this is all over the internet with legitimate and not so legit explanations. But essentially canards take advantage of the negative pressure gradient on the surface of a splitter. They are falling out of favor on high end WTAC cars because their L/D ratio is always horrendous compared to a front diffusor for example, but you use them when you are desperate for front downforce. A splitter has a huge surface area in comparison to canards. Canards will do nearly nothing without a splitter.

I’m working on a front DF package for my mustang, and I’m testing out some concepts on the simple model I shared above. Baseline is the body with no aero except the rear wing at +5AOA, then I added a huge front flat blade splitter, and then added endcaps/endplates to the splitter, then removed them and added “generic” 4 large canards.

As you can see in the graphs, the splitter provides a huge boost in CL (downforce) at a negligible drag penalty, while endcaps add a massive amount of DF (more than I expected, I but I suspect that the swirling wake kept air out of the wheel well and ameliorated the tire squirt a bit) with a good sized drag penalty, while canards added a bit of DF, but with a worse L/D ratio (higher is better).

SimpleFront.png
 
Last edited:
I'm not sure why you'd say it's lacking the gurney flap...I see it more as not needing one. It's a low production CF piece. They could have made it whatever shape they wanted.

It seems to me that with the uplift at the hood, that aero balance kept the wing with lower downforce than possible. Otherwise it would push more at high speeds than the non-R.

Gurney flaps are a great add-on when other restrictions limit the size height, mounting height, chord length, etc. of the wing. I don't think that's relevant to the R. Just my pedestrian view.

Um, isn't it a street car? Why add race specific pieces to a street car? The cost/benefit ratio isn't there. Where do you draw the line? There is always a hierarchy of values when making these decisions.
 

Grant 302

basic and well known psychic
If you could do this, you would be a millionaire.

Er...No comment. ;) :D

The flow structures generated by complex bodies cannot be visualized a priori, they need to be calculated or evaluated in a simulation environment like a wing tunnel.

I generally agree, hence my previous statement that "It's hard to understand what we can't immediately visualize."

Despite that, I actually *think* I have the answer or at least in part. Not only from casual observation, but from identifying it as an alternate solution for a desired flow which I produced with a slightly different element. It (the GT4 canard) is a more elegant way to skin the cat.

I’m sure multimatic was solving a problem with those canards
So am I. And I think it's one of the problems discussed often in this thread...

and it might have had advantages further down the body of the car than the strictly at the nose.

:D Exactly. And I was *hoping* somebody with enough skill and computing power *or* direct access to relevant Ford/Multimatic engineers could chime in! ;)
 

Grant 302

basic and well known psychic
Um, isn't it a street car? Why add race specific pieces to a street car? The cost/benefit ratio isn't there. Where do you draw the line? There is always a hierarchy of values when making these decisions.

That's an interesting argument for a car with ~$13,000 CF wheels. I don't think a gurney flap would have been a problem to include regardless of incremental cost to the wing and infinitesimal cost to the program, if desired.
 
Er...No comment. ;) :D

Despite that, I actually *think* I have the answer or at least in part. Not only from casual observation, but from identifying it as an alternate solution for a desired flow which I produced with a slightly different element. It (the GT4 canard) is a more elegant way to skin the cat.

OK, care to share your thoughts?

:D Exactly. And I was *hoping* somebody with enough skill and computing power *or* direct access to relevant Ford/Multimatic engineers could chime in! ;)

I'm sure Multimatic engineers have NDA's signed to give their life away if they spoke about details. But how about this, if you find me a solid 3D model (.IGES, .STEP, or even a meshed STL) of a GT4 mustang, i'll dedicate a week's worth of compute time to run the analysis with and without canards.

This is my go-to for CAD models- https://hum3d.com/?s=mustang but no GT4 mustang.
 
That's an interesting argument for a car with ~$13,000 CF wheels. I don't think a gurney flap would have been a problem to include regardless of incremental cost to the wing and infinitesimal cost to the program, if desired.

Carbon wheels are not racecar parts. Even LMP1 cars have this rule:

· Metallic homogeneous.
· Produced as integral part mandatory
· Without welding
· Without cavities

Then, even if CF wheels did find themselves on racecars, I’m pretty sure they would give tire engineers quite the headache, as keeping the tire carcass hot depends to a degree on the convection of heat from the brake rotor to the metal (magnesium or aluminum) wheel. I don't have my reference source handy right now, but it's in Jorge Segers's book on racecar data acquisition. When i'm able to dig it up i'll search for the data.
 

Grant 302

basic and well known psychic
OK, care to share your thoughts?
Yeah, when the 'challenge' is tackled.

I'm sure Multimatic engineers have NDA's signed to give their life away if they spoke about details. But how about this, if you find me a solid 3D model (.IGES, .STEP, or even a meshed STL) of a GT4 mustang, i'll dedicate a week's worth of compute time to run the analysis with and without canards.

This is my go-to for CAD models- https://hum3d.com/?s=mustang but no GT4 mustang.

Thanks for the offer, but I doubt anybody is building a GT4 model for the fun of it. I think the GT350R is as good a place to start as any. Though I'm not sure this would show up using a simplified model anyway. Without flow through the engine bay, wheels, underbody, etc., might all be moot. For all we know, it might have been designed and developed completely from wind tunnel data.

And this is part of why I think it's not *my* 'secret' to share, even if I did figure it out. If somebody else sees it too, then there's no longer a secret to keep.

Carbon wheels are not racecar parts. Even LMP1 cars have this rule:

· Metallic homogeneous.
· Produced as integral part mandatory
· Without welding
· Without cavities

Then, even if CF wheels did find themselves on racecars, I’m pretty sure they would give tire engineers quite the headache, as keeping the tire carcass hot depends to a degree on the convection of heat from the brake rotor to the metal (magnesium or aluminum) wheel. I don't have my reference source handy right now, but it's in Jorge Segers's book on racecar data acquisition. When i'm able to dig it up i'll search for the data.

Agreed. I have barrel temps in my logbook. I don't always take it, but it gives a quick read on the relative tire temp. Good data if you aren't bothering with the tread profile temps, but only useful between sessions. Great subject, but a bit off-topic here. I only mentioned the wheels in cost context regarding the R program.
 
More likely that the gurney flap would break rules about radiused corners for safety reasons and would have to made out of rubber and therefore nowhere near as useful.
 
What about splitter end plate? Is it something worth to do?

I have wheel spat to cover the tire and RHR canards
IMG_3710.JPG

I was thinking of adding of adding a piece of metal about 2" tall and 3-4" long to the side of the splitter. I can run it all the way to the wheel spat
IMG_3713.jpg

Or leave a gap ~1"
IMG_3712.jpg
 

Grant 302

basic and well known psychic
My guess is that the closed condition will generate the most downforce, unless the gap is designed and shaped to produce a beneficial downstream flow. Any gap should have a tiny bit less drag overall, but I wouldn't use one unless incorporating an angled element like the notch shown earlier in one of AJ's wing endplates.
 
Thanks for the offer, but I doubt anybody is building a GT4 model for the fun of it. I think the GT350R is as good a place to start as any. Though I'm not sure this would show up using a simplified model anyway. Without flow through the engine bay, wheels, underbody, etc., might all be moot. For all we know, it might have been designed and developed completely from wind tunnel data.

And this is part of why I think it's not *my* 'secret' to share, even if I did figure it out. If somebody else sees it too, then there's no longer a secret to keep.


Oh, it's a guessing game. Thanks for reminding me why i mostly stopped visiting forums.

re: simplified model, this is a normal & routine part of the CFD process. Things like bolt heads, mustang logo's, minor fillets and things of that nature are routinely removed, as for example, the mustang logo on my S197 CAD model has 10,000 face (ten thousand) and the aerodynamic impact is negligible and and can be modeled with a simple flat face, instead of driving the meshing software crazy trying to snap hex's to the logo.
 
I still believe that s550 will benefit a lot with a flat floor, and side skirts ( no splitter).
What you guys think?
Also, I don't find anywhere those things..

Sorry, but you would be mostly wrong. You can't find YES/NO, black and white answers, because everything impacts everything. You need to understand some theory, and then go out and apply that theory by testing it (in the digital domain or other). Here is a good primer, but you need a good background in math. Math is everything when it comes to car dynamics, aero or otherwise.

https://www.amazon.com/Understandin...119967511/ref=mt_hardcover?_encoding=UTF8&me=

Side skirts add a bit of drag and no DF, in some cases lift, due to rake angle, GENERALLY. However, designed with some strakes near the end of the skirt towards the rear wheel, can ameliorate the tire squirt into the diffusor (if you have a diffusor). They also tend to shift the CoP forwards due to their coefficient of moment. Just like placing the battery in the trunk of your car impacts the F/R weight balance and polar moment of inertial in yaw conditions, placement of aero devices will impact the aero moment on the CG.

Flat floor- yes, can add up a good amount of DF- like about as much as a simple flat plywood splitter- ONLY IF YOU HAVE A SPLITTER. The idea of the floor is to keep the air under the car accelerated and low in pressure. Then, if you have a diffusor (and i mean a real diffusor, not the show diffusors on factory cars, but the kind you see on GT3 and GTLM cars) a flat floor can boost their effectiveness (effectiveness, not efficiency, mind you).
 

Grant 302

basic and well known psychic
Oh, it's a guessing game. Thanks for reminding me why i mostly stopped visiting forums.

re: simplified model, this is a normal & routine part of the CFD process. Things like bolt heads, mustang logo's, minor fillets and things of that nature are routinely removed, as for example, the mustang logo on my S197 CAD model has 10,000 face (ten thousand) and the aerodynamic impact is negligible and and can be modeled with a simple flat face, instead of driving the meshing software crazy trying to snap hex's to the logo.

Great. So the volume of air going through the radiator or engine bay don't make a significant difference? I'm not talking about lack or inclusion of a grille emblem. :rolleyes:
 
My guess is that the closed condition will generate the most downforce, unless the gap is designed and shaped to produce a beneficial downstream flow. Any gap should have a tiny bit less drag overall, but I wouldn't use one unless incorporating an angled element like the notch shown earlier in one of AJ's wing endplates.

My advice is to close it, keep it simple. Without modeling the gap or testing it, you'll likely end up with vortex streets/shedding from the gap and the vortex generated over the cap. Lose the canard, without a negative pressure gradient it probably won't do anything, except adversely interact with the vortex from the endcap...

But i could be wrong, and you may want to test it both ways.

The other endcap iteration that might be better is to raise the side of the end cap about 3/4" above the splitter, and then chamfer the splitter under the cap, such it has a knife edge. This promotes a little acceleration under the endplate, and generates a finer vortex on the edge of the splitter.

This is all really little stuff though. If you are going to max front DF and are running a plywood splitter, the best thing you can do is make the leading edge thicker (like double up the plywood) and then give it a leading edge shape. A simple B-spline shape will do the trick. Look at the blue streamlines here from RHR's blog post about the Crusher splitters:

crusherside3.png


and see how they get sucked under the splitter

ref:

https://en.wikipedia.org/wiki/Coandă_effect

you can model this in photoshop or 2D CAD, and run it through Javafoil for fun.
 
Last edited:
Great. So the volume of air going through the radiator or engine bay don't make a significant difference? I'm not talking about lack or inclusion of a grille emblem. :rolleyes:

Yes, of course they do, my point was that "simplification" of a CAD model means removal of logo's and bolt heads, bleeder screws on brake calipers (in my CAD model, 68 faces per screw) and things of that nature, not flow paths.
 

TMO Supporting Vendors

Top