Presented by Nine Lives Racing
1 2 3 4
sleepyhead
sleepyhead HalfDork
1/9/18 8:03 a.m.
GameboyRMH said:
loosecannon said:

I hired an aerodynamicist to analyze and model my car in CFD. Going from no aero to good aero made to the limit of EMod rules would get me a 2% decrease in lap times on a course with an average speed of 90 km/h. This is with an 1800 lb car and the lighter the car, the more effective aero is. So, to answer your question, aero will probably only drop tenths of seconds for you, not 2 seconds.

Well you did already have a splitter, spoiler and diffuser before you got CFD done. This car has none of those.

Hey loosecannon... do you have a downforce value for that spoiler at 50mph, and the downforce without the spoiler (same speed)?

loosecannon
loosecannon Dork
1/9/18 8:42 a.m.

In reply to sleepyhead :

90 lbs of downforce but that is spoiler and diffuser together 

freetors
freetors New Reader
1/9/18 9:09 a.m.

While the quest for more downforce and less drag is always a worthwhile endeavor I don't think it's the root problem here. I'm going to go out on a limb here and suggest an entirely different theory. The sn95 mustang has a four link rear suspension with the upper links triangulated, converging to the rear. This should be enough to control to allow the rear suspension to be kinematically free to move up and down and roll while constraining its fore/aft and side to side movement.

 

Now you've stated that you added a panhard bar to control side to side movement. This should not be needed and is technically over constraining the movement of the rear suspension. In addition to that you have added presumably stiffer bushings into the links that are less tolerant of moving axis. I would be willing to bet that your suspension is binding significantly! This is the enemy of good handling and grip. You probably don't notice this most of the time but I think it's highly likely this is what's happening here.

 

I would first ditch the panhard. Then convert as many of those bushings to spherical bearings or rod ends as you can.

PS, I know that in the mustang community the triangulated four link system gets a bad rap for binding and so many owners convert to something else. The ONLY reason it binds is because of rubber bushings. This type is setup is kinematically free with bearings that can rotate freely!

 

PPS, I think it would be very enlightening to you if you pulled out the rear springs and cycled the axle through it's travel. I bet it doesn't move freely for very far ;)

jj
jj HalfDork
1/9/18 12:26 p.m.

In reply to Tom1200 :

Tom, that is very interesting point about watching the damping in the slalom part of the video.  So it looks like I may be over damped on one end.  I will have to try taking a bit out and see what happens there. 

I don't have a way to set corner weights though since I don't have coilovers.  I can't imagine that its very good though.

Apexcarver, you are right about those seats!  They are very poor, expecially the leather ones.  I put in a fixed back race seat bolted straight to the floor, and it is light years better.  Although not very comfortable.

Freetors.  I gave a lot of thought to what you are saying regarding the rear binding with the PHB.  I do not think it is a problem in this case for 2 reasons.  One is that the LCA have sperical bushings.  Two, the UCA are stock and have very sloppy and worn out bushings.  I think these two things would allow enough movement to not cause binding with the PHB.  Here is a quote from Maximum Motorsports about that: "As the suspension allows the chassis to move when bumps are encountered and as the body rolls during cornering, the rigid Panhard Bar keeps the chassis centered over the rear axle, something the stock four-link design fails to do. The opposing angle of the Mustang upper control arms causes the chassis to move in an odd path when the body rolls, which tries to make the chassis shift sideways relative to the axle. Rubber upper control arm bushings deflect enough to allow the chassis to follow the path determined by the Panhard Bar, and do so without significant suspension binding."

Again thanks for all the responses. 

Apexcarver
Apexcarver PowerDork
1/9/18 2:17 p.m.

I'll back JJ up on the 4 link issue, yeah, its commonly referred to as 'quadrabind', but you are likely underestimating the size of the rubber bushings involved to allow the interference.  Many many cars are running PHB's with stock uppers in competition environments with no trouble, the stock rubber bushings allow enough movement that its not a problem. 

 

replace those sloppy upper bushings with urethane or stiffer (or egads, speherical bearings) and you WILL break parts or rip the floor.  I have witnessed aftermarket UCA with poly fail (fracture) at an autocross.   

 

Driven5
Driven5 SuperDork
1/9/18 2:32 p.m.

The original (worn out) rear uppers does bring up another issue.  Excess slop in the upper links could still be allowing for undesirable axle movement, even if it's not lateral movement thanks to the Panhard bar.  I would consider replacing the rear upper link/bushings, possibly on just the passenger side, with new OE quality rubber replacements.

rslifkin
rslifkin SuperDork
1/9/18 2:45 p.m.

In reply to Apexcarver :

Spherical bearings shouldn't bind in a 4 link.  Poly will bind horribly, however.  

Apexcarver
Apexcarver PowerDork
1/9/18 2:57 p.m.
rslifkin said:

In reply to Apexcarver :

Spherical bearings shouldn't bind in a 4 link.  Poly will bind horribly, however.  

wrong.   theres interference in a direction that a spherical constrains

 

 

the uppers are double duty on this also doing lateral constraint.  they dont do it well, hence aftermarket panhards.

rslifkin
rslifkin SuperDork
1/9/18 3:02 p.m.

In reply to Apexcarver :

What direction is that?  If the links are needing to change length, something is wrong with the basic design.  

Driven5
Driven5 SuperDork
1/9/18 3:04 p.m.

I think one of you may be talking about sphericals with Panhard, as is currently in this particular application, and the other reverting to sphericals without Panhard.

Apexcarver
Apexcarver PowerDork
1/9/18 3:06 p.m.
rslifkin said:

In reply to Apexcarver :

What direction is that?  If the links are needing to change length, something is wrong with the basic design.  

there is, its from a 1978 ford fairmont...  

 

 

Yes, they do have to change length somewhat and the bushings do allow it.

 

consider the image that the (canted) uppers are all the lateral constraint there is stock, then consider what the arms have to do in roll. 

 

sleepyhead
sleepyhead HalfDork
1/9/18 3:07 p.m.
loosecannon said:

In reply to sleepyhead :

90 lbs of downforce but that is spoiler and diffuser together 

so, I finally opened up my Race Car Aerodynamics, 2nd Edition.  And all the data it has is for up to a 100mm high spoiler that's 20deg from vertical, 1510mm long and 1172mm back from the rear axle (indeterminate distance from the rear sedan window)... showing a Cl increment of 0.4.  (Data is from a 1977 SAE research paper).

Running the numbers from an assumed width of 60in, and length of 153in, at 50mph and "standard day conditions"... Cl = 90 / (.5 * .002377 * (73.3^2) * 63.75 ) = 0.22

there's not a lot of other info on these types of spoilers in there, other than a recommendation that "there's a lot of info in W.H. Hucho's Aerodynamics of Road Vehicles"...  I might have to pull out my two volumes of Hoerner and see if there's anything in there to draw out.

I find that a bit interesting, since some digging indicated educated guesses that your spoiler was around 10" high, and I'd guess 15deg from vertical, loosecannon?  It's possible the difference is because the SAE paper test was conducted at higher speeds... although, with it being published in 1977, 55mph could be valid?  I'm not an SAE member, so all I'm finding is it being referenced in various books/papers out there.

*however*... assuming that JJ can only make 50% of loosecannon's "downforce increment"... the 40% greater area of the Mustang means that 65lbs of downforce is available ( not including the "downforce of the elements added weight cheeky ).  This is only valid, though, if you consider these spoilers working similar to a "split flap"... with their affect being applied to the whole area of the car.  Seems valid to me, but figured I'd throw that caveat out there.  Similarly, if JJ can make 70% of loosecannon's "downforce increment" then they'll be at equivalent downforce levels... although that would almost certainly require diffuser construction to be realistic. (?)

Anyone ever build a multi-element slotted spoiler (I'm going to guess those aren't allowed in SCCA)?

Apexcarver
Apexcarver PowerDork
1/9/18 3:13 p.m.
 

 

 

 

Sleepyhead, I refer you to my earlier post showing a national championship winning A-Mod car.

 

Though we could be treading the differencing between wing and spoiler in definitions.  I dont think any production car classes allow that (other than XP or SM maybe?)

sleepyhead
sleepyhead HalfDork
1/9/18 3:30 p.m.

yeah, that image is the genesis of that comment... although I've been coming at it from a "grm challenge" perspective... these speeds are slow enough that plywood would probably suffice for building something like that (and challenge is probably the only venue "open enough" to allow something like that in).  And the speed/Rn of the element should be low enough that "flat plate is valid"-ish

re:wing vs. spoiler... sometimes it all comes down to terminology used by the sanctioning body / people... I don't think the wind cares that much wink... and I'm open to "creative interpretations" for what a spoiler is.

then again, JJ alluded to the fact that this group wasn't "SCCA Class oriented"... so the aero limitations might be a bit more free?

anyone got a recommended "GRM data collection" setup?  In case I find some motivation to try building my "Venetian Spoiler"?

 

jj
jj HalfDork
1/9/18 3:30 p.m.

Maybe Im going to build a small wind tunnel and test on THIS.  Couple of THESE too.  

Driven5
Driven5 SuperDork
1/9/18 3:33 p.m.
Apexcarver said:

Yes, they do have to change length somewhat and the bushings do allow it.

 

consider the image that the (canted) uppers are all the lateral constraint there is stock, then consider what the arms have to do in roll. 

Ok, so I was wrong in optimistically thinking you were still referencing it 'with Panhard'.  

No, a triangulated 4-link should have NO BIND in roll when arms utilizing spherical joints at BOTH ends are used.  Rubber bushings will have some bind, but not much compared to non-compliant bushings. If you think about it though, OE style fused rubber bushings even 'bind' in pure axial rotation as well. 

The only reason the arms would need to 'change length' in a pure triangulated 4-link, is that rubber bushings are used in the stock configuration rather than sphericals.  What the rubber bushings, and spehericals for that matter, are really doing though is allowing for multi-axis misalignment.  The change of length noted by Maximum Motorsports is only necessary when a Panhard is added to the mix, otherwise over-constraining the system.  Plenty of vehicles have used triangulated 4-links with sphericals on both ends of all 4 links over the years, demonstrating zero bind in roll.

One common mistake that might cause people to erroneously blame the sphericals, is people who assume that a spherical on one end of a link means a non-compliant bushing can be used on the other end.  However, this still doesn't allow for the true multi-axis misalignment that is required.

Or is there something else specifically about the Ford design that I'm unaware of which makes it substantially different than pretty much every other full-spherical triangulated 4-link live axle application, from racecars to rock crawlers?

sleepyhead
sleepyhead HalfDork
1/9/18 3:37 p.m.
jj said:

Maybe Im going to build a small water tunnel and test on THIS.  Couple of THESE too.  

FTFY

at that scale, you'll get better results correlation with water... although I'll have to crunch some numbers to be sure... and give you water flow rates

jj
jj HalfDork
1/9/18 3:45 p.m.

So a scale 50mph wouldn't correspond directly because the viscosity of air can't be scaled.  Is that right?  Is that what the Reynolds number is?

Apexcarver
Apexcarver PowerDork
1/9/18 3:56 p.m.

Driven5, 

I understand if you dont want to trust me.   Here it is from another source.  The mustangs design is different in that most 4 link designs retain some other form of lateral constraint. Underneath the skin, every mustang from 79-04 is the same as a 1978 ford fairmont (except the 99+cobra IRS, which has its own issues).  

 

http://www.maximummotorsports.com/tech_rear_susp_panhard.aspx

Upper Control Arm Bushings
Whether or not your Mustang is equipped with an MM Panhard Bar, it is very important to have rubber upper control arm bushings. This is one application where the compliance of a rubber bushing is a benefit.

A four-link is over-constrained, a situation that requires the upper arms to physically change length as the suspension moves. Obviously, the metal control arm cannot change length. But its effective length, the distance between the control arm's two pivot points, can change because of the inherent compliance of a rubber bushing. If the ability of the upper control arms to change their effective length is hindered by a noncompliant bushing material, the suspension will bind up, and not move freely. The only way for the suspension to move is for the metal control arm mounting brackets to bend. The resulting restriction in the ability of the rear suspension to freely articulate will cause poor handling; the car will have a tendency to oversteer, and it may do so in a sudden and unpredictable manner.

Retaining the rubber upper control arm bushings is a necessary compromise for acceptable handling.

Apexcarver
Apexcarver PowerDork
1/9/18 4:02 p.m.

it isnt just a making up for the panhard thing...  go model the rear suspension off one if you want to, but in roll it does need it with or without a panhard or watts. 

 

 

Some people do make do with a spherical upper, by only running a single upper control arm after a panhard or watts is installed (commonly referred to as a "poor man's 3 link")   Theres tons of information out there and has been done to death since at least the late 80's.

 

 

sleepyhead
sleepyhead HalfDork
1/9/18 4:55 p.m.

yes... although it's more that density and viscosity can't be scaled...

Rn = rho * V * l / mu

the problem is 50 * 25 = trouble with mach... and that's discarding the whole "you don't have a fan powerful enough" aspect of things.

Rn for the mustang at that speed is around 7million.  If you stayed above 1million, things should be "close-ish"... still, that's gonna need 350mph, and you've gotta generate that without adding heat to the flow (otherwise you lose density... adding to your problems)... and hopefully not ticking off your neighbors wink

then again, it *looks* like you'll need 32ft/s of water to get a 1/25 model to make 1million Rn in water... which also seems... challenging... although, it's possible I have the math wrong... I'm used to working in english units, don't have the coefficients for water "easily at hand" like I do air.

Driven5
Driven5 SuperDork
1/9/18 4:56 p.m.

In reply to Apexcarver :

MM appears to have misapplied a description the cause of bind in a more traditional/conventional parallel 4-link, which requires a 5th (over-constraining) lateral link, to a self-locating and fully-constrained triangulated 4-link application.  A triangulated 4-link is NOT inherently over-constrained, and does not inherently require (at least one of) the links to be able to change length. 

Even IF the Mustang specific 4-link does actually have some other geometric issue creating bind when used with sphericals at all 8 locations, that particular widely parroted false explanation is not accurately describing the cause...Nor does it apply to triangulated 4-links in general.

As such, I'll generally trust the people writing suspension/racing engineering texts and those who have built bind-free spherical jointed triangulated 4-links, over the marketing team at Maximum Motorsports and the warnings of their followers who have never actually tried it before. 

Apexcarver
Apexcarver PowerDork
1/9/18 6:10 p.m.

Well, I will agree to disagree.   I have personally seen ripped floors and broken parts on cars that replaced the uppers at a rate far exceeding trouble with ones who stay with the stock bushings in my 15 years of toying with SN95 mustangs. 

APEowner
APEowner HalfDork
1/9/18 6:56 p.m.
Driven5 said:

In reply to Apexcarver :

MM appears to have misapplied a description the cause of bind in a more traditional/conventional parallel 4-link, which requires a 5th (over-constraining) lateral link, to a self-locating and fully-constrained triangulated 4-link application.  A triangulated 4-link is NOT inherently over-constrained, and does not inherently require (at least one of) the links to be able to change length. 

Even IF the Mustang specific 4-link does actually have some other geometric issue creating bind when used with sphericals at all 8 locations, that particular widely parroted false explanation is not accurately describing the cause...Nor does it apply to triangulated 4-links in general.

As such, I'll generally trust the people writing suspension/racing engineering texts and those who have built bind-free spherical jointed triangulated 4-links, over the marketing team at Maximum Motorsports and the warnings of their followers who have never actually tried it before. 

It is possible to design a fully constrained triangulated four link that's bind free with spherical joints at all 8 locations as witnessed by the many vehicles that run them.  For some reason Ford didn't do that with the Fairmont/Fox Body Mustangs and getting rid of the rubber bushings will cause them to bind.  I don't know why they made that choice.  I believe that the SN95 Mustangs also have the same characteristic but I don't have first hand knowledge of that as I've only worked with the fox body cars.  On those the inherent bind in the system would cause the rear suspension to stiffen as roll increased kind of like a non-linear variable roll bar.  Stiffening the springs and/or roll bars to keep body roll in the range that didn't bind (or bound less) was one way to deal with the situation but the axle housing would move significantly side to side with sticky tires.  I never tried a panhard bar or watts link to prevent that but I think there would be some danger of it contributing to the binding situation.  

freetors
freetors New Reader
1/9/18 10:09 p.m.
Driven5 said:

In reply to Apexcarver :

MM appears to have misapplied a description the cause of bind in a more traditional/conventional parallel 4-link, which requires a 5th (over-constraining) lateral link, to a self-locating and fully-constrained triangulated 4-link application.  A triangulated 4-link is NOT inherently over-constrained, and does not inherently require (at least one of) the links to be able to change length. 

Even IF the Mustang specific 4-link does actually have some other geometric issue creating bind when used with sphericals at all 8 locations, that particular widely parroted false explanation is not accurately describing the cause...Nor does it apply to triangulated 4-links in general.

As such, I'll generally trust the people writing suspension/racing engineering texts and those who have built bind-free spherical jointed triangulated 4-links, over the marketing team at Maximum Motorsports and the warnings of their followers who have never actually tried it before. 

I agree with this. You have a choice, you can choose to believe the guy that wants to sell you stuff, or you could build a small model and test it out and see for yourself. It doesn't even need to be to scale or dimensionally accurate.

But if you already have your rear suspension bushings as sphericals so it would be pretty easy to change them all over to sphericals.

In doing research about this topic I did actually find a few people stating that they had ALL the rear bushings replaced with sphericals, one of them even did road racing. They say it works even better than the other options and is more supple and bind free. But then there are an awful lot of people that have never tried it but say it can't or won't work. Kinda funny how that works...

I'm not going to say the Mustang's four link suspension is the pinnacle of rear suspension but I don't think it deserves the criticism it gets.

1 2 3 4
Our Preferred Partners
aDj89EGzJcShCtn8dClecCle6cCcnJyImaPihlNqWLjyGhs7MuXzbBOMOCTr0N6e