3 link with a Y upper

[Motorcitydak]

I am going to put a 3 link in the rear of my '68 Charger. Ill be using a custom full floating 9 inch rear and coilovers. I was thinking about using a triangulated upper link for it because if I do 1 straight bar it will have to mount way up on top of the diff where if I do a Y with 2 mounts on the axle I can move em farther down and move the lower link mounts down too. Plus a Y link will control axle movement side to side and that will save me on weight since I will not have to run a track bar with all the supporting brackets that would require. I would also not have to worry about the track bar always being perfectly level or causing the car to jack up or down when I am turning.

The 2 mounts on the axle would mount to it at least 12 inches apart and I would be using a very strong joint on the frame side since it will see a lot of side loads

[The WidowMaker]

theres no reason that the upper would be too high. thats how most are set up. you can also check out billys site and check out an offset 3 link. but either way, id run a traditional 3 link with a watts.

[Eric Howell]

You could also run an off set three link with a track bar. Billie Shope (hope I spelled it right) is gonna come in and tell you the same thing. After reading about it for years I changed my mind.

Couple of sketch up pics.





Billie's calculator.
http://www.racetec.cc/shope/tim.17.htm

[BillyShope]

There are many, many ways to skin a cat. Yes, I have a lot of 3link information at my site, but you'll also find some other designs. I consider the last one I posted (page 43) something to be considered. It would be relatively easy to fab and doesn't require a Panhard. I'd use a ball joint at the rear of the link.
http://www.racetec.cc/shope

[Motorcitydak]

I can't say I like that design. I cannot adjust the AS at all. I have seen setups with a Y link in the lowers, I just figured that a symmetrical setup would be better so running a Y as the upper makes to most sense.

[BillyShope]

Why would you want it symmetrical? That would mean tire loads that are far from symmetrical as you accelerate. With an asymmetric suspension, you have the choice of dynamically cancelling some or all of that driveshaft torque effect.

And, of course, any setup can be made adjustable if you like. But, the "why" question reappears. For dragracing, 100% antisquat provides the minimum of oscillatory loading while, for autocross or road course, it is only necessary to find some reasonable value (less than 100%) which provides enough vertical force for cancellation but not so much that it will trigger wheel hop during braking.
http://www.racetec.cc/shope

[Motorcitydak]

So then a setup like the one posted about would be better for me? Does it make a difference if the upper arm is on the drivers or passenger side? I am trying to get thru all the info on your website but, as you know, there is a lot. I really appreciate all your help here tho. I am not trying to be hard headed, just discussing some different options.

[BillyShope]

...since it will see a lot of side loads

Sounds like this is an autocross car. Since you have a strong "anchor" for a Panhard, I'd use a 3link with Panhard, but with an offset odd link. (The Page 43 setup is for a full competition tube chassis car.) I'd suggest staying south of 80% antisquat.

Take a look at Pages 40, 18, 4, 5, 6, and 37.
http://www.racetec.cc/shope

[Ripper]

Does it make a difference if the upper arm is on the drivers or passenger side?

Yes. It should be in the center, or to the right (passenger) if you want to decrease tourqe steer.
Just remember, when you move it from center to the right, the car might pull slightly to the left (am I thinking correct now? or will it be right?) during braking.

How far to the right you should mount it depends on the engine and how much flex there are in bushings, rear axle, etc.

[BillyShope]

Just remember, when you move it from center to the right, the car might pull slightly to the left (am I thinking correct now? or will it be right?) during braking.

This would be true ONLY as the wheels lock up. The hydraulic fluid provides equal torque to each wheel pair up to that point. Keep in mind that the C-Type Jaguar won Le Mans with an offset upper link.

How far to the right you should mount it depends on the engine and how much flex there are in bushings, rear axle, etc.

The offset can be determined analytically and is the basis for the spreadsheets at my site. Determinants are the tire radius, wheelbase, center of gravity height, percent antisquat, and axle ratio.
http://www.racetec.cc/shope

[Ripper]

This would be true ONLY as the wheels lock up. The hydraulic fluid provides equal torque to each wheel pair up to that point. Keep in mind that the C-Type Jaguar won Le Mans with an offset upper link.

It would be the last problem after you locked up your wheels. After all - you don't have very much control over the vehicle with four wheel lockup. ;)

I don't have any practical experience of it since I've never had any track time with a three link suspension, but I imagine that it rather has with unegual deflection in the axle. This would probably give different toe changes on right and left wheel, which will pull the car to the left.

The offset can be determined analytically and is the basis for the spreadsheets at my site. Determinants are the tire radius, wheelbase, center of gravity height, percent antisquat, and axle ratio.
http://www.racetec.cc/shope

Where on that site can I find that info? I'm currently building a dedion-suspension for my Firebird and it would be interesting to see how much it should be offset.
I haven't found any info on how to find the offset in Milliken&Milliken.

[BillyShope]

It would be the last problem after you locked up your wheels. After all - you don't have very much control over the vehicle with four wheel lockup. ;)

I was thinking here more of rear wheel lockup, which is just as undesirable. I'm certain the driver of the C-Type never locked up his rear wheels during the 24 hours of Le Mans.

I don't have any practical experience of it since I've never had any track time with a three link suspension, but I imagine that it rather has with unegual deflection in the axle. This would probably give different toe changes on right and left wheel, which will pull the car to the left.

Leave that defenseless little nit alone and stop picking at it! Yes, the asymmetry would leave the left side a bit weaker, but, again, the C-Type indicates deflections can be successfully controlled.

Where on that site can I find that info? I'm currently building a dedion-suspension for my Firebird and it would be interesting to see how much it should be offset.

You don't need asymmetry on a De Dion; driveshaft torque has no effect on rear wheel loading during forward acceleration.

I haven't found any info on how to find the offset in Milliken&Milliken.

I just checked and "asymmetry" is not mentioned in the index of Race Car Vehicle Dynamics. I used RCVD as a text when I was teaching and I don't recall a discussion of asymmetry, though I could be mistaken. [Yes, I am mistaken. It's found in the last paragraph of page 653. The last sentence says, "Under braking there would be a slight tendency to roll to the right, however." It should be understood that the preceding sentence was addressing the fact that, with full cancellation of the driveshaft torque, there would be NO body roll. In other words, the subject of the sentence is body roll and NOT a "pulling" of the car during braking.] I did supply a problem involving asymmetry which is a part of the student workbook which accompanies RCVD and the results of my derivation are there. It's also briefly mentioned in the Millikens' Chassis Design.
http://www.racetec.cc/shope

[David Pozzi]

I've noticed with a centered 3 link rear, on right hand turn exit, there is a greater loss of rear traction compared to left hand turns. I've felt it on Road Course slower turns in second gear, and autocrosses which were also in second gear. But if I were to make the 3 link Asymmetric would the car lean more in left hand turns?

Here's an Asymmetric 3 link in a Trans Am car, it's adjustable.

[JEFFTATE]

Oh my Goodness !
That's beautiful David....

[BillyShope]

I've noticed with a centered 3 link rear, on right hand turn exit, there is a greater loss of rear traction compared to left hand turns.

Yes, I'd expect this. In the first case, the driveshaft torque effect is added to the unloading due to the turn; in the second case, it's being subtracted.

But if I were to make the 3 link Asymmetric would the car lean more in left hand turns?

It might be better to ask: Would asymmetry provide more reliable and repeatable curve exit performance? And, the answer is "yes." With asymmetry that provides 100% cancellation (that's 100% cancellation, NOT 100% antisquat), the car would exit a left turn curve the same as a right. This is, of course, why Jaguar used it on the C-Type.
http://www.racetec.cc/shope

[-DA]

So, under braking you no longer need to dynamically counter driveshaft torque and, like you said Billy, each rear brake will get equal braking torques. But for some reason I'm still thinking the asymmetry would induce a brake steer issue. ?? I must be over thinking this...

If it were an issue though, a fourth sliding-link could be beneficial. Place it symmetrically on driver side with the capability to lengthen but not compress. During acceleration it does nothing as the upper links are in tension; however, once the brakes are applied the link goes into compression and prevents any asymmetric braking issues. Quite possibly a great solution to a problem that doesn't exist though.

[BillyShope]

I must be over thinking this...

Consider a free body diagram of the entire car and it should become obvious.

Quite possibly a great solution to a problem that doesn't exist though.

The Ramchargers organization was originally composed of engineers and technicians. Each of the two groups was necessary for success. The engineers made sure F still equaled Ma and the technicians were around to remind us that it's "lefty-loosey" and "righty-tighty." But, at times, problems arose. This was the case when some of we engineers proposed the asymmetric 3link. There were those who thought the car would certainly make an abrupt turn with brake application. So, we ended up with 4 links, the upper left being exactly as you describe it.
http://www.racetec.cc/shope

[Motorcitydak]

After looking at your site, it shows that the right lower link sees about 2x the loads of the left side. Should I screw around with the mounting locations until I can get a more balanced load on the 2 lower links? Like moving the axle mounts toward the center?

[BillyShope]

...until I can get a more balanced load on the 2 lower links?

If this is a concern, I would suggest that you minimize the offset to minimize the load difference. If you want equal loads, you must rely on different angles on the lower links (see Page 40).
http://www.racetec.cc/shope

[BillyShope]

Hold the phone!! That last question caused me to "play around" with the inputs to the spreadsheets on Pages 18 and 40. Just for kicks, I decided to determine the vertical force components in the links and check to see that I got 100% cancellation. I didn't! Panic!! Well, I was able to quickly find the problem and it's been fixed. If anybody fabbed to the previous output, you got the antisquat you wanted, but you didn't get complete cancellation. You're still better off than you would have been with a symmetrical setup, and...most likely...you're happy with the result. But, you don't have complete cancellation.

Here's what happened: The original solution relied on a derivation based on an equation set with enough equations that the algebraic manipulations were a real headache! When I decided, a couple of months ago, that I wanted to provide different angles with the symmetrical links (so the odd link wouldn't necessarily have to be offset), I looked for a way to avoid that equation set. I figured out a way to iterate in on the solution and was then able to post Page 40. At the same time, I changed Page 18 to an iterative solution. Well, somehow, one term in the torque balance equation received a plus sign when it should have had a minus sign.

What can I say? This is very embarrassing. I could have made the change and just kept quiet, but I'll undoubtedly make another mistake in the future, so you might as well be prepared.
http://www.racetec.cc/shope