3 link Tech

[Twentyover]

I’ve been looking for 3-link design criteria now for a while both here and on CC. I guess I’m looking for target values of % antisquat and typical link length (try saying that fast 5 times) that yield good handling. I see how to calculate the antisquat values, but I really don’t know what a good target value should be,and why that value is good. Can anyone point to a good resource so I can read up on finding out why one selects one value over another?


Appreciate the need to reinforce shock mounts going to coil-overs from semi-elliptics, that aspect shouldn’t be too much of a challenge.

[exwestracer]

I’ve been looking for 3-link design criteria now for a while both here and on CC. I guess I’m looking for target values of % antisquat and typical link length (try saying that fast 5 times) that yield good handling. I see how to calculate the antisquat values, but I really don’t know what a good target value should be,and why that value is good. Can anyone point to a good resource so I can read up on finding out why one selects one value over another?


Appreciate the need to reinforce shock mounts going to coil-overs from semi-elliptics, that aspect shouldn’t be too much of a challenge.

Greg,
This is one of those subjects that will likely generate a LOT of different answers. I'm sure there are a lot of people who have 3 links "figured out"... but for what application, tire design, engine weight and power, etc.? That's why we usually see a lot of holes in the chassis brackets! When setting up a 3 link on a "multi-purpose" car, I always try to keep the lower links level to the ground, and shoot for 100% A/S (as a STARTING point). You will find that those 2 constraints will pretty much lock in the instant center. I use 100% because we need to load the tire pretty hard getting off the corner, but any more can lead to chatter under heavy braking. Keeping the lower links level minimizes the effect of forward thrust force on A/S percentage change.

I also like to make the upper link as long as possible to minimize pinion angle change. Normally the floor pan will be the limiting factor there.

[dontlifttoshift]

What Ray said plus my own limited experience.

On mine I ran the lower links up hill towards the front. For my car it packaged better. What I found is a little bit of roll steer from the rear helped the car get through quick transitions better on tighter autocross courses. On a roadcourse I don't think that is so desirable.

I have four mounting holes on my upper link in the front and two on the lowers in the rear. I can adjust anti aquat from approximately 110% down to 50%. I think I am currently running about 80% but I would have to check my notes. I have yet to experience any brake hop but I haven't given it a 6000 rpm downshift yet either.....that's good for exposing weak links.

As far as link length, my lowers over twice as long as the uppers. The actual numbers are scribbled somewhere, I will see if I can find them.

Building suspension is a negotiation. You go to the car with what you want, the car tells you what you can have, and then you argue about it from there until it drives.

[ace_xp2]

Hey Ray, do you find tarmac cars respond to 100% well? Not in terms of braking of course, but in power application? I ask because I'd long aimed to do a rear suspension with that much, but I've read that modern TA cars run as little as 10/15% with the number generally decreasing as the power goes up.

[exwestracer]

Hey Ray, do you find tarmac cars respond to 100% well? Not in terms of braking of course, but in power application? I ask because I'd long aimed to do a rear suspension with that much, but I've read that modern TA cars run as little as 10/15% with the number generally decreasing as the power goes up.

Ace, just want to make sure we're comparing apples to apples. There are actually 2 different anti-squat calculations. The one racers with stick axles play with most is the force resisting pinion climb. This is what I was referring to setting at 100%. and yes I've found it is a very good compromise on a modified street car. Typical I/C numbers would be in the area of 50" out and 8-10" up, which is about where you would find a torque arm mount in a lot of cases.

We also get anti-squat from the angle of the links on a birdcage torque arm or IRS system. You will find diagrams out there which look very similar, but have the I/C and 100% line in different locations. This is anti-squat purely from forward thrust; basically the axle or hubs trying to drive out from under the car and lifting the chassis in the process. The percentage numbers for these calculations are usually much lower.

The only example I'm familiar with where we would see pinion numbers as low as 10% would be an OEM triangulated 4 link system. Can't see why you would want that small an A/S force. But who knows...I'm not up on the latest trends in a lot of areas

[exwestracer]

duplicate...

[ace_xp2]

Yeah, I was surprised by that figure as well. I found it in a lengthy thread at nissan road racing, posted by someone named MadBill who didn't seem to have much more to say than that. Seemed like a big enough discrepancy though to make me start wondering.

As a Nissan site the car they were talking about was IRS. However, MadBill was making reference to a Trans Am car he was working. From what I can quickly find online Trans Am is basically to scca GT-1 spec, and that used to be solid axle.
link:
http://www.nissanroadracing.com/showthread.php?t=2296&page=17

[ace_xp2]

Well, I just found the GT-1 rules, and it seems that so long as the car had IRS from the factory you can use one. Oddly, you can backdate a car to live axle, but you can't update a live axle to IRS, even though the rear suspension appears completely custom in either case...
Nonetheless, I suppose you can disregard. Though it is interesting that a drop from AS in the teens to AS in the tens seemed to help the car.

[Norm Peterson]

If you've got birdcages, don't you have to consider the links coming off those in combination with the 3rd link or pullbar and also run the anti-squat construction line from the contact patches through the SVIC? It's still a stick axle and you still have the axle shaft moment causing pinion climb to deal with, which the birdcages cannot resist all by themselves.

An IRS takes the axle shaft torque reaction out directly through the diff housing into the chassis, leaving you with only the forward force acting at hub bearing height.

10% - 15% sounds a bit low for any stick axle suspension unless the lower trailing arms are inclined significantly downhill going toward the chassis.

If you go back to the first post in that nissanroadracing link and use the comparison of tan(thetaR) as shown in the sketch (which was taken directly from RCVD) to h/L you avoid any problems deciding where to draw the anti-squat line (so that you can use its height at the front axle line and go from there back to figure A/S from the CG height).


Norm

[exwestracer]

Norm,
Good point...I was only trying to point out that there are 2 "separate" forces at work. With birdcages, the anti-rotation device (torque arm, etc.) handles ALL the pinion reaction, while the links handle all the thrust reaction. There is obviously a "sum of" these 2 forces and 1 resulting reaction on the chassis.

With IRS, the chassis IS the anti-rotation device, and there is NO anti-squat effect available from pinion rotation, because there is no direct connection (axle housing tube) between the pinion and wheel hub.

[sccacuda]

If you have the room, look into a decoupled 3-link with torque arm. By decoupling, you can run a shorter torque arm. You can get north of 125% A/S without the brake hop issue. The upper link is used under braking only and moves the I/C high and forward. This loads the rear under braking.

[exwestracer]

If you have the room, look into a decoupled 3-link with torque arm. By decoupling, you can run a shorter torque arm. You can get north of 125% A/S without the brake hop issue. The upper link is used under braking only and moves the I/C high and forward. This loads the rear under braking.

http://www.unbalancedengineering.com/Product_Development/Development%20of%20the%20Unbalanced%20Engineering% 20Decoupled%20Torque%20Arm.pdf

[sccacuda]

I've seen the unbalanced setup, and for very tight quarters and those unwilling to put buckets in the backseat, it's probably the way to go. I use a modified pullbar on the top of the diff and the torque arm below. I preload each with bushings of different durometers to get the "balance" right. My rear "floats" between the upper and lower links, with the trailer arms being the constant.

Correction on my last post. The I/C moves low and forward under braking. High and forward on acceleration.

[Bryce]

I built my own 3-link with watts, take a look at my build thread.

[MonzaRacer]

http://www.amazon.com/How-Make-Your-Car-Handle/dp/0912656468/ref=sr_1_1?s=books&ie=UTF8&qid=1339918812&sr=1-1&keywords=how+to+make+your+car+handle
From Marcus the GURU at SC&C : http://www.amazon.com/Make-Your-Muscle-Handle-Performance/dp/1934709077/ref=sr_1_2?s=books&ie=UTF8&qid=1339918854&sr=1-2&keywords=how+to+make+your+car+handle