Compression only Upper Link

[Motorcitydak]

Im building a rear 3 link for my car with the upper link set to the right side. That will perfectly balance out the acceleration aspect so the car goes straight but what about when I hit the brakes? It will be a symmetrical load going to an unsymmetrical setup. Ive been thinking about building a 4th link that would telescope out as far as needed but will bottom out to act as a 4th link in compression only. It would be just the same as the upper on the other side in an attempt to balance out the load in braking only. Should I do it? I just do not want to car to have brake-steer

[Bryce]

I dont think you will have brake steer until you start locking up the rear brakes. But that is just my initial thought.

The other idea would work also take a look at the decoupled torque arm with an adjustable length 3rd link. same concept as what you are thinking.

[David Pozzi]

You have an offset upper link?
I changed the name of this thread to better reflect it's topic.
David

[Motorcitydak]

Falcon, I was thinking the brake steer would not come up very much since no one else ever talks about it. Also knowing how I think 30% of the braking is done by the rear where 100% of the acceleration is. I figured that is why I would pose this question. I cannot build a torque arm, Im already commited to the 3 link so im just going to work with it.

Dave, yes my upper link is offset 5 inches to the right to eliminate torque steer. I used a billy strope worksheet to design the system. As he wrote to me, 'why use a symmetrical syspension system to control a non-symmetrical force'

[LowBuckX]

I have an offset upper also but the centered upper link has been proven to be a winner. I did mine offset per Billy but also for packaging.

[silver69camaro]

The newer 3-link Mustangs seem to do OK with an offset UCA. I think it would be interesting to try out.

[Bryce]

Matt,

Newer Mustangs? 2005 - 2009 had centered UCA, at least that is what I thought. Did the 2010 switch to an offset UCA? Anyone have pics?

Going to google right now!

[silver69camaro]

You're right, 05 did have a centered link...I must be thinking of something else!

[Norm Peterson]

Factory Five's 3rd link upgrade kit for Cobra replicars with the stick axle, maybe?


Norm

[PhillipM]

Becuase you followed Billy's spreadsheet hopefully he will chime in and help out with some answers. I will be interested to know as I am just about ready to start in on my 3 link in which I plan to do an offset upper per Billy's spreadsheets.

[ace_xp2]

Hey Motorcity. Just a quick question for you.
When I plug through Billy Shope's equations, getting torque cancellations with the other (i.e. the non offset) arms parallel to the ground usually requires an offset in the 20" range, with the offset increasing as the anti squat and the cg height decrease.
Have I been doing the calcs wrong? Or are you going with less than full cancellation? Or maybe accepting some roll steer?

[PhillipM]

Which equation are you using? How do you know if your getting full cancelation? I'm working on mine which I think I've done it properly, my offset is at 14".

[Mean 69]

Keep in mind where the forces are coming from in the acceleration versus braking. Under acceleration, the forces are coming from the driveshaft, through the differential (turning the torque 90 degrees....), through the axles and to the tires. Under braking, the forces come from the tires (uniformally, unless loaded in a hard turn where there will be differences) to the axles, and immediately fed into the HOUSING via the rotors and brake calipers. I don't see loading as a function of offset upper link as an issue under braking for this reason, "maybe" some funny business could happen in extreme offset situations, but more from material/strength considerations? As mentioned, braking forces are significantly lower than acceleration forces as well.

This logic, as well as packaging issues on my current project (first gen convertible, with a power top pump located right where a centered upper link would go) have led me down the path of an offset upper link as well. I have a lot of confidence that it will work, and should be murdering cones in the reasonably near future to find out in the practical, rather than theoretical, world.

For any link system, pay special attention to the overall kinematics, especially if packaging requires short links. It is possible that assymetries from an offset upper could create some unwanted challenges that overwhelm the pro's of the drive torque reduction.

Perhaps one of the smart folks on the site can check to torque application logic I described: Norm/Dave, what are your thoughts on this? Regardless, mine's done so I'll find out for better or worse!!!

Mark

[ace_xp2]

Heh, Yeah, sorry I meant just using his web pages. No math skills of note I'm afraid. I've been using his earlier page, as it's been easier to deal with:
http://www.racetec.cc/shope/tim.17.htm

With a 26" tire radius, 3.73 gears, 18" cg height,and a 108" wheel base. I need about 20" offset to get my front mount for the left and right links the same height as the axle mounts.
According to the website, the values you get will completely cancel drive shaft torque, I'm guessing moving the odd link mounts left or right from the prescribed amount while keeping the other positions equal would decrease or increase cancellation. Maybe I'm off base there, but that would make sense to me.

[Motorcitydak]

Here are the figures I used for the design:

Distance forward to IC 70
Tire radius 12.5
Axle ratio 3.50
Car weight 3000
Weight of axle 225
CG height 16
Wheel base 117
Anti-squat 60%

First link (Upper)
Distance forward from center 0
Vertical height from ground 18
Offset from center 5
Length 14

Left lower link (Second)
Distance forward from center 0
Vertical from ground 9
Offset -22
Length 22

Right Lower link (Third)
Distance forward from center 0
Vertical from ground 8
Offset 22
Length 22

And the results for the frame side mounts

Vertical from ground
First 15.59
Second 6.19
Third 8.45

IC Height 5.74

Link loads at 1G acc

First 2788 tension
Second 2469 Compression
Third 3073 Compression

I had to shorten my upper to only 8 inches long (I know...) which moved the mount height to 16.62 by the way. I drew it out and seemed to not be too bad as far as axle movement or pinion angle changes

[Norm Peterson]

Mark - All that matters to the links as far as wheel torque reactions go is that the axle is under some sort of torque about the Y-axis. Neither the direction (from acceleration or braking) nor where it is specifically applied (pinion loading/pinion bearings/pumpkin or axle ends, respectively) matters as far as link loads are concerned. Detail stress evaluation of the axle itself is where the location of applied torsion loads matter; for example, in the acceleration case the axle tube to pumpkin joints would be more critical than axle ends at the backing plates. Vice versa for braking.


dak - Using a compression only upper link as you describe might work well enough for braking in a straight line. But I'm a lot less comfortable about what's going to happen when the car is rolling in a corner. I'm assuming that the upper(s) won't always be perfectly horizontal, either by design intent or due to vehicle loading. Without having drawn anything out on paper, I suspect that when cornering in one direction the compression gap will open up and that "link" won't do anything at all and in the other direction the gap will close up solid and bind the suspension. I suppose this could be mitigated via a spring or a bushing with the right compliance and a bit of on-car test-n-tuning similar to what you go through with UE's decoupled torque arm (which is another arrangement that has a different element active under braking than acceleration).

If you're really concerned about brake steering, it's not like you only have the choice between 0% and 100% acceleration effect cancellation. No more so than having to choose either 100% antisquat or 30%. Just that as far as I know, Billy's sheet doesn't offer acceleration effect cancellation as an independent variable that the user can select. I'd expect a brake steer concern to be greater if for any reason you are using an unusually heavy rear brake bias (drag strip car with set-back engine and skinnies up front? Rear-mid Toronado conversion?).


Norm

[Mean 69]

Norm, I don't think I explained my point correctly. Agreed on all counts with regard to the loading on the individual links, what I was pondering was the actual torque reaction in the two very different cases.

The offset upper link is purported to reduce the right/left tire loading induced by the torque from the driveline: i.e. driveshaft. This torque originates along the length of the vehicle and then turns 90 degrees from the differential. Two examples of this that I can think of are drag cars that launch really hard on leaf springs, like the good old Mopar drag cars, they often do one wheel "wheelies" due to the massive engine and driveline rotational torque. The crankshaft, clutch, flywheel, driveshaft all add up to a significant amout of rotational mass, the resultant reaction has to manifest somewhere. Another example is a shaft driven motorcycle, relative to a chain driven bike. If you have riden both types, at least the ones I had, the shaft driven bike exhibits significant vertical lift under acceleration which is not nearly as detectable with a chain driven bike.

In the braking example, there is no longitudinual torque of significance, I don't see how the offset link could induce a left to right inbalance in this case.

All of the above is not to be confused with any kinematically driven (or worse, execution errors in construction!!!) axle steer issues that may arise from the layout of the links, etc, in a left versus right roll situation.

Thanks,
Mark

[ace_xp2]

Here are the figures I used for the design:

Are you sure that won't lead to roll steer with axle movement?
From what I see there you've got the left link pointing down and the right pointing up. So under two wheel bump the left link will pull the axle forward while the right will push it back. under rotation I think both will move the same direction, but in differing amounts.

[Norm Peterson]

Norm, I don't think I explained my point correctly. Agreed on all counts with regard to the loading on the individual links, what I was pondering was the actual torque reaction in the two very different cases.

The offset upper link is purported to reduce the right/left tire loading induced by the torque from the driveline: i.e. driveshaft. This torque originates along the length of the vehicle and then turns 90 degrees from the differential. Two examples of this that I can think of are drag cars that launch really hard on leaf springs, like the good old Mopar drag cars, they often do one wheel "wheelies" due to the massive engine and driveline rotational torque. The crankshaft, clutch, flywheel, driveshaft all add up to a significant amout of rotational mass, the resultant reaction has to manifest somewhere. Another example is a shaft driven motorcycle, relative to a chain driven bike. If you have riden both types, at least the ones I had, the shaft driven bike exhibits significant vertical lift under acceleration which is not nearly as detectable with a chain driven bike.

In the braking example, there is no longitudinual torque of significance, I don't see how the offset link could induce a left to right inbalance in this case.

All of the above is not to be confused with any kinematically driven (or worse, execution errors in construction!!!) axle steer issues that may arise from the layout of the links, etc, in a left versus right roll situation.

Thanks,
Mark
I don't think equalizing the wheel loading under acceleration is as simple as the lateral offset of the odd link. Note that for 100% traction equalization with a conventional single upper 3-link has the two lowers at slightly different side view inclinations.

Yes, for the acceleration case you have an Mx to resolve that is not present under braking.

You also have the upper (odd) link applying a force under braking that is offset from both the vehicle sprung mass CG and from the resultant brake force (from all four tires). If the sprung mass is to be held from yawing, lateral forces have to exist at the contact patches (all four). The tires will then operate at some slight slip angle rather than purely longitudinal slip. These lateral forces and slip angles aren't going to be much (I'd guess on the order of 100 lb/tire at 1g decel for a car with dak's numbers above), but they do exist.

That's assuming equality of forces in the two lowers, which isn't the case during forward acceleration when tire loading is fully equalized; If they aren't identical under braking, there would be a yaw moment generated from them as well (and probably opposite to what the upper link force is trying to do).


Norm

[Norm Peterson]

Are you sure that won't lead to roll steer with axle movement?
From what I see there you've got the left link pointing down and the right pointing up. So under two wheel bump the left link will pull the axle forward while the right will push it back. under rotation I think both will move the same direction, but in differing amounts.
You always have roll steer with any of the "simple" stick axle locating arrangements, so it's not whether you have it or not here.

It's that it will be slightly asymmetrical as far as left turns vs right turns is concerned. There are plenty of cases where this is done intentionally (roundy-round racing). There might even be a case for using it to offset some other chassis asymmetry for a car that has to turn both directions "in anger" (I don't know that there is such a condition, but I can't dismiss the possibility out-of-hand either).


Norm

[ace_xp2]

It looks like it would be there even under a two wheel bump though. Not the end of the world, but depending on its severity I'd think it'd be a little disconcerting. Though, I guess you'd get used to it.

Also, I thought that having lower arms parallel to the ground was generally the goal, as it would pull one side in as much as the other so long as each side was compressing/extending the same amount. I guess the same amounts bit is where it all goes wrong?
In other words, it's rare that each side is moving equal amounts in opposite directions, so roll steer is still around.

[exwestracer]

Assuming rigid axle housing and rigid chassis, there shouldn't be any issues with straight line braking.

Real world hard braking may result in some off-side wheel hop due to axle tube wrap, especially with light, small diameter tubes.

Braking while the chassis is in roll (trail braking) is another issue. With the top link centered, brake torque reaction is applied more or less equally to both rear springs. Depending on the amount of top link offset, the possibility exists to have more brake anti-squat (or squat, depending on how you look at it) in left turns than in right turns.

Say for example the top link is offset 6" to the right. During right turn trail braking, the chassis is rolled 2" left, and the top link has just passed horizontal. You would begin to see brake anti-squat at the right rear, with no equal effect at the left rear. In a left turn, with equal roll, there would be no anti-squat, as the top link would be downhill to the chassis with the right side compressed.

One thing I can say for sure, from experience, is NEVER angle the top link to the side for any reason. If you have to offset it, make sure it remains parallel to the chassis centerline.

[Motorcitydak]

Thanks for all your input guys. I think I am going to run with the 3 link I have setup, throw a watts link at it and move on to other aspects of the car. I will have to see how it goes when I finally get it on the road in a few years