Angle of upper bars on triangular 4 link setup

[jonesmoto]

I know a guy who is building his own triangular 4 link setup and was wondering what is the minimum angle that the upper bars need to be at so there is no side to side 3rd member movement. He told me he was fabricating them to be at 30 degrees. I said I think they need to be no less than 45 degrees. Can anyone shed some light on this. Do not really care who is right or wrong, just want the facts.
Thanks.

[silver69camaro]

30 degrees will work great.

[Norm Peterson]

30° is only a couple of degrees less than the angle of the uppers in the Fox Mustang.

Side movement in an actual installation depends on things like the stiffness of the bushing material and bracket deflections. With any given combination of bushings and bracket stiffness, the 30° arrangement will always permit slightly greater lateral axle movement, but it does not have to be an excessive amount.


The one thing to keep in mind is that the link load due to cornering with the links at 30° is going to be about 41% higher than if they were at 45°. This shows up in the bushing and bracket deflections as well as in the size of the load that must be resisted.

Link loads from acceleration and braking will be slightly less with the 30° links.


Norm

[jonesmoto]

Thanks for the replies.

[Jasons72]

So if you run the higher angles does that eliminate the need for a panhard bar?

[LowBuckX]

So if you run the higher angles does that eliminate the need for a panhard bar?
Yes

Wouldnt the more angle be better in terms of the rod ends being stressed in shear instead of in bending???

[Norm Peterson]

So if you run the higher angles does that eliminate the need for a panhard bar?
Theoretically, a panhard bar is not kinematically necessary with any variation on the converging 4-link. That's assuming rod ends and minimal bracket/etc. deflection on the arms that are more sharply inclined in plan view. In practice, an installation at 45° with compliant bushings would benefit less from adding a PHB than one at 30°, which in turn benefits less than a 15° setup would.





Wouldnt the more angle be better in terms of the rod ends being stressed in shear instead of in bending???
Rod ends are not loaded in either bending or shear when the load is radial (in the plane perpendicular with respect to the bolt axis through the ball). The ball is assumed to be a frictionless bearing, meaning that it will not resist bending. If there is no bending at either end of the arm and there are no loads applied in the middle of the arm (such as in the Fox Mustangs with LCA-mounted springs), there can be no shear anywhere along it either.

There are a couple of general exceptions related to poor practice/maintenance - if you run the rod end out of angular travel, or if the rod end somehow gets frozen (rust, crud, mechanical damage, etc.), the ball ends will resist moments.

Loading a rod end along the bolt axis will absolutely put the rod end in bending, but you'd have to go well out of your way to design them to load a rod end in a stick axle rear suspension arm or link in this manner.

Where you can see rod ends loaded in this fashion is in independent suspensions, most commonly in small formula cars and (IIRC) in some of the Lotus 7 replicars.


Norm

[silver69camaro]

Don't forget packaging. Wider angles require either narrower axle housing mounts (which can contact the center section) or wider frame rails. You have to compromise somewhere.

In my experience, with our triangulated 4-bar, the housing will move laterally about .100" at about 0.9g. Hardly something to worry about.

[Bryce]

Out of curiousity. is there a difference if the bars angle in or out. The chassis mounts being closer together than the housing mounts would be angled in (fox mustang). Chassis mounts being wider than the housing mounts would be angled out (art morrison triangulated four link)

From pure geometry i dont see a difference, just packaging?

[formula]

provided they were at the same relative angle, and outside of all the forces on the links being opposite, the big issue seems to be packaging. Both will work, that's for sure.

[Rhino]

Out of curiousity. is there a difference if the bars angle in or out. The chassis mounts being closer together than the housing mounts would be angled in (fox mustang). Chassis mounts being wider than the housing mounts would be angled out (art morrison triangulated four link)

From pure geometry i dont see a difference, just packaging?

The difference between the differing convergence (front vs rear) is something I've been wondering about lately. It seems that, for my goals, it works out much better to have the upper bar frame mounts further toward the center. To figure out exactly why this is, I'll need to dig out my notes.
I haven't seen many other set ups run like this, so I wonder if there's something I'm missing.

[Norm Peterson]

Out of curiousity. is there a difference if the bars angle in or out. The chassis mounts being closer together than the housing mounts would be angled in (fox mustang). Chassis mounts being wider than the housing mounts would be angled out (art morrison triangulated four link)

From pure geometry i dont see a difference, just packaging?
Roll center height and axle roll steer will be different because you've moved the convergence point of the uppers.


Norm

[Jasons72]

So which would be better?

[-DA]

That depends on where you want your roll center height and the slope of the uppper and lower links.

Anyone please correct me if i'm wrong here (Norm, Billy) but here's the way I've come to understand it...

As Norm said, the roll axis and roll-steer values will differ with the links converging at the frame vs axle. If only the top links converge (lowers are parallel to each other) then the roll axis would run parallel to the lower links and pass through the theoretical intersection point of the converging upper links. Keep in mind we're only talking about the characteristics of the rear suspension here and disregarding the front.

So... while looking at your car from a side view, let's assume your upper links are sloped down towards the front of the car and the lowers are parallel with the ground (small of anti-squat value, btw). If the links converge behind the axle (narrower at axle) then your roll axis height will be parallel to the ground and above the axle, as it must pass through the link intersection point. now with the same side profile, if the uppers converge on the frame side (narrow in front) then your roll axis will still be parallel to the ground but will pass through the much lower link intersection point.

The slope of your roll axis ties in with the roll-steer issue. If the lower links are still parallel to each other but sloped towards either the front or rear then, in roll, the rear axle will induce understeer or oversteer accordingly. The steeper the slope the more dramatic the roll steer effect. No slope equals neutral steer.

Choosing between the links converging forward or rearward depends on how high you want your roll axis. Just don't forget the importance of the slope of your lower links, too, as they'll determine the slope of the roll axis. All this said, I think I'd want a neutral or slightly understeering effect as it would seem to promote stability with body roll. If I want a roll center height near the wheel center then, with my lower links parallel to each other and the ground, this would place my upper links converging forward, towards the frame with a slight slope down towards the front.

[formula]

Roll center height and axle roll steer will be different because you've moved the convergence point of the uppers.


Norm

oh man, totally didn't think about that.

Norm's right. Everybody listen to norm instead of me, haha!

[Jasons72]

I'm so glad i joined this site lol, so lower links parallel to each other and the ground, uppers slightly pointed downwards towards the front of the car and angled at 45 degrees towards the center of the car (wider at the axles and narrower in front of the axles) do i have that right? Then i won't need a panhard bar? I was thinking of using a either a rod end or johnny joint on one side of the link and a poly bushing on the other.

[Norm Peterson]

All it takes to eliminate the necessity of a PHB/Watts link/other lateral locating device is to have either the uppers or the lowers skewed either wider at the axle ends or wider at the chassis ends by enough to limit lateral movement of the axle. For this part of the design solution, it does not matter which ends are wider. Beyond that, you probably need to run some roll center and roll steer math to see what you've got.

Should you choose to run them wider at the axle ends, keep in mind that you need to maintain a little clearance to the driveshaft for the links and all added structure necessary to support the chassis-side brackets under any possible suspension movement or position.


Norm