So I played around in Solidworks for a while yesterday just in a sketch, and with a combination of constrained angles, lengths, driven dimensions, and rough fiero dimensions, I have found it is possible to fix the bumpsteer problem with Tie rod location.
Ok first let me define the problem. Well the pre 88s toe in when the suspension is compressed which is fine, but was is not fine is that upon rebound it hits a point where the wheel starts to toe out. And that is not fine because coming out of a turn the rear end basically forces you into crazy oversteer and unpredictale handling. Now if you have a super stiff suspension it is not that big of a deal (like 350# springs) because you are never going to get to this point where the toe switches, but with the stock suspension it is possible.
So I found after taking down many points and doing many perpendicular bisectors that although there is not one instant center, there is an area where at normal suspension movement the centers are all in the same area. If you put a tie rod attached right there, you get decent toe but it swtches very little back and forth because the movement of the tie rod attachment point on the knuckle is not a consistant arc. You can fix this by making the arm the same angle that it is when at this point, but by making it longer. When the tie rod is longer it changes angle less and so it changes it's x component distance less. This causes the car to consistantly toe in when the suspension compresses and drops (like the 88s).
Here is a crappy paint picture of what I mean.
So the problem with this easy(maybe not super easy) super cheap and real solution is that the tie rod would hit the cradle. So what I was thinking was like a right angled type piece with a nut in the end essentially that you thread the tie rod into. It would bump up the tie rod above where it would hit the cradle yet the effective link would remain the same. It could be made really stiff too so It would not bend. If you used a conventional rod end it would be able to rotate on its axis and then smack everything it could come in contact with. So a joint like this that ccfiero is using to limit movement to just X and Y direction. They would attach on the side of the rear most cross member that faces the front of the car. I think it is a pretty decent Idea and if there are no significant issues with my idea then I will measure the actual fiero dimensions and make it in solidworks until I do the modification to my car.
Zach
Oh and with 3 inches of suspension movement (1.5 both ways) there was a toe in of .00156 inches on compression and toe in of .0066 inches on rebound. And the longer the tie rods the more it will toe in. ( I measured toe in by making the length of the tie rod driven and then writing down what the length was at different heights of suspension movement. I know when the tie rod length grew smaller that the knuckle would toe out (because obviously it is not actually going to grow smaller)
[This message has been edited by zkhennings (edited 07-20-2012).]
Yea of course, my point was that bump steer isnt all bad like on the 88s and you can redirect the toe changes to all be toe in with the pre 88 and this is how. It would be way more predictable and stable and there would be no problems exiting fast turns or hitting big potholes. But you would have to weld to the cradle but hopefully no cutting would be necessary. You could remove the current attachment points if you wanted. But yea this is not a "get rid of bumpsteer thread" it is just to fix the main issue of the toe change from toe in to toe out.
I have an idea, that totally get the tie rod and linkage and mounting points, gone.. no toe change at all.. it's so simple, and the parts are already in use and have been for 70+ years.. in vehicles..
just have to cad it up, test it and then patent it.. as every irs would/could use this and still have adjustment..
How much of a difference does that make? I finally experienced rear bump steer, when I really flogged my car, and it's just a minor wiggle. - It required full suspension travel in the hands of a race driver at a fast pace, and then even as a passenger it was MINOR. - When I'm driving I don't even notice it. - Maybe I instinctually correct for the wiggle, or it's even more minor when I'm at the wheel.
That said, how does the handling change with this relocation? Any interesting characteristics?
Most of the 84-87 cars' bad habits come from big gooey cradle and control arm bushings as well as improper (lack of) tire stagger.
Once you go to hard cradle and control arm bushings and get wide tires in the rear, then bump steer starts to show up as an uneasiness or lack of precision in placing the rear end of the car.
[This message has been edited by Will (edited 07-21-2012).]
Do not confuse bump steer w/ maintenance and design problems. Most owners read car rags and internet, worse people trying to sell you expensive parts, saying Fiero has bad "bump steer" is hog wash at best. see my cave, bump steer in suspension section.
------------------ Dr. Ian Malcolm: Yeah, but your scientists were so preoccupied with whether or not they could, they didn't stop to think if they should. (Jurassic Park)
People please. I have new kybs all around. New ball joints, new poly all around, new solid aluminum cradle mounts all around, a shop alignment, and new tires.
This upgrade would be next to free, not affect the hubs at all, not affect wheel size you run at all, offers an actual solution unlike RCC kit, keeps the car very close to stock, adds next to none unsprung weight wise as well. I just cant justify doing something like the Held kit when I could swap an 88 cradle and when it adds all that unsprung weight. I notice bumpsteer the most when I am sliding in a turn and then then the wheels hook up as I straighten out. It throws the car the other way. It feels different than the center of gravity pulling the car around. Also, super fast lane changes, the back end wants to dance around. Also on rough roads it is pretty bad.
Mathematically everything works out. You would no longer toe out. Of course an 88 cradle swap or held kit would remove bumpsteer all together but but both are a more involved job and more expensive. Also I want to stay away from the 88 rear hubs because aren't they virtually non existent as replacements except from china? Anything else is not really in the budget right now.
The main problem with this method is that the height of the attachment point to the cradle would have to be adjusted if height was adjusted.
Originally posted by zkhennings: Also I want to stay away from the 88 rear hubs because aren't they virtually non existent as replacements except from china?
The 88's use the exact same wheel bearing in the rear as the 84-87's use... only the fronts are different (and somewhat hard to come by).
The rear suspension setup is just one of the many reasons I don't bother with 84-87s anymore and stick with 88's only.
Ok that is good to know. Yea it is just not in the budget to do a cradle swap or held kit right now so I was looking for a good solution. Once I make this modification I will report back. I will maybe either do a cradle swap or maybe even just a knuckle swap and recreate an 88 suspension eventually, but maybe not if this modification works well.
And another reason to do the modification is because of heavy braking instability. With the amount the pre 88s dive on braking, the rear suspension drops enough for the wheels to start to toe out. But mathematically this solution would fix that.
And fieroguru I really like what you did with lowering the attachment point of the rear links on the knuckles I do not know if something like that is possible with the pre 88 knuckle, so I have considered a really wide tubular A arm to clear the transmission and attach to new attachment points higher up on the cradle (to effectively increase negative camber gain on compression) Brackets could even be made to stick out from the cradle and be raised up a little and maybe the stock arms could be reused.
[This message has been edited by zkhennings (edited 07-21-2012).]
The behavior you've described is the result of having a rear bar with rear struts that are not stiff enough.
When you have a bar, you have two different spring rates: one in jounce/rebound and a stiffer rate in roll. The struts can only match one rate. When the struts are stiff enough to match the springs but are not stiff enough to match the springs and the bar, the car can have roll oscillations that cause it to fishtail coming off a slide. My Formula does this with stock rear bar/springs and Konis set full stiff (also had rod end lateral links).
Ditch the bar, stiffen your springs and struts to compensate and you'll reduce/eliminate that tendency.
I removed the poly on my sway bar end links front and back put rodney's end links and I have a nice tight handling car. No twitch, stock with kyb shocks. When I hit 100 my my front end squats and handles nice. Good luck chasing your problems
This may be a moot point but, unless I'm mistaken, the Held/Arraut(sp?) bump steer kit ties the toe link back to the A-arm. There were some issues with the production pieces (I saw a toe link connection actually break ) but it sounds like a good idea on the face of it.
Edit - It's a little more complex than that, but the point is, it remains fixed, in relation to the LCA.
Edit again - I see that the knuckle is not in the same plane as the ball joint, so it would still screw things up without the additional hardware. I'll leave this here anyway, unless someone considers it a blatant hijack.
[This message has been edited by Raydar (edited 10-01-2014).]
I was thinking more of a DIY fix instead of buying a complicated solution.
I bought a book a while back where the writer just moved the attachment point of the tie rod and use a different size tie rod to minimize the bump steer. Can't find the book or remember specific details.
So I was wondering a bout such a fix or something similar.
Unfortunately once I took the measurements and saw where the tie rod would need to be attached, it would need to be curved so it clears the cradle, which is fine kinematically, but realistically the tie rod would bend like crazy being as oddly shaped as it would need to be for this to work.
I don't think there is a good DIY fix for this yet that doesn't require some serious fab work. I have looked into lowering the tie rod attachment point at the knuckle by drilling out the tapered section, using a steel cylinder to drop the attachment point to level with the control arm/ ball joint center. You would then need to use a rod end instead of a tie rod end. Then this obviously needs to be supported so the moment on the steering arm doesn't break it off/bend it so I looked into using a longer stud instead of the pinch bolt on the ball joint so it would serve two purposes, one being that it holds the ball joint in the knuckle, the other being that it could be used to bolt to a supporting feature that would support the cylinder to lower the moment on the steering arm. Then the other end of the new tie rod would need to be mounted to the control arm itself on the plane between the ball joint center and the control arm bushing axis.
I have not done any FEA or anything like that on this nor have I mocked up any prototypes because I decided if I am going to do this much fabrication to solve a problem that isn't huge I might as well go all out and redesign the rear suspension completely since it is not optimized at all for our car. I'm planning on an ecotec swap so I decided I was just going to fab up a new cradle, raise the inner pivots to raise the roll center, and make the control arms as long as possible. I will probably go with corvette knuckles all around the whole car front and rear to get bigger and better and lighter brakes (than my grand am calipers) better bearings, bigger ball joints, and better wheel choices since a lot of the 5x100 have a high offset for the VW and Subie people. I will be fabricating tubular control arms and will probably move to a delrin bushing, I am going to try and avoid rod ends except for maybe the upper front control arm for super adjustability of caster and camber and also I might use rod ends for the trailing link I have planned.
In my oval track car, I run the HELD kit with aluminum cradle bushings, a rear sway bar and 88 strut tower adapters. Until I had a bad shock, bump-steer was a non-issue.
Yes the HELD kit is expensive but probably more so because it's not volume manufactured.
Originally posted by zkhennings: I will be fabricating tubular control arms and will probably move to a delrin bushing, I am going to try and avoid rod ends except for maybe the upper front control arm for super adjustability of caster and camber and also I might use rod ends for the trailing link I have planned.
Use rod ends/spherical bearings throughout... think of the maintainability. When a rod end wears out, screw a new one in... When a delrin bushing wears out, turn a new one on a lathe. I learned that the hard way.
Use rod ends/spherical bearings throughout... think of the maintainability. When a rod end wears out, screw a new one in... When a delrin bushing wears out, turn a new one on a lathe. I learned that the hard way.
I was thinking that delrin bushings would outlast rod ends on the road, I have seen rod ends with seals on them but I have no idea how well they function. If from experience you have seen that lifetimes are similar, then I would prefer to go with rod ends for everything.
I don't think it cold flows I am pretty sure it is crystalline or glassy (forget which term I'm looking for), it has some lubricating properties as well though which you won't get with UHMWPE but I bet UHMWPE is more durable than the delrin.
Honestly it would be easier and result in better performance if I get rod ends, if they have similar lifetimes to bushings then I don't see a downside. Do you happen to have the seals on them and can comment on how effective they are?
Originally posted by Will: My UHMW bushings wore out surprisingly quickly. That material also apparently cold flows (not sure about delrin, which is a trade name for nylon)
UHMW, aka UHMWPE, is a subset of polyethylene, like HDPE and LDPE. (Chines soup container (HDPE) and lid (LDPE) and many other uses.)
quote
Originally posted by Raydar: This may be a moot point but, unless I'm mistaken, the Held/Arraut(sp?) bump steer kit ties the toe link back to the A-arm. There were some issues with the production pieces (I saw a toe link connection actually break ) but it sounds like a good idea on the face of it.
It is an attempt for older car to be 88 rear. It has a number of stress points that can bend or break. Rod Ends that connect to knuckle bracket is a big one but not only fail point.
In Most cases... many can reduce or eliminate "bump steer" and/or "torque steer" by replacing struts w/ premium gas units, using poly/metal cradle bushing and replace dieing/dead CA bushings.
(Here's the specific post: http://realfierotech.com/ph...php?p=145114#p145114 ) The RCC kit is bad design and a crappy product. Yes, Delrin and UHMW are different materials, but the same wear and loading cautions apply to both. I used to be a proponent of UHMW bushings, but I've found mine wore too quickly for my liking.
quote
Originally posted by zkhennings:
I don't think it cold flows I am pretty sure it is crystalline or glassy (forget which term I'm looking for), it has some lubricating properties as well though which you won't get with UHMWPE but I bet UHMWPE is more durable than the delrin.
Honestly it would be easier and result in better performance if I get rod ends, if they have similar lifetimes to bushings then I don't see a downside. Do you happen to have the seals on them and can comment on how effective they are?
If you're designing from scratch, use these:
Steel on steel, sealed and engineered for automotive applications, 30 year production history and CURRENT USE by Mercedes.
Originally posted by Will: That material also apparently cold flows (not sure about delrin, which is a trade name for nylon)
Certainly not the point of the discussion, but Delrin is the trade name for acetal, not Nylon (which is actually a trade name itself for something else).
They talk in that document that Delrin has "good resilience and resistance to creep", but I don't know "compared to what". The details start on page 21.
) but it sounds like a good idea on the face of it.
