Bump steer purgatory - solved

[Rdone585]

I thought I was knowledgeable on this topic, but I’m quickly finding out I fall way short of understanding how to address my bump steer issues. I know this is a big post and a lot of info, thanks for your patience.
MK II, originally built in 2002
FFR tubular lower (standard) control arms
SN95 spindles (94-95 version with the dogleg)
SAI mod, pioneered by David Borden
Unisteer manual rack with extenders, and solid offset bushings
Maximum Motorsports bolt-through outer tie rod bump steer kit
Notes:
• The steering rack is adjusted to the lowest position and centered as specified in the directions
• The floor and frame are level for bump steer measurements
• Ride height, corner weights, set (although they don’t impact bump steer)
• Camber -2.0 degrees, Castor +5.0 degrees
• Static toe is -1/16th inch
• Front sway bar and coil over disconnected for bump steer measurements
• I have only measured driver side so far
142862

PROBLEM:
I had about 2” of total bump steer present before starting to work on this. I ordered a new manual steering rack (with rack extenders and solid offset bushings) and the bump steer kit to fix this problem. I also purchased Longacre scales, alignment tools and their bump steer gage to measure my progress. I’ve used all of the adjustment range for the outer tie rod bump steer kit - all the way up to the spindle’s steering arm. I still see almost 1” of total bump steer with the outer tie rod below the steering arm. I’m seeing 9/16th inches toe out on compression and 7/16th inches toe in on rebound / droop.
So, I figured I was improving… what happens if I go above the steering arm? Tried that. It gets better but in the opposite direction, still with 0.5” of total bump steer. The toe profile changes to 1/8th in toe in at compression, and 5/16th of toe out on rebound.

ANALYSIS:
I took to Excel to try and find where the outer tie rod wanted to be for a theoretical 0.0” total bump steer. Turns out it is with the bottom surface of the outer tie rod being ¼” below the upper surface of the steering arm. I confirmed this by taking multiple bump steer curves at various locations and plotting the results.
Then I got to thinking and reading, and reading, and trying to figure out how this could be. Most of the bump steer issues come from mixing the wrong parts together, or trying to do something the car wasn’t designed to do. So, I took a close look at the SAI mod and how it fit with my parts versus what I see on the forum. I noticed picture of the upper control arms have the ball joint perpendicular to the control arm. But mine is slanted.
142863

POSSIBLE SOLUTIONS:
1. If my upper control arm ball joint geometry contributes to bump steer – get new upper steering arm components and restart the bump steer effort (need some experienced help here so I don’t spend time and money if not warranted).
2. I could install extended length ball joints. The problem is they would have to be 1” extended. I can only find up to 0.5” extended length ball joints for K8259 ball joints that fit into the FFR LCA. Also, this would tend to raise the tire relative to the frame and reduce clearance with the body.
3. I could raise the steering rack and move the outer tie rod above the steering arm. This might solve the issue but will mean I have to make a custom brake duct mount at the spindle. This might be the best option if it’s successful.
4. I could make custom steering rack brackets to further lower the steering rack. While initially appealing I would have to alter the existing brackets to allow the required adjustment.
5. Gordon Levy has a kit that lowers the rack, but it would interfere with my current radiator, oil cooler, and sway bar positioning.
6. I could switch to 96-04 SN95 spindles. These spindles have a straight steering arm. The outer tie rod would attach above it. The issues with this are that it moves the tire out from the frame about 3/8th inch from the frame on each side. I have 4 sets of wheels and tires that would no longer fit because I would need a different backspace for the front wheels.
7. I could move the outer tie rod above the steering arm and add 4 degrees of castor. That would put me at 9 degrees of castor with manual steering. I fear the steering effort would be too much.
8. I could install 0.5” extended length ball joints, and reduce castor to 1 degree. This would get it close but I fear the car would be unstable on fast straight at the track with high speed.
9. I could attempt to design custom length steering rack extenders. I haven’t thought of this to much, but it’s an option.
Here is where you guys come in. I’m hoping you can shed some light on this and help me on this journey. Perhaps others will learn something in the process, I certainly have.

The edit simply corrected David's name.

Thanks,
Robert

[rich grsc]

I know exactly what I would do if it was my car. Take all the stuff you added to "fix" the bump steer off the car, build it back to the way FFR designed it. Go drive the car and see how it feels. I know on my old car, rack extended just made it feel worse. I just think you threw way too many changes to know what helps and what doesn't.
I just never had enough bump steer to be an issue.

[tonywy]

Robert, I think your upper control arms are mounted wrong. If I see your pic correct they are mounted on the side of the mount. I have a MKII as well and they are mounted on top of the mount. If you want I can send a pic. Also my car is using 13819 FFR spindle adapters.

[Rdone585]

tonywy, please send me a pic. I can send other pics of mine from different angles if that helps.

[Rdone585]

rich grsc, I know exactly where you are coming from, never make more than one change at a time. I wish I could go back to a known good setup. The problem is this all started to become very evident after changing my lower control arms. The FFR arms are slightly longer than the ones I had. But I had no other logical choice since I had to replace my old Pro shocks, they were leaking. The Konis wont mount in my old LCSs. I'm also not certain that I haven't had a bump steer issue for a long time. I've also considered hacking up the FFR arms and modifying them with rod ends (so they would be adjustable) like some others have done. But I wanted to look into other options first since that doesn't really take me back to a known good setup.

[mosh1999]

I believe your upper ball joints are on reversed sides.

[Gromit]

on the newer MkIV I think the upper a arms are flipped over so the ball joint points out to the knuckle not in to the center of the car. not sure which way it goes on a MkII


I believe your upper ball joints are on reversed sides.

[tonywy]

Robert,this the left front.142874

[GThompson]

When FFR went from the Mk2 to the Mk3 one of the changes they made was moving the steering rack from being mounted directly above the front pivot of the lower control arms (between the 2 plates the front pivot was also mounted between) to being mounted behind the rear plate of the front pivot and against the front of the X-brace in the front suspension. My understanding is that this improved Ackerman on them, did it also reduce bump-steer? I made this change to my gen 1 Daytona but haven’t driven it enough to know if it’s good.

Also, it’s David Borden, not Daryl.

[rich grsc]

Good catch guys, I missed that. Yes the ball joints are installed upside down. The angle of the mount is wrong, tonywy shows the correct installation.

[Rdone585]

Thanks so much. I thought this might be the case but wanted to check here before just trying it out. It does make sense that this could be the source of my issues. In step 12 of the SAI installation instructions (found on the other forum with a title search "SAI Mod install Instructions Within", by David Borden) it states "In order to get the maximum range of motion, it may be best to flip your upper control arm over so that the angle of the ball joint is reversed top to bottom changing the angle of the ball joint". So tonywy, do your have the SAI mod kit in you setup? I can't tell for sure with the pic you sent.

George, thanks I don't know how Daryl got in my post, I always knew it was David. I'll attempt to edit my original post so it doesn't confuse anyone.
It may take a few days but I'll post what I find out by flipping the upper control arms.

[Rdone585]

GThompson, it's my understanding, from the reading I've done, that ackerman will not add to or remove bump steer. It will however (intentionally) change the turning radius between the right and left sides. In some cases this can be a good thing, in others not as much. From what I've gathered ackerman can be beneficial for autocross, but maybe not so much for road race track events. There are many tutorials on ackerman online if you're interested in learning more. Many of the tutorials use go-karts to illustrate the concepts. There are instructions that came with the steering rack kit that tell how to drill into the X-brace of the chassis and move the rack mount rearward in order to add ackerman. So if the holes drilled into the X-brace (mind you this technique is MKI and MKII specific) were drilled lower, then while introducing ackerman one could also improve bump steer if done properly. I just hesitate from drilling the X-brace because I would have to remove so much (oil cooler, radiator, fan, and shrouds) to gain access to drill.

[Hoooper]

GThompson, it's my understanding, from the reading I've done, that ackerman will not add to or remove bump steer. It will however (intentionally) change the turning radius between the right and left sides. In some cases this can be a good thing, in others not as much. From what I've gathered ackerman can be beneficial for autocross, but maybe not so much for road race track events. There are many tutorials on ackerman online if you're interested in learning more. Many of the tutorials use go-karts to illustrate the concepts. There are instructions that came with the steering rack kit that tell how to drill into the X-brace of the chassis and move the rack mount rearward in order to add ackerman. So if the holes drilled into the X-brace (mind you this technique is MKI and MKII specific) were drilled lower, then while introducing ackerman one could also improve bump steer if done properly. I just hesitate from drilling the X-brace because I would have to remove so much (oil cooler, radiator, fan, and shrouds) to gain access to drill.

ackerman will be present on most high performance road cars with track potential. Porsches, Ferraris, etc. It can be done to varying levels and I would expect a lot of ackerman would be better for autocross and just *some* ackerman for road course, but still an amount of ackerman that would be considered aggressive for a road car is good for the road course. Not relevant to this thread of course but just throwing that out there

[CraigS]

RDone your UCAs are mounted to the frame correctly. With the old Fox spindles the UCA was mounted on top of the plate. The purpose of the SAI mod was to move the upper balljoint outward to reduce the SAI. It was designed by Borden and he used the SN95 spindle since it was shorter than the Fox spindle. It needed to be shorter so the ball joint could move outward. Other wise the ball joint would hit the rim. So BJ moved out (for SAI) and down (because of shorter spindles). Therefore, to keep the UCA angle in front view similar to what it used to be, the frame end of the UCA also needed to move out and down. Hence the UCA is mounted on the vertical plate at the frame. This also made it easier and less expensive to do the SAI mod because you re-used the same UCA.
Your UCA ball joint plates are upside down. The BJ tapered shaft should be pointing down and out. That is mainly done so the BJ has a range of motion w/o it binding because it can't move far enough. But I am not sure it would affect bump steer. The crucial part of the ball joint is the spot inside that it pivots around. I don't think that would change much by flipping the plates if any. Think this way. Imagine the spindle/SAI block is locked into a position. It is in that position because it gives you the alignment specs you want. The ball inside the BJ is also locked into a position since it is fixed into the SAI block. Flip the plate, or flip the whole UCA. Unless you want to change your camber caster specs, you will adjust the UCA sleeves slightly so it all goes back together and the spindle/SAI block/BJ hasn't moved. BTW you can fix the plate by removing the plates from the UCA, unscrewing the BJs, screwing them in from the other side, and possibly needing to swap the plates from driver side to passenger side UCAs.
Back to bump steer. There is another option for you. Traditionally it has always been best to lower the rack as much as possible which you have done. Mostly that was best because the outer tierod end has a taper, which fits in the steering arm w/ it's taper so there was no choice of where the outer tierod end is going to go. The bump steer kit changes that since it uses a straight bolt. So now your have the situation where your outer rod end needs to be about the same place the end of the steering arm is. It might be that moving the rack as high as possible and putting the rod end on top of the steering arm would work out. Another variation is to go back to centered rack mount bushings. So right now you could move the rack from low to high. Let's say that moves it 5/8 inch. If being high doesn't work, a centered bushing would move it back down by 5/16 inch. So you have 3 possible rack mount heights. Be aware that if you move the rack up you may need to shorten the short shaft at the bottom of you steering shaft. Also be aware that you may need to notch the square tube that forms the large X. (On my MkII I had to notch it when I moved my rack back against the vertical square tube to do the Ackerman mod. But mine is a power rack so I think it is larger in that area than a manual rack.) So if you decide to try raising the rack, I'd check one side bumpsteer as soon as you move the rack to see if this direction will help or not. But before you do all that my slow moving brain has two other options;
1- This is a quick no $ experiment. Try increasing your caster to +8 to 9deg. That will move the end of the steering arm up. If it fixes the bumpsteer, installing power steering (I'd go electric so no need to change your rack) may be the easiest of all your options.
2- Do some research. Far enough back in my memory that I am not sure I am correct but....I think the rack extenders were for power racks only.
BTW you may find that getting a perfectly straight bumpsteer line will never happen. I always work more on the compression side of the curve figuring the droop side is less important. In a turn compression is the heavily loaded outside tire vs droop is the much less loaded inside tire.

[Rdone585]

CraigS, lots of good info in your post. It is much appreciated.
I will flip the ball joint plates, and reassess bump steer before moving forward again.
1) It is not possible for me to convert to power steering. I have a dry sump motor and the 3 gal tank is in the way. That's true for both a pump and electrical. We worked real hard to try to make room, but it just didn't happen. If there was an electric version that mounted (and fit) where the rack is, then that might work. I haven't seen a solution that I can use yet. Based on my calculations, increasing caster to 9 degrees (from 5) would give me almost 0.5" of extra space improving the bump steer as you said. But it would probably still leave me with about 1/2" of total bump steer. Still better but with manual steering it would be tough. It's kinda hard to see but this a pic from under the car where the steering column passes by the tank on it's way to the rack.
142916
2) The steering rack kit vendor indicated the extenders were needed to improve bump steer. So here is a question, if I measure the angle of the LCA and the angle of the outer tie rod, will that allow me to then calculate the ideal amount of rack extender needed? I'm assuming the two should ideally be the same.

[CraigS]

Fast Freddie used to sell a PS kit using a Toyota electric hydraulic pump. Some have mounted it to the X frame. It can go anywhere though only limited by how long your hoses need to be to get to the rack and reservoir.
"2) The steering rack kit vendor indicated the extenders were needed to improve bump steer. So here is a question, if I measure the angle of the LCA and the angle of the outer tie rod, will that allow me to then calculate the ideal amount of rack extender needed? I'm assuming the two should ideally be the same."
This is a close approximation but not 100% accurate. One would think that if the tierod and the LCA were at exactly the same angle and exactly the same length there would be no bump steer but....that ignores the UCA. We need the spindle to follow the exact same path as the outer tierod end but the UCA is shorter, and usually at a different angle, so it is moving the spindle also. If the outer steering arm tierod end were at the exact same height as the lower BJ then the LCA would have 100% control of bump steer. But it isn't. The outer tierod end is maybe 1.5 inches above the lower BJ so the UCA has some effect on bumpsteer.
BTW, when measuring bump steer be sure to figure out where the suspension is at normal ride height so you can put that into your drawing. So you have a curve w/ normal ride height at zero bump steer, and then compression travel is above zero and droop is below zero. As I mentioned compression is the more important part of the curve. Also note that you would ideally have zero bump steer in compression but if there has to be some bump steer you want it to move toward toe out in compression. This way, if you hit a bump in the middle of a turn, you get slightly less steering input than you put in at the steering wheel. This will be much more natural to compensate for than if the bump causes the car to tighten the turn.

[Rdone585]

Again, I learn more with each new post. I understand most of the concepts but different discussions put them in the appropriate light for me to see how these things contribute to what I'm trying to accomplish. I'll go measure the passenger side and report the difference since I haven't played with the suspension on that side yet. BTW, I did start the bump stee measurements at ride height geometry for the suspension. I measured the height of the lower BJ relative to the frame with the care off the lift and settled by rolling backward then forward a couple of times. I replicated this difference as the starting point when measuring bump change with the gage.

So why am I so concerned with the bump steer in my car? I installed an new motor a few years ago and attempting to pay more attention to the setup of the car. I noticed the front LCSs were binding and the urathane bushings were missing chunks between the LCS and the mounting ears on the frame. So I opted to replace the bushings with the sperical bushings FFR provides. The great thing is the suspension moves freely now. The bad thing, when you have bump steer as bad as I did, is the phenomena I'm about to describe. If I get a little heavy with th go pedal, the front suspension actually goes in extreme droop mode. This causes the maximum toe in situation. When I finally let the go pedal rest, the weight of the car puts the front suspension in compression... maximum toe out. The car will dart side to side until the suspension settles, and later, the pilot. Fast gear shifting doesn't give the suspension long enough to show the effects. It's when the "run is over" that it's noticed. I've learned now to let off the go pedal slowly and this helps the situation quite a bit. So, this is the condition I am striving very hard to eliminate. With the bushings in the LCA this wasn't noticed. I guess the suspension wasn't responsive enough. I also know that a difference in castor from side to side can be a player. When I last had my car professionally corner weighted and aligned, the castor was different side to side by 0.2 degrees. I've read how left to right castor difference can cause similar things to happen. However, I think the difference needs to be greater than what I (use to) have. I now have all the tools to do corner weighting and alignement myself and will eliminate this as a possible contributor also.

[Rdone585]

Turned out it was easier to restore the driver side to the original tie rod location, just below the spindle steering arm with no spacer.
Total bump steer with this geometry was: 0.980 inches, 0.560 toe out at 2" compression, 0.420 toe in at 1.5" droop.
At ride height...
UCA @ 88.9 degrees
Tie rod @ 83.2 degrees
LCA @ 82.3 degrees
I understand what can be done with these measurements, but I'm not able to analyse this data because I don't know the relative locations and diminsions for all the pivot points. I also don't know if it's worth the effort.
Here are bump steer measurements at different tie rod attachment locations.
142986
My next step is to flip the UCA and check / restore alignment, then reassess bump steer.
Next I will raise the rack, move the tie rod above the steering arm, reset toe, and reassess bump steer.
It might take a few days to get back with an update.

[totem]

Here are bump steer measurements at different tie rod attachment locations.
142986
My next step is to flip the UCA and check / restore alignment, then reassess bump steer.
Next I will raise the rack, move the tie rod above the steering arm, reset toe, and reassess bump steer.
It might take a few days to get back with an update.

"No spacer above " is fairly close. If you raise the steering rack by inverting the offset bushing, you’ll be near perfect. Note that 2in is a lot of travel.

Since the compression bump steer is the equal opposite of the rebound, it means that the tie rod is the right length, that you have the right rack extender.

As a rule:
1. Toe-out under compression and toe-out under rebound: shorten tie-rod by installing rack extenders.
2. Toe-in under compression and toe-out under rebound, raise rack or lower outer tie-rod end.
3. Toe-out under compression and toe-in under rebound, lower rack or raise outer tie-rod end.

T.

[Rdone585]

totem, I agree and it's my next step to raise the rack and take measurements with the tie rod above the steering arm with no spacer, add spacers if needed.
I'm not sure what 2" of compression with a tire looks like in the wheel well. Here is a pic from a little over 2 years ago (previous color and number for my car). Not sure what that is for compression on right and rebound on left, but it felt like close to maxed out. I've had other pics where the inside tire was completely off the ground. This was with my previous motor, the new one has more get-up-and-go.
143011

[CraigS]

I agree that 'no spacer above' is looking good. You have a really nice rig for checking bumpsteer but it could almost be said to be too nice. By that I mean that reading to .001 inch is, I think, more accurate than you need. This is my technique. Clamp a laser to the rotor and shoot it at the cardboard box.
https://live.staticflickr.com/4835/46116534764_ed9f73872e_z.jpg (https://flic.kr/p/2dgai7m)IMG_20130428_162158_880 (https://flic.kr/p/2dgai7m) by craig stuard (https://www.flickr.com/photos/152454123@N04/), on Flickr
Of course this magnified the measurement. I remember starting out w/ maybe 3 inches of total laser movement and getting it down to a under an inch with most of the inch being in droop.
You have a lot of roll. For reference I had 800 front springs, 400 rear, and a 1 inch front bar from Forte.

[Carl Johnson]

I have a MK2, I installed this years ago, very happy with my steering. it moves the rack forward and down.
http://lrclassicsllc.com/store/competition-bump-steer-street-kit/

[Rdone585]

Carl, are you able to share pics of the mounts and new location of the rack. I'd like to see if I have room. I did notice that as an option but had space constraint concerns since my car is already built. I fear my sway bar and radiator are in the way. If that's true then my shrouds and oil cooler would also have to be moved. I would need to know how far forward and how far down the rack moves to make an educated guess.

[Carl Johnson]

143270

[Rdone585]

Carl, thanks for the pic. It looks like the relocation has the rack covering most of the front face of the chassis tubes. As I feared, I'm pretty sure my sway bar and the routing for my lower radiator hose are a problem, possibly the radiator angle as well. I suspect your lower radiator hose is now routed above the rack?

[Carl Johnson]

143279
Carl, thanks for the pic. It looks like the relocation has the rack covering most of the front face of the chassis tubes. As I feared, I'm pretty sure my sway bar and the routing for my lower radiator hose are a problem, possibly the radiator angle as well. I suspect your lower radiator hose is now routed above the rack?

[Rdone585]

Hmmm, that sheds a new light on my thoughts. I'm already in the process of moving the rack up and rechecking bump steer. If I can't dial it in to something reasonable I may give Gordon a call and discuss my situation and components and how they might work with his kit.

[Rdone585]

Craig, sorry didn't respond to your earlier post. I'm running 500# front and 350# rear with 7/8" front bar and VPN rear bar. The roads hear in WI are so bad I don't think I could stand going much heavier on springs without risking loosing some teeth :rolleyes:.
I agree measuring to 0.001 isn't beneficial. When I'm measuring it's really to 0.01 most of the time, and that's because this gage will do that. It also helps to notice if the toe stalls or moves the other direction slightly during compression or droop. If it does that's diagnostic information that is help in deciding what the best solution is. Example, if during rebound (droop) the toe changes from in to out, the tie rod may need to be shortened.

[Rdone585]

I finished flipping the upper control arms and realigning. I wound up increasing castor from 5.0 to 5.2 during the realignment. The castor change along with flipping the control arms did result in a minor change to the bump steer. But oddly enough only when the tie rod is above the steering arm.

I then raised the steering rack and rechecked the bump steer. It did improve when the tie rod is above the steering arm. The change puts it very close to where it should be, 0.040 toe in on compression, 0.140 toe out on droop. It could be better but I'm good with something in that range. However, this was with no spacer at all between the tie rod and the steering arm. The tie rod binds up on the steering arm at about 1" of droop with no spacer.

So, I plan to lower the castor to 3.5 to 4 degrees in hopes that I'm able to use the 1/8" or 1/16" spacer so the tie rod doesn't bind. Since I believe the end is in sight I'm going to reset the ride height to where I originally had it because that didn't help like I had hoped. Then I'll realign with the lower castor setting, and recheck bump steer. Somewhere along the way I'll also corner weight the car again. I'll give an update with the final results when done.

[CraigS]

Glad it is working out for you. You could consider different rod ends w/ a greater allowable angle. This is a standard rodend where the body around the eye is about the same thickness as the threads.
https://www.summitracing.com/parts/qa1-hmr8ht
Here is one where it is much thinner around the eye. Note I chose these to link just because of the angle that their picture was shot at makes it easy to see the difference.
https://www.summitracing.com/parts/qa1-pcyfr8t

[Rdone585]

The HMR8HT looks much like the rod end in the Maximum Motorsports tie rod. The PCYFR8Trod end looks like it has the spacer as part of the ball. The Maximum Motorsports tie rod end is 5/8" and attached to the steering arm with a 5/8" bolt more like:
https://www.summitracing.com/parts/hal-pcymr10-12t.
It's hard to tell what the rod end height is since they don't publish that dimension on Summit. I found the QA1 catalog and those rod ends are 1 3/8" thick, that's 5/8" thicker than the Maximum Motorsports tie rod end. So while it would not bind it also would limit the bump steer correction possible near the steering arm.

It appears that I could file away about 1/32" off the end tip of the Maximum Motorsports rod end head and there would not be a binding issue. But this would weaken it slightly and I don't want to do that if I can find another way. Still planning and hoping that reducing the castor from 5.2 to about 4 degrees will free up at least 1/16" for a spacer.

[CraigS]

Could you grind a little off the interfering corner of the steering arm?

[Rdone585]

Yes I could. That's probably a better option if necessary since it's a much stronger and larger mass of metal.

[Mark Reynolds]

Another thing you can do is mount the upper control arm to the inside surface of the angle iron. This effectively makes the upper control arm radius longer which reduces the amount the upper ball joint moves in and out when the suspension moves up and down. This may require grinding the 3/4" tubes to clear the rear pivot and getting longer control arm swaged tubes.

[Rdone585]

Thanks for all the help from everyone, it really was instrumental in formulating a plan and following through. I wanted to come back and share what worked for me. Possibly this can help others working on bump steer.
I flipped the upper control arm to properly align the ball joint.

As a side note I also rechecked bump steer at this point, There was a small change, about 0.25" improvement when below the steering arm.
I raised the steering rack using the offset bushings.
I moved the tie rod ends above the spindle steering arm with no spacer.
I reset toe to 1/16" in.
Using a bump steer gauge I re-evaluated the amount of driver side bump steer present.
I found 0.205" total bump steer from 2" compression to 1.5" droop.

Suspension - Bump steer
2" (compression) - 0.025" toe in
1.5" - 0.040" toe in
1.0" - 0.040" toe in
0.5" - 0.030" toe in
-0.5" (droop) - 0.040" toe out
-1.0" - 0.100" toe out
-1.5" - 0.180" toe out

At this point I knew I was on the right track.
I noticed a slight amount of binding with the tie rod end and the steering arm. I ground away a very small portion of one edge of the steering arm and the bind was eliminated.
I changed the alignment to provide a slight lowering of the steering arm, and thus improving bump steer adjustment range with the tie rod end above the steering arm.
camber -1.8 degrees, castor 4.0 degrees, 1/32" toe in.
I then re-evaluated the amount of driver side bump steer present.
With no spacer I found 0.025" total bump steer from 1" compression to 1" droop, and 0.090" total bump steer from 2" compression to 2" droop. Note as mentioned earlier, +/- 2" of suspension travel is an extreme amount.

Suspension - Bump steer
2" (compression) - 0.065" toe out
1.5" - 0.050" toe out
1.0" - 0.020" toe out
0.5" - 0.000"
-0.5" (droop) - 0.008" toe out
-1.0" - 0.025" toe out
-1.5" - 0.050" toe out
-2" - 0.090" toe out


I then checked the passenger side bump steer present.
I found that a 0.25" spacer on the passenger side provided the optimum bump steer solution.
I found 0.050" total bump steer from 1" compression to 1" droop, and 0.130" total bump steer from 2" compression to 2" droop.

Suspension - Bump steer
2" (compression) - 0.120" toe in
1.5" - 0.080" toe in
1.0" - 0.040" toe in
0.5" - 0.015" toe in
-0.5" (droop) - 0.005" toe out
-1.0" - 0.010" toe out
-1.5" - 0.010" toe out
-2" - 0.005" toe in


This solution is much better than the prior condition. When I started this process I saw about 1" of bump steer per side, now it's almost nonexistent. I'm now in the process of reevaluating the corner weight, that should be done in a couple of days. I can't wait for the weather to warm up a little so I can do a test drive.

[mosh1999]

Glad to hear you are on the right track. Looking forward to the results after you drive it.

[CraigS]

An excellent ending. I would like to see the passenger side go to toe-OUT on bump but, since it is such a small amount compared to what you started with, it probably won't be noticeable.

[Rdone585]

Craig, I agree. If the bump steer were any more I would have sought to make it toe out at the cost of a little more toe change.

It was a very nice day today. I was able to do a test ride for the first time since it was parked for the winter. The car drives so much better, even than any of the professional alignments and corner weighting I had paid for. All the alignment and steering issues are gone. No more having the car try to decide which way it wants to go, I feel like I'm in control now - finally. I can't wait for the next track event to see how much better it handles.

[CraigS]

That is great news. It is a fantastic feeling when you do lot of work and the end result is what you were hoping for isn't it? Congratulations!

[J R Jones]

Rdone, It appears that you have a solution. Reading through the thread I became concerned that the problem was over-thought creating paralysis through analysis. The beam of hope came from CraigS and the laser.
I too am near MKE. In my racing days Carroll Smith's Prepare To Win was my bible, and current technology makes his advice more easily implemented. Building a car from scratch is a "Greenfield" of solutions, working with an existing package is harder.
I have a Series One Excalibur (50's Studebaker chassis) that I resto-roded with Lexus IS300 parts. My front suspension/steering is a hybrid of the two and I had limited freedom of part placement.

Most simply the solution is a parallelogram between control arm and tie rod.
Ideally the tie rod and lower control arm have matching length (and vertical position) so that they rotate in matching arcs.
Like CraigS I use a laser, a gun site laser, clamped to the hub or rotor pointing forward (not lateral) at a screen. Sweeping the hub through travel limits (without shock or spring) demonstrates bump steer on a large scale. My target field was about six feet out.
My task was to make the tie rod arc match the lower control arm arc. If the steering rack and tie rods are positioned above the lower control arm plane, it gets more difficult, as the upper control arm contributes.
In my case the Lexus steering rack was too wide and the tie rods were too long. I replaced the tie rods with an intermediate rod out front of the rack which gave me a "greenfield" for tie rod inner pivots, vertical position and width.
My development tie rod assembly was threaded rod and rod ends. I clamped the inner rod end to the intermediate rod, bolted the outer tie rod to the steering arm and bump steered the laser dot path.

If you do not have a perfect parallelogram to start with, compromises in height and rod length are tried empirically until an acceptable solution is achieved. Then fabricate real tie rods to match the development part. Looks like you are there.

[Rdone585]

J R,
The laser method you describe is a good method I've read about many times. It will lead you to the same end result. I suspect using that method does mean you have to pay particular attention to a few more control variables. For example, laser placement and alignment as well as consistent distance from the "reflective screen" used to view the results. The bump steer gauge also has variables to control but can be easily zeroed out between uses so that results are comparable at later points in time. The gauge also allows one to know the magnitude of the bump steer that is present. Both methods can be used successfully.
The challenge I was faced with was working with a finished build. The components used are not as adjustable in areas when you might want to make adjustments (e.g. steering rack position). With every design change there is a compromise to be made. For example, changing the tie rod end to above the steering arm caused me to have to redesign the bracket holding the brake cooling ducts, and changing the upper control arm orientation caused the need to pay particularly close attention to brake caliper clearance, raising the steering rack resulted in shortening the steering linkage with little capability for that adjustment. So, attention to detail throughout the process is needed.
You have an interesting project that you shared. Over the years I've participated in multiple Cobra builds from various vendors. I'm intrigued with different projects because I also help a neighbor with his figure 8 cars that he runs at WIR. We build a new one each year in an attempt to build a better, faster car that will also entertain the crowd. Last year we built a small school bus based on a metric frame. All the body was custom fabricated just for that car. It was a hit and crowd pleaser. This year we have something new ready for the events. It will be unveiled next weekend at the Motor Mania indoor car show. My Cobra will also be there. I plan to be there as much as I'm able on Saturday.

[J R Jones]

Rdone, In the seventies I raced a Shelby GT350, Boss 302 and a Tiger in SCCA Nationals. The basic platforms required a lot of development. With age comes wisdom. Now I utilize more proven platforms, well the Excalibur was not, the Lexus is. My other project is an Avanti body on a custom C4 Corvette chassis. I built the 109in WB frame to utilize the C4 parts. BTW if you have Chevy buddies I have the Corvette 327/300hp engine and trans available. The C4 L98 350ci and automatic are also available. I am near Oconomowoc.

[CraigS]

JR, the problem w/ pointing the laser forward is that the wheel moves in and out through suspension travel. So that moves the laser dot left and or right which could be interpreted as toe change. I wish I could remember where I first saw the idea of pointing the laser to the side so I could give credit where it is due. Unfortunately all I remember is that it was one of those light bulb moments for me.

[J R Jones]

Craig, I understand, I don't remember considering that two years ago. My lower control arm is 22 inches pivot to laser and the tie rod is 26 inches. That trigs out to 0.2 toe out at the extremes I will measure with your method.
I use the Studebaker control arm mounts. The Studebaker uprights had king pins and trunnion pins. I built the upper control arms with poly pivot bushings and QA1 ball joints. I converted the lower control arms to the Lexus ball joint. I shortened the Lexus upright ~7 inches. I replaced the recirculating ball steering box, drag link, bell crank & pivot housing and 26 inch tie rods with the modified Lexus R&P. The car is now a resto-rod replica of a 1929 Mercedes SSK. I added six crossmembers to the frame because Excalibur deleted the "X" member. Thanks for the tip. My paternal family is from Vienna Va.

[J R Jones]

Craig, With rumination I realized I have a further complication, making a lateral bump steer beam problematic.
If you study the photos I posted you will see that my control arms are not 90 degrees lateral. They are swept back at 12-13 degrees. A 90 degree beam would describe an arc, more extreme than forward facing. My intuition is that I would have to fixture the laser perpendicular to the control arm pivots. I am not certain what is gained by the swept back configuration, perhaps anti-dive. GM used a similar design on C1 Corvettes and Chevies.

[Rdone585]

Make sure the rotor is firm against the hub, use a couple of lug nuts to hold it in place. A 90 degree beam from the rotor should give you what you need. You can keep the rotor from turning with a small clamp. With the bump steer gauge a plate is mounted to the wheel lugs. The fixture measures the distance (with a micrometer) that plate moves in / out as the suspension moves the spindle vertically.

[CraigS]

Yep, as I picture the movement in my mind w/ your 'not 90 deg' pivot axis I think that will affect bump steer measuring. I guess it will depend on how bad yours is. You could attach the laser and do a quick experiment w/ both front and side orientation. My first time ever dealing w/ it as in the pic in my post #21, I had > 2inches movement of the beam. When I got it down to a little under 1/2 inch I was happy. Later w/ different components I improved on that a little. As I mentioned to Rdone585 way back, I think it is possible to get too analytical w/ this. IE: I am well aware that high end race cars, and probably production cars at the engineer test phase, measure bump steer to .001 inch. But I think that is way, way beyond what we need to do on an FFR. I am in the let's measure it, let's understand it, let's do the best we can with it, and then let's move on camp.

[CraigS]

Let me ad one more possibility. Computer analysis. For several years I had this program on my PC. The Roll Center Plus version.
http://performancetrends.com/rc.htm
I wasn't working w/ bump steer then and the software was lost when I upgraded to Win 10 so I am not sure it deals w/ bump steer. There are several versions so you would need to look at the details of each.
I found the $140 price at the time was well worth it to me. Of course it's $170 now.

[J R Jones]

Craig, My day job was product development in the recreation industry. We adopted Pro Engineer in the late eighties, and it was in development too. Simulations like FEA took a lot of power. Better now I assume. J Schenk at FF evaluates this I assume. I looked back at my 2017 notes, I started with 1.0 in toe change at 5 feet in droop. (subject to re-evaluation) I had -2.0 degree camber change at full jounce. This neo-classic will have 5.5 wide by 30 diameter Coker tires on wire wheels145959, so world class suspension geometry is a moot point.

145963