Aftermarket ball joints that change the roll center?

[RWD4G63]

And so is adding lowering/performance springs. You guys are already lowering the pivot point for the ball joint and the tie rod end because of your lowered suspension, it seems only natural that one would prefer to raise the inner control arm attachment points and then lower the tie rod end on the upright to avoid bump steer. This would be a better option. It retains the stock ball joints, doesn't require modification to the lca's, doesn't apply a large bending load through your now taller ball joint, and can improve the caster at the same time. Also avoids an wheel clearance issues there as well.

You're failing to take into account a couple things. 1. There's very little room to raise the rear pivot point on the control arm. without doing that you're not actually raising the roll center. 2. The tie rod and control arm are not the same length, which means just because they're parallel doesn't mean there isn't bumpsteer, especially when you're actually steering the car around corners.

 

[802Hoop]

You're failing to take into account a couple things. 1. There's very little room to raise the rear pivot point on the control arm. without doing that you're not actually raising the roll center. 2. The tie rod and control arm are not the same length, which means just because they're parallel doesn't mean there isn't bumpsteer, especially when you're actually steering the car around corners.

I can understand the first issue you pointed out, but I don't buy you second reason. I'm aware that the tie rod and control arm are no the same length, but that just means that this issue is present no matter what you do with the suspension, raising the control arm pivot points or not. This actually brings up another reason why a custom k-member would be beneficial... you could make the steering rack attachment points adjustable with shims.

Honestly, for my purposes, I'm looking at this from strictly a race car perspective, so street use isn't a factor for me. I'm building a 1g for ChampCar out of a salvage car with no-title, so I'm no appose using a hammer to massage some clearance to raise the control arm points. For the steering, one would ideally use a longer steering rack body to get them closer to being equal length, but then you have to worry about changing the ackermann geometry. For racing, I'm okay with reducing the amount of ackermann, but lengthening the steering rack body would do the opposite, so I'd have to position it further forward. Oh the rabbit holes we must follow sometimes...

 

[RWD4G63]

I can understand the first issue you pointed out, but I don't buy you second reason. I'm aware that the tie rod and control arm are no the same length, but that just means that this issue is present no matter what you do with the suspension, raising the control arm pivot points or not. This actually brings up another reason why a custom k-member would be beneficial... you could make the steering rack attachment points adjustable with shims.

Honestly, for my purposes, I'm looking at this from strictly a race car perspective, so street use isn't a factor for me. I'm building a 1g for ChampCar out of a salvage car with no-title, so I'm no appose using a hammer to massage some clearance to raise the control arm points. For the steering, one would ideally use a longer steering rack body to get them closer to being equal length, but then you have to worry about changing the ackermann geometry. For racing, I'm okay with reducing the amount of ackermann, but lengthening the steering rack body would do the opposite, so I'd have to position it further forward. Oh the rabbit holes we must follow sometimes...

The most simple solution is to measure. Get your roll center correction where you need it, then measure and adjust the correction on the tie rod end to minimize bumpsteer. Sure, you may be able to do this with control arm pickups or adjustable rack position or whatever, but you have a far more likely chance of screwing things up than making them better, especially if you're actually racing the car around a circuit. It's unlikely you need to go to all the trouble and complication of adjusting every aspect of the front suspension to have an effective setup. Always measure first.

Also for everyone in general, keep in mind that when you get longer ball joint shanks, you are increasing the ackerman of the steering because of the angle of the ball joint.

 

[tstkl]

Agreed. Just need ride height or a basic understanding of roll center height. COG is going to be about crank centerline.



Hopefully not. That's a good way to induce bumpsteer.



Don't do this for a couple reasons. The first being that it won't actually do anything to adjust roll center. The second being that increasing anti-lift/anti-dive (which is what you would be doing, not decreasing) is not necessary as that should be controlled by the spring. 1G's don't need more anti-lift, they need more anti-squat, which is rear suspension related. Weight transfer to the front under braking is good for AWD cars as it helps rotation and the front brakes do the work.

I'm sorry but it most certainly would adjust the roll center. It also most certainly would decrease the anti-lift, which is most certainly not controlled by the spring. The needs of anti-lift or anti-squat for any platform are a matter of opinion depending on driver preference as well as the type of racing that occurs. In any event, anti-lift and anti-squat are not used to impact weight transfer, although having a lot of anti-dive will result in a less stable rear end under trail braking.

The link below is to a photo that shows how roll centers are calculated on a McPherson strut suspension vehicle. As you can see the inner suspension mounting point is used to determine the instant center which is then used to determine the roll center. Although the roll center is actually the point at which the line connecting the left side contact patch to the left side instant center intersects the same line from the right side, not the vehicle centerline. This ends up being the vehicle centerline when the vehicle is in the static position (and actually symmetrical).
https://rqriley.com/wp-content/uploads/2018/08/fig-14-1.gif

It is fair to say that this would result in a net change of the bump steer characteristics of the vehicle. However I assume anyone who is using a tubular K member with custom suspension points would be able to figure out a way to adjust their bump steer. If you're clever you might even try knuckles off of a 1g FWD turbo car.

Is this aluminum on steel or steel on steel? Either way that's not going to work. That joint needs to rotate and under track usage it will definitely wear the sh** out of whichever is the weaker metal, making the hole oblong and lead to all sorts of issues. You need a spherical bearing in there.

I'm pretty sure it's steel based on the machining marks. The factory control arm is also steel. Given how much bearing area there is and that this is a racecar that won't see too many miles I would recommend machining the hole larger and pressing in a teflon sleeve or some other self lubricating hard plastic. If you're going to use a bearing the best solution would be to use a needle roller bearing. Otherwise I would just use Delrin/UHMW as the compliance wouldn't be that much less but the mass would be significantly less.

The most simple solution is to measure. Get your roll center correction where you need it, then measure and adjust the correction on the tie rod end to minimize bumpsteer. Sure, you may be able to do this with control arm pickups or adjustable rack position or whatever, but you have a far more likely chance of screwing things up than making them better, especially if you're actually racing the car around a circuit. It's unlikely you need to go to all the trouble and complication of adjusting every aspect of the front suspension to have an effective setup. Always measure first.

Also for everyone in general, keep in mind that when you get longer ball joint shanks, you are increasing the ackerman of the steering because of the angle of the ball joint.

I don't see this being a significant contributor. The angle of the arm with respect to the ground plane should remain relatively constant through the travel of the steering rack assuming no change in ride height (bump or droop). So long as the ball joint is positioned vertically above or below the mounting point, moving this point up or down should not change Ackerman angles beyond the fact that it would change the toe which should be corrected to restore the original Ackerman angle from before the modification.

 

[RWD4G63]

I'm sorry but it most certainly would adjust the roll center. It also most certainly would decrease the anti-lift, which is most certainly not controlled by the spring. The needs of anti-lift or anti-squat for any platform are a matter of opinion depending on driver preference as well as the type of racing that occurs. In any event, anti-lift and anti-squat are not used to impact weight transfer, although having a lot of anti-dive will result in a less stable rear end under trail braking.

The link below is to a photo that shows how roll centers are calculated on a McPherson strut suspension vehicle. As you can see the inner suspension mounting point is used to determine the instant center which is then used to determine the roll center. Although the roll center is actually the point at which the line connecting the left side contact patch to the left side instant center intersects the same line from the right side, not the vehicle centerline. This ends up being the vehicle centerline when the vehicle is in the static position (and actually symmetrical).
https://rqriley.com/wp-content/uploads/2018/08/fig-14-1.gif

It is fair to say that this would result in a net change of the bump steer characteristics of the vehicle. However I assume anyone who is using a tubular K member with custom suspension points would be able to figure out a way to adjust their bump steer. If you're clever you might even try knuckles off of a 1g FWD turbo car.

I'm pretty sure it's steel based on the machining marks. The factory control arm is also steel. Given how much bearing area there is and that this is a racecar that won't see too many miles I would recommend machining the hole larger and pressing in a teflon sleeve or some other self lubricating hard plastic. If you're going to use a bearing the best solution would be to use a needle roller bearing. Otherwise I would just use Delrin/UHMW as the compliance wouldn't be that much less but the mass would be significantly less.

I don't see this being a significant contributor. The angle of the arm with respect to the ground plane should remain relatively constant through the travel of the steering rack assuming no change in ride height (bump or droop). So long as the ball joint is positioned vertically above or below the mounting point, moving this point up or down should not change Ackerman angles beyond the fact that it would change the toe which should be corrected to restore the original Ackerman angle from before the modification.

It would effect roll center yes, but only half the amount you raised the pickup point. I'm very familiar with how roll center is calculated, the "pivot points" include both front and rear pickup points on the subframe. Therefore if the pickup points are parallel you can draw a line from the ball joint through the centerline of both pickup points. If they are not parallel, you need to draw a line from the balljoint and split the difference between the height of the front and rear pickup point. Therefore moving only one pickup point on the subframe is relatively ineffective at changing roll center height. You're much better off with longer ball joint shanks. It's unfortunate that the 1G knuckle is cast iron, as cutting and spacing the ball joint pickup and steering arm are much more difficult to do properly vs a forged steel part.

Forgive me, I was incorrect about anti-lift, I had it mixed up. You are correct that it would decrease anti-lift and anti-dive. However, let me clarify: anti-lift and anti-dive motion are controlled by the spring, and generally race vehicles with relatively stiff springs don't need as much of either as street cars. So this may be helpful.

Anti-lift/dive and squat are not driver preference items. There is an ideal amount of each depending on tire compound, track surface, and braking and accelerating forces. Geometry change during these events must also be taken into account. The 1G has pro-squat geometry, and could desperately use more anti-squat in the rear when used in anything besides gravel rally. I'd wager less anti-lift and stock rear geometry is not a good combo for front traction under power.
A needle bearing is actually terrible at dealing with forces not directed in a perfectly radial direction to itself. A large spherical bearing is a better option. I would agree that a delrin or Teflon liner would also probably work though.

Ackerman in the steering has nothing to do with the angle of the control arm relative to the ground plane. It has everything to do with how far inboard the steering arm pickup on the knuckle is from the balljoint. Let me give you a diagram.

You must be logged in to view this image or video.

You must be logged in to view this image or video.

As you can see, as you space the ball joint down and out, you increase the ackerman of the steering. I'm not saying whether this is good or bad, as I don't have experience with how the front tires wear on a 1G while doing roadcourse work, but generally high ackerman is good for autocross and bad for roadcourse work.

 

[tstkl]

It would effect roll center yes, but only half the amount you raised the pickup point. I'm very familiar with how roll center is calculated, the "pivot points" include both front and rear pickup points on the subframe. Therefore if the pickup points are parallel you can draw a line from the ball joint through the centerline of both pickup points. If they are not parallel, you need to draw a line from the balljoint and split the difference between the height of the front and rear pickup point. Therefore moving only one pickup point on the subframe is relatively ineffective at changing roll center height. You're much better off with longer ball joint shanks. It's unfortunate that the 1G knuckle is cast iron, as cutting and spacing the ball joint pickup and steering arm are much more difficult to do properly vs a forged steel part.

Forgive me, I was incorrect about anti-lift, I had it mixed up. You are correct that it would decrease anti-lift and anti-dive. However, let me clarify: anti-lift and anti-dive motion are controlled by the spring, and generally race vehicles with relatively stiff springs don't need as much of either as street cars. So this may be helpful.

Anti-lift/dive and squat are not driver preference items. There is an ideal amount of each depending on tire compound, track surface, and braking and accelerating forces. Geometry change during these events must also be taken into account. The 1G has pro-squat geometry, and could desperately use more anti-squat in the rear when used in anything besides gravel rally. I'd wager less anti-lift and stock rear geometry is not a good combo for front traction under power.
A needle bearing is actually terrible at dealing with forces not directed in a perfectly axial direction to itself. A large spherical bearing is a better option. I would agree that a delrin or Teflon liner would also probably work though.

Ackerman in the steering has nothing to do with the angle of the control arm relative to the ground plane. It has everything to do with how far inboard the steering arm pickup on the knuckle is from the balljoint. Let me give you a diagram.

You must be logged in to view this image or video.

You must be logged in to view this image or video.

As you can see, as you space the ball joint down and out, you increase the ackerman of the steering. I'm not saying whether this is good or bad, as I don't have experience with how the front tires wear on a 1G while doing roadcourse work, but generally high ackerman is good for autocross and bad for roadcourse work.

I'll start off by saying that you are indeed correct about the Ackerman angle being impacted. Since the steering axis of the 1g is not vertical a vertical change in the ball joint will change the ackerman. The point would move by the cosine of the steering axis angle, which will end up being very small.

Regarding the way roll center is calculated, what you have stated is not quite correct. The two inner pivot points create an axis of rotation. The roll center of the front axis will be found by connecting the ball joint with the shortest distance to this axis (right angle). Given the geometry of the 1g front control arm this point actually ends up slightly in front of the front pivot point. So moving the front pivot point (either up or down) would give a slightly larger magnitude movement of the roll center than if one were to move the ball joint by the same corresponding amount.

I stand by my statement that anti-lift and anti-dive are driver preference driven. There are certainly certain tracks that would greatly favor one setup over the other, but to a certain extent if a driver is or is not inclined to trail brake that will greatly determine their anti-dive preferences. There are a lot of ways of achieving the same "balance", so you can always "work around" some bad behavior.

Regarding a needle bearing being a bad idea for the rear suspension mounting point I will assume you intended to say radial and not axial. Given that, how do you propose it would see forces that are not radial? How would you package a spherical bearing into the factory mount? For clarification I am not talking about a needle thrust bearing, but a needle roller bearing. That being said using a needle thrust bearing on the front face to take some of the braking loads would further increase the robustness of the design so that the front suspension point doesn't have to carry all of the braking loads. As you can probably tell, I disagree that a spherical bearing would be a better choice and strongly disagree that a needle bearing would be a bad choice.

 

[RWD4G63]

I'll start off by saying that you are indeed correct about the Ackerman angle being impacted. Since the steering axis of the 1g is not vertical a vertical change in the ball joint will change the ackerman. The point would move by the cosine of the steering axis angle, which will end up being very small.

Regarding the way roll center is calculated, what you have stated is not quite correct. The two inner pivot points create an axis of rotation. The roll center of the front axis will be found by connecting the ball joint with the shortest distance to this axis (right angle). Given the geometry of the 1g front control arm this point actually ends up slightly in front of the front pivot point. So moving the front pivot point (either up or down) would give a slightly larger magnitude movement of the roll center than if one were to move the ball joint by the same corresponding amount.

I stand by my statement that anti-lift and anti-dive are driver preference driven. There are certainly certain tracks that would greatly favor one setup over the other, but to a certain extent if a driver is or is not inclined to trail brake that will greatly determine their anti-dive preferences. There are a lot of ways of achieving the same "balance", so you can always "work around" some bad behavior.

Regarding a needle bearing being a bad idea for the rear suspension mounting point I will assume you intended to say radial and not axial. Given that, how do you propose it would see forces that are not radial? How would you package a spherical bearing into the factory mount? For clarification I am not talking about a needle thrust bearing, but a needle roller bearing. That being said using a needle thrust bearing on the front face to take some of the braking loads would further increase the robustness of the design so that the front suspension point doesn't have to carry all of the braking loads. As you can probably tell, I disagree that a spherical bearing would be a better choice and strongly disagree that a needle bearing would be a bad choice.

Well thank you for educating me on that aspect of roll center calculation. Just proves that you always have something to learn eh? That actually makes more sense. I will still stand by my statement that bumpsteer should be measured while moving the pickup points to make sure things are not being made worse.

What I was trying to say was that needle bearings are great at handling radial load but not combined load (I'm going to edit my paragraph above for clarity). Braking, cornering, and acceleration all put load on the bearing in a combined direction. They are also terrible at dealing with dirt and corrosion. If you cut off the bushing "bar" and welded on a suitably sized rod, it could be secured through a spherical bearing (with the correct spacers of course) and then bolted into the factory bushing position with some modifications to the mounting bracket.

 

[pagosa dsm]

Is this aluminum on steel or steel on steel? Either way that's not going to work. That joint needs to rotate and under track usage it will definitely wear the shit out of whichever is the weaker metal, making the hole oblong and lead to all sorts of issues. You need a spherical bearing in there.

the steel bushing has been working fine for year with no wear. Ray Peters gave me the idea.

 

[Ludachris]

Just to revive this thread and get some clarification - to anyone who changed out the OEM ball joint for a longer one, was it the Howe K772 that was used? It looks like at least one or two people had done this, but not everyone who was participating in the discussion is still active on the site anymore (probably don't even own their DSM anymore). Hoping we can get someone to confirm which ball joints were used. Thanks!

 

[getto822000]

Just to revive this thread and get some clarification - to anyone who changed out the OEM ball joint for a longer one, was it the Howe K772 that was used? It looks like at least one or two people had done this, but not everyone who was participating in the discussion is still active on the site anymore (probably don't even own their DSM anymore). Hoping we can get someone to confirm which ball joints were used. Thanks!

Yep it was the K772. I did this to my car back in 2017. Here are the part numbers from when I ordered the bushings.

You must be logged in to view this image or video.

 

[cioc]

Yep it was the K772. I did this to my car back in 2017. Here are the part numbers from when I ordered the bushings.

You must be logged in to view this image or video.

What length did you order? I assume if you lowered the car an inch over stock you want the 1" extensions?

 

[pagosa dsm]

here is what I did to a LCA a number of years ago. still with no problems

You must be logged in to view this image or video.

 

[Ludachris]

here is what I did to a LCA a number of years ago. still with no problems

You must be logged in to view this image or video.

What length did you go on the stem? And how did that relate to the amount the car was lowered? Me and Rory were trying to figure that out.

 

[pagosa dsm]

What length did you go on the stem? And how did that relate to the amount the car was lowered? Me and Rory were trying to figure that out.

I installed the threaded ball joint from Speedway motors. try to have a slight downward slope from the inner bolts of the LCA. so, you will negative camber gain.

 

[pagosa dsm]

here is what I did to a LCA a number of years ago. still with no problems

You must be logged in to view this image or video.

You must be logged in to view this image or video.

 

[pagosa dsm]

here is what I did to a LCA a number of years ago. still with no problems

You must be logged in to view this image or video.

You must be logged in to view this image or video.

 

[cioc]

here is what I did to a LCA a number of years ago. still with no problems

You must be logged in to view this image or video.

PAGODA!!! I was hoping you would chime in. Speedway has a couple different lengths the longest being a full inch. Howe I believe has a 1.5". My hamster brain was trying to equate lowered amount to how much of an extended ball you need but somehow I don't think it's that simple. Looking at the pic it looks like you picked the full inch. Can you confirm?

 

[pagosa dsm]

You must be logged in to view this image or video.

PAGODA!!! I was hoping you would chime in. Speedway has a couple different lengths the longest being a full inch. Howe I believe has a 1.5". My hamster brain was trying to equate lowered amount to how much of an extended ball you need but somehow I don't think it's that simple. Looking at the pic it looks like you picked the full inch. Can you confirm?

I built these 5 years ago. my car may be lowered about 1/2 inch. If I remember I used the .5inch longer. If you lower your car too much you will have problem with the rear suspension. you are looking for 0 caster in the rear. as I said you want to have a slight downward slope to the LCA. that will maintain a negative camber gain.

You must be logged in to view this image or video.

 

[pagosa dsm]

1 more thing you must do if you are planning to use longer lower ball joints. must adjust your bump steer. if you notice Evos have extended ball joints with extended tie rod ends.

 

[cioc]

1 more thing you must do if you are planning to use longer lower ball joints. must adjust your bump steer. if you notice Evos have extended ball joints with extended tie rod ends.

Yes! Thanks. I ordered a kit from Bobby at FIBA.