Dual Ball Joint Strut Modification

[JMortensen]

On 6/25/2018 at 1:40 PM, jpndave said:

So hypothetically you put two new joints/your fabbed knuckle/arms in that suspension - spread to intersect and give the zero scrub. What will that do to the kingpin angle? Does the wheel/tire end up flopping back and forth because it's steep? Does having the two joints add to a steering load especially the further they are spread or just act like a single longer arm pivoting out at the "imaginary" intersection? Looking at that photo your car has a serious scrub radius. Is the steering super heavy at low speeds? Sorry if I'm missing some technical terms here just trying to grasp how all this relates together.

The way the wheel moves is the part that kinda blows my mind. It moves back and forth a little bit, and you would think that would be really bad for steering feel or make the handling wonky, but new Mustangs are really good autoxers, so they must have something right. Same with the single ball joint BMWs. Many (all?) of the newer ones have the ball joint offset forward from the bottom of the strut. Again, this is going to mean that the wheel isn't going it pivot around that lower ball joint, so that should cause some weirdness in the feel, but they're very successful autoxers.

I have power steering so I can't tell you how heavy it would be without. I know it was pretty hard to turn at low speeds with 10" slicks on it, so it's probably not good with 14s... 

If I was thinking this there would be two viable options. One, transplant a suitable suspension from another car or, two, get a suspension program that allows you to explore in a practical way what does what. Unless it can be done geometrically on paper a good program is the only real option for this DIY stuff. When designing the trailing arm lateral links rear suspension for my Z31 project a double wishbone (think about it) program was used, no way would any alternative method be considered.

There is a lot more to transplanting another strut system than it seems like. Angle of the struts and spindles, steer knuckle length, brakes, etc. If I were going to do that I'd rather figure out SLA. Probably same effort and if done correctly, better end result.

[tube80z]

On 6/29/2018 at 7:02 AM, jpndave said:

Any suggestions for the "good program" anyone - particularly one that would allow the double joint lower? I've played with the cardstock models that the suspension design references suggest and they help to get an idea of 2 dimensional dinamics but 3d is lacking and changes to pickup points and lengths are tedious. I'd rather not drop thousands on a single car design but would certainly pay a resonable amount.

 

I'd recommend Susprog3D.  It's nothing fancy but gives you all you need to solve the DIY pick-a-part suspension designs.  That's what I'm using to see if we can figure out a simple approach to reduce scrub on Jon's car.

 

Cary

[260DET]

The latest version of Susprog caters for the Holden (G8) front suspension? Looks like it.

[jpndave]

I ordered up the Susprog3d which allows 2 computers to be used so JD @69Honda and I can use it to coordinate our suspension. We're close enough on builds that combining efforts really makes sense. I'll definitely look into the double joint lower at least in the front. Not sure if we'll need it but it might buy that extra little bit of distance to zero out the scrub should that be a problem.

Edited July 4, 2018 by jpndave

[JMortensen]

I told Cary I would get some more measurements. The rack on mine is 5.5" in front of the LCA pivot. Tried to measure the height of the inner and outer pivots but car is on jackstands now and I didn't want to take it off just for that measurement. Stock steer knuckles and rack moved forward and up and tie rods are using .9125" bumpsteer shim. Bumpsteer is minimized.

Hopefully that gives you enough to go on. If not it will have to wait until next week most likely.

[74_5.0L_Z]

I think you would be better off creating an SLA front suspension rather than the double ball joint set-up.

 

Alternatively, have you considered putting narrower front wheels / tire on the car and using longer lower control arms to push them out to the width of your fenders.  You would of course have to relocate the top of the strut outboard by a distance similar to the added length of the lower control arms.  The factory LCA is ~11.3 long center to center.  You currently have 15x14 wheels with essentially a 4" backspace.  I would go to a 15 x 11 with a 5.5 inch backspace and make some LCAs about 16 inches center to center.

 

Which ever avenue you pursue, I am watching eagerly and would be willing to provide any help I am able.

[JMortensen]

That's an interesting idea Dan, but I'd have several problems:
1. The cage is built to support the stock strut tower, and I'd be moving it over quite a ways.

2. Car would be way wider than it needs to be
3. Don't really want staggered tires. Always had issues getting the front end to hook up, and that's going to be exacerbated with narrower front tires
4. Tie rods will interfere with the wheels. Possible that a 16 x 10 or better, a 16 x 12 might work, but one of the reasons I went with the 15x14 is the easy availability of the FA tires

5. Doesn't really fix the scrub radius issue. Obviously would be less with a narrower rim, but might still be 4" if I couldn't do the 5.5" backspace.

I like the dual ball joint thing because it could potentially be easy to implement, we'll see what the computers say.

If the dual ball joint doesn't work and the car doesn't hook up with my new tires then I think I'll go one of two ways; either SLA or go to a 3" flare and run a 15 x 11 with the 275/35/15 Hoosier or 16 x 12 if I can find an appropriate slick to fit.

 

[TUME]

I think you should consider diffrent spindles. I have in my car S13 front spindles and ofc. S13 front coils though.

 

With "only" 9" wheels ET35 i have nearly zero scrub.

 

[JMortensen]

That's a really interesting option, thanks. How would one go about figuring out which spindles to try? Assuming some Nissan stuff would have the same bolt pattern so might be able to bolt the existing suspension up to it, thinking would need new front brakes, or at least new caliper brackets. I know Nissan used the Set 2 and Set 6 bearings up through the 300ZX, so that part might be the same if I stayed Nissan.

[TUME]

Take a look into my build and you get an idea. Wheel bolt pattern is same as S30, mine has 5-lug with more options with wheels.  

 There are plenty of aftermarket spindles, so you can play almost everything with them. And ofc. control arms, brakes, etc.

[JMortensen]

I checked it out and it's easy to see why the scrub is reduced. This actually looks like a pretty easy solution. I'm guessing there is a 12.2" Wilwood setup for an S13 that's readily available too.

Still want to see what can be done with the dual ball joints, but thanks again, very helpful!

 

Edited July 8, 2018 by JMortensen

[caperix]

I looked at your build as I am looking into different front suspension options.  Is the added front track width from your custom control arms or is the s13 just wider? It looks like you reangled the steering arms on the knuckles, how is that holding up, I have always been nervous about welding cast pieces.  The only other write up of a s13 swap I found was on zilvia, he used a rhd skyline rack for due to swapping the knuckles to front steer, are you still on the s30 rack? How is bump steer with it. Sorry for the thread jack.

[jpndave]

3 hours ago, caperix said:

I looked at your build as I am looking into different front suspension options.  Is the added front track width from your custom control arms or is the s13 just wider? It looks like you reangled the steering arms on the knuckles, how is that holding up, I have always been nervous about welding cast pieces.  The only other write up of a s13 swap I found was on zilvia, he used a rhd skyline rack for due to swapping the knuckles to front steer, are you still on the s30 rack? How is bump steer with it. Sorry for the thread jack.

I think the added width is tires and wheel offset which brings with it a horrible scrub, hence the thread looking for solutions which the dual ball joint potentially offers. If it works, tie rod outer points and their attachment at the knuckles would need to be located correctly to allow proper Ackerman and zero or minimize bump steer (assuming things are correct on the inner points to start with). Caster shouldn't really be affected. Dynamic camber probably would at least a little but maybe not enough to cause issues. I'm intrigued by the possibilities. Advantages to SLA or strut would be similar.

[jpndave]

On 7/6/2018 at 7:23 AM, 74_5.0L_Z said:

I think you would be better off creating an SLA front suspension rather than the double ball joint set-up.

 

Alternatively, have you considered putting narrower front wheels / tire on the car and using longer lower control arms to push them out to the width of your fenders.  You would of course have to relocate the top of the strut outboard by a distance similar to the added length of the lower control arms.  The factory LCA is ~11.3 long center to center.  You currently have 15x14 wheels with essentially a 4" backspace.  I would go to a 15 x 11 with a 5.5 inch backspace and make some LCAs about 16 inches center to center.

 

Which ever avenue you pursue, I am watching eagerly and would be willing to provide any help I am able.

While I believe your are right in the "ideal" setup and intend to go that route on my car (it's a mess up there from multiple impacts and rust plus the engine swap so all the structure is getting replaced anyway.) For that to really work correctly wheels large enough to hold the whole outer end would be needed and a full inner structure with all new arms, rack, knuckles, coilovers, etc. Would be necessary. A build like this where the existing setup is already there and very sound - tearing the whole front end apart for a complete redesign is probably not worth the effort. At best the entire upper structure has to be redone most likely lowers and rack too.

 

Longer lower arms should help but bring all the same issues and more. This double joint idea is simply a long virtual arm. Advantage would be getting the effective long arm without messing with crazy brake/hub/knuckle offsets or maybe impossible parts. Potential disadvantages I see are extra leverage on the steering to move the compounded outer end and not sure what the thing would be like dynamically. Just thinking out loud here, hopefully the discussion helps anyone considering this myself included.

[jpndave]

On 7/7/2018 at 12:19 PM, JMortensen said:

I checked it out and it's easy to see why the scrub is reduced. This actually looks like a pretty easy solution. I'm guessing there is a 12.2" Wilwood setup for an S13 that's readily available too.

Still want to see what can be done with the dual ball joints, but thanks again, very helpful!

 

Very nicely done! I really like the sway bar implementation and will "borrow" that idea if I can make it happen. Clean quality work. I'll enjoy following your build TUME.

[JMortensen]

1 hour ago, jpndave said:

I think the added width is tires and wheel offset which brings with it a horrible scrub, hence the thread looking for solutions which the dual ball joint potentially offers. If it works, tie rod outer points and their attachment at the knuckles would need to be located correctly to allow proper Ackerman and zero or minimize bump steer (assuming things are correct on the inner points to start with). Caster shouldn't really be affected. Dynamic camber probably would at least a little but maybe not enough to cause issues. I'm intrigued by the possibilities. Advantages to SLA or strut would be similar.

My control arms are very slightly wider than stock, but yes, the main thing is the 15 x 14 wheels. I was surprised at the relative lack of difference that rack position had on Ackerman. When I set up the ps rack I had to move it forward. Prior to that I had a manual rack which I moved backwards as far as possible. Moving the stock rack back about 1.5" yielded another degree or two of angle change at full lock, so in the normal driving zone had almost no effect. Moving the new rack forward about 2" from stock basically turned it into a parallel steer car. 

As you say, the dynamic characteristics is the interesting variable. I don't know if there is a way to get this setup to have better Ackerman characteristics or not, and bumpsteer was pretty well managed before, so it will be interesting to see how that works out. I did have a friend with a 510 who went from stock knuckles to ones that were bent for Ackerman and the turn in on his autocross car was vastly improved. I'd like to have more Ackerman if possible, but I think it's really hard to do with the stock spindles.

Cary has been super busy but is thinking he'll have something modeled in a couple weeks.

Edited July 21, 2018 by JMortensen

[jpndave]

Ackerman itself is pretty simple to sort. The other thread had some good posts and "Chassis Engineering" has a great write up on it as well as the rest of the chassis. I have and have read a bunch of them and Herb Adam's is hands down the best for basics. Carroll Smith's series is next on the list for advanced stuff. All the rest are a few good points here and there all IMO as a disclaimer. I'll see if I can get a chance to post some of that basic information. I highly recommend Adam's book as a must have for basic understanding from this stuff to cage design and chassis stiffness. Not a how-to but all the sound basic theory to get you there. 

 

You'll have to watch as steering points get moved for ideal Ackerman to not upset bump steer. Up and down doesn't really matter on the Ackerman just getting it right in relation to the kingpin pivot line relative to the center of the hypothetical rear axle. So a line from outer steering joint through kingpin pivot ideally lands at the center of the rear differential (if it's centered) outboard is less that ideal past center is more. Slow speeds I think "perfect/ideal" is the goal. Other applications might mess with it to band-aid issues. I wiil design for center of rear or ideal which minimizes scrubbing any tire for turning a different radius as both front tires are now turning the same angle as vehicle travel. If you start to push, get loose or drift that is no longer true since the car isn't turning the same as steering. If the car is going to spend a lot of time with one of those conditions setting Ackerman to match conditions is possibly a better option. 

 

If I can time to get that program loaded up entering the points would be simple enough.

 

I'd probably start with the lower arm coordinates then move to the arm location to optimized Ackerman and steering.

 

Hopefully the program will show any dynamic changes. Steering effort issue will likely be try and see but with the power rack will probably be a non-issue for you. I'm not planning on a power rack so it could be a problem for me. 

Edited July 21, 2018 by jpndave

[JMortensen]

I've got Adam's book. I actually ended up with a roll cage that was inspired by the balsa testing he did. Fig 12-14 on p90 if you're interested. I thought Staniforth's Competition Car Suspension had more info on how much to use and why, but YMMV.

In any event, here is the thread where I showed what I measured when I moved the rack back and forward, used longer and shorter steer knuckles, etc: 

 

[jpndave]

17 hours ago, JMortensen said:

I'll have to go through Staniforth's book again with a fresh set of eyes. In his defense it was one of my last reads and after Adam's Basics and multiple Smith advanced books much seemed redundant.

 

I'll review the link thanks for posting it up. 

Edited July 22, 2018 by jpndave