Roll Center design theory

[Salt Racer]

...I could be in my new shop as early as next week...

I inquired some specs of HRP's version 2 knuckles with centerlock hubs. There's fair amount of offset from hub face to steering axis, like 8" at spindle centerline. KPI is 10 degrees. This obviously ain't gonna work with rim width we're most likely using on street cars, but with 12~13" wide rims they use on front of T/A cars, it's possible to get near zero scrub radius. I think I'll stick with C5 or S10 4x4 hubs for my Riv. I doubt I can fit anything wider than 10" rims up front, unless I want to live with turning diameter of a semi truck.




OK guys, just to clarify some things...

The primary reason why Dennis and I started his spindle revision was his very low ride height. LCAs had excessive angle, causing 0.5" side scrub at 2" bump! Spindles only had to be slightly taller than what he had in terms of FVSA length, but both UBJ and LBJ had to be lowered. Then the project snow-balled, just like any other projects of this nature.

We kept raising the UBJ in an effort to decrease RC lateral migration and bumpsteer (FVSA length was too long relative to tie rod length). My final design had 110" FVSA, which I felt was good compromise for many purposes, but I guess Dennis made it even shorter for further reduction in RC migration and bumpsteer.

I don't know how we are getting so much descrepancies in FVSA length. Unless Dennis screwed up the measurements, the numbers we came up are correct for what he has.

BTW Wally, 77" FVSA is normal for road race cars. But unless arms are really long, there'll be more side scrub than a similar suspension with longer FVSA. Thus my comment "depends on what your object is". You can drive it w/o any problem whatsoever, but it could be better for certain other applications. Whatever you do, I'd still recommend making the FVSA adjustable (multiple UCA chassis mount holes, slots with offset bushings, etc).

[Blown353]

Hey Dennis, what's your front tire OD and the distance from the bottom of your x-member to the ground? I'm trying to compare ride height / suspension working angle between our two cars, with those two numbers I should be able to get a good grasp comparing how our two cars sit up front. If things are similar I would be inclined to give a set of your spindles a whirl. Right now I'm at 26.1" tire OD and my x-member is 5" off the pavement. I should probably take a measurement to get a difference from the inner LCA pivot point to lower balljoint, that would be a better comparison.

What UCA's are you going to run? Are you going to keep your Pole Positions? I'm wondering if I would be able to get away with my current GW's with the different spindles. My guess is with the revised KPI and the limited adjustability provided by the factory "shim-n-pray" method I probably can't shim them nowhere near enough to get things to work.

I think I've got a lead on a good Elky frame, if it pans out I'm going to pick it up and start on a rear 3-link...

Troy

[wally8]

Denny,

#5 almost made me spit water through my nose :-)


1. I agree with that. That's a real problem.

2. It is a lot of fun. We cure all those problems though. The RCH is actually 4" since we relocate the UCA pivots. The steering ratio is moot since we use a quick box (2:1 quickener) and bumpsteer is easily fixed by lowering the steering box and idler arm mounting points and shimming the outer tie rod with a heim stud adapter.

Salt Racer,

I'm still missing how raising the UCA BJ pivot point by virtue of a taller spindle while maintaining the stock UCA inner pivot would lengthen the FVSA. Again, this could be because I'm completely missing something and it's getting late but from a geometric standpoint it doesn't make sense to me. What sort of angle are you talking about for the LCA? Maybe that's the issue. Mine is level at normal ride height.

Talk to me about side scrub as well if you would. I'm drawing a blank on what that is. Maybe it's a difference in terms between the two worlds.

Oh yeah, one more thing: You said FVSA was too long compared to tie rod length. What's the problem with that? You just have to find the right location for the tie rods so that they follow the same arc as FVSA, right? This is something that I'm recalling from a book so that could be wrong too. It's not clear to me right now. Good thing to add to the discussion though.

Wally

[dennis68]

Katz, I'll have to recheck some things, I may have transposed some numbers wrong.


Troy, Right now front tire OD is 25.75 but will be 25.00 in the end. Right now crossmember height is 4.5" but will wind up at about 2" all said and done. The new spindles are basically 3" drop spindles plus whatever spring drop I dial in.

I don't run Pole Position arms, they are Speedway and yes I will stick with them. If they neded be shorter or longer that is easy enough to order new arms.

[Blown353]

2" crossmember height?!?!? Egads, man! Youse crazy!

I'd love to slam my car that low but that's too low to be safe and practical out here in my opinion. Potholes and road irregularities would eat me alive! That, and I'd get tired of buying new headers every 3 days as I scrape them to oblivion getting into and out of my own driveway.

Guess I need to get to work on my own spindles as our parameters are quite a bit different. I definately want to improve on my current B-body spindles but I'm going to have to make some geometry changes and some compromises to keep the car road "friendly" at least by my definition (aka, higher ride height.) Custom spindles for less than $500, as you said, is a very worthwhile deal.

Have you thought about a custom, longer LCA so you can run a wheel with more backspacing for a better scrub radius? Or have you managed to find a solution that works well with stock length LCAs?

Sorry guys for semi "thread-jacking" and introducing another subject. If it gets too disruptive, I'll start another thread.

Troy

[dennis68]

The spindles Katz and I have designed will work fine for you, my ride height is going to be a result of spring drop.

I was able to acheive 1 3/4" of scrub radius as is so I am not going to invest in building new lower control arms altough it would be nice to get under an inch of scrub by simply adding an inch to the LCA's.

Hmmmm good job Troy, now I'm thinking again.

[Blown353]

LOL, sorry Denny. Guess I need to stop thinking, it appears to be contagious. :icon_razz

1.75" scrub radius isn't too shabby at all given you're working within the limits of the stock LCA. I personally think it will handle quite well. But you're right, there is the possibility for improvement with longer LCA's. Is it worth the extra cost and effort? That's up to you. My gut is saying you'll be fine with the 1.75", but sometimes it's worth the extra effort (and cash outlay) just to get rid of that gremlin in the back of your head which constantly nags "I know it works as-is, but don't you wish you would have done it that other way?" I HATE THAT LITTLE GUY!

I think I could live with the extra 1.5" drop your spindles would give me over the current 1.5" drop in the B-body spindles. If not, I could just stuff some slightly longer springs in there to raise things back up. Or use this as a good excuse to switch to coilovers up front. LOL.

I was going to ask what lower balljoint taper you were going to have them made for (standard or B-body lower taper), but I suppose if I ordered a set I could get them with whatever taper I wanted.

Troy

[dennis68]

The ball joint taper is Chrysler screw-in design. I'm not sure what taper it is. Anybody know what the taper is of the Chrysler screw-in design?

[zbugger]

The ball joint taper is Chrysler screw-in design. I'm not sure what taper it is. Anybody know what the taper is of the Chrysler screw-in design?

It's what ever chrysler made it to be.


Ok, that's the last outta me.....

[dennis68]

[yody]

underdog!

[Salt Racer]

Katz, I'll have to recheck some things, I may have transposed some numbers wrong...

Den,
I had you triple-check the numbers so I don't think there's an error, but it's still a good idea to re-check everything before you order the spindles. One-off custom parts are too expensive to make mistakes on.


Wally,
Dennis initially had 858" FVSA length (yes, eight-hundred fifty eight). IC was still towards the chassis side. LCA inner pivots were 7.5" off the ground, now it's 7.0". LCAs were pointing up towards LBJs by 13 degrees. So I dropped LBJ down to 7.0" so LCAs are level, and adjusted UBJ height for 110" FVSA.

There is something wrong here. Maybe (big maybe) there is some differences in chassis pickup points between '64-'67 and '68-'72 A-bodies. Could you do us a favor and post your height measurements? Dennis had LCA inner pivot height of 7.5" (front & rear), and 17.75"/16.50" UCA pivots, front/rear respectively.

Side scrub is track width change at the contact patch when suspension moves up and down. Maybe it isn't a big deal on dirt track, but you definetely want to minimize it on pavement.

Last but not least... Incorrect tie rod height is not the only thing that can cause bumpsteer. Let's say you got zero bumpsteer on suspension. Imagine what would happen if you swap in a wider drag link and 5" center-to-center tie rods. This causes toe-in bumpsteer (on front steer) on both bump and rebound even though the tie rod height is correct. This is what I was talking about. It is rarely talked about b/c most OE suspension have reasonably close-to-ideal tie rod length, and it's not quite as sensitive as the tie rod height. For example, my car currently has 1/16" toe-out bumpsteer at 2.25" bump and rebound. This is an indication of correct tie rod height but incorrect tie rod length. It's rear steer, so tie rods are too short. To get rid of this last 1/16" (actually 1/32" in GM's language), I need to extend each tie rod by 2"!! The car has good Ackermann, and tie rods are right against the pan at full lock, so I'll live with 1/32" bumpsteer for a while.

[wendell]

HA. Some days I love this web site!

Dennis,
Performance Trends it is.

Dennis, Katz et.al,
I'm presently doing 2D sketches in Solidworks to get my design started. My figuring is that eventually I can flesh out the structural design in 3D. Is this a good idea? I'm mostly conserned with being able to share information effiecently. Thanks for your time. I'm enjoying this conversation.
J

[dennis68]

Katz, I think in the last evaluation I added the tall ball joints and lowered the chassis height an additional inch just for kicks. It improved RC migration and camber curve but at the expense of FVSA. It shows how little change change to x,y,z numbers results in substantial change of different angles. IE, just an inch of chassis drop can totally change the handling characteristics.

[wally8]

Salt Racer,

Looks like we have the same type of measurements. Currently my LCA inner pivot is 7 3/8" and my LCA BJ is 8" with stock spindles and the tire size that I anticipate. I'm using different figures though since I'm anticipating a higher pin height on the B- body spindle (+1in I think) and I'm going to use a taller LBJ as well. When it's all said and done I should be level on my LCA at about 7 3/8 or 7 1/2.

My UCA inner pivots are very similar to Denny's as well. Slightly lower just like the LCA.

So in my case, using the numbers I have which could be off due to not having the exact spindle, I have to move the UCA inner pivot up to get a longer FVSA. Raising the spindle more and more will shorten it. This would be the general trend for any setup given this case, no?

You're right that the A bodies start out with an atrocious number.

Thanks for the side scrub explanation. I remembered what it was as I was going to sleep last night. And you're right, it's not that big of deal on dirt with this particular suspension combo. The left front is rarely on the ground and the overall load transfer is very radical already.

In fact, I once saw one of the nationally ranked mod drivers lose his left front tire due to loose lugnuts. He completed the entire heat (10 laps on a 1/2 mile) and kicked everyone's butt doing it. Yeah, he's good. He had some dirt packed in his LCA though :-)

Very good point on the bumpsteer also. I take that for granted because we all "know" which drag link to use and where to relocate the mounting points. That's the advantage of working within the parameters of a highly developed combo versus starting from scratch I guess. Another important note is that we don't get to 0 bumpsteer, just tolerable limits.


Wally

[Mean 69]

In fact, I once saw one of the nationally ranked mod drivers lose his left front tire due to loose lugnuts. He completed the entire heat (10 laps on a 1/2 mile) and kicked everyone's butt doing it. Yeah, he's good. He had some dirt packed in his LCA though :-)

Oh, that's cool! I saw a guy run a practice session in AI with no brakes once!!! How's that for needing new underwear?!

Den, you slay me. Great humor!

I think there was a question of KAI for a typical AI car previously. The overwhelming majority of AI/X cars are late model Mustangs, which use mac strut front suspension. As such, it really isn't too relevent to compare. AI/X is a pain from the rules standpoint, as I have gone off on before. Basically they are forced into making consessions for the whining Mustang crowd that makes up the majority of the field. Everytime they suggest that they may open up the rules and allow thing like aftermarket subframes, etc, a bunch of guy pee and moan and they retract the idea. It is really frustrating. They may come around eventually, but you'd be nuts to build an AI specific car, in that the rules will probably bite you sooner or later. Eh, Sean?

Mark

[CoryM]

Good thread guys.
Spindle reamers for chyrsler screw in balljoints are 1.5inch/ft and GM are 2inch/ft so the actual taper will be close to that.
When you guys talk about RC migration how much do you personally like to see on a full body car?

[dennis68]

OK, the Chrysler uppers share the same taper and size as the GM truck, which is what most of the round-round arms run. The lower is off by itself, I think Howe will build a precision ball joint with a GM body and Chrysler stud reasonable enough.

The less RC migration the better, no compromise here. RC migration evolves to unpredictable handling very quickily. I don't know that there is a "number" that could be assigned as a reasonable amount. Each chassis is going to respond differently to change.

[Mean 69]

Good point on the migration issue, hard to put a number on it, but I'd say anything more than a few inches is something that I'd tend to look a bit deeper into (how many is a few?? I dunno, four, five?). As importantly, and it would be hard to imagine if the absolute numbers were pretty small to begin with, is the rate of change. If it goes ballistic at some specific point, I'd be alarmed.

Again, to reinforce, the migration is probably more important for folks that are doing higher speed stuff, such as open track/road racing. I just feel very strongly that you don't want any funny business going on when you are really moving out. Imagine being loaded laterally in a fast turn, and having to chase the car due to erratic behavior over small bumps/reises, etc, it would seem to me to be a bit unsettling.

Great thread folks, nice to get the brain working once in a while!

Mark

[David Pozzi]

chrysler and GM up to 1969 used 7 deg tapers, GM 1970 up with cast knuckle, used 10 deg tapers. Nova and it's cousins with forged knuckle and bolt on caliper brackets, used the 7 deg thru 1974.