View Full Version : minimizing scrub, high KPI or lots of BS?

05-28-2006, 06:53 PM
Hi guys,

I'm getting ready to tear back into my Datsun 240-Z project. I'll soon be buying a suspension design program to help me figure things out, and then I'll begin the front snout of the spaceframe.

Been reading over material on scrub radius and have a question. I searched the archives and didn't find anything about this.

I may be missing something obvious or my suspension theory may not be 100%, so let me make sure I know what I'm talking about first off... Scrub is the distance at the ground between the wheel centerline and the intersect point of the imaginary line running between the two ball joints yes?

So, most people solve the problem by using a steeper KPI (Herb Adams' book recommends 7-9*) to make the imaginary ball joint line more angular... thereby making the intersect closer to the wheel centerline.

That is fine, but what I don't see people doing is using a spindle with 0* KPI and just moving the steering knuckle further into the wheel. With today's larger diameter (and wider) wheels, it seems like this would accomplish the same thing. If you have a short BJ to mounting surface length, (say 4'') you could use a 10'' rim and have the ball joint centerline directly in line with the wheel centerline.

Maybe this just isn't practical, but is there any other reason not to do it this way (aside from space considerations inside the wheel)?

Thanks for any comments. At present, I'm planning to probably go with C5 front knuckles and will build the control arms around that.

05-28-2006, 07:24 PM
Running zero KPI would be fine if it were possible to space the wheel such that you could obtain near zero SR. It is very unlikely on a typical street car to obtain near zero SR with no KPI. This would involve running wheels with as much as 7" back space on an 8" wheel.

05-28-2006, 09:24 PM
wouldn't zero scrub radius provide very little steering feedback or feel?

05-29-2006, 07:43 AM
Usually when you run anything less than 1/2" of SR you eliminate the P/S. At true zero the steering would have no feel and you could one finger the wheel without assist. I would not recommend less then 1/2" of SR.

05-29-2006, 03:48 PM
Thanks guys! I'm sure I'll have more questions as I go.

Speaking of which... I've downloaded the demo of Suspension Analyzer from Performance Trends, and have started modeling some potential designs.

So, here's another question if anyone is reading... how much roll center migration is acceptable? My initial modeling is just trying to get my RCH in the ballpark. Again, Herb Adams' book recommends static height between 2'' above ground and 1'' below. If I start within this range, how far out of it should I safely go between roughly 3.5'' of bump and jounce.

I'm playing with a level lower control arm that is fairly long and a slightly shorter upper arm that is angled downward at the center ever so slightly (giving me an intersection point between the two at between 150'' and 200'' away). RCH still migrates quite a bit. Any suggestions for a beginner?

05-29-2006, 06:53 PM
It will take months of trial and error to get RC migration nailed down and still fit all the suspension under the car. My suspension took about a year to figure out on paper and RC migration both lateral and vertical is less than 1/2" at max roll/dive. Good luck, shot for less than an inch.

David Pozzi
05-29-2006, 07:56 PM
Start by figuring where the tires will sit in the car (width), then figure out wheel offsets and suspension from that.
KPI is needed to provide self-centering or steering return. 0 KPI would stay turned when you turn the wheel.

I'd look at a roll center height near 3" front. Ideally the roll center height would move fairly closely with the chassis. Good camber curve, and minimal side scrub, and scrub radius of .5 to 1.5"

05-30-2006, 09:57 AM
Wow Dennis! How did you manage to get the RC migration so minimal? I'm not worried so much about fitting it under the car. The Datsun project has lots of space to play with (it is about 74'' from fender lip to fender lip) and I'm building the frame from scratch to accomodate the suspension.

David, thanks for the confirmation. That is how I've been trying to do it. I really need to get my hands on a spindle though so I can really start getting some definite starting coordinates.

Any other tips for design are greatly appreciated. I hope to start building by August or so. I'll be poking through the archives some more!

05-30-2006, 04:07 PM
My suspension took about a year to figure out on paper

05-30-2006, 04:25 PM
Gotcha ;)

I meant to ask if there were any tips you have with respect to how to go about getting RC migration so tight so that I might not have to go through quite so much trial and error.

05-30-2006, 05:04 PM
No, trial and error is the only way. You have to consider so many things; don't concentrate only on eliminating RC migration. RC height, SVSA, control arm individual lengths and spindle height all have to be considered. In my application changing the LCA length is not really an option at this point so everything was built around that. When you get close you will find that ride height has a substantial effect on RC movement during dynamic conditions.

05-30-2006, 06:26 PM
Thanks Dennis. I've been playing with the program more tonight to see what I can do. Sure is tricky to get two different parameters to fall into appropriate ranges simultaneously!

Man, this is too much fun... I'm supposed to be studying for the bar exam... = )

05-30-2006, 08:04 PM
Also remember that even though all the numbers on paper are perfect; it may handle like crap. Sometimes things that shouldn't work, do and things that should work perfect, do not.

06-06-2006, 04:49 PM
Well, here's a brief update.

I got my C5 spindles, around which I plan to build the suspension. Actually, I bought a whole C5 front suspension clip because I got a good deal on it... We'll see what parts I end up taking to put together this frankenstein suspension. I'll sell the others and make back a few bucks.

I'm still plugging away with the Suspension Analyzer, trying to make things work well. I've managed some improvement. Will keep you all posted. As I further update, I'll probably put any pictures and such in the project updates section.