Grassroots Motorsports

Register Log in to GRM
- Community
- You have 11 items in your cart.
Birthdays
- carlosmessi
- mickmack22
Recently on the Classic Motorsports forums

Subscribe get a full year for only $19.95 Sample get a free issue—no strings attached

Login to post Forums » Grassroots Motorsports » McPherson roll center/SAI issue

1 2

Jan. 5, 2009 9:34 p.m. GameboyRMH SuperDork

Well I've finally saved up enough to get those coilovers, and now I'm taking care of all the "prerequesite" parts. I finally found a factory front strut brace after 4 years of searching, so now it's just the rear strut brace and roll center spacers.

http://www.d2racing.com/content/products/RS_coilovers

You're not supposed to lower a car past the point where the lower arms are level to prevent getting a positive camber curve on compression. But the thing is the arms are level from the factory on this car. In fact, if you draw a line from the lower arm mounts to the pivot point on the ball joint, it's higher at the wheels, and I still have a negative camber curve. I've heard they still handle better when lowered as-is but I'd rather play it safe and get spacers anyways. I'm looking at reducing the ride height 1"~1.5" (6" to 5.5" ground clearance) and maintaining the stock rake angle. I think I'll go for roughly 20% stiffer than stock all around to start out with. I intend to increase the stiffness of the suspension until the car becomes too skittish and unforgiving or totally unstreetable, whichever comes first.

But anyhow the puzzling issue is the rear suspension. I can't make roll center spacers because it uses a pair of tubular bars on each side for lower arms. The only way to get the same affect would be custom uprights which equals a major PITA. The lower arms are normally angled downwards so lowering by 1"~1.5" would make them level (and I suspect would -decrease- (angle inwards) the rear camber when level..if that turns out to be an issue I'll solve it with the camber plates).

So this issue is that if I use roll center spacers on the front and not the rear, I'm tilting the roll axis inclination further to the front which means more oversteer...I'm happy with the handling balance now and I intend to keep it pretty much the same. Custom rear uprights are the surefire solution but mean a lot more time and money. Running no roll center spacers could really screw things up. What do you guys think?

edit: D'oh I don't know why I put SAI in the title when I was thinking RAI...
Jan. 5, 2009 9:50 p.m. AngryCorvair Dork

direct quote from a friend who spent a couple years designing suspension in F1 at the turn of this century (LOL, that sounds like a long time ago, doesn't it?):

berkeley roll centers. spring and bar the car properly, and take care of bump steer.

and that's not just his personal opinion, but a pretty common thread throughout the paddock.
Jan. 5, 2009 10:03 p.m. GameboyRMH SuperDork

That's another problem...right now the tie rods descent a bit towards the wheels, so the front wheels toe in under compression. at -1~1.5" the tie rods will be close to level which means bump-out in either direction.

Hmm you know rod end spacers could fix that
Jan. 6, 2009 1:21 a.m. RedS13Coupe New Reader

AngryCorvair wrote:

direct quote from a friend who spent a couple years designing suspension in F1 at the turn of this century (LOL, that sounds like a long time ago, doesn't it?):

berkeley roll centers. spring and bar the car properly, and take care of bump steer.

and that's not just his personal opinion, but a pretty common thread throughout the paddock.

While I am not the one to tell anyone how to set up a car... Things are a bit different when you got a car with THAT much aero. F1 cars have tiny fractions of an inch of suspension travel and sprung so stiffly that their tires make up the majority of compliance... Suspension geometry can sacrifice some mechanical grip to get the thing down low enough to suck it to the track.
Jan. 6, 2009 6:30 a.m. 44Dwarf Reader

Get "front end pro" from AUTOWARE. Take all the data and plug it in it will tell you bump steer, roll center, caster gain, camber gain, for and aft movement. and once you have the data in you can click on it and change it in the softwear and see what it did before you do it in real time.

44
Jan. 6, 2009 6:32 a.m. GameboyRMH SuperDork

44Dwarf wrote:

Get "front end pro" from AUTOWARE. Take all the data and plug it in it will tell you bump steer, roll center, caster gain, camber gain, for and aft movement. and once you have the data in you can click on it and change it in the softwear and see what it did before you do it in real time.

44

I live in Barbados, we don't have anything like that. I have a rock, a stick and some coconuts

(Okay seriously it's not quite that bad. I can get the car laser aligned and corner balanced, and if I'm super duper wealthy, dyno'ed, but that's about it.)
Jan. 6, 2009 6:42 a.m. petegossett Dork

When you say the lower arms are already level, do you mean the arms themselves, or the line between the balljoint and the lower control arm pivot? That imaginary line is all you really need to be concerned with, and while it's true your camber curve will go to hell once that line goes past the point of being parallel, if you're only needing a small amount of suspension travel, you could probably live with the minimal camber gain you'll have.
Jan. 6, 2009 6:54 a.m. GameboyRMH SuperDork

The arms themselves are level, the "imaginary line" rises from the frame to the wheels. If you take a look at the pic of the car in a hard corner in the ride profile, the wheel camber is about the same as when level, or maybe a bit more (on both the front and rear).
Jan. 6, 2009 8:38 a.m. Tom Heath Production Editor

Here's what we used with the Berzerkeley; you can even download a copy online. Not quite as cheap as we'd all like it to be, but accurate and simple. Worth 80 bucks if you feel like you need the information.

http://www.performancetrends.com/rc.htm
Jan. 6, 2009 8:54 a.m. GameboyRMH SuperDork

Hmm I might buy that app...the price is a bit steep but I'd pay it for something that does all that.
Jan. 6, 2009 9:18 a.m. Tyler H Dork

I think a little toe-out in the front won't be the end of the world. I've been to (and driven in) Barbados. I rented a Daihatsu Cuore and drove all over the place. I didn't encounter too many freeway-like stretches of road like we have here.

I think the ability to negotiate traffic in a roundabout would be what I would optimize the suspension for.

My experience with coilovers has been that they require almost constant tweaking. Take your base guess at a baseline and start driving. Then tune them where they feel best to you. On an MR2, the front is about 1" higher than the back. If you adjust it level, then the front roll center is messed up and the car understeers worse. I didn't give the roll center much thought until I experienced this first hand.

The dampers will soften up a lot before they settle in, so you'll probably be tweaking the stiffness a lot at first.

I had a lot of fun driving in Barbados, and saw a lot of cool Japanese imports we don't get here..
Jan. 6, 2009 9:30 a.m. 44Dwarf Reader

It's a computer program. If you have a micosoft computer it will work you take measurements your self its easy to fallow the pictures.

http://www.auto-ware.com/software/asgp/asgp.htm

It does stut type too so not to worry
Jan. 6, 2009 10:17 a.m. carguy123 Dork

44Dwarf wrote:

It's a computer program. If you have a micosoft computer it will work you take measurements your self its easy to fallow the pictures.

http://www.auto-ware.com/software/asgp/asgp.htm

It does stut type too so not to worry

It looks as if it only works on Windows 95 or 98.
Jan. 6, 2009 10:35 a.m. RandyS New Reader

I relocated the attachment points for the lower control arm 1" higher. The rear of the car became much more stable in autocross situations after that.

Some of it probably had to do with the rod-end attachments eliminating the "dynamic steer" of the rubber bushings but I believe most of the improvement came from the more aggressive camber curve. Because of the curve I was also able to run much less static neg camber which improved braking and lessened snap spin events from a better inital turn-in set

http://www.solomiata.com/images/rod-end-rear-control-arms.jpg
Jan. 6, 2009 11:43 a.m. GameboyRMH SuperDork

carguy123 wrote:

44Dwarf wrote:

It's a computer program. If you have a micosoft computer it will work you take measurements your self its easy to fallow the pictures.

http://www.auto-ware.com/software/asgp/asgp.htm

It does stut type too so not to worry

It looks as if it only works on Windows 95 or 98.

No problem, I have a retro box running Windows 98SE.

An attachment point relocation like that wouldn't be possible on the AE92 (at least definitely not on the rear, I'd have to take a look at the front but I don't think there's enough room).
Jan. 6, 2009 4:34 p.m. ae86andkp61 New Reader

Tyler H wrote: My experience with coilovers has been that they require almost constant tweaking. Take your base guess at a baseline and start driving. Then tune them where they feel best to you.

I think it is a really good idea to explore theory and get a good idea of the potential variables and issues you will be dealing with, and it seems like you have a handle on this. As someone who can get so wrapped up in this process that I spend months or years working on the theoretical "best" setup, I would have to say that my one regret is all the times I stood on the sidelines watching my buddies race or drive while my project was still up on blocks awaiting the magical combination to come together.

So don't stop reading, thinking, learning, asking questions, and dreaming up new setups, but at the same time, don't get hung up on the stuff that will be really hard to work out or work around. Do what you can with relative ease, skip the tough stuff for now, keep the potential shortfalls in mind (like rear roll center,) try it, and do further tuning or mods from there.
Jan. 6, 2009 6:23 p.m. 44Dwarf Reader

it works on all windows. I've had it on my XP desk top and on my short lived vista lap top...$700 POS.....
Feb. 17, 2009 11:47 p.m. Alex_W

Tyler H wrote: My experience with coilovers has been that they require almost constant tweaking. Take your base guess at a baseline and start driving. Then tune them where they feel best to you. On an MR2, the front is about 1" higher than the back. If you adjust it level, then the front roll center is messed up and the car understeers worse. I didn't give the roll center much thought until I experienced this first hand.

I also have an MR2, a 91. It is primarily a street / autocross car, and I have have done a lot of suspension analysis on it using optimum K.

The car is currently lowered 1", which based on my analysis causes some (more than I would like, but not horrible) rear roll center migration, but MASSIVE front migration, and in the opposite end of the rear. I have been looking into the parts necessary to raise the front roll center, which would significantly reduce the movement (and more or less confine it to the same side of the car as the rear movement), but would raise the front roll center above the level of the rear one.

The research I have done indicates that a roll axis higher at the front than the rear is quite rare (although not unheard of). I think the effects of my raised front roll center would be very similar to your 1" higher front ride height. I'm really curious to get input as to how this might effect handling.
Feb. 18, 2009 5:50 a.m. Tyler H Dork

I'm no suspension expert, but here is my seat-of-the-pants experience. I initially set the coilover ride height (via shortening the length of the whole assembly) to put the car at about 1.5" lower than stock, and at an even ride height. I assumed that the OEM rear springs were sagging with age and the added weight of the Gen3 swap, hence the rear was lower than the front and I 'corrected' it.

The car handled well under most situations, but seemed to have less mechanical grip in the front at the limit, no matter how I set it up.

I was complaining about this with one of my autox buddies who has a lot of MR2 seat time and he explained the roll center thing with these cars and told me to jack the front of the car up. I used the spring perch adjustiment to raise the car ~.75~1" in the front and noticed an immediate improvement. It would grip better in sweepers and I could get on the gas sooner coming out. Once on the gas, the car also kept a tighter line. (BTW, this car had about 300whp.)
Feb. 18, 2009 6:49 a.m. Jensenman UltimaDork

When you have millimeters of suspension travel, roll center migration is no biggie. When you have 6 or so inches of travel, yes it can become a problem. With a strut suspension, the roll center is pretty much controlled by the lower control arm angle. It's pretty easy to plot since you don't have the added complexity of the upper control arm angle.
Feb. 18, 2009 9:35 a.m. GameboyRMH SuperDork

Just to update, my plan is to keep it at stock ride height until I can get roll center spacers for the front and custom uprights for the rear.
Feb. 18, 2009 9:36 a.m. Alex_W New Reader

Tyler H wrote:

I'm no suspension expert, but here is my seat-of-the-pants experience. I initially set the coilover ride height (via shortening the length of the whole assembly) to put the car at about 1.5" lower than stock, and at an even ride height. I assumed that the OEM rear springs were sagging with age and the added weight of the Gen3 swap, hence the rear was lower than the front and I 'corrected' it.

The car handled well under most situations, but seemed to have less mechanical grip in the front at the limit, no matter how I set it up.

I was complaining about this with one of my autox buddies who has a lot of MR2 seat time and he explained the roll center thing with these cars and told me to jack the front of the car up. I used the spring perch adjustiment to raise the car ~.75~1" in the front and noticed an immediate improvement. It would grip better in sweepers and I could get on the gas sooner coming out. Once on the gas, the car also kept a tighter line. (BTW, this car had about 300whp.)

Thanks for the input. Seat of the pants feel is what I'm looking for at this point, I can read theory and plot graphs till I'm blue in the face and still not know how it will really effect things. I have definetly found myself wanting more front grip for corner entry, although it can get a little loose on exit sometimes. Like you I am at about the 300RWHP point, which can have a big effect on rear grip ;) I also have a relativly small tire stagger (225 - 245), so that may be part of it. I'm planning to increase that here soon.

My biggest concern is that I will "fix" the low speed / autocross handling, but make it too twitchy to be safe on the streets and windy back roads. However the way you describe it this sounds like it improves everything:) Better front grip for turn in, but also better rear grip for putting power down later in the corner.
Feb. 18, 2009 9:55 a.m. Jensenman UltimaDork

Raising one end or the other will also change weight distribution.

The thing I find hardest about fixing something goofy in the handling: many times it can happen so quickly that you aren't 100% sure what really occured. It took me several incidents with my 1st gen RX7 to figure out exactly what brought on the oversteer; once I figured out it was mostly a large transition thing rather than small transition or steady state the number of possible causes and their possible fixes got MUCH smaller.
Feb. 18, 2009 9:56 a.m. Wally UltraDork

Do your rear arms mount directly to the floor or is there a sub frame they bolt to and the subframe bolts to the floor? Perhaps you could cut an inch out of the subframe to move the mounts up.
Feb. 18, 2009 10:02 a.m. GameboyRMH SuperDork

There's a subframe but there's no space to move the mounting points up.

1 2

You'll need to log in to post.