L8 APEKS

Hey gang,

I can't afford to revamp the suspension at the moment, but I'm wondering how one might go about lowering the car to get the COG down, *without upsetting the spring rates.*

Is this possible with a very light front spring, and just adjusting the torsions? If so, does anyone make a lowered spring that stays close to the OE spring rate? Thanks!

Dmitry S.

What balance? The car understeers from the factory. You can adjust the rear torsions down, but you'll have to get new springs in the front to lower it. If you get new springs in the front, the car will tend to understeer more.

Danno

"I can't afford to revamp the suspension at the moment, but I'm wondering how one might go about lowering the car to get the COG down, *without upsetting the spring rates.*"

Let's say you lower the car so that you've got 3" of suspension travel now rather than 4" before. How will you make up for that loss of suspension travel? You'd want to have the car bottom on 3" of travel with the same bump force as before that used to take up 4" of travel...

Also the length of the moment-arm that determines the amount of body-roll is between the COG and the instantaneous roll-center. The body rolls/pivots on the roll-center and the amount of lean is based upon the lateral cornering-force pushing on the COG. Contrary to popular myth, the car does not pivot on an imaginary point on the ground between the front-tyres.

On a MacPherson strut front-suspension, the roll-center is about 4-6" off the ground. If you lower the car by 1", it actually lowers the roll-center by 2-3". This increases length of the moment-arm between the COG and the roll-center. What this means is that for the same lateral-cornering force as before, a 1" lowered car wth stock suspension will actually roll MORE than before... This makes the outside tyres gain more positive camber and overwhelm its outer edge...

.... there's a picture I posted on this a couple years ago... It had a picture of Dave's car in an autocross. It was in response to someone saying something to the effect of stickier tyres would move the COG laterally so much that it would cause the car to lift its inside tyre and possible roll over... Pretty funny picture I can up with to illustrate this...

MTM

You better believe it!


Danno, do you still have the picture you described? I'd love to see it!

Here's another good one (in this case I think it's just a high center of gravity, not really sticky tires):


And these two are taken on the same corner of Thunderhill. Apparently it's like a sharp turn right after a hump or something:



It doesn't necessarily have to be sideways force:



And this one is not really related. I just think it's pretty neat.

Campeck

mmmm. rally.

Danno

Matt, those are awesome photos! All of those pictures shows cars that, except for the autocross MB, got air under their tyres due to a vertical upward force, not cornering forces leveraging the COG over the outside tyres. The 968, 951 and NSX hit an inside berm at a good enough speed that the ramp-up rate pushed the inside of the car upwards. The body is still moving vertically after the berm drops down to normal road-surfaces, thus the air (you can do this in a straight line too). This is a common technique to preserve speed through a corner by running over the berm to keep the line as straight as possible, although in Rick White's case with the 951, it was a little extreme.. This corner is a sharp left-hander on top of Magic Mountain and you'll usually sacrifice that corner a little bit to set up for the fast sweeping right-hander down the hill. Here's some more photos I took from that day at Thunderhill:


This EVO somehow managed to roll on a straightaway:

Supposedly, there's an on-board video clip showing this roll. Apparently the 1st roll has a tonnes of crashing noises and yelling & screaming. Then it's airborne and everything's quiet... Then crashing and yelling & screaming again... Then silence as it catches air.. etc... If anyone has this footage, please IM me with the URL...

Some more rolled cars...





The cause of these rolled cars, including the MB you posted, wasn't due to COG height, but rather the phenomenon of "catching an edge". It involves a left-right transition such that the body of the car is turned 90-degrees to the directon of travel and momentum. This puts all of the momentum and weight transfered to the forward tyres, which aren't turning anymore since they're traveling at 90-degrees to rotation. ALL of the weight of the car ends up loading up the foward tyres, they grip even more and the car rolls.

The sudden left-right or right-left transition that turns the car 90-degrees to its direction of travel typically is caused by noobies apexing early and running off the outside of the track coming out of a turn. They panic and turn the wheel quickly to get back on the track quickly. But... the outside tyres are in the dirt and dirt doesn't give much traction to steer. So they reflexively turn the wheel even more. So now, we've got a car traveling at 80-100mph with its outside tyre turned 20-30 degrees trying to get back onto the track. The scrubbing tyre on the dirt eventually does steer the car back onto the track, however, when that outside front-wheel that's turned 20-30 degrees touches pavement, guess what? It really grabs and immediately spins the front end of the car... and fast... by this point, you're not gonna catch it.. .way too late. The car then spins back onto the track (remember that idiot McClure in ALMS?), and the driver makes a 2nd mistake, he tries to correct his mistake by steering in the opposite direction and aim the car back down the road. The timing is perfect to stop the car's spin such that it's perfectly perpendicular to the direction of travel, the tyres dig in, and you end up rolling. In the case of that mustang, it rolled THREE times! The instructor said that on the 1st roll, the roof caved in 1/2 way and he immediately grabbed the seat-lever and laid back the seat. The 2nd roll caved the roof all the way down and the 3rd roll pinned him against the flattened seat...

Here's a picture of the suspensoin geometry I talked about last time. Let's take a stock MacPherson strut suspension:

The roll-center is determined by the geometry and is the point where the centerline of the car intersects with the line between the strut's pivot axis on the ground (near the tyre's outer edge usually), and the instantaneous center. The distance between the roll-center and the C.O.G. is the lever-arm that pushes the car sideways when there's a lateral weight-transfer due to cornering-G forces.

Here's what happens when you lower a car with MacPherson struts:

Note that the minimal amount that you lower the C.O.G. actually results in a much larger lowering of the roll-center. This increases the length of the moment-arm between the roll-center and COG. Meaning that for the same lateral-weight transfer at the same cornering-G, the car will roll MORE than before the lowering.

This has the detrimental effect of causing even MORE positive camber-gain on the outside tyre, causing it to ride on its edge even more and lose grip. The actual amount of reduction in lateral-weight transfer is minimal when lowering the C.O.G.; most of the weight-trasnfer is due to cornering-G, the body-roll is just in reaction to the weight-transfer.

So the trick is to minimize the negative effects of the positive camber-gain on the outside tyre when the body rolls. To keep the tyre flat on the ground, you can:

- add more static negative camber to account for body-roll, about -4.5 to -5.5 is optimal with stock suspension
- add stiffer springs and/or swaybars to reduce body-roll for the same amount of lateral weight-transfer
- upgrade to double-wishbone suspension which can maintain a vertical outside tire and flat contact-patch regardless of body-roll amount

Notice that the ride-height has nothing to do with this. So I get Sean's intentions and results, just that the path he wants to use to get there may not be the best way to get those results. He'd be best off getting stiffer springs and/or swaybars. The goal is to keep the tire vertical and contact patch flat for maximum cornering-grip as the body rolls in response to cornering (well with a little negative camber to account for tyre squirm).

Here's a picture I drew up to illustrate the silliness of thinking that extra-grippy tyres would 1) cause a great lateral-shift in C.O.G and 2) would put the C.O.G. outside of the tyre's contact patch and cause the car to roll:



There's no way you're gonna get tyres that are grippy enough to stand a car up like that in order to roll it under cornering. You gotta do some crazy Z-shaped transition and spin the car to do that...

xsboost90

put some weltmeister 1.5" drop/ 200lb springs-$100- and lower the rear eccentric bolts.

Macfreak007

^^^^ and a turbo rear sway, thats exactly the setup I am going for...

Campeck

with serges droplinks. lol!

StoogeMoe

Thanks Danno for the diagrams. You've alluded to this in another post and now I finally understand what the hell you are talking about

It looks to me like the problem with lowering the car is that the lower a-arm is no longer parallel to the ground. If you could raise the mounts somehow, then the geometry would be maintained.

Say, I'm in the mood for some heavy duty suspension knowledge. Is there any good books or web sites you would recommend? I want math and equations. Lay it on me. I'm especially interested in the new car rear suspensions which seem to be all lumped into the multi-link category. I know they aren't all using the same techniques.

____________________________________
1986 944 NA - Guards Red - 78k miles
1984 911 Carrera Coupe - Guards Red (gone, but not forgotten)
"If you're not living on the edge, you're not living"

adrial

Travis @ Rennbay's geometry correction balljoints solve the issue that Danno mentioned by lowering the lower, outboard suspension point.

Danno

Yup, Rennbay's balljoints will lower the outside pivot on the A-arm to restore geometry back to stock levels when you lower the inside pivot. Another way to do this that's common on the hot-rod muscle-cars is a drop-spindle.

"Say, I'm in the mood for some heavy duty suspension knowledge. Is there any good books or web sites you would recommend? I want math and equations."

Here's my reference list in order of complexity:

Secrets of Solo Racing - Henry Watts
Performance Handling - Don Alexander
Tune to Win - Carroll Smith's (along with all the other "To Win" series)
How to Make Your Car Handle - Fred Puhn
Race Car Vehicle Dynamics - Miliken <-- this last one's the "Bible" of suspensions

"I'm especially interested in the new car rear suspensions which seem to be all lumped into the multi-link category. I know they aren't all using the same techniques."

All the multi-link suspensions try to control the wheel's movement and have the tyre stay flat on the ground when the suspension compresses under body-roll. I think BMWs have the best suspension really, they give the best wheel-control and great performance without resorting to using super-stiff spring-rates. These multi-link suspension are effectively the same double-wishbone suspensions you find on cars designed for performance from day one, like Ferraris, Corvettes, 996-GT2, even the SupraTT.

Due to packaging constraints, you may have to snap the wishbone and attach it differently, hence "multi-link", but the effect is the same. The common trait shared by all these cars with double wishbone suspension is gain in negative-camber as the suspension compresses and the body leans in the corner. Unlike cars with MacPherson struts which gains positive-camber as the suspension compresses and body leans.

The stiffening we do to our suspensions is to keep the body from leaning to make up for this bad behavior of struts but it also introduces other detrimental effects like poor ride-quality and poor bump-absorption under cornering. Here's a picture of the progression of the suspension upgrades I've done. The very first step was to install a set of TurboS sway bars:



Didn't have a picture, so this is one of a stock TurboS (left). You'll notice that even then, it's got a lot of body-roll and the outside tres have positive camber. Then I went to a set of 968 M030 bars with the rare 21mm rear bar. Also added camber to -4.5 degrees and we have the picture on the right. This was worth about 8-seconds per lap at Willow Springs over stock TurboS suspension (on same crappy SP8000 tyres).

Then I went to my current configuration with Weltmeister swaybars. Not only are they beefier than the 968 M030 bars, they have infintely smooth adjustment increments. With the 3-hole rear bar on the 968, I find that the middle position had too much understeer, yet the stiffest setting had too much oversteer. Also added 400lb-springs + 27mm torsion bars, corners flat as a pancake to keep the tyres flat on the round:



However, you'll notice that the E46 M3 with its latest multi-link suspension may look like it's got the most body-roll compared ot me or the E30 M3, but it outcornered all of us! With similar 140-treadwear street tires, but an +800-lb penalty, it was still 5-10sec/lap faster than both of us! And it takes you on 500-mile road-trips with total comfort too!

Serge944

Why would you put stiffer front springs on a car that already understeers? If you want a minor improvement, go with better swaybars with solid chassis mounts. Just to point out how awful the stock ones are, I could barely tell the difference between no swaybars at all and a 20/14 stock setup. That will improve your car's response much better than some rice springs.

Matt O.

If Porsches were meant to have larger sway bars, they would have come that way from the factory. The stock sway bars are the same used on the 1963 Diesel F108 Tractor (14 HP thank you very much), and if they are good enough for farm work they are good enough for the measley track. Anything larger then the stock sway bars is overkill.

StoogeMoe

Thanks Danno. That info is gold. Gold! I'll have to check some of those refs out.

The gain in negative camber is counter the roll, right? So the end effect is a tire that is vertical and has the maximum contact patch on the pavement?

____________________________________
1986 944 NA - Guards Red - 78k miles
1984 911 Carrera Coupe - Guards Red (gone, but not forgotten)
"If you're not living on the edge, you're not living"