spring/bar/weight/diff setups for FWD beam axle racers

[sambb]

Track guys might find this interesting. I found this article by Chassis guru Mark Ortiz. Its a series of questions and answers regarding how to best setup a balanced FWD macpherson strut/beam axle race car (not autocross). In other articles he goes into a lot more depth on the drawbacks of lowering the front of mac strut cars and why stiffer front ends work well at killing understeer (which goes against conventional wisdom) and explaining rear roll centres on beam axle FWD'ers.

April 2000

In this article he touches again on the need for stiffer fronts and getting the rear to 3 wheel through corners. Scroll down a bit.

August 2014

[seangti]

thanks for sharing, so much information to take in. My car suffers on the longer 3rd gear openning radius corners as I can't get on the throttle till it's quite straight ahead. Have been trying to run such corners in 4th to control spin, whilst it's quicker than spinning through 3rd, it's not the quickest way.

I also ran the rear sway bar hard on the weekend. On cold tyres it was a bit slippery under trail braking, but once I had heat into it, was all good. Still had a very marginal hint of acceleration induced understeer in the high speed corners, but adjustable in the lower speed. I wasn't going to back off mid-corner to see what that did, keep the foot burried and it kept it's composure.

I would like to get some more front end grip, but I don't think that'll come by stiffening the rear.

[kaanage]

In other articles he goes into a lot more depth on the drawbacks of lowering the front of mac strut cars and why stiffer front ends work well at killing understeer (which goes against conventional wisdom)

Do you have a link to an article that explains this one? I can't fathom it unless it's to do with camber control when using EXTREMELY grippy (ie very wide, full slick, soft compound) tyres.

Edit: OK, I had a look at the 2nd article which mentions using total weight transfer off the inside rear (full 3 wheeling) where the stiffer front won't have the normal effect of unloading the inside front so you retain more camber control without losing the grip on the inside (plus the note he has about this having the greatest benefit on a dropped Mac strut car where the front camber curves are compromised). I seriously feel this would make our cars unstable in a throttle liftoff situation, however, which is damned scary at high speed.

[sambb]

yeah it is pretty heavy on. He's done a tonne of articles on just each singular aspect, but then here he's sort of thrown it all together so its a lot to digest. I like how he makes it quite simple - eg don't go all spring or all bar but 50 percent from each; go stiff enough to cock the rear wheel but only so that it is just in the air; get it to cock the rear wheel so that you get front traction under power to get on the throttle earlier; and that front lowering = bad unless you muck around with strut base lengths and/or extended ball joints to keep control arms level or even slightly lower at the ball joint.

Where I'm stuck is that I've measured the front control arm bush pivot and the ball joint and the control arms are more or less level now with stock springs. I have some weitec springs to go in but am a bit shy of changing those angles by introducing any degree of drop at all. But then if I go fine i'll keep the stock springs and stiffen the front with a bigger anti roll bar then surely you move away from that 50/50 spring/bar ratio he speaks of. I think for a street car, the latter might suit me better.

This Ortiz guy seems to agree with alot of what the Shine racing guys said - that at the rear the roll centre is so high that you need to play with ride height to adjust rear roll centre down without going overboard with springing/bars as you do, and that you actually need to go stiff at the front without going lower because the roll centre needs to stay high there to keep bump steer and camber gain in roll under control.
I can dig out the other articles that relate if anyone wants them.

[kaanage]

These would make life a lot easier but I'm not sure if they'd be road legal here - http://www.vwwatercooled.com/forums/...ice-26265.html

And these would be even better H2Sport Spindle. Improved handling and geometry correction for lowered cars.

[sambb]

"I seriously feel this would make our cars unstable in a throttle liftoff situation, however, which is damned scary at high speed."

I think the reason why going stiff at the front won't induce nasty lift off oversteer surprises, is because of what you are doing with the whole package eg balancing out weight transfers F:R by taking from the rear and adding to the front, and flattening out the roll axis.

The lowering at the rear that eg Shine says should accompany it is the first bit. From the bits that I've read our car due to its beam axle has a high inherent roll centre (RC) at the rear. Total Roll resistance is made up of geometric roll resistance (roll centre influence) and elastic roll stiffness (springs/bars). So our cars even if they have only moderately stiff rears elastically, still have high total roll resistance because of the very high RC. During cornering, weight transfers move through roll centres and since ours is high coupled with the fact that our car has a narrow track and a high COG then thats alot of weight transfer happening at the rear and up high. Thats why those "does your VAG handle?" articles has Shine racing saying that there is only so much you can do at the rear - namely lower it because that drops your COG some and takes your rear RC with it.
Now if you drop the rear RC you will decrease rear geometric roll resistance. Then you have a choice: if you want to retain the old balance the car had before lowering ie maintain the old total roll resistance you could compensate by increasing elastic roll resistance ie install off the shelf lower springs which usually have a higher spring rate. If you do that you benefit from a drop in COG and roll centre but total roll resistance is still unchanged and is maintained elastically so you get a choppy ride that is harder to tune dampers for. But for example if you have coilovers or get original/moderate rate springs reset lower, you can drop rear ride height without changing spring rate and that will achieve what the shine guys advocate - getting the roll centre and COG down which lowers the height at which rear weight transfer happens, but doing it with a drop in rear total roll stiffness. This is what will give a predictable ass end.

But there is only so far you can go by dropping rear roll centres with beam axles. Ortiz explains that the lowest practical beam axle roll centres end when you risk bottoming out and that beam axle roll centres end where independent rear end roll centres begin. So there is a limit to what you can do at the back of our car to lower the roll centre.

So to compensate you build more weight transfer into the front by increasing its total roll resistance. But at the front because its a mac strut and you don't want to change geometry (lower control arm angles matter alot) you have to leave the roll centre where it is. On a mac strut if you lower, the COG drops a little but the roll centre drops exponentially which can stuff the handling.
So your only option for increasing front total roll resistance is to increase elastic roll resistance (springs and bars).

So basically our car has more weight transfer at the rear atleast relative to the weight that it carries (ortiz) and you have to take some from the back (geometrically with rear lowering and not going nuts with springs and bars) and give it to the front (elastically with spring/bars) to square things up. Additionally our car has a steep roll axis ( a drawn line from the front roll centre to the high rear roll centre about which the car can theoretically be thought to roll). By lowering the rear we lower the rear roll centre as much as we can (keeping the front one untouched) which flattens the roll axis. This drops the COG slightly too. Therefore a slightly lower COG moves around/through a flatter roll axis. Theoretically squaring up the weight transfers and flattening the roll axis means you can be much stiffer at the front than you'd think you theoretically should be without the rear getting scary.

Note : this is race car stuff - no factory would do this because they want a car that tends toward understeer. eg Despite our stock high rear roll centre and lots of rear weight transfer which would normally tend to cause oversteer, our front end rolls too much causing masses of camber gain that makes it an early understeer.
If you are getting it to three wheel I'd imagine that yes, going to narrower rears or an inferior/different compound at the back would be incredibly dodgy.
Also in case your wondering I'm a total geek for this stuff - I understand and can relay the theory but don't race myself so it'd be interesting to hear track guys experiences if they try to implement any of it. Like I said none of this is mine I'm just trying to put it in easy terms.

[Blue9N3]

Sean, there is probably only so much the suspension can do and it will take an LSD to stop the inside wheel spinning. No matter how good the setup is, the inside front wheel will always be less loaded than the outside one.


Sent from my iPhone using Tapatalk

[sambb]

These are for MK4's but are the extended ball joint principle that would allow you to lower the front for COG advantages but maintain a high roll centre.

Roll Center/Bumpsteer Correction Kit : USRT, Usually Sideways Rally Team

maybe Eddy's guy could fashion up some of these to bolt onto the end of our control arms.

[kaanage]

I don't THINK that lowering the rear will have enough of an effect on the roll centers to make this work. I used to run perches under my rear springs which gave my car a Polo Cross look when I removed the rear seat and spare wheel. The car was very adjustable on the throttle but I also spun it at Turn 1 at Phillip Island with only a small degree of throttle lift on turn in.

Removing the perches dropped the rear by about 20mm which has made the car more stable but less agile.

When I added the rear anti-roll bar, I wanted to maintain the balance of the car so I also added a stiffer front bar. I didn't find any noticable reduction in understeer, although it did make the car a touch more responsive to steering and throttle inputs.

If I had shorter rear springs, I would try going lower at that end.

[sambb]

Yeah I agree with the first bit you said. Its hard to believe it would have such a pronounced effect. I had a bit of a look at our beam axle and compared it to this Ortiz article that tells you how to calculate beam axle RC and actually references VW rear ends.

April 2000

It seems that lowering the rear has a linear relationship with lowering the COG but that RC will never move as far. Perhaps the car is just very sensitive to lowering the RC.
But unless I'm wrong it sounds like it did work in your car or initially atleast. You played with rear ride height for a fixed spring rate, which means you isolated rear RC (geometric roll resistance) as the thing being tested. In the first instance you had a higher rear ride height (higher RC). So if geometric roll resistance higher + elastic roll resistance unchanged = higher total roll resistance. It was looser to the point of looping it.
Then you lowered the ride height (lower RC). So geometric roll resistance lower + elastic resistance (springs) still unchanged = lower total roll resistance. It was more stable but didn't point as well because the rear had been given more grip.
I think if you believe the theory, if you'd stopped at that point and added ONLY a front bar then you'd have been approaching the setup that this Ortiz guy mentions.
But you added front AND rear bar. For sure I can see why you added front and rear bar - because you liked the now stable rear and wanted to maintain that balance but just go stiffer overall. I'd have done that too. You maybe didn't notice a reduction in understeer because any advantages you may have won at the front by going stiff were just simultaneously counteracted at the rear ie adding the rear bar (introducing elastic roll stiffness) just increased the rear total roll resistance back toward where it was when you started. This moved some of the weight transfer bias back toward the rear of the car from the front where (atleast Ortiz and shine) think it may have performed well. I'm still hurting my head over it. What do you think?

It'll take a while before its all in but I think what I'll do now that I'm getting all my bits together is install the 27.5 Nm/mm weitec fronts that apparently have stock ride height and get the stock rears reset an inch or so lower. It'll have koni sport dampers. That'll stick to their formula of decreasing rear geometric roll resistance and adding front elastic roll resistance, so i'll give it a test.

[sambb]

This article boils it all down nicely:

TDIClub Forums - View Single Post - What is Handling?

notes d), e) and f) explain how roll centre height (geometric roll stiffness) influences the weight transfer distribution in the car and the last 3 paragraphs summarise how springs (elastic roll stiffness) acheive the same. Its just that you do the former at the back, and the latter at the front.

And here is another article, a june 06' ortiz newsletter article. While he's talking about a RWD he talks about how spring stiffening when you have mac strut front/beam rear goes against conventional wisdom when the biggest issue you are facing is poor camber gain properties at the front end. Its only the opening couple of paragraphs but sums it up well.

April 2000

[seangti]

Sam, are you tracking your car? If so, what's your local track? I can't see how such finite adjustments are going to make much/any difference on the street, and it's only when on the track at the limit chasing 1/10ths that such adjustments will become evident (for better or worse).

Dare I say with my current set-up, I'd be travelling backwards through turn 1 at Philip Island too. It's highspeed, heavy brake to tip in down the hill, if I got it tipped in and back on the throttle it'd be right, but currently it'd be twitchy doing that. At Lakeside though, the high speed corners are travelled at WOT, so the rear follows as the weight transfer is loading the rear under acceleration. QR doesn't have the elevation changes and braking is pretty much all straight.

My point: depending on the track you are running on will influence the set-up relating to roll. I wouldn't be changing camber, castor or toe for different tracks, though would at least be considering to adjust roll via sway bars.

[sambb]

No mate I have neither the time or dollars to track the car - barely enough to even work on it. Which isn't to say that I don't punt the thing along a lonely road when I get the chance and would prefer that it was setup appropriately for that scenario. I just can't bring myself to do anything to a car unless I understand exactly what/why I'm doing it rather than just going out and buying the whole catalogue. You can't get much simpler than running a set of relatively stiff front springs, and resetting stock springs a bit lower especially if then I can get away without having to use a rear bar.
I'm keen to try the setup because it coincidentally suits my situation perfectly. I commute 90km a day across the most ****e roads sydney has to offer so want an acceptable ride, so getting the rear sorted by simply going to lower yet still soft springs appeals. A thumping stiff sprung rear beam will do my head in especially if I know its not needed.
I also duck through a short rough section of national park on the way home with all its undulation/ruts/jumps where I get to pretend I'm in the WRC and have a driveway like a half pipe so retaining front ride height (although stiffer) is another perfect fit.
So if I can get it to handle by doing these relatively simple things (I don't think they are fine at all) while not compromising much else I'd rather atleast try it. To me it seems like there's plenty of evidence to support it eg. while those articles reference race cars, I don't think anything about those setups make them un streetable and its what the Shine racing guys advocate for 'GT' (read street) cars anyway.

[seangti]

yeah fair enough, gotta respect that approach. Great to understand the workings and the why's so you can optimise the gear you have. There's a greater sense of achievement when you get the desired outcome.

[sambb]

yeah well that last bit remains to be seen! When it drives like a pig and I go quiet on the forum you'll know how things really worked out. ha ha. Nah just kidding - if it doesnt work out, i'll be honest and let you know.