If you dial out the lift off oversteer, you may regret it, since it will be harder to rotate the car. More throttle will increase the rear grip. Decreasing radius and 911s are not a good combination.

I fought the toe out on jounce for years. To fix it you need to go with stiffer torsions, stiffer shocks (Koni's stiffen more in jounce than bounce) and adjust your bumpsteer to minimum.

No LSD. That was only the 2nd event that I've had wheelspin issues like that. While I would love an LSD, it would quickly spiral into one of those "while you're in there" things which is out of the budget, at least for now.

Instead I'm going to try going a little too much rear torsion bars and ditching the rear sway. I'm hoping it will give me the roll rate I need while keeping that inside rear planted, with the added benefit of dropping a few lb. At least that's the theory.

In that .gif file I found later my settings had slipped and I had rear toe out. :eek: I moved to 1/8" in and the car settled down. Just for poops and giggles, I jacked the rear end up to full jounce the other day and remeasured toe - it was down to 0. Not nearly as much difference as I expected.

Stupid rules. 15s aren't going to get it done, as there aren't any legal tires wide enough. 225s are as big as they go in 15. I'm already giving up 60mm in front to the S2k and lighter NC. On the bright side, I don't think I'm too far down in low end grunt. While it's not surprising that I was gaining 0.2 sec in the first 60ft at the pro, even with the dreaded 915 1-2 shift I was easily beating the 50hp more s2ks to the first turn by a good margin. That was up to about 60mph.

As far as weight reduction, there's not a whole lot left I can do. I figure I saved maybe 10lb with the exhaust, 5lb with the seat. I can still do an Odyssey type battery, so that could be a good 25lb+. Some light forged wheels and fancy pants brakes could maybe save me another 40lb of total, unsprung, and rotational weight. I was at 2680lb before the exhaust swap with a half tank of gas, and my car has a sunroof.

For comparison, full prep s2ks are averaging 2650, NCs are around 2350.

I don't know what year AC became standard, but I'd like to see an early 911SC with no AC, slicktop, stripped out, and built to the limit of the rules would weigh in at. I think if one could come in under 2400 it could be a pretty serious weapon.

Adding rear roll stiffness (that is, increasing the proportion of total roll stiffness contributed by the rear) will only make your inside wheelspin problem worse. It won't matter if it is from sway bars or torsion bars. The only difference comes when you have a high frequency bump to only one wheel. In that case sway bars transfer weight faster for the same roll stiffness.

I upped the rear torsion bar size on my car and it doesn't push anymore but I can't get on the gas as early. So you will have to make up time by coming into the turn faster and waiting longer to get on the throttle- which is more 914 than 911.

Ooh, interesting. Thank you. Could this be helped with a big enough front sway to lift the inside front? There still has to be some percentage on the inside, if all of it is the rear that's the most you can do, right?

If you increase the front percentage of roll stiffness you will add understeer but it will keep more load on the inside rear so that you can get on the throttle sooner/more with an open differential.

You are correct in that these guidelines only apply up to the point that one wheel lifts in the air.

That said, the benefit to stiffer torsion bars as opposed to sway bars for a given roll stiffness is that they also add pitch and warp stiffness so you will speed up transitions and help keep the inside front closer to the ground although you can't really change the amount of load on it. The mass center of the car being above ground means that roll will cause a shift in the mass center out of the turn but this is generally small- it is related to the sine of the roll angle and the cg height (I would guess 20"-22").

The best way to go about this is to lower the mass center. If you have a weight limit then by all means get as much underweight as you can then add ballast as low as you can. This will reduce the total load transfer (in any direction) for a given lateral acceleration, keeping more weight on the inside rear while maintaining the desired handling balance. The only downside to a lower cg is you have less rearward weight transfer on acceleration so unless you are currently popping a wheelie you won't be as well off at the drags.

By the way, a clutch type LSD will add stability under braking while a torque-biasing one will be freer. On power, the troque-biasing one will actually add an oversteering moment if the wheels are turning different speeds. The clutch-type will try to match speeds and so add understeer.

Last event of the season was a zany one yesterday:

YouTube video

http://i43.tinypic.com/2cffq1l.jpg
http://i40.tinypic.com/r9ii2v.jpg
http://i44.tinypic.com/2hqtc9h.jpg
http://i42.tinypic.com/wahpus.jpg
http://i41.tinypic.com/acpgds.jpg

Also got a ride in this gorgeous 912E:

http://lh3.googleusercontent.com/-xS...h431-no/13+-+1
http://lh5.googleusercontent.com/-N5...o/IMAG0360.jpg

Subscribed. I too am trying to prep my car for autox and still sorting out best setup. Currently running: 21/30mm, stock f&r sways, strut bar, 205/50/15:225/50/15, -1.5f, -2.0r, toe also set. Pushes badly in corners and rotates too easily and rapidly on the slightest lift. Please keep posting your progress. Ty. -John

Thanks. I'm in offseason mode right now, already stir crazy.

I've already swapped to softer front torsions (23mm -> 21mm). I'm also going from the stock rear 18mm sway to a 22mm one. I'm hoping that will help the mid corner balance.

I know I'll never be able to completely dial out the lift throttle antics, but since it's a transitional thing I think it can be somewhat controlled with shocks. I had ERP rebuild / revalve my shocks last year with their off the self settings. Looking at my charts, I have a ton of rear rebound and not much else. On lift throttle, the inside rear is in rebound and the outside front is in compression. To reduce oversteer in that phase, you should want to soften the rear rebound and firm the front compression.

I've got a local guy who is an OEM dynamicist, an avid autocrosser, and revalves Bilstein struts. I may ask if he'll take a test drive of my car and take a crack at his valving.

In the meantime, I'm also going to try reducing my rear camber and dialing in a bit more rear toe-in. I'm getting less and less confidence in my self-done alignments every time I remeasure, so I also may bite the bullet and get one done professionally.

Not a lot of action on this thread, so I'll just update the blog. I probably won't get the car back between cones until late April at the earliest, so it'll be pretty dry until then.

I have good advise, and it makes basic common sense, that adjusting roll should be for spring rate not shock rate. Shocks (dampers) just should be used to reduce spring movement and fine tune handling along with tire pressure. If the shock is so stiff it reduce body roll it is way to stiff to help handling, infact it hurts handling.

Determining the correct srping rate should be the basic step that is known for each course. Do you have a good spring combination that does not increase oversteer?

Good to hear you are dialing it in. Shocks and alignment are critical in AutoX. Spend your money there next.

No, not yet. I'm getting wildly different handling in each phase of a corner. There is crosstalk, but springs & sways control balance mid corner. Shocks control transition.

On a track a car sees relatively little time in transition. If you want, you even can separate out each part of a turn into separate bits: Brake. Turn in. Hold steady through turn. Accelerate.

In autocross, you are almost constantly in transition. Ripping the wheel like a crazy person, and often without time or space to avoid a comparatively violent simultaneous wheel and brake input. On top of that, they'll stick cones in whatever bumpy broken pavement lot they can get their hands on. Some are nice and smooth airport runways, but many are old lots in disrepair. Overspringing a car is detrimental in these cases.

As such, nearly every fast street based autocross car today uses a digressive shock valving strategy. For low shaft velocities, they can be set very stiff to act as a helper spring and speed transition. Then for higher shaft velocities (bumps) they can blow open to a second stage of valving which is closer to an old-school "properly damped." It's very cool.

That said, first order of business should have been to get spring rate right for good mid corner balance. I was far from correct last season. Went way too front stiff and/or rear soft. I just ordered the biggest solid Sway-A-Ways they made figuring bigger is better. Autocross to win suggests 2.2-2.5Hz, with the rear slightly higher. I'm dropping down to still front stiff but closer to equal rates - ~2.2Hz front / ~2.0Hz rear. I'm hoping this will give me better balance mid - corner and after. I'll worry about the rest after I feel I get step one covered.

Of course, it's all theory. A 17th out of 23 at my first National event last year shows that me, the car, and/or my setup have plenty of opportunity for growth. I'll let you know better this season if I'm on the right track or talking out of my ass.

I am going to go to stiffer TBs but will try 22/29 hollow bars to see how that works and try better sway bars. I am not convinced that stiffer is going to help as much as balance. Also reducing my wheel movement will help reduce sudden problems with balance...slow hands.

That's too stiff in the front versus the rear......

Not that I know what does work, but I agree. I know that 23/30 was way too front stiff on my car. See this chart I made. The weights I used are my actual corner weights minus what I estimated my unsprung weights might be.

Subscribed SmileWavy

My car weighs around 2500lbs with about equal reduction in F/R weight (~200lbs). I am going to get my car corner balanced again after the weight reduction. Your chart is very helpful. How did you calculate the wheel rate? Was this from another source? Is the unsprung weight listed or total weight for each corner?

This is an interesting chart. Knowing torsion bar spring oscillation rate must be an important factor in selecting the correct hardware for balance between TB's, anti roll bar and shocks/dampers.

But how?

That chart is labelled incorrectly. Unsprung weight should be around 40 pounds front and maybe twice that rear. But you ignore unsprung weight and focus on the sprung weight when calculating ride frequencies. He has sprung weight listed there, or more likely just the load on the wheel, which is sprung plus unsprung weight.

I waiting to see how this plays out. What torsion bar sizes will the calculations recommend to be used?

I'm guessing 22 front and 29 rear. For shocks, probably Bilstiens w/custom valving to match the t bar size and the car's weight.

Keep us updated.