Excessive wheelspin on corner exit!!!

  • Thread starter Thread starter Gupster74
  • 38 comments
  • 8,949 views
The Higher the numerical setting on extension, the stiffer (harder to pull apart) the setting. Therefore, the slower the weight transfer is when the shock is seperating.

The higher the extention, the more stiff extention is. So, the more fast is your weigth transfer.
The higher the compression, the more stiff compression is. So, the more fast is your weigth transfer.
The higher your spring are, the more stiff springs are. So, the more fast is your weigth transfer.

It's quite easy.

@rotary : I told a lot of bs'bs about the curve after the longest straigth line before, it's strange you didn't noticed it :)
 
The higher the extention, the more stiff extention is. So, the more fast is your weigth transfer.

Yes.

The higher the compression, the more stiff compression is. So, the more fast is your weigth transfer.

The faster your initial transfer is, yes. But overall the transfer will be slower (you get a first "hit" as you first accelerate or decelerate, then the shock continues to compress causing slightly more weight transfer to occur).

The higher your spring are, the more stiff springs are. So, the more fast is your weigth transfer.

Yep.

@rotary : I told a lot of bs'bs about the curve after the longest straigth line before, it's strange you didn't noticed it :)

I had quoted you before you edited it in :p. I'll edit this post with my thoughts on it.

Edit, here we go:

I used 47/53, as all spring devices says, by not tracking weigth reduction distribution for the initial tests. I need a few more day on the car before being sure, my tuning entry won't be finished.

I think real lvl3 weigth is between 47/53 and 50/50. So "less/more" or "even/even", that is quite different of "more/less" that you choosed. (that's why it's a pig btw)

The factory springs never match weight distribution though... And that's my point there.

C5 Corvette has 50/50 weight distribution, does not have 50/50 springs on any suspension kit.

I've also found that weight distribution takes most of what weight it saves away from wherever the engine isn't... So in a front engine car, it takes weight mostly off the rear of the car, in a rear or mid engine car it mostly comes off the front, etc etc. Therefore putting the weight balance towards whichever end the engine is at.

As for it being a pig... It is incredibly fast everywhere I've taken it so far. It's improved massively over stock settings.


I don't really understand. If you used 60/40 you should have set a-roll bar like that : more/less. Are you saying that less/more or even/even is better ?

In the case of anti-roll, the Impreza seems to like less/more as I said earlier. ~4mph difference in certain high speed corners at full throttle.

Then again, I run extremely strong differential settings and 50/50 or 45/55 torque bias... So my setup for this car looks very similar to my FWD setups, just with a bit more rear spring rate.

About the dampers, i've got 178km/h with a flat SR/RH @ 47%/53%, and flat dampers and a-roll bars : 5/5/ - 5/5 - 4/4 and 184km/h with my classic dampers, so no, I won't change a line in my moto : that's 3.5% more speed for me, or a chrono @ 1'40 vs a chrono @ 1'36.5 :)

I think you're doing too much testing in 100% controlled conditions and not enough in more "real-world" situations. Your style of dampers when used on the Impreza did help solid-state cornering to an extent (cornering with the same amount of steering input all the way through) but entry was compromised... And at Trial there is a certain amount of curb jumping required to be fast (not corner cutting, just jumping the curbs with 2 tires) through the first sector... And the "inverted" dampers upset the car over said curbs... Also found more understeer on corner entry as I let off the brakes; the result is a car theoretically faster but slower in practice.


You are holding the curve at more speed, so you are quicker around the track (less needs to brakes and more speed after curving), maybe what you saw is speed problems.

Theoretically, yes. In practice, no. If you can enter the corner faster and exit at a higher speed you'll be faster than holding a higher speed in the middle. AWD's main advantage is traction; the disadvantage is weight and therefore steady cornering speeds. It's better to exaggerate advantages than to minimize disadvantages.

Braking, even less, at higher speed is more violent than braking at lower speeds but you enter into the "torque function" differently, more smoothly, with low torque @ high rpm (torque for engine brakes) going to high torque at middle range rpm than @ midrange rpm with high torque allready going to low torque at low range rpm then low torque @ high rpm. Watch your decel settings. I guess they are set low just for that curve.

Wha? I apologize but I simply can't make sense of what you're trying to say here.

Hence first 0.3 second braking stability in first curve after the long straigth line. So people won't shoot the wall there. And low ext help a lot braking too.

Braking stability is not a problem with this Impreza. Well, actually, it is... There's too much of it. :lol: With extension set lower than compression, this becomes even worse (understeer).

Aero set @ 0/20 is the fundamental problem there. :D aka "understeer land". :D

0/20 is better than 0/5 if you can compensate for the aerodynamic imbalance; more grip is more grip. If you ABSOLUTELY CANNOT get the rear tires to use their full grip at 0/20 then decreasing downforce is beneficial... But if you can compensate, do. The car will be faster.


For susp/aero settup-ing default lsd is "l'enfance de l'art", as we say in french.

Children's art hm? :lol: 5/5/5 on all differentials will have the absolute least affect on handling possible, though it will cause wheelspin on the inside tires. (With AWD it tends to be the inside front unless torque bias is heavily towards the rear).
 
Last edited:
5/5/5 on all differentials will have the absolute least affect on handling possible

I thought 60/5/5 would result in the most 'open' differential possible?
Based on the most recent 'LSD settings help' thread, Where it was stated that 5,60,60 results in the most locked differential possible. It was implied that the 'initial' is how much torque difference between the 2 tires is required to lock the axles?

You don't need to try and explain LSD, I'm just curious if the goal of 'least effect on handling' means as open as possible?
 
The faster your initial transfer is, yes. But overall the transfer will be slower (you get a first "hit" as you first accelerate or decelerate, then the shock continues to compress causing slightly more weight transfer to occur).
It won't compress on my setups, since it's set 2 or 3 more clicks than anything else...

The factory springs never match weight distribution though... And that's my point there.

C5 Corvette has 50/50 weight distribution, does not have 50/50 springs on any suspension kit.

I've also found that weight distribution takes most of what weight it saves away from wherever the engine isn't... So in a front engine car, it takes weight mostly off the rear of the car, in a rear or mid engine car it mostly comes off the front, etc etc. Therefore putting the weight balance towards whichever end the engine is at.
I agree somehow, but this car is difficult. All tests I made put the weigth at less/more (from flat 6kg/14kg to 14kg/6kg). This method is only an indication. On some car it works really great, on some others it just don't.

In the case of anti-roll, the Impreza seems to like less/more as I said earlier. ~4mph difference in certain high speed corners at full throttle.
From 6/4 I'm now at 3/4 or 3/5 (can't decide yet). It follow what I though on the weigth. I made a pretty dogmatic setup, so whatever. It'll be a test.

Then again, I run extremely strong differential settings and 50/50 or 45/55 torque bias... So my setup for this car looks very similar to my FWD setups, just with a bit more rear spring rate.
I've got 32% front for torque. And accel like 40/50.

I think you're doing too much testing in 100% controlled conditions and not enough in more "real-world" situations. Your style of dampers when used on the Impreza did help solid-state cornering to an extent (cornering with the same amount of steering input all the way through) but entry was compromised... And at Trial there is a certain amount of curb jumping required to be fast (not corner cutting, just jumping the curbs with 2 tires) through the first sector... And the "inverted" dampers upset the car over said curbs... Also found more understeer on corner entry as I let off the brakes; the result is a car theoretically faster but slower in practice.
Yes but it works "somehow". I just don't understand how you do low 1'27. Either I can't drive, or you're cutting by the grass, which will be a no-no for time trials if it's like this. With aero @-/20, you just can't pass it clean. You need no aero contribution to your car just for that turn. hence a -/14. i will post my car very soon, I'm fed up with her now.

Wha? I apologize but I simply can't make sense of what you're trying to say here.
yup, that's the bs part :D

Braking stability is not a problem with this Impreza. Well, actually, it is... There's too much of it. :lol: With extension set lower than compression, this becomes even worse (understeer).
This is because you messed up initial/decel settings :)

Actually, I use that for flat LSD :
Initial : (front torque / 5) | (rear torque / 5)
Accel : (front torque / rear torque) * 100 | 100-front accel
Decel : (front weigth% / ~2) | (rear weigth % / ~2)

Guess what I used ^^, and guess how is constructed the default setup of that car, when you put the torque 30/70 (default LSD, by testing should be one of a 30/70 car).

For the op, that would give :
Initial : 2/18 (muahaha 10/90 is a fail)
Accel : 11 / 89 (muahaha 10/90 is a fail)
Decel : didn't studied the car.
 
It won't compress on my setups, since it's set 2 or 3 more clicks than anything else...

You don't understand what I'm saying methinks.

A stiff damper will take more time to compress but it will eventually compress to the same height as a soft damper. Spring rate is what controls the amount of forward/rearward pitch, dampers merely control how quickly it happens. Stronger compression will make the initial weight transfer happen slightly sooner as they won't compress fully instantly then slowly transfer a bit more as they compress further, causing the center of gravity to shift slightly more.


I agree somehow, but this car is difficult. All tests I made put the weigth at less/more (from flat 6kg/14kg to 14kg/6kg). This method is only an indication. On some car it works really great, on some others it just don't.

As I said, I'm running a 60:40 front/rear spring rate ratio. It works for me at least. :D

From 6/4 I'm now at 3/4 or 3/5 (can't decide yet). It follow what I though on the weigth. I made a pretty dogmatic setup, so whatever. It'll be a test.

It shall indeed be interesting.

I've got 32% front for torque. And accel like 40/50.

Heh, I'm a small bit tighter on differentials... My front initial is 4x my rear initial, accel values are within 3 of each other, decel on both ends is fairly low.

Yes but it works "somehow". I just don't understand how you do low 1'27. Either I can't drive, or you're cutting by the grass, which will be a no-no for time trials if it's like this. With aero @-/20, you just can't pass it clean. You need no aero contribution to your car just for that turn. hence a -/14. i will post my car very soon, I'm fed up with her now.

No grass cutting here. I do jump the curbs quite a bit (2 tires over them at the inside of both of the first two corners, first part of the final chicane) but always have two tires on the track or exactly on top of the curb.

This is because you messed up initial/decel settings :)

Actually, I use that for flat LSD :
Initial : (front torque / 5) | (rear torque / 5)
Accel : (front torque / rear torque) * 100 | 100-front accel
Decel : (front weigth% / ~2) | (rear weigth % / ~2)

Guess what I used ^^, and guess how is constructed the default setup of that car, when you put the torque 30/70 (default LSD, by testing should be one of a 30/70 car).

For the op, that would give :
Initial : 2/18 (muahaha 10/90 is a fail)
Accel : 11 / 89 (muahaha 10/90 is a fail)
Decel : didn't studied the car.

I don't entirely understand these calculations :lol:

10/90 is the torque split set in the VCD though, not the difference between front/rear initial/accel.
 
You don't understand what I'm saying methinks.

A stiff damper will take more time to compress but it will eventually compress to the same height as a soft damper. Spring rate is what controls the amount of forward/rearward pitch, dampers merely control how quickly it happens. Stronger compression will make the initial weight transfer happen slightly sooner as they won't compress fully instantly then slowly transfer a bit more as they compress further, causing the center of gravity to shift slightly more.
I understand what you're saying, I use that thing for braking grip.

I set comp>ext to prevent hysteresis moves of the suspension AND to brake better.
If ext are set higher, front train will bump, vibrate a lot and loose grip, mainly because of the spring that goes the same way and comp don't absorb ext bump reaction. If comp is set higher, this phenomenon is absorbed because a slow/strong comp absorb the spring better than a quick/strong ext. Plus I've got extra braking grip. Everything is better, at least in front.

I make a semi-reverse use of ext and comp : the lower the ext mean lots of bumps absorbed, the higher the comp mean high grip in curves (to a point). I know it's strange but if GT5 have worked better on that vibration, I'd use them as everybody.

And the funny thing is it works quite very well. I'm thinking on hybrid damper setup theory now :) (front ext < comp and rear comp< ext)

I don't entirely understand these calculations :lol:
Initial follows the torque split distribution.
Reason is if all the power is in front, the highest torque threshold must be in the front and vice-versa.
Decel is easy : amount of engine brake must follow the weigth distribution.
Accel follow a strange law but for me it works. :)
 
Last edited:
I have been messing with the whole inverted damper thing today. There are certain benefits to that setup but the track surface has to be entirely smooth and predictable, not a lot of bumps to soak up, which when used sets impressive lap times on Grand Valley or Laguna Seca. However, try the same setup on Deep Forest and you find it doesn't work there about 20 seconds into the first lap.

I kind of had to make me an inverted chart to help get my head around the idea. The number in parenthesis (-) being the inverse as how many people think it should be or assume it is. The ones not in parenthesis how you set it in menu.

(-)-Rebound--------------------Bound----
-------------------------------------------- fast end of the spectrum
(1)-------10------------------------1------
(2)--------9------------------------2------
(3)--------8------------------------3------
(4)--------7------------------------4------
(5)--------6------------------------5------
(6)--------5------------------------6------
(7)--------4------------------------7------
(8)--------3------------------------8------
(9)--------2------------------------9------
(10)-------1-----------------------10-----
-------------------------------------------- slow end of the spectrum

Like so if for example, I tune my springs on Deep Forest using a neutral and standard bound equals rebound setup, then I could easily invert the setup by going -1 rebound and +1 bound and generally the result was stable, I could up that another level by another increment in opposite directions and the pair would usually be stable (assuming you're trying a smooth track like Grand Valley).

Basically, this rule of thumb kind of fit most of the time I tried it. I just take my setup as is for Deep Forest with say DE 5/5 and DC 5/5 and find DE 4/4 and DC 6/6 was a usable combination and the ride was more often than not balanced. Then if I wanted to try another increment, DE 3/3 and DC 7/7 would normally be stable as well. DE 2/2 and DC 8/8 also worked sometimes, DE 1/1 and DC 9/9 usually didn't, but sometimes DE 2/2 and DC 9/9 worked out. Much was dependent on spring rate in ratio increments or adjustments to anti-roll bars if DE 2/2 and DC 8/8 didn't work, it might work at the next spring rate tweak. Getting them to add up somewhere around 10, I know it sounds silly, seems to work - as a quick rule of thumb on my tunes.

Basically each increment either moves both towards the slow spectrum end or the fast spectrum end depending on how you want it to respond.

The other rule of thumb, "bound should be one setting below rebound" now also makes sense here since it one notch above the corresponding rebound (-) in the spectrum. So in theory now, bound 4/4 and rebound 6/6 can work for a lumpy track and bound 6/6 and rebound 4/4 works for a smooth track, etc.

Just to let you know how well this works, I just ran a 1:13.005" lap followed by a 1:13.006" on Deep Forest using a rebound of 10/10 and a bound of 2/2 on my Camaro Z/28 RM '69, it's like driving a stick of butter now... which actually adds up to 12, so bad example. DE 8/8 and DC 2/2, much better... now I want some stiffness back on my rear wheels for better acceleration but handling at the front, so DE 8/6 and DC 2/4... this makes tuning so stupid easy, can't believe I didn't see it until now.
 
Last edited:
I have been messing with the whole inverted damper thing today. There are certain benefits to that setup but the track surface has to be entirely smooth and predictable, not a lot of bumps to soak up, which when used sets impressive lap times on Grand Valley or Laguna Seca. However, try the same setup on Deep Forest and you find it doesn't work there about 20 seconds into the first lap.

I kind of had to make me an inverted chart to help get my head around the idea. The number in parenthesis (-) being the inverse as how many people think it should be or assume it is. The ones not in parenthesis how you set it in menu.

(-)-Rebound--------------------Bound----
-------------------------------------------- fast end of the spectrum
(1)-------10------------------------1------
(2)--------9------------------------2------
(3)--------8------------------------3------
(4)--------7------------------------4------
(5)--------6------------------------5------
(6)--------5------------------------6------
(7)--------4------------------------7------
(8)--------3------------------------8------
(9)--------2------------------------9------
(10)-------1-----------------------10-----
-------------------------------------------- slow end of the spectrum

Like so if for example, I tune my springs on Deep Forest using a neutral and standard bound equals rebound setup, then I could easily invert the setup by going -1 rebound and +1 bound and generally the result was stable, I could up that another level by another increment in opposite directions and the pair would usually be stable (assuming you're trying a smooth track like Grand Valley).

Basically, this rule of thumb kind of fit most of the time I tried it. I just take my setup as is for Deep Forest with say DE 5/5 and DC 5/5 and find DE 4/4 and DC 6/6 was a usable combination and the ride was more often than not balanced. Then if I wanted to try another increment, DE 3/3 and DC 7/7 would normally be stable as well. DE 2/2 and DC 8/8 also worked sometimes, DE 1/1 and DC 9/9 usually didn't, but sometimes DE 2/2 and DC 9/9 worked out. Much was dependent on spring rate in ratio increments or adjustments to anti-roll bars if DE 2/2 and DC 8/8 didn't work, it might work at the next spring rate tweak. Getting them to add up somewhere around 10, I know it sounds silly, seems to work - as a quick rule of thumb on my tunes.

Basically each increment either moves both towards the slow spectrum end or the fast spectrum end depending on how you want it to respond.

The other rule of thumb, "bound should be one setting below rebound" now also makes sense here since it one notch above the corresponding rebound (-) in the spectrum. So in theory now, bound 4/4 and rebound 6/6 can work for a lumpy track and bound 6/6 and rebound 4/4 works for a smooth track, etc.

Just to let you know how well this works, I just ran a 1:13.005" lap followed by a 1:13.006" on Deep Forest using a rebound of 10/10 and a bound of 2/2 on my Camaro Z/28 RM '69, it's like driving a stick of butter now... which actually adds up to 12, so bad example. DE 8/8 and DC 2/2, much better... now I want some stiffness back on my rear wheels for better acceleration but handling at the front, so DE 8/6 and DC 2/4... this makes tuning so stupid easy, can't believe I didn't see it until now.

So, correct me if your wrong but the formula your suggesting goes something like more rebound than bound for bumpier courses and more bound than rebound for smooth tracks to a formula of a total of 10 for both?? I am in admiration for the work you guys put in to get these results!! I have the kno-how but unfortunately no where near enough time so your research is interesting and time saving. I took your advice, lowered my rear camber to 1.0, set the lsd and vcd(which interestingly was set at 20/80!) to standard and hey presto, problem solved!! Muchos Grazias!!!
 
Back