I don't believe that unequal shaft lengths causes torque steer

[pushrod]

In FWD or applicable AWD configs, people always say that torque steer is caused by unequal length drive axles on each side, which affects the quantity of torque delivered to each wheel. I see that as a condition that exists temporarily for a single instant, as any shaft torsion differences would be recovered with the turning of the spider gears in the differential.

Instead, I think that almost all torque steer effects are from differences in scrub radius and tire deflection both static and induced from road surface.

I'd like to be proven wrong, if I am wrong.

[Thorpedo]

[Tcoat]

Well now Push when did you subscribe to the Ubersuber school of ultimate statements?
It isn't theory that unequal shafts will cause torque steer it is well documented fact. As they got better at designing the unequal drive systems they pretty much eliminated it through bushings and shaft composition but they did most certainly cause it. There are probably another half dozen or so things that can also cause it but to say "most" is tires may be a stretch.
Now, if your statement included the statement "in current vehicles" I would then have to agree with you.

[pushrod]

Originally Posted by Tcoat

It isn't theory that unequal shafts will cause torque steer it is well documented fact.

I've never succeeded in finding real documentation of this "fact". It's like searching for ab exercises or probiotics; there is no getting through the nonsense.

I can easily picture bushing deflection, shaft elasticity and other compliance creating the conditions temporarily, but again, if both tires have grip, the diff should make up basically all of it.

Also, scrub radius would determine the amount of actual steer. If there was 0 scrub, there'd be no moment to create steer at all.

[Tcoat]

Originally Posted by pushrod

I've never succeeded in finding real documentation of this "fact". It's like searching for ab exercises or probiotics; there is no getting through the nonsense.

I can easily picture bushing deflection, shaft elasticity and other compliance creating the conditions temporarily, but again, if both tires have grip, the diff should make up basically all of it.

Also, scrub radius would determine the amount of actual steer. If there was 0 scrub, there'd be no moment to create steer at all.

I can ask the engineers at work if they have anything on this. I bet they do. Mind you it will mean I have to listen to them for an hour or so.

[justatroll]

Originally Posted by pushrod

In FWD or applicable AWD configs, people always say that torque steer is caused by unequal length drive axles on each side, which affects the quantity of torque delivered to each wheel. I see that as a condition that exists temporarily for a single instant, as any shaft torsion differences would be recovered with the turning of the spider gears in the differential.

Instead, I think that almost all torque steer effects are from differences in scrub radius and tire deflection both static and induced from road surface.

I'd like to be proven wrong, if I am wrong.

I actually agree with you.
"In an ideal system" the torque applied to each side of the car due to unequal length axles should not be different.
An ideal axle has zero flex, and length makes no difference.
And even if it IS different, the only difference should be in the flex in each, but as you say, differences in flex should not cause a difference between the two sides under constant acceleration, but only at the very beginning (like first few milliseconds).

So it is likely something that makes the system differ from an "ideal system".
My guess is that it is drivetrain loss from one side to the other.
More likely is that even a tiny difference in tire pressures or temps is amplified by a FWD car.

[pushrod]

Originally Posted by Tcoat

I can ask the engineers at work if they have anything on this. I bet they do. Mind you it will mean I have to listen to them for an hour or so.

If you do, I will give you all of the internets I've got.

[Tcoat]

Originally Posted by pushrod

If you do, I will give you all of the internets I've got.

I recall reading about the topic back in the pre internet days when things were actually on paper that the shafts were the major cause of the problem back then. Since then they have found work around so the issue isn't as bad. I will see if I can get something "official" in writing (if nobody here beats me to it).

[Tcoat]

Originally Posted by justatroll

I actually agree with you.
"In an ideal system" the torque applied to each side of the car due to unequal length axles should not be different.
An ideal axle has zero flex, and length makes no difference.
And even if it IS different, the only difference should be in the flex in each, but as you say, differences in flex should not cause a difference between the two sides under constant acceleration, but only at the very beginning (like first few milliseconds).

So it is likely something that makes the system differ from an "ideal system".
My guess is that it is drivetrain loss from one side to the other.
More likely is that even a tiny difference in tire pressures or temps is amplified by a FWD car.

When talking things like the 84 Citation (that torque steered like a bastard) there is very, very, very little "ideal" systems to be found!!!!

[DarkSunrise]

Here is the best explanation I've seen so far on torque steer:

Equal-length half-shafts help, but not because of torque distribution. The shorter shaft does not "get more torque."
What happens is that the pivot length of the shafts is different, which messes with the suspension geometry resulting in slightly different suspension movement on each side.
If you put an extra CV joint in the middle of the longer half-shaft, you can eliminate the different pivot lengths and eliminate the problem associated with unequal-length half-shafts without having to make the shaft lengths exactly equal. An example of what I'm talking about can be found here (though the text is in German, the illustrations are clear): http://www.golfmk6.com/forums/showth...t=36475&page=2
The real reason for torque steer is the fact that the steering axis (the imaginary point around which the wheel pivots when you steer) is not the same as the tire's center. When you apply torque through the tire, it imparts a force that behaves as though it is originating from the center of the tire's contact patch with the road. Since the contact patch is not in the same spot as the steering axis, the tire is always trying to "pull itself around" the steering axis. When both wheels have the same amount of grip, the force of each tire cancels the force of the other tire (since one will be trying to pull the steering left and the other will be trying to pull the steering right). However, when one wheel has more grip than the other, the forces don't cancel and you feel a tug on the steering wheel.
A more thorough explanation with good illustrations can be found here: http://www.caranddriver.com/features...ined-tech-dept

Taken from this thread:
https://www.reddit.com/r/explainlike...t_is_but_what/

[Talus1]

Originally Posted by Tcoat

When talking things like the 84 Citation (that torque steered like a bastard) there is very, very, very little "ideal" systems to be found!!!!

Try a Saab Viggen. We test drove one back in the day while waiting for the wife's base 9-3 to be serviced. That Viggen combined massive torque steer - way worse than the base 9-3 - with ridiculous turbo lag and a clutch bite point about 1/2" inch off the floor. The thing would spontaneously change lanes when the boost came on in 2nd gear.

[NahumCC]

Originally Posted by Talus1

Try a Saab Viggen. We test drove one back in the day while waiting for the wife's base 9-3 to be serviced. That Viggen combined massive torque steer - way worse than the base 9-3 - with ridiculous turbo lag and a clutch bite point about 1/2" inch off the floor. The thing would spontaneously change lanes when the boost came on in 2nd gear.

OH yes the beloved viggen. Turbo lag wasn't bad at all so my guess is you had a bad example but the torque steer was indeed terrible. Hence all the Viggen Rescue Kit (VRK) braces that were manufactured by the tuner. Even the 9000 had its dosing of torque steer but its mass helped keep it in check.

[gravitylover]

Ooh the Citation, I learned to drive in a 1980 Pontiac Phoenix. When you put your foot on the gas you better have had a good grip on the steering wheel and if you went into a corner power on the force better have been with you. Somehow I managed to keep that car on the road through some mighty questionable maneuvers.