Torque Steer


Is there a mathematical explanation for the phenomenon called "torque steer" affecting front wheel drive vehicles having unequal length half shafts? Although this was first recognized at least 20 or so years ago, the SAE has no mention of it in the 1999 edition of SAE Standard J67e Vehicle Dynamics Terminology. My research has only been able to unearth several very superficial explanations of this well documented phenomenon, some of which are obviously absurd. Any help you can lend in steering ( ? ) me in the right direction is greatly appreciated.


Yours is the second question on torque steer so I thought I would look into it. In researching the term on the web via AltaVista I even found contradictory definitions of the problem.

My first instinct was to go eliminate the angular displacement inherent in long shafts under torque. The twist of the shaft, if you will. In any reasonable length, reasonably stiff shaft this transient effect would almost certainly go un-noticed by a driver. Also on the web at I found the following:

transmission In 1984 Chrysler came out with its first turbo 4 cylinder; and it was good. There was one small problem though, the car under boost would tend to torque steer. This is a tendency of a car to "pull" in one direction while under power. Its caused in part by one axle being longer then the other and when the engine is under power the longer axle becomes a lever which pulls the car in that direction. Ma Mopars fix was to make the drive axles the same length by adding a intermediate shaft the was perfectly straight (in effect becoming part of the transmission) till the axle on the right side was the same length as the left. Axles are even interchangeable left to right. Most cars that do not have this system can be retro fitted to it as most 2.2-2.5 engines have the bosses cast into the block.

The notion of one the half shafts becoming a lever an physically pulling the car in one direction strikes me as iffy. I include a picture from the above referenced web site an attachment to this email. In that you can see that the introduction of an intermediate shaft would aggravate the twist issue, further substantiating my idea that time delay between engine and wheel is not the culprit.

I get the impression that torque steer is a catch-all term for anything that might transmit force back to the driver through the steering mechanism during acceleration. Suppose for example that the steering axis, that axis about which the wheel is turned in steering, did not pass through the center of the contact patch between the tire and the road. Then there would be a moment arm in the horizontal plane through which the reaction force of the of the road against the tire would act to tend to steer the vehicle during acceleration. This seems to me to be a likely candidate for the torque steer phenomena.

What is not clear to me is how the steering axis depends on the length of the half shafts. Possibly the steering axis shifts a bit during acceleration and that shift is mitigated by the added stiffness of a short shaft. Perhaps these ramblings will trigger something in your mind. If you have any question about what I have said, or would like to enlighten me further please respond.