Jump to content


Photo

What causes torque steer?


  • Please log in to reply
12 replies to this topic

#1 Jammy

Jammy

    Moved Into The Garage

  • Members
  • PipPipPipPipPipPipPipPipPipPipPip
  • 25,397 posts

Posted 30 May 2007 - 09:05 AM

Its commonly thought that torque steer is caused by having one driveshaft longer than the other, however, I think this following article raises some interesting points.

Thanks to Geehawk for finding it.

Technobabble: September 1999
Scrub Radius and the Dave Point

By Dave Coleman

I've been told that in the mountains of Alaska there are grand majestic peaks that don't even have names. That's hard to imagine living down here in the lower 48, geographic features as unexciting as freeway overpasses are often named after some great overpass luminary. The same is true of science and engineering, where designs, mechanisms, and even equations are named after those who figured them out first. A solid axle rear suspension, for example, typically has either a Watts linkage or Panhard rod, unless it's a new Nissan, in which case it will have a Scott-Russel link. (I guess it took two people to figure that one out) It is really a combination of skill and luck to have some mechanism bear your name, after all, a Panhard rod is hardly a stroke of genius, but Panhard must have been an opportunistic little bugger to tack his name to a simple little bar.

I've never really felt any desire to have a linkage, or a bar, or a pivot named after me, but an opportunity just came up and I had to take it. You see, I was trying to figure out the best way to explain what causes torque steer (which is really supposed to be the point of this story, but I'll get to that in a minute) and I couldn't find a name for the point where the steering axis intersects the ground. Without a name, I had to continually refer to it as the point where the steering axis intersects the ground. So I write about the point where the steering axis intersects the ground this, and the point where the steering axis intersects the ground that, and pretty soon I realize there is an opportunity here. I could save myself a lot of grief and ensure eternal fame by simply naming the point after myself. So from this day forth, I declare the point where the steering axis intersects the ground to be the Dave Point.

So back to torque steer. First, what is it and why do you care? Torque steer is simply the tendency of a front-wheel drive car to try to steer itself when you are accelerating. Why do you care? If you are like me (except you don't have a point named after yourself) torque steer really pisses you off, and while understanding where it comes from won't make it *yellow human water* you off any less, it may give you something to think about while you're fuming. The cause of torque steer is really quite simple.

When you turn the steering wheel, it turns about an axis that is, surprisingly enough, called the steering axis. (I told you this was simple.) As with anything that turns about an axis, if you push on it, it will probably turn. The simplest example is a door; it rotates about hinge, and if you push on it, it will rotate.

So what's pushing on the wheel? You and your right foot, actually. The engine's torque output moves the car by pushing on it right at the contact patch between the tire and the ground. When the tire is sitting flat on the ground, that contact patch is basically a rectangle, and you can assume that all the force is being applied right in the middle of it. Now, the Dave Point (remember, this is the point where the steering axis hits the ground) is the point that the contact patch rotates about, so if the Dave Point is also in the middle of the contact patch, the car can simply accelerate in a straight line. If you try to open a door by pushing on the hinge pin, you won't go anywhere. Similarly, if the Dave Point is right in the middle of the tire, the car should accelerate in a straight line.

Of course, as you have probably guessed, it is very difficult to put the Dave Point in the middle of the tire. The steering axis, on cars with MacPherson strut front suspensions (like Nissan Sentras, Subaru Imprezas, and just about anything cheap), is defined by the line that connects the top of the strut and the lower ball joint. Since the strut can't pass through the middle of the tire, and the ball joint has to battle with the brake rotor for real estate, it is quite difficult to get the Dave Point into the middle of the tire. Other suspensions define the steering axis different ways, but virtually all of them present some sort of challenge to proper Dave Point placement.

In most cases, the Dave Point is slightly inboard and slightly in front of the center of the contact patch. The distance between these two points is called the scrub radius because it defines how far the contact patch has to be scrubbed across the pavement when the steering wheel is turned. Now, if you push on the tire at middle of the contact patch, and the Dave Point is off to the side somewhere, the tire will try to steer around the Dave Point. The higher the scrub radius, the more steering torque will be generated, and the more pissed off you will be.

Now, those of you who are really on the ball have probably already picked up on this next fact. In most cases--pay attention here--there are two front wheels, and unless something has gone horribly wrong, the suspension geometry on one side is a mirror image of the other side. So while the right wheel tries to turn left, the left wheel tries to turn right, and they cancel each other out and go straight. Unfortunately, in the real world, that seldom works out.

The contact patch, you see, is a dynamic thing. It flexes and bends over uneven surfaces, and if there is an imperfection in the road, the center of the contact patch will move, and the steering torque on each side of the car will no longer cancel out. The same can happen if one wheel has less traction than the other. If one wheel is pushing with more force, it will exert more steering effort, and you will get pissed off.

There is very little that you, as a non-car manufacturer, can do to reduce torque steer, but there are things you can do to make it worse (or not do to not make it worse). Wider tires have wider contact patches, and therefore more latitude for the scrub radius to move around when you encounter uneven surfaces. Low profile tires with stiff sidewalls are less willing to conform to those surfaces, and so can actually move the patch around even more. Having the wrong offset on your wheels can increase the scrub radius by moving the contact patch. Even extreme alignment settings can increase the scrub radius by moving the Dave Point around.

All of these things you want to avoid are things that make your car handle better, so you have to decide which is more important, reducing torque steer or handling well. Duh! Live with the torque steer. Here's a solution: keep both hands on the wheel.



#2 mini13

mini13

    Up Into Fourth

  • Members
  • PipPipPipPipPipPipPipPipPip
  • 3,794 posts

Posted 30 May 2007 - 02:50 PM

It is my opinion that rather than the "dave point" needing to b in the centre of the tyre contact point, It actually needs to be at the most heavily loaded point of the tyre contact patch which depends heavily on the camber of the wheel. This point is known as the "Joe point" :proud:

#3 MRA

MRA

    Previously known as 'mra-minis.co.uk'.

  • Banned
  • PipPipPipPipPipPipPipPipPipPip
  • 7,607 posts
  • Location: Due to move again....

Posted 11 June 2007 - 06:48 PM

One way of helping to overcome this is to fit a full set of rose jointed tie-bars and bottom arms then set the geometry with near neutral settings except camber 1 to 2 degrees and castor about 3 degrees - what does this do you may ask ???? Well quite simply it stops the wheels from moving around (something you get with rubber bushed suspension) think of it this way as one wheel gains grip it pulls back on the rubber bush, now if you draw an imaginary line through the wheel centres you would find that it has the same effect as that old go-kart you made as a kid with scrap wood etc....... now the other wheel gains grip and off you go the other way. If your suspension is rose jointed A) it doesn't flex so much and B) as you have less toe out 1/4 to 1/2 degree works well the wheels are almost straight ahead anyway so even if it does lose a little traction (as it will) it will still go straight ahead, ok so with 200bhp and sticky tyres you will get torque steer but it is manageable the upside to all of this is that braking from speed is less ragged the front end doesn't dart from pothole to pothole....... :)

#4 camp freddy

camp freddy

    Definitely Not Camp Or A Freddy

  • TMF+ Member
  • PipPipPipPipPipPipPipPipPip
  • 3,696 posts
  • Local Club: Wreake Mini Wanderers

Posted 07 September 2007 - 10:57 PM

I think torque steer has more to do with the fwd differential than geometry,
when you wheel spin the diff allows one wheel too spin due to the fact that when cornering
the wheel nearest to the bend needs to be able to rotate more slowly than the outside wheel.
therefore if you accelerate hard one wheel will be driving harder (ie left wheel) thus
causing the car to steer to the right. try it out.
if you fitted an lsd or locked diff as in a drag car you won't get this affect because both wheels
are getting equal torque but you would find it harder going round a bend as the car would
want to go straight on, not that bad in a mini but worse in a rwd car.

#5 Ethel

Ethel

    ..is NOT a girl!

  • TMF Team
  • PipPipPipPipPipPipPipPipPipPipPip
  • 25,379 posts
  • Local Club: none

Posted 07 September 2007 - 11:04 PM

I think torque steer has more to do with the fwd differential than geometry,
when you wheel spin the diff allows one wheel too spin due to the fact that when cornering
the wheel nearest to the bend needs to be able to rotate more slowly than the outside wheel.
therefore if you accelerate hard one wheel will be driving harder (ie left wheel) thus
causing the car to steer to the right. try it out.
if you fitted an lsd or locked diff as in a drag car you won't get this affect because both wheels
are getting equal torque but you would find it harder going round a bend as the car would
want to go straight on, not that bad in a mini but worse in a rwd car.



A diff is a torque equaliser

#6 camp freddy

camp freddy

    Definitely Not Camp Or A Freddy

  • TMF+ Member
  • PipPipPipPipPipPipPipPipPip
  • 3,696 posts
  • Local Club: Wreake Mini Wanderers

Posted 08 September 2007 - 09:40 PM

I think torque steer has more to do with the fwd differential than geometry,
when you wheel spin the diff allows one wheel too spin due to the fact that when cornering
the wheel nearest to the bend needs to be able to rotate more slowly than the outside wheel.
therefore if you accelerate hard one wheel will be driving harder (ie left wheel) thus
causing the car to steer to the right. try it out.
if you fitted an lsd or locked diff as in a drag car you won't get this affect because both wheels
are getting equal torque but you would find it harder going round a bend as the car would
want to go straight on, not that bad in a mini but worse in a rwd car.



A diff is a torque equaliser


A diff does deliver torque evenly to the driven wheels as you say, but its main purpose is to allow the
wheels to rotate at different speeds when cornering.

http://web.mit.edu/2...fferential.html

#7 Scruffs

Scruffs

    Mini Mad

  • Members
  • PipPipPipPip
  • 252 posts

Posted 09 September 2007 - 03:49 PM

This is an interesting article, but I think it misses a major point somewhere along the line as it puts torque steer down to unevenness in the road. A torque steering car will torque steer on a billiard-table smooth surface aswell as on a standard issue rough-as UK road. Wayward steering on a rough road is going to be as much to do with bump steer and steering kickback as it is to do with torque steer, if not more so.

As I understand it there are many different causes of torque steer, one of which is differing scrub radii. It's also to do with the different angles the driveshaft axes form with the steering axes, which results from uneven driveshaft lengths. Uneven driveshaft lengths also give different angles of twist giving different wheel torques in transient acceleration. It can also be down to the diff physically moving in the car. With limited slip/locking diffs then things can get even worse through different driveshaft torques. All of these can explain torque steer when trying to drive straight on a perfect surface.

if you fitted an lsd or locked diff as in a drag car you won't get this affect because both wheels
are getting equal torque


I'm afraid thats not true! :withstupid: The big advantage of semi and fully locking diffs (by whatever method they lock) is that they can give different torques to either wheel on that axle. On front wheel drive they can give HUGE amounts of torque steer when the left and right wheels have different levels of grip.

An open diff can only ever deliver the same torque to each wheel by design. That is why, with an open diff. you go practically nowhere with one wheel spinning on ice, even though the other wheel may have a sticky tyre in glorious sunshine. Of course these things are never perfect and slightly more torque may go to the slower wheel, but nowhere near as much as with an LSD/locking diff.

blah blah blah I could go on forever, I have a bit of a sad obsession with vehicle dynamics! >_<

#8 Ethel

Ethel

    ..is NOT a girl!

  • TMF Team
  • PipPipPipPipPipPipPipPipPipPipPip
  • 25,379 posts
  • Local Club: none

Posted 09 September 2007 - 04:19 PM

Interesting point on the driveshaft angles...

even it the torque is equal as it leaves the diff it may not be the case that it's equal when it gets to the tyres. Maybe the mini's unequal shafts can actually work in it's favour in certain situations?

#9 Scruffs

Scruffs

    Mini Mad

  • Members
  • PipPipPipPip
  • 252 posts

Posted 09 September 2007 - 05:24 PM

even it the torque is equal as it leaves the diff it may not be the case that it's equal when it gets to the tyres


That's a good way of putting it! A component of the driveshaft torque is also put directly into the upright steering axis as a result of the angles between the driveshaft, wheel axis and steering axis.

It's a bit of a headache :withstupid:

#10 miniSLO

miniSLO

    Starting My Mini Up

  • Noobies
  • Pip
  • 4 posts
  • Local Club: Mini Slovenia

Posted 28 November 2008 - 11:36 AM

One way of helping to overcome this is to fit a full set of rose jointed tie-bars and bottom arms then set the geometry with near neutral settings except camber 1 to 2 degrees and castor about 3 degrees - what does this do you may ask ???? Well quite simply it stops the wheels from moving around (something you get with rubber bushed suspension) think of it this way as one wheel gains grip it pulls back on the rubber bush, now if you draw an imaginary line through the wheel centres you would find that it has the same effect as that old go-kart you made as a kid with scrap wood etc....... now the other wheel gains grip and off you go the other way. If your suspension is rose jointed A) it doesn't flex so much and B) as you have less toe out 1/4 to 1/2 degree works well the wheels are almost straight ahead anyway so even if it does lose a little traction (as it will) it will still go straight ahead, ok so with 200bhp and sticky tyres you will get torque steer but it is manageable the upside to all of this is that braking from speed is less ragged the front end doesn't dart from pothole to pothole....... :lol:


Hi

I have same problem with steering. I have Pro-motive R1 kit FWD with Quaife ATB limited slip differential and 10 inch tyres. What angles should i use for camber, castor and toe and also for rear wheel i have adjustable camber?

#11 fastcarl

fastcarl

    Speeding Along Now

  • Members
  • PipPipPipPipPip
  • 344 posts

Posted 28 November 2008 - 10:37 PM

set 1 deg camber , 5 degs castor, and set toe 30 mins out , you really need rod end bott suspension, although you shouldn't suffer with any torque steer with an R 1 enginse as thay don't produce any,lol.


carl

#12 mike.

mike.

    Crazy About Mini's

  • Members
  • PipPipPipPipPipPipPipPipPipPip
  • 6,176 posts

Posted 18 February 2009 - 02:57 PM

hehe Speaking of torque steer:

0RlktUKNJq0

#13 Ethel

Ethel

    ..is NOT a girl!

  • TMF Team
  • PipPipPipPipPipPipPipPipPipPipPip
  • 25,379 posts
  • Local Club: none

Posted 18 February 2009 - 03:23 PM

He was just trying to warm up his tyres >_<

nice spot of post resurrection though :-




1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users