47 replies to this topic

[meb58]

I recently had the opportunity to drive my old 2005 JCW Mini back to back with a current version of the same car. I came away shaocked by the differences in steering feel and torque steer. The older car has pretty good steering feel nearly no torque steer. The newer version has better steering feel right up until torque steering destroys any of those self aligning torques...and all goes knumb. The newer car's turbo devlops 15 addition lb ft of torque...I cannot imagine 15 lb ft cause such a dramatic difference in torque steer.

My old car had slightly different axle lengths but the longer axle was supported by half shaft bearing making both ends identical in length from the half shaft to the center of the wheel bearing.

Is this purely the difference between degrees of offset (less positive offset) and or scrub radius?

[Greg Locock]

I recently had the opportunity to drive my old 2005 JCW Mini back to back with a current version of the same car. I came away shaocked by the differences in steering feel and torque steer. The older car has pretty good steering feel nearly no torque steer. The newer version has better steering feel right up until torque steering destroys any of those self aligning torques...and all goes knumb. The newer car's turbo devlops 15 addition lb ft of torque...I cannot imagine 15 lb ft cause such a dramatic difference in torque steer.

My old car had slightly different axle lengths but the longer axle was supported by half shaft bearing making both ends identical in length from the half shaft to the center of the wheel bearing.

Is this purely the difference between degrees of offset (less positive offset) and or scrub radius?

There are those who claim that torque steer is thoroughly understood, those who think you can get rid of most of it with appropriate actions, and those who think it is black magic. I'm in the second camp. That is, once you exceed a certain torque to weight ratio (or something) the car will be squirrelly even if its less powerful relative isn't in the slightest.

My guess is that the contact patch is doing some odd things so your scrub radius isn't stable (if it were equal on both sides it should have no effect). Narrower tires might help.

[meb58]

Both cars are pretty light relative to their competition...~2,600 lbs. So that combine with the extra 15 lb ft torque lower in RPM band could aggrivate torque steer. I also think that turbo has a temporary over-boost option. I hadn't given more narrow tires a thought.

[Magoo]

Black magic indeed.

In some cars torque steer can be produced simply through an unfortunate stackup of assembly tolerances. Loosen up all the drivetrain and cradle mounts, let everything sorta dangle and get all casual and relaxed where it wants to be, then retorque everything to specs. Voila, torque steer goes away.

Extreme case: I owned a series of front-drive Cadillacs with the 300hp Northstar transverse V8 that would drive nice on new rubber, but once the tires got around 18,000 miles on them, these cars would torque steer like a bastard. GM fixed them by softening the engine calibration, essentially. It wasn't dangerous or disturbing, only annoying.

[gruntguru]

Another thing to consider is differential action. An open diff guarantees identical torque in each axle and consequently similar longitudinal force at each contact patch. Throw in a LSD and things suddenly change. Perhaps the turbo car has a different differential setup?

[Greg Locock]

Another thing to consider is differential action. An open diff guarantees identical torque in each axle and consequently similar longitudinal force at each contact patch. Throw in a LSD and things suddenly change. Perhaps the turbo car has a different differential setup?

Good one. Thinking about how the friction in a diff affects the load split, perhaps there is a simple explanation after all.

[bigleagueslider]

Another thing to consider is differential action. An open diff guarantees identical torque in each axle and consequently similar longitudinal force at each contact patch. Throw in a LSD and things suddenly change. Perhaps the turbo car has a different differential setup?

Excellent point. Think about how much understeer/oversteer effect an oval track car with a spool and tire stagger experiences, just from driving the rear wheels.

[Lee Nicolle]

I will probably start a furore but all front drive cars are inherently dangerous. The majority are ok in normal circumstances. But as Magoo says when the tyres get old they drive like crap, and sometimes fall off the road.I drive them but will never like them. or drive them very hard.
Interestingly I recently bought an 04 IFS Landcruiser. Compared with the beam axle one it torque steers badly, and the self centreing on the steering is poor though I think the toe is out. This is a common front drive trait though.

[Magoo]

I will probably start a furore but all front drive cars are inherently dangerous. The majority are ok in normal circumstances. But as Magoo says when the tyres get old they drive like crap, and sometimes fall off the road.I drive them but will never like them. or drive them very hard.
Interestingly I recently bought an 04 IFS Landcruiser. Compared with the beam axle one it torque steers badly, and the self centreing on the steering is poor though I think the toe is out. This is a common front drive trait though.

Lee is what you call old school. One of these days we need to get his views on overhead valves and tubeless tires.

[Magoo]

Lee is what you call old school. One of these days we need to get his views on overhead valves and tubeless tires.

One of the things that makes this forum great is the wide variety of knowledgeable perspectives. People don't always agree and that makes it interesting. And useful. And stimulating of further thought.

[meb58]

Yes is does and my old car did not. I know very little about LSD variables...I know what they are 'supposed' to do or 'sold' to do. Both cars also have traction control and I believe the brakes are used to limit loss of traction. So lots going on there.

Lee Nicolle,

I've driven lots of front drivers on the track and the Mini was by far the most predictable and stable of the bunch. It's 'cute' looks aside, it was a pretty good accomplice on track. I often had an opportunity to trade off with a model year 2000 BMW M3 (full roll cage), back to back with the Mini. I found it much more fluid and predictable at somewhat higher limits. Still, you'd have to be a sleep to cause either to do something unintentional. I would take the M3 if I had a choice, to your point.

Another thing to consider is differential action. An open diff guarantees identical torque in each axle and consequently similar longitudinal force at each contact patch. Throw in a LSD and things suddenly change. Perhaps the turbo car has a different differential setup?

Edited by meb58, 06 March 2013 - 15:40.

[sharo]

...I cannot imagine 15 lb ft cause such a dramatic difference in torque steer.

Maybe the gearing and engine speed/torque characteristics also contribute to amplifying the effect?

[Dipster]

One of the things that makes this forum great is the wide variety of knowledgeable perspectives. People don't always agree and that makes it interesting. And useful. And stimulating of further thought.

I do not know Lee. I have read many of his posts though. I suggest that he is a bit like my old Dad. That is experienced and knowledgeable. I have been around and worked on cars for many decades. But I will never know what my Dad forget and certainly do not have his machining skills. It appears to me that Lee has experience perhaps similar to those of my Dad.

My point is that, in my opinion, it is always useful to listen carefully to the utterings of those with experience. They may not always be right, but often are.

I agree that many FWD cars are horrid to drive. Torque steer can be downright dangerous. But I have also driven FWD cars that were a delight. But they do, in my experience, due to their comparitive (to RWD) complication of their front ends, tend to get a bit sloppy with age.

Personally I prefer 4WD on road vehicles (and tubed tyres when I can!).

[Powersteer]

I wonder if more castor will help as it might take away the pivoting effect of one side of the tyre to the other while also using the other side's castor hold steering.

[meb58]

I thought about that, but caster can also reduce what I might incorrectly call 'mechanical leverage' during turn-in. And the torque steering is so bad in the newer JCW Mini on simple highway ramps that adding enough castering effect might not allow it to turn at all:)

Edited by meb58, 06 March 2013 - 17:39.

[Catalina Park]

Lee is what you call old school. One of these days we need to get his views on overhead valves and tubeless tires.

Also disc brakes and radial tyres as well.
After we sort out that mess we could move on to separate chassis vs unit construction.

[Lee Nicolle]

Yes I do use radial tyres, without tubes preferably. And disc brakes too. Though as a rear brake on a road car they are perfectly fine. A drum brake is inherently more efficient, the same amount of work for a far smaller volume of fluid. If they could just get rid of heat!!
Front drivers dont like LSDs, they dont want to turn with them,, pretty good off the corner though. As for stability, hoo boy. Mid corner you have to keep your foot on the gas or they go straight ahead. Many years ago I raced against some very good Minis [real ones not BMWs] Cross flows trick rear axles etc and got clouted a couple of times,, and basically worried them off too. They did not seem to like caster either, made them torque steer worse. No caster, no camber, heavy steering and no rear tyre heat. Great balance for motorsport!
I damn near killed myself many moons ago, torque steering early Lasers [Mazda 323s] They do not go where they are placed, partly my fault but a dangerous thing at high speed. Like 140kmh!
Many modern cars are not much better, I drove a late model 2 litre? Corrolla manual recently and when I gave it a squirt it went right and when I jumped off the gas it went left. I guess I will have to retire to boredom machines. Camry Autos!! Oh what a feeling,,, snore!

[Greg Locock]

Corrolla manual recently and when I gave it a squirt it went right and when I jumped off the gas it went left.

That's classic straight line torque steer and only exists because of bean countery or incompetence or 'just don't care' these days. Given the large differences in how cars are developed, and their targets, at different companies, it is often a puzzle to work out whether Toyota are smarter than us, or just more cynical, or I suppose, both. If the guy buying the car new doesn't give a monkey's about torque steer in a straight line from a relatively low performance car, why spend money preventing it? That's 20 dollars (guessing they've got unequal length halfshafts)!

Far more exciting is what the Americans call skate, as you go round a corner with a rough surface the live rear axle walks towards the outside of the corner as you apply power (or even without). Very similar vehicles from different companies will display markedly different tendencies, depending on whether they've done enough work on it and got the architecture right.

[kikiturbo2]

well, modern european hot hatches, such as the renault megane / clio, focus RS, etc, and some not so hot FWD's, have really trick front suspension designs... mainly to separate the steering geometry from front mcpherson struit geometry.... mainly to tame the torque steer with stable small scrub radius..
Personally, I like to have really good steering feel, which is something I am missing these days... and for good steering feel I will tolerate some torque steer, which can provide good info, IMHO..
One thing I do not like are FWD's with plate diffs, such as Mazda 3 MPS (turbo 2.3) which can be really dangerous if you are not commited all the time..

Helical diffs in renault and ford are much nicer..

[gruntguru]

That's 20 dollars (guessing they've got unequal length halfshafts)!

From the first time i saw a FWD arranged to equalise the half-shaft lengths, I have wondered how this fixes/improves torque steer. MOI is still uneven side-side and the difference is minute anyway. So what's the theory behind equal length half-shafts?

[Lee Nicolle]

Odd length drive shafts is a product of cut price engineering. It is easier and cheaper to build the cars as such. Then you have to have a mount, usually hanging off the engine to try and equalise the driveshafts. It does work to a degree. And can also cause more greif when having to work on the driveshafts. A pet hate, those stupid wire circlips that often stop the driveshafts coming out of the gearbox or the carrier, That can turn a 1/2 hour job into a filthy dirty epic. That and torn CV boots, again a filthy greasy job.

All these jobs that you do not get with a rear drive car.

The life of a outer CV on a front drive is maybe 80000km. If the driver nudges the gutter regularly halve that. The life of a CV on a rear drive IRS is usually life time unless the driver has a heavy foot, and or hits gutters regularly. Personally I have never done one, or a CV boot either. Front drive it is just part of a service!

Front drive = cheaper to build, far dearer to maintain. Far less stable to drive, as the vehicle is inherently unbalanced. The front tyres do all the work and the rears just follow. Usually coupled with the dreaded cheap and nasty McPherson struts just adds to the tyre wear. And very premature old age. The only maybe plus is good traction, at least on the front in inclement conditions, that is until the toque steer spits you off the road.

[gruntguru]

You missed the key benefit of FWD - packaging. Compare the Mini to the Morris Minor and it becomes obvious.

[Kelpiecross]

From the first time i saw a FWD arranged to equalise the half-shaft lengths, I have wondered how this fixes/improves torque steer. MOI is still uneven side-side and the difference is minute anyway. So what's the theory behind equal length half-shafts?

I have to agree - it makes no sense to me either. But so many car makers do arrange the half-shafts to be equal length - surely there must be some advantage in doing this. Or are they just copying each other without really knowing why they do it?

[Kelpiecross]

Odd length drive shafts is a product of cut price engineering. It is easier and cheaper to build the cars as such. Then you have to have a mount, usually hanging off the engine to try and equalise the driveshafts. It does work to a degree. And can also cause more greif when having to work on the driveshafts. A pet hate, those stupid wire circlips that often stop the driveshafts coming out of the gearbox or the carrier, That can turn a 1/2 hour job into a filthy dirty epic. That and torn CV boots, again a filthy greasy job.

All these jobs that you do not get with a rear drive car.

The life of a outer CV on a front drive is maybe 80000km. If the driver nudges the gutter regularly halve that. The life of a CV on a rear drive IRS is usually life time unless the driver has a heavy foot, and or hits gutters regularly. Personally I have never done one, or a CV boot either. Front drive it is just part of a service!

Front drive = cheaper to build, far dearer to maintain. Far less stable to drive, as the vehicle is inherently unbalanced. The front tyres do all the work and the rears just follow. Usually coupled with the dreaded cheap and nasty McPherson struts just adds to the tyre wear. And very premature old age. The only maybe plus is good traction, at least on the front in inclement conditions, that is until the toque steer spits you off the road.

I drove Minis of various types for many years (including a small amount of racing) and I like FWD handling characteristics. I think FWD is far more stable and forgiving - if you arrive at a corner too fast - you back off the accelerator, the car goes a bit sideways and washes off speed - then back on the power to hold the car in that attitude and maybe wind on the lock to counter the understeer. RWD drive is the opposite (so they tell me) - arrive too fast, back off power, then opposite lock etc. - opposite lock I always thought was a bit "unnatural" in having to turn the steering wheel in the opposite direction to the corner.

I suspect these self-correcting/foolproof handling characteristics gave Minis their remarkable competition record. Even though RWD is actually ultimately probably faster in cornering speed, FWD's more foolproof handling allowed the power to be kept on for a greater percentage of time in a race than RWD and thus be faster overall - whereas the RWD driver always had to be more circumspect in attacking corners etc. to avoid going off/spinning etc. especially in the wet/gravel etc.

As for FWD straightline traction - I always found it woeful, presumably due to weight transfer when accelerating. I have never seen another 55 horsepower car (as my Mini was) lay down so much rubber (and go nowhere much) in a full-RPM-drop-the-clutch type of start.

I think that four-wheel-drive is probably the best handling solution - same safe FWD foolproof handling combined with an excess of traction.

Having said all the above - when had my first real drive in a RWD car (a Datsun 1200 Coupe) I was amazed to drive a car that actually went where you asked it to without arguing/darting all over the place - but overall the Mini was more predictable in handling.

[Greg Locock]

I suspect the main benefit from equal length halfshafts has nothing to do with length, and everything to do with 3 dimensional angles. The layshaft means that the outboard CVs sees the same angles.

If you put a torque through a CV with an angle something has to supply a non axial torque to make the shaft torque go round the corner.

There's also the slightly mysterious requirement to make sure that the torsional stiffness of each of the two sides, from diff to wheel, is the same. I've nver done the experiment. That means the layshaft would ideally be very big, if the halfshafts are the same OD and ID and length.

[gruntguru]

I suspect the main benefit from equal length halfshafts has nothing to do with length, and everything to do with 3 dimensional angles. The layshaft means that the outboard CVs sees the same angles.

If you put a torque through a CV with an angle something has to supply a non axial torque to make the shaft torque go round the corner.

Thanks Greg. I hadn't considered that.

[Lee Nicolle]

I suspect the main benefit from equal length halfshafts has nothing to do with length, and everything to do with 3 dimensional angles. The layshaft means that the outboard CVs sees the same angles.

If you put a torque through a CV with an angle something has to supply a non axial torque to make the shaft torque go round the corner.

There's also the slightly mysterious requirement to make sure that the torsional stiffness of each of the two sides, from diff to wheel, is the same. I've nver done the experiment. That means the layshaft would ideally be very big, if the halfshafts are the same OD and ID and length.

Sounds very feasible. The different CV angles would not improve stability at all, hence less torque steer.

Front drive cars all suffer dramatically from mounts that are less than satisfactory, everything walks around. Drive over speed humps and you are sure everything up front is ready to fall on the ground.That was my first test when I used to buy a lot of cars, drive them overthe inevitable local humps,often on the premises, a hard right and left turn. And then price accordingly.

[Lee Nicolle]

I drove Minis of various types for many years (including a small amount of racing) and I like FWD handling characteristics. I think FWD is far more stable and forgiving - if you arrive at a corner too fast - you back off the accelerator, the car goes a bit sideways and washes off speed - then back on the power to hold the car in that attitude and maybe wind on the lock to counter the understeer. RWD drive is the opposite (so they tell me) - arrive too fast, back off power, then opposite lock etc. - opposite lock I always thought was a bit "unnatural" in having to turn the steering wheel in the opposite direction to the corner.

I suspect these self-correcting/foolproof handling characteristics gave Minis their remarkable competition record. Even though RWD is actually ultimately probably faster in cornering speed, FWD's more foolproof handling allowed the power to be kept on for a greater percentage of time in a race than RWD and thus be faster overall - whereas the RWD driver always had to be more circumspect in attacking corners etc. to avoid going off/spinning etc. especially in the wet/gravel etc.

As for FWD straightline traction - I always found it woeful, presumably due to weight transfer when accelerating. I have never seen another 55 horsepower car (as my Mini was) lay down so much rubber (and go nowhere much) in a full-RPM-drop-the-clutch type of start.

I think that four-wheel-drive is probably the best handling solution - same safe FWD foolproof handling combined with an excess of traction.

Having said all the above - when had my first real drive in a RWD car (a Datsun 1200 Coupe) I was amazed to drive a car that actually went where you asked it to without arguing/darting all over the place - but overall the Mini was more predictable in handling.

A 69BHP 1200 is hardly a sports car, in fact handle like a wet sponge on their tiny 12" tyres and weak Mac struts. They can be made acceptable but that means replacing most of the suspension, then finding 80hp.
But yes they do at least handle without all the twitches and wobbles of a front driver.
The old Minis competition record is because the Brits had such a woefull collection of cars it made the Mini look respectable,, and a Cooper S is no slouch even now. But the only real competition it had was a Mk 1 Cortina in at least 1500GT standard. Once the twincam arrived it was all over for BMC.

[murpia]

I suspect the main benefit from equal length halfshafts has nothing to do with length, and everything to do with 3 dimensional angles. The layshaft means that the outboard CVs sees the same angles.

If you put a torque through a CV with an angle something has to supply a non axial torque to make the shaft torque go round the corner.

There's also the slightly mysterious requirement to make sure that the torsional stiffness of each of the two sides, from diff to wheel, is the same. I've nver done the experiment. That means the layshaft would ideally be very big, if the halfshafts are the same OD and ID and length.

I don't think the torsional stiffness requirement has been met on any euro FWD car I have seen.

Then again, I don't believe it to be necessary with a differential in the driveline.

But the equal length / equal angle driveshafts is important. Once we had a cost-reduced 'hot hatch' on the rolling road (with unequal-length / unequal angle driveshafts). We had to wind the windows down and ratchet strap a 2" x 4" to the steering wheel, braced against the door tops, to keep it on the rollers...

Regards, Ian

Edited by murpia, 08 March 2013 - 13:43.

[rory57]

The FWD Lotus Elan M100 has been said to be as good as it gets for front drive steering/handling: any clues there? With the wide and ultra-low profile tyres now common, there may be a migration of the effective contact patch across the tyre width under roll. This might change the "leverage" felt between steering wheel and tyre slip angle. Bring back centre point steering? Time for real, powered, roll cancellation?

[Lee Nicolle]

I don't think the torsional stiffness requirement has been met on any euro FWD car I have seen.

Then again, I don't believe it to be necessary with a differential in the driveline.

But the equal length / equal angle driveshafts is important. Once we had a cost-reduced 'hot hatch' on the rolling road (with unequal-length / unequal angle driveshafts). We had to wind the windows down and ratchet strap a 2" x 4" to the steering wheel, braced against the door tops, to keep it on the rollers...

Regards, Ian

Not an uncommon thing with front drivers and some 4wd cars too. My local dyno shop reckons it is harder to tie down a 200hp front drive than a 600hp rear drive.
There was something like that somewhere on You Tube, a kiddy car twisted itself off the rollers, conversly Pro Stock cars seemed quite stable. There is a whole pile of dyno runs on You Tube and a few disasters too, cars escaping and exploding. A good way to spend a bored hour!

[Kelpiecross]

[quote name='Greg Locock' date='Mar 8 2013, 14:55' post='6156534']
I suspect the main benefit from equal length halfshafts has nothing to do with length, and everything to do with 3 dimensional angles. The layshaft means that the outboard CVs sees the same angles.

If you put a torque through a CV with an angle something has to supply a non axial torque to make the shaft torque go round the corner.


Just what do you mean here (in words of one or maybe two syllables)? A CV at an angle tries to straighten itself out? A CV at an angle causes a force at 90 degrees to the first angle (and thus a self-steering force)?

It is difficult for me to picture how a difference of a few degrees from side-to-side in the CV angles could cause such powerful self-steering effects. I have driven FWD cars with rock-hard suspension (felt like none at all) where the CV joint angle variations from side-to-side must be really minimal yet they still darted about on acceleration.

What about that F1 car (a Williams perhaps) with the very steep permanent uphill angles on the driveshafts (presumably there is a CV joint at each end of the shaft) - what direction would this try and twist the wheels?

Edited by Kelpiecross, 09 March 2013 - 03:38.

[Catalina Park]

Can you learn to quote properly? Just type your dribble after all the stuff in the reply box and don't chop off the quote tag.

[275 GTB-4]

and anyway, apart from issues with maybe unsprung weight, and ground clearance...why do we need CV's? Couldn't universal joints be made just as good? (you might get a little binding on tight turns )

[Kelpiecross]

Can you learn to quote properly? Just type your dribble after all the stuff in the reply box and don't chop off the quote tag.

Up yours, Knobhead.

Maybe you can explain all about the CV's Knobhead?

Edited by Kelpiecross, 09 March 2013 - 10:13.

[Tony Matthews]

Up yours, Knobhead.

"Up yours, Knobhead." A new series of culture programmes, featuring the latest in the Arts, Literature and Music. Hosted by a number of intellectual giants in their field, it will concentrate on the essence of progress in Mankind's seeking forArtistic Truth. See "Pick of the Day", page 47.

[Kelpiecross]

"Up yours, Knobhead." A new series of culture programmes, featuring the latest in the Arts, Literature and Music. Hosted by a number of intellectual giants in their field, it will concentrate on the essence of progress in Mankind's seeking forArtistic Truth. See "Pick of the Day", page 47.

Thank you for that. I will look forward to seeing this new programme.

To explain slightly further my annoyance - if CP uses the somewhat insulting word "dribble" to describe my questions about CV's/torque steer etc. - this implies to me that he is so intelligent and well-informed that it should be a simple task for him to explain in plain language the relationship between CV angles and torque steer etc.

Edited by Kelpiecross, 09 March 2013 - 12:01.

[rory57]

and anyway, apart from issues with maybe unsprung weight, and ground clearance...why do we need CV's? Couldn't universal joints be made just as good? (you might get a little binding on tight turns )

CV joints are used rather than UJs because UJs cause a variation in the relative velocity of the shafts either side of the joint four times per revolution. Constant Velocity joints do not do this which is why they are always used.
This cyclic variation would cause unacceptable vibration in the FWD outboard position because A) the angles across the joint can be quite large and B) because the inner joint angle is not the same as the outer joint angle. UJs are ok in pairs on rear axles because the inner and outer angles can be arranged to be the same and in any case are less extreme than at the front. The inner and outer UJ must be positioned at 180 degrees to one another so as to cancel out the cyclic velocity variation. -----

Edited by rory57, 09 March 2013 - 13:39.

[gruntguru]

The old Minis competition record is because the Brits had such a woefull collection of cars it made the Mini look respectable

So the Monte Carlo rally and Bathurst 500 (Minis won outright in the sixties) were restricted to British cars were they?

[gruntguru]

Just what do you mean here (in words of one or maybe two syllables)? A CV at an angle tries to straighten itself out? A CV at an angle causes a force at 90 degrees to the first angle (and thus a self-steering force)?

If you draw a free body diagram of the CV (ignoring any forces) there is an input torque (along the half shaft) and and an output torque (along the wheel axis). If the CV is operating at an angle the two torques will not be equal and opposite therefore there must be a third torque equal to the negative of the vector sum of the first two torques. (ie the three torques cancel)

If the first two are both perpendicular to the steering axis, the third will also and there will be no torque steer component. In practice this will not be the case so turning the steering wheel and increasing the angle between the half shaft and the wheel axis, increasing the magnitude of the third torque and its component about the steering axis.

Now - back to Mat's wedge problem.

[Kelpiecross]

If you draw a free body diagram of the CV (ignoring any forces) there is an input torque (along the half shaft) and and an output torque (along the wheel axis). If the CV is operating at an angle the two torques will not be equal and opposite therefore there must be a third torque equal to the negative of the vector sum of the first two torques. (ie the three torques cancel)

If the first two are both perpendicular to the steering axis, the third will also and there will be no torque steer component. In practice this will not be the case so turning the steering wheel and increasing the angle between the half shaft and the wheel axis, increasing the magnitude of the third torque and its component about the steering axis.

Now - back to Mat's wedge problem.

Thank you for the explanation. I wouldn't say that I can fully picture in my mind the various torque vectors etc. - but I am at least starting to get an inkling of how it could work.

[Kelpiecross]

and anyway, apart from issues with maybe unsprung weight, and ground clearance...why do we need CV's? Couldn't universal joints be made just as good? (you might get a little binding on tight turns )

I think the prewar "traction avant" Citroens used ordinary universal joints.

[275 GTB-4]

CV joints are used rather than UJs because UJs cause a variation in the relative velocity of the shafts either side of the joint four times per revolution. Constant Velocity joints do not do this which is why they are always used.
This cyclic variation would cause unacceptable vibration in the FWD outboard position because A) the angles across the joint can be quite large and B) because the inner joint angle is not the same as the outer joint angle. UJs are ok in pairs on rear axles because the inner and outer angles can be arranged to be the same and in any case are less extreme than at the front. The inner and outer UJ must be positioned at 180 degrees to one another so as to cancel out the cyclic velocity variation. -----

Thanks Rory...nice explanation

[Kelpiecross]

So the Monte Carlo rally and Bathurst 500 (Minis won outright in the sixties) were restricted to British cars were they?

!966 the Minis won - first half-dozen or so places. '66 was the only year I ever went to Bathurst - it is much easier to understand what is going on when you watch the race on TV - but you lose a lot of the atmosphere of the place. Downhill through the Dipper etc. is unbelieveably steep when you see it in real life.

Back to the CV problem: Is it safe to say that a shaft at an angle causes the CV at the wheel try to "straighten out" under load and the caster angle on the wheel then causes a component of this "straightening " force to act around the steering pivot thus causing torque steer? Without a factor like the caster angle being introduced I don't see how a torque component around the steering axis can arise.

[carlt]

That's classic straight line torque steer and only exists because of bean countery or incompetence or 'just don't care' these days. Given the large differences in how cars are developed, and their targets, at different companies, it is often a puzzle to work out whether Toyota are smarter than us, or just more cynical, or I suppose, both. If the guy buying the car new doesn't give a monkey's about torque steer in a straight line from a relatively low performance car, why spend money preventing it? That's 20 dollars (guessing they've got unequal length halfshafts)!

Far more exciting is what the Americans call skate, as you go round a corner with a rough surface the live rear axle walks towards the outside of the corner as you apply power (or even without). Very similar vehicles from different companies will display markedly different tendencies, depending on whether they've done enough work on it and got the architecture right.

I remember 'skate' in a hot Renault 5 Gordini - the only FWD car I've ever enjoyed driving on the limit - Great Fun

[carlt]

"Up yours, Knobhead." A new series of culture programmes, featuring the latest in the Arts, Literature and Music. Hosted by a number of intellectual giants in their field, it will concentrate on the essence of progress in Mankind's seeking forArtistic Truth. See "Pick of the Day", page 47.

[gruntguru]

Thank you for the explanation. I wouldn't say that I can fully picture in my mind the various torque vectors etc. - but I am at least starting to get an inkling of how it could work.

Torque vectors align with the axis of rotation. So one points along the input shaft, one along the output shaft and the third is in the same plane and closes the vector triangle if all three are arranged "nose to tail". Note that vectors one and two both point towards or away from the CV joint because they act in opposite directions.

Back to the CV problem: Is it safe to say that a shaft at an angle causes the CV at the wheel try to "straighten out" under load . .

No - apart from friction in the joint which may have that effect.

. . and the caster angle on the wheel then causes a component of this "straightening " force to act around the steering pivot thus causing torque steer? Without a factor like the caster angle being introduced I don't see how a torque component around the steering axis can arise.

If the "third" vector is perpendicular to the kingpin axis there will be no torque steer applied via the CV since it has no component acting along the KP axis.

[Kelpiecross]

Torque vectors align with the axis of rotation. So one points along the input shaft, one along the output shaft and the third is in the same plane and closes the vector triangle if all three are arranged "nose to tail". Note that vectors one and two both point towards or away from the CV joint because they act in opposite directions.

No - apart from friction in the joint which may have that effect.

If the "third" vector is perpendicular to the kingpin axis there will be no torque steer applied via the CV since it has no component acting along the KP axis.

Thank you for the additional information (and some assistance from Google) - I see how it works now. "Torque vectors" are not something I have ever encountered before.
Presumably front driving beam axles (as on some 4WD's) don't suffer from "torque vector" problems?