Car slightly pulls to the left

[string]

You’ve lost me here, I don’t follow what you’re trying to say (but, I suspect I might disagree if I understood you clearly…).

The effect is similar to that of a car in a cross-wind (i.e. an applied lateral force at the centre of gravity). Our example car is traveling down a flat road with a wind blowing from right to left.

The more understeer a chassis is, the more it will yaw to the left in the cross wind. An oversteer chassis will yaw to the right in the wind, and a neutral steer chassis will not yaw at all.

If you take an understeer chassis and instead apply a lateral force further towards the rear of the CG, it will yaw less than if the force was applied at the CG. If you go far back enough you will reach a point where you get zero yaw.

If you instead apply the force ahead of the CG, the chassis will yaw even more to the left than the first example. Go far enough forward that you are now at the front track and imagine that instead of the cross-wind applying a force, the front tyres do due to geometry changes under power application. The greater the car is understeer, the greater the yaw (felt as torque steer) from the tyre force.

If you modify your understeering FWD's suspension such that the rear track's slip angles are greater per unit of lateral acceleration, and the front track's slip angles are less per unit of lateral acceleration*, the point at which a side applied force gives zero yaw (the neutral steer point) is moved forward towards the CG. To get the same yaw to the left as the understeer example, you need to apply the force much further forwards towards the front of the vehicle - the same lateral tyre force at the front track results in less lateral acceleration and less yaw. Forces at the front track have a harder time upsetting the vehicle in this configuration.

Summary: The more understeer your chassis, the more torque steer.

I hope I've made myself more clear. Feel free to pick up my mistakes. Automotive dynamics is not my direct field of study, simply a hobby.

* There are many ways to do this. It is not impossible, or even difficult to produce a neutral steer chassis with a forward CG - this configuration does not have to mean perpetural understeer.

[JohnL]

Awesome, so mine could be simply my tyres then i think... mines front wheel drive but had new tyres put on the front when i got it but not the back cos they were alright and the pulling wasnt that bad and wasnt wheel alignment... but since then its gotten worse and my back tyres are now pretty ****ed.. would that effect it even tho my car is front wheel drive and its the back tyres that are ****ed?

You say your steering pulls left under accelerative forces, and pulls right under braking forces. This doesn't sound to me like the problem is at the rear end. It does very much sound like it is a front end problem. If it were a rear end problem then there would be no difference between coasting (no acceleration or braking) and accelerating. Braking could cause a rear end problem to cause a steering pull if the braking force caused some part of the rear suspension to dynamically misalign (which would be caused by a worn bush or bushes).

Note too that a pull such as you describe might be caused by worn suspension articulations in the front end.

If your front camber is significantly unequal then this could cause the effective scrub radius to be different enough that the tractive forces at the contact patches would be leveraged differently enough that you might feel a steering pull when accelerating and an opposite directioned pull when braking.

If I'm barking up the right tree, then in your case (with a left pull under accceleration and a right pull when braking) I would expect to find that the left side had more neg camber than the right, and as a result the left front would have greater effective scrub radius then the right side. Note that this would be negative scrub radius, not the more usual positive SR. Note too that even if the nominal SR is positive, the effective SR can easily be negative (especially when more neg camber is used), but also that the effective SR can fluctuate substantially and rapidly, and more so the wider the tyre's tread is (and the stiffer the tyre casing, the higher the tyre pressure, and the more uneven the road surface, which is why wide stiff sidewalled tyres at a higher pressure on a rough road cause a lot of very variable forces to be felt at the steering wheel).

Note also that the wider the tread and the stiffer the tyre casing etc, the more sensitive the car is likely to be to unequal camber angle.

Lets imagine a car with an extreme case of this same unequal front camber on each side. For ease of visualisation lets assume very high tyre pressures (that disallow any significant vertical tyre flexure with loading), and let's also assume that the right side has zero camber and SR (in theory creating zero steering pulls related to SR).

In this case the right side will be carrying it's load across the full width of the contact patch and on very smooth surfaces no tractive forces would be leveraged into the steering axis on that side (i.e. acceleration and braking pulls would be zero on that side).

If however the left front has substantial neg camber then it will be carrying most of it's load on the inner edge of the tread, and will have an effective SR equal to nearly half the tread width (extreme hypothetical case).

In this case an accelerative tractive force will cause the centre of the contact patch (this centre being a point defining the average centre of contact patch grip, as well as loading) to attempt to rotate around the steering axis, which it can only do by attempting to move forward in an arc centred on the steering axis.

This will be felt by the driver as a pull to the left under acceleration because it isn't being countered by an equal but opposite 'pull moment' from the right side (which remember has an actual and effective SR of zero). The affect is the same but opposite under brakes, i.e. no affect at the right side but a right pull moment from the left side as the contact patch tries to move backward in an arc centred on the steering axis.

Now, the neg camber on the left side will also create a seperate tendency to pull to the right because of the 'rolling cone' effect. This pull will be a constant, i.e. it will be there all the time regardless of tractive force direction (forward or backward or zero) with any other pull moments being overlayed upon it.

A car that pulls left when accelerating and right under brakes because of unequal camber (creating unequal efffective SR), is not unlikely to have at least a slight pull to the right (from the rolling cone camber effect) when not accelerating or braking. Note that even just travelling at a constant speed requires engine power that might be enough to cause some left pull (through unequal effectiveb SR) that might or might not be enough overcome a tendency to pull right due to the rolling cone effect...

Now, this is all theoretical, it should hold true in practice but the real world is a complex place. Keep in mind that I've been using extreme examples to illustrate the principles involved, examples where we have zero affects on one side and simple strong clear affects on the other. In the real world we will have constantly fluctuating forces acting on both sides, any predominant pulling affects as felt at the steering wheel being created by inequalities in these forces left vs right.

[JohnL]

Awesome, so mine could be simply my tyres then i think... mines front wheel drive but had new tyres put on the front when i got it but not the back cos they were alright and the pulling wasnt that bad and wasnt wheel alignment

Swap front tyres side to side to see if this makes any difference. Don't worry if you cause directional tyres to rotate 'backwards' as this is only an experiment and you can change them back if it makes no difference (but if it does make a difference then the pull could be related to uneven front tyre wear).

Note that bias steer seems to get worse the more the tyres wear (bias steer being to do with how the steel belts are laid up in the tyre).

... but since then its gotten worse and my back tyres are now pretty ****ed.. would that effect it even tho my car is front wheel drive and its the back tyres that are ****ed?

A rear problem is I suspect more likely to cause the car to 'crab' as it travels in a straight line, more than it's likely to cause a pull (not meaning to implyn that it can't cause a pull). I would expect any pull caused by the rear end to be potentially not as strong as a front problem, and to be a farly constant thing, not particularly changeable with acceleration / braking.

Try swapping rear tyres side to side. Try swapping front to rear and rear to front...

Either way im getting a full new set going up to 17'' so hopefully that sounds like it will fix it?

Personally I think 17" is too big. Not so much that the wheel is too big, but the tyres that are likely to be fitted to them will probably be too low in profile (i.e. not enough sidewall). Too much sidewall is not a good thing, but this does not necessarily mean that as little as possible is...

Keep in mind that If my speculation related to uneven camber and effective SR is correct (if you even have uneven camber / ESR...), then a very low profile tyre will have limited sidewall flexure that may well make the problem worse than it is already is...?

For a road car (and most race cars), I personally think that tyre profiles less than 50 are not a very good idea...

[FastFwd]

crap i really need to get mine looked at hey...going onto the freeway this morning applied some boost got upto speed and then let off the gass and the car felt like it wanted to turn a heavy right on me all of a sudden while it was wet/raining and im doing 100-105.

And on certain roads when they are extemely wet it feels like the front end is just sliding...as if your aqua planing.

[JohnL]

crap i really need to get mine looked at hey...going onto the freeway this morning applied some boost got upto speed and then let off the gass and the car felt like it wanted to turn a heavy right on me all of a sudden while it was wet/raining and im doing 100-105.

Wide tyres? High psi? Stiff / low profile casing? Uneven tyre wear? Slight camber inequality? Uneven road sufaces? Predominant left inclined road camber? A soft bushing (etc)? All factors that might be causing problems for you.

And on certain roads when they are extemely wet it feels like the front end is just sliding...as if your aqua planing.

You might well be...

[FastFwd]

Yes Wide tyres, Yess High PSI, Yes extremely stiff suspension, Not that much uneven tyre wear, yes slight camber, yes uneven road, No left incline road camber, yes maybe soft bushings.

But there has to be a way to get my setup to feel like its ok to put the foot down and it doesnt want to go off into different directions.

Will i be able to achieve this with my setup? or will i always have an issue due to stiff suspension, wide tyres, camber etc.

[string]

What are your spring rates?

[JohnL]

Yes Wide tyres, Yess High PSI, Yes extremely stiff suspension, ................... yes slight camber,

Slight neg camber is a good thing, but not too much for a road car, especially if you're using wider tyres. Mac Strut suspended cars tend to require quite a lot of neg camber, but Hondas with SLA suspension don't (IMO it's counter productive to use substantial static neg camber angles with Honda SLA suspensions, far better to use lots of caster angle, e.g. my Accord has -1° static camber and 6+° caster).

But, what you really need to know is whether or not the camber is equal side to side (or at least very close to it), to know this you need the numbers an alignment will give you, but before you align the steering and suspension angles you need to replace any suspect bushes or joints (otherwise there is no point).

I strongly suspect that a car with wide tyres with stiff casings (sidewalls in particular) at high pressure will tend to be more sensitive to unequal (side to side) alignment settings than narrower tyres with softer cases at lower pressures. Stiff spring and damper rates are not unlikely to add to this sensitivity.

Have you increased the caster angle? I also suspect that greater caster angle also tends to make the car more sensitive to unequal alignmnet settings.

(Note that an increase in caster angle is very likely to introduce significant to substantial bump steer that should be rectified because it causes 'roll oversteer', which makes the car unstable from turn-in to mid corner, at least it did with my CB7 until I corrected the bump steer.)

yes uneven road, No left incline road camber, yes maybe soft bushings.

When you pay attention to it, you'd be surprised how often the road is at least slightly inclined to the left. You need to replace any suspect suspenion components. You shouldn't really be complaining about steering pulls etc unless you are sure that your suspension is in good order re bushes etc.

But there has to be a way to get my setup to feel like its ok to put the foot down and it doesnt want to go off into different directions.

See above...

Will i be able to achieve this with my setup? or will i always have an issue due to stiff suspension, wide tyres, camber etc.

You should be able to, but you need to know that the basics are good before you start trying to fix a problem that might be caused by something rather simple. Make sure all your articulations are good and know your alignment numbers, then get back to us if the problem still exists...

[string]

Caster comes at the cost of steering feedback. With 6 degrees can you even feel what your tyres are doing?

[JohnL]

Caster comes at the cost of steering feedback. With 6 degrees can you even feel what your tyres are doing?

Yes, I find very much so compared to the 'feedback' I had at a much lesser caster angle.

Caster creates steering feel in it's own right, but it does tend to overwhelm another source of steering feel. This other source is the reduction in steering 'weight' that occurs as the front tyre nears it's limit of lateral adhesion, this loss of 'weight' being caused by a reduction in 'pneumatic trail' as the grip falls away.

"Pneumatic' trail is created by rolling tyre deformation, whereby the contact patch actually moves backward relative to the wheel (and the rest of the tyre) in elastic response to the rolling resistance inherant in the tyre. Note that this reduction in pneumatic trail is less significant with stiffer tyre casings (lower profile / stiffer sidewalls) or wider treads (because a wider contact patch will be a shorter contact patch), or at higher pressures (which effectively increases casing stiffness).

Pneumatic trail is very significant with cross-ply tyres because their tread isn't very stiff (and has high rolling resistance), but much less significant with radial tyres because their tread is so much stiffer (with much lower rolling resistance). This is why old cars originally designed to use cross ply tyres typically require a significant increase in 'mechanical trail' when fitted with radial tyres (achieved with a caster angle increase), otherwise the steering tends to be vague and wandery because total trail (i.e. pneumatic + mechanical) is insufficient for steering stability (due to the lesser rolling deformation of the radial tyre).

Anyway, this 'at the limit' steering feel is caused by an effective reduction in pneumatic trail as the tyre starts to lose grip, and it's 'trail' that is the operative word here. This kind of feel occurs at / near the limit of adhesion, but the steering has feel / feedback / communication well below this and this is where the feel created by the trail associated with caster largely occurs.

Note that the feel created by caster angle is actually (mostly) to do with the 'mechanical trail' that is created by caster angle. Trail is in effect a lever arm that feeds the lateral forces generated at the contact patch into the steering axis (as a 'torque' around the steering axis), and vice versa.

When the driver steers the car he/she is in effect pushing against the contact patches through the 'lever arm' of trail, and this causes the contact patch to move in an arc around the steering axis (which it wouldn't do if trail were zero), which is resisted through the trail and thus creates 'weight' in the steering that is felt as the tendency to 'self-centre'. (steering 'weight' is also created by caster angle causing the car mass to be lifted above ride height with any steering input, but this is a more or less seperate thing).

So now, when trail exists and more so with greater trail, because it's the contact patch that has the grip, steering inputs alone (i.e. lateral accelerations created by force and inertia are not yet heavily involved) cause a lateral force to act upon the chassis (i.e. causes the contact patch to exert a lateral force against the chassis through the trail), and the front of the chassis will then move inward in the steered dirction (this action actually sharpens up steering response with higher caster / trail). Then, in response to the onset of lateral acceleration (i.e. lateral 'G force') the inertia of the car mass then 'pushes' back against the contact patch through the trail lever arm, and the more lateral accelerartion there is the heavier the steering will become, until we approach and then exceed the limits of lateral adhesion (these effects are stronger with greater trail).

Now, when we reach / begin to exceed the limit's of adhesion the grip starts to diminish, and so too does the lateral acceleration. This means that the steering will start to lighten up (because the force being fed into the steering axis is reducing), providing some degree of 'on the limit' steering feel, i.e. the driver detects a lightening of the steering that is interpreted as increasing grip loss, in much the same manner as the driver interprets lightening of the steering due to reducing pneumatic trail as increasing grip loss.

Note that this effect operates with a fixed trail and a change in the force leveraged through that trail, whereas the pneumatic steering feel operates through a changing effective trail value as well as a change in contact patch force as leveraged through that changing trail.

So, we have (at least) two sources of steering feel here. At higher levels of trail the 'mechanical feel' is stronger and tends to dominate the pneumatic feel, making subtle changes in the steering weight caused by changes in pneumatic trail harder for the driver to detect.

In some instances the mechanical feel may be quite good so it may be no great loss to lose much of the pneumatic feel (i.e. 'lost' in the stronger background feel coming from high levels of mechanical trail). In other instances the mechanical feel may not be so good (i.e. might be too abrupt), so losing the pneumatic feel may cause the steering to become too un-communicative approaching the limit of adhesion.

Steering feel is a hugely complex interplay of forces / resistances, and a very great deal of what any change in the geometry might have will depend on the tyres that are being used....

[string]

Now, when we reach / begin to exceed the limit's of adhesion the grip starts to diminish, and so too does the lateral acceleration. This means that the steering will start to lighten up (because the force being fed into the steering axis is reducing), providing some degree of 'on the limit' steering feel, i.e. the driver detects a lightening of the steering that is interpreted as increasing grip loss, in much the same manner as the driver interprets lightening of the steering due to reducing pneumatic trail as increasing grip loss.

The tyre aligning torque from pneumatic trail peaks well before the maximum lateral grip does - with pneumatic trail the driver can know about the grip limit well before he or she exceeds it.

[JohnL]

The tyre aligning torque from pneumatic trail peaks well before the maximum lateral grip does - with pneumatic trail the driver can know about the grip limit well before he or she exceeds it.

Yes, so assuming that you do have tyres that create a significant degree of pneumatic trail as they roll, then they should give plenty of user freindly warning of impending grip loss. Such a tyre would tend to have a softer casing, be run at lower psi, and have a wider but longer contact patch.

This isn't to imply that a tyre with a stiffer casing, higher pressure and a wider but shorter contact patch won't exhibit pneumaitc trail characteristics, but they would be significantly less. Note that it's typical for high performance and racing tyres to provide substantially less 'warning' to the driver than 'low' performance tyres.

Keep in mind that with increased caster angle (i.e. increased mechanical trail), any pneumatic trail affects don't actually disappear (or even lessen in absolute terms), they just become less pronounced to the driver against the larger unchanging constant of the mechanical trail. The driver may need to be more sensitive / attuned to feel the less pronounced loss of steering 'weight' from the lessening pneumatic trail approaching the limit, but it's still there and still useful to the driver.

A great many racing cars use substantially high caster angle because of the desirable steered camber changes it provides, and the improved steering response. I've heard figures quoted (bandied about?) of up to 20° caster for some V8 Supercars. Though I'm not certain as the veracity of this number, these cars do use a spool differential and so may need huge caster angle to assist in jacking the inside rear tyre, like a kart does (if so then this is outside the scope of this dicussion). The cars are however still driveable on the limit, so must retain adequate steering feel despite the huge caster angle (a lot less than 20° is still huge...).

Note that while my car now has 6° of caster angle, despite this being nearly 3X the stock caster it isn't really a great deal relative to some cars (10° isn't unknown). Golf GTIs and Porsche 911s have about 7° from memory. I find my car's increased caster angle has substantially improved steering feel / communication at sane cornering speeds (i.e. not at the limit of adhesion), as well as on centre feel and steering response. On the limit the feel is fine, though it's hard to say if it's any better or worse than it was at the stock caster angle (I don't spend much time at 10/10ths...).