Scrub radius versus grip limit feel

[sierra4000]

Hello All ,

How scrub radius affects steering feel when car approach lateral grip limit?
driver complains for very bad tire response

is FWD car which have really big wheel spacers but only 2 degree caster and 14 degree KPI

he suffers also strong torque steer under acceleration..........is low caster value contribution?

Thank You for opinions

Radek

 

[BrianPetersen]

What is the car? What IS the scrub radius? How much mechanical trail? What tire size? What tires? What's the camber? Open differential, or does it have a limited-slip diff? Equal-length front driveshafts, or not, and if not, how much angle on the CV joints?

I am assuming MacPherson suspension judging by the high KPI angle.

2 degrees of caster is low. If the steering axis (line between lower ball joint and upper strut mount) passes closely through the centerline of the front wheel (which is the usual case) the low caster means you will only have about 10mm of mechanical trail (assuming tires of fairly normal sizes) and that's not much, and *that* will mean there's not much feedback of lateral load through the steering.

Most front-drivers nowadays have small but non-zero scrub radius. Large scrub radius (in either direction) is asking for bad steering reactions if there is unequal grip left to right (and a limited-slip diff would contribute). Too much camber on too-wide tires with insufficient sidewall height (stiff sidewalls) is also just begging for torque reactions.

 

[sierra4000]

Thank You Brian,
I will try more detailed,

is Honda Type R EP3 - MacPherson - with limited slip diff,
tire slick Michelin 17 X 8 low aspect ratio
have equal driveshafts but seems unequal KPI (estimate 1 degree difference), although both caster is same -3 degree.........so I think torque steer start there, from different CV joints angles
also seems some engine inclination

scrub radius value unknown , but 30mm spacers over stock

 

[sierra4000]

excuse my nonsense,
different KPI will affect unequal scrub radius, but CV joints angles is same with same camber.......always
yes?

 

[BrianPetersen]

So non-stock very wide wheels of unknown offset (or known, but you haven't told us).

Figure out the rolling diameter of the wheels that you have and compare that to stock tire size for that car, make sure they are close.

Figure out the net offset including the effect of that wheel spacer and compare that to stock.

High scrub radius (in either direction) + limited slip diff in a front-drive application is gonna be trouble.

Low profile tires are not going to like camber, either. You haven't told us what the camber is.

Why is the KPI different left to right? That's bad. Is something bent?

I have no specific knowledge of that car; the above are generalities of the "good/bad" nature.

 

[sierra4000]

Its not my car but I try find his problems,

Google says: OEM wheel offset 45mm......................now is 38mm -30mm spacers, so in fact 37mm change,
tire diameter is close OEM,
camber is -3 degree and caster +2 sorry for confusion,
KPI is different probably from knuckle/damper connection.............I I see 10mm difference on camber plate (top mount)setting,

also seems some toe compliance problems, that toe changed under longitudinal load
(is usual EP3 steering rack wear problems when tie rod inclination causes some steering rack rotation under load)

so is possible that creates some steering wheel angle change under acceleration to keep straight line?
due to toe change under asymetric load from torque steer?

also I think, toe changes will do bad feel during cornering if acceleration/deceleration changes...........It will given uncertainty for driver that multiplied with bigger scrub radius. yes?

 

[BrianPetersen]

In other words, you are dealing with a helluva mess that is probably somebody else's hack job.

I don't know what the stock scrub radius is but it's probably pretty slightly negative, that's common on front-drive cars nowadays. Going 37mm towards the positive direction is an enormous change.

If it has camber plates for the top strut mount and they're set differently left to right and it doesn't correspond with the actual camber of the wheel, something's bent. Or maybe they've opened up holes where the strut attaches to the spindle or installed different bolts and messed up the way the spindle attaches to the strut.

I suspect that with those wide and low-profile tires, 3 degrees camber is too much. Mind you, that would move the center of contact pressure with the road back in closer to where the steering axis is ... kinda sorta maybe compensating for the excessive positive scrub radius, variably depending on what the road surface is doing.

Bump steer is bad.

Fore/aft compliance of the hub isn't necessarily bad, if the geometry of the lower control arm and the tie rod is correct - which minimizes the steering effects of the fore/aft compliance. (There was another recent thread about why lower control arm bushings are mounted the "wrong" way - read up on that.) But I'm guessing this car has probably been lowered, and there's a good chance those bushings have been replaced with aftermarket, the net result is probably that the geometry and the bump steer and the compliance effects have gotten all messed up.

The Civic Type R is a good car. Honda put plenty of engineering into getting them to work properly straight off the showroom. Then the hacks get involved ...

 

[sierra4000]

Yes Brian, is usual mess and big mix non-standard parts that do not cooperate
includes low ride height with roll center "compensators" so previous bumpsteer was more than +25 degree per meter!!! and car feel very unstable on bracking and transitions behavior

now bumpsteer was corrected, is close zero and car stable under braking and quite good in transitions, but uncertainly steering feel close lateral grip limit and crazy torque steer under acceleration and hard gear shifting up

(although bumpsteer zero, tie rods still maintain some inclination which creates steering rack rotation so tie rod vertical movement on rack)

lower arm have rose joints without any usual compliances..........all main compliance happens only in rack rotation

Yes ,seems spindle attaches to the strut does problem with mutual KPI

tire tread wear is quite equal............previous bumpsteer correction was catastrofic outer side

 

[sierra4000]

for clarification
I think this toe compliance we can call real clearance which changes toe quite significantly already with small longitudinal forces, then spacers will affect probably many

https://www.youtube.com/watch?v=kfqCJcEsnRQ

still more and more I think these wheel spacers is main problem with some others things contribution
spacers works like "enhancer" so after spacers removing problems will be minimize

interesting is also that on wet tires all problems are significantly weaker

 

[BrianPetersen]

Yeah that steering rack is shot. Fixing broken stuff is first on the to-do list.

How much clearance have you got between the inside edge of the tire and the strut?

In my previous car (which had a properly developed Bilstein kit and much wider front wheels than standard) there was perhaps 3 - 5 mm clearance between the tire and the strut, but clearance is clearance, and all you need is for it to not hit!

If the way the strut is attached to the spindle allows camber change (either via eccentric bolts or simply clearance in the bolt holes) then lessening negative camber will move the top of the wheel outward and improve that clearance (and allow wheel spacers to be thinner or hopefully completely eliminated).

If those adjustable upper strut mounts allow fore-aft adjustment (or can be re-engineered to allow fore-aft adjustment instead of side-to-side adjustment) you have a way of getting more caster and therefore more mechanical trail.

If there's a way to shift the rear lower control arm attachment point outward, it'll shift the ball joint forward, that will give you more caster and more mechanical trail.

Less camber, more caster, hopefully no wheel spacers, probably higher ride height (MacPherson doesn't generally like being lowered) or at least make sure the instant centers are in a decent place, hopefully no bump steer, definitely sort out that steering rack.

 

[Mr168]

Brian - My understanding is that increased caster promotes "self centering" of the steering and increased steering effort. Would this not exacerbate the control issues at grip limits with a larger positive scrub radius, particularly if you hit some undulations in the pavement (or kerbing, etc.)? Not being contentious here, genuinely interested in a response as I have this very same issue, albeit with a rear wheel drive car (large positive scrub radius going from a 225 factory tire to a 315 series tire, wheel moved outboard via adapters to clear the coilover springs, caster maxed out as much as caster/camber plates would allow, and the car wanted to rip the wheel out of your hands through bumpy turns or when landing after catching a little hang time).

 

[BrianPetersen]

No argument at all. Increased caster does promote self-centering and that inherently means higher steering effort. Self-centering comes from several distinct phenomena. If there is no significant cornering load - at parking lot speeds, for example - there is a complicated geometric relationship between the KPI, the caster, the scrub radius, etc that leads to a "jacking" effect of the whole car when the steering is turned to the side. Obviously gravity doesn't want that to happen, so this leads to a self-centering torque around the steering axis. At higher speeds, with appreciable cornering load, more caster (with "normal" positioning of the lower ball joint) leads to more mechanical trail. The difference in side-view distance between the pivot axis and the tire contact patch multiplied by the cornering load gives a torque around the steering axis (Mz) that opposes the cornering ... this is the "Steering feel" that the original poster was looking for. (Inherently more Mz means more steering effort!) Also, when turning a corner, the tire itself wants to self-center because the inside has to travel ever so slightly less distance than the outside of the tire, and the friction between the tire and the road opposes that. This is "road feel". Wider tires will drastically increase this effect.

With the exception of the parking-lot self-centering, the scrub radius hasn't entered into this ... until you hit a bump. Now look at the front view.

Let's take the concept of "scrub radius" to the extreme and look at what happens with an old fashioned horse cart with a simple center-pivoted axle if you hit a bump on one side. The bump pushes the wheel up but it also pushes the wheel back. There is a huge distance for this force to act upon (between the wheel at the side of the cart, and the steering pivot at the center) which leads to a huge steering torque. At the other extreme, look at zero-scrub-radius steering. A bump acting on one wheel imposes a backward "whack" but it's centered around the steering axis ... no net torque reaction, the bump may be felt by your buttocks but not in the steering wheel.

Wide, low profile tires have the additional complication that this bump might not have the same effect on the inside edge of the tire as it does on the outside edge of the tire. If the bump is more towards the inside/outside edge, the bump reaction will try to steer the wheel inward/outward. Essentially the center of the pressure on the tire contact patch is momentarily not in the centerline of the tire any more. If there is too much camber (which may be part of the original poster's situation), that may be the case all the time.

Additional self-centering torque from more caster might not have much effect on the initial bump hit but it might help settle things down afterward.

As with any such thing, there can be resonant frequencies involved (e.g. "death wobble" that can show up on solid-front-axle vehicles) and that's where the math gets far too complicated for my small mind to understand!

 

[sierra4000]

Thank You for ideas Brian,
makes sense all for me,

moreover today I found engine more than 0,5 degree transverse inclination,
(he use non standard PU mounts)
I think this is real problem for torque steer so in connection with steering rack clearance creates steering wheel angle change under acceleration/deceleration.

 

[Mr168]

Thanks for the reply Brian, certainly some interesting points in there I hadn't considered!