bump steer

[carenthu]

how bump steer is minimised with the help of steering arm? does change in the relative lengths of controls arms will also be an effective measures?

[gruntguru]

This is the "Engine, transmission and controls" forum.
There is another one called "Aerodynamics, chassis and tyres".

[Greg Locock]

Not too sure about terminology, but yes, changing the steering arm length, or the wishbone lengths, or the rack height, or the rack width, etc etc etc will change the bump steer curve.

[PlatinumZealot]

Bump steer can be desirable or undesirable.

[strad]

Bump steer can be desirable or undesirable.

Not sure when it would be desirable

[Greg Locock]

Production cars use bump steer (well actually roll steer but you can't always separate them) as part of the understeering budget.

https://en.wikipedia.org/wiki/Bundorf_analysis

The example given is made up and a bit ludicrous, tho it must have been an IRS on crossplies, radials don't give much camber thrust.

On a modern FWD in production tune you often end up with as much as 10 deg/m of bump steer on the front, as the only (?) other cheap way of getting understeer that doesn't annoy the driver is to increase the front bar size yet again, which causes ride problems. You can't add compliance steer, it makes the car feel wobbly, you can change the tires, which will get you shot, or the tire pressures, which will get you shot by different people. Given the choices I prefer to aim at 50% camber recovery, highish front roll centre, and 2-4 deg/m of roll steer, and no compliance steer. But reality often intervenes.

[strad]

WE must be speaking about something different.
Bump steer is when your steering angle of the front wheels change
Usually mid corner and often as the name implies when some uneven surface is encountered, and that's just not a good thing.
It really nice when the wheels maintain the path the driver sets.
Raised or lowered vehicles are the extreme examples where the arc the wheel travels thru changes the wheels steering on it's own .

[Greg Locock]

No it is the same

Bump steer is when the toe angle, or steer, of a wheel changes as it moves through its vertical travel. It is typically expressed in degrees per metre. It is a kinematic property of the suspension. What you are describing is an outcome of that, it is the same mechanism. Wheel goes up, it toes out (typically on the front axle). Wheels go up by different amounts and it steers.

Ideally we'd set it so that there was zero bump steer for single wheel lift, to minimise sensitivity to road irregularities, and still gave us roll steer on the whole axle for understeer. There is some smarty pants stuff we can do to get some of that but it is not perfect, and has significant downsides, in particular that for two wheel lift, aka changes in vehicle loading, you'll be pumping in a lot of toe.

What we actually do when we can is somewhat different to the above. If you imagine a vehicle driving on a straight road and encountering single wheel bumps, the objective is to keep the vehicle tracking true, not worrying about kinematics, So we can use things like longitudinal compliance steer, toe/Fx, to help minimise Fy inputs for a bump. So, define a road surface with some bumps in it, then run different geometries and bush stiffnesses down it, and define success (eg minimise steering wheel torque for a straight path). Real cars with that sort of geometry and so on are noticeably more stable on rough roads.

The various horrible things people do to raised or lowered cars aren't really worth worrying about, amateurs (in a broad sense) do what amateurs do, which is fine if they learn from it.

[Tommy Cookers]

for trackdays one might hope to have chosen a car with little roll steer
and once might have imagined the absence of roll steer being part of the mysterious qualities of a supposed 'real' sports or GT car

the place I found roll steer was in a fast S bend
trying to minimise the waste of track by going asap from eg full right cornering to full left, ie quickly reversing the steering input
really showed how much of the wheel steer angle came and went from roll steer and how seemingly little came from the steering wheel

though roll steer was not obvious with standard understeery setup
it became apparent after stiffening the rear springs 30% to give a notionally neutralish setup
yielding more lateral g and so proportionately more roll
but disproportionately ? more roll steer combined with less steering wheel input
(ride heights were standard throughout)

of course a neutralish setup mid-corner on a track is a danger on corner entry
that's why road cars must have an understeery set up

Greg, your explanation is much appreciated here, because it makes sense of my bits of findings as above
but doesn't modern rear suspension clever (compliance ??) steer reduce the amount of understeer required (for safety) from other sources ?
by adding on vigourous throttle-off/braking/downhill corner entry a pro-understeer steering term ?
improving safety but not in principle improving roadholding in expert hands ?

[PlatinumZealot]

WE must be speaking about something different.
Bump steer is when your steering angle of the front wheels change
Usually mid corner and often as the name implies when some uneven surface is encountered, and that's just not a good thing.
It really nice when the heels maintain the path the driver sets.
Raised or lowered vehicles are the extreme examples where the arc the wheel travels thru changes the wheels steering on it's own .

You can use a small amount of bump steer to your advantage to get more turn-in on the front wheels when the attitude of the car changes. Or for the rear wheels you can use it to get more toe-in or toe-out. Just depends on what you want the wheels to do in the corner. Not saying it is all good though, it is bad when it is significant or even say on a rally car or street car where suspension travel is large, as we all know bump steer can throw you into the bushes.

[strad]

The various horrible things people do to raised or lowered cars aren't really worth worrying about, amateurs (in a broad sense) do what amateurs do

Jeeez
I know I'm an amateur by some, perhaps your standards, but let me pose a real life problem to you.
I have lowered my 2004 Mustang Cobra with IRS on Eibach springs and bars as a kit.
Now Eibach is internationally renowned suspension people that claim to have designed it just for this car.
Now when it was dropped it of course developed a case of bumpsteer mid corner, but the well respected SCCA champions at Steeda provide a bumpsteer "kit" that takes care of that to a tee.
NO more bumpsteer, much improvement in cornering.
Now especially with the new Koni adjustable shocks I felt I had to go to the softest compression and rebound setting to achieve more compliance. To me it is too still (springs? Bars?) too stiff and feels like a go kart. To which my my friends who are even more amateurish and stupid than I, say, well isn't that what you want?
I think you'll agree it is not.
However when you spend money on long distance calls to experts and this is what they say to do, what is a naïve amateur like myself to do.
It seem very much like the Guldstrand set up I put under my 67 Vette and that worked great but was also too stiff in my opinion.
SO if you have answers that will work you be sure to let me know, but be assured you're saying out front that you know more than Eibach or Steeda.

[Greg Locock]

For non aero cars:

Bear in mind that their audience equates stiff suspensions with handling, and to a certain extent, that's true (Chapman's quote comes to mind). But if we start with the idea that we want to exploit the friction circle continually, then that is made much easier if all 4 tires are touching the ground at all times, and if they are roughly equally weighted. So a more compliant suspension is likely to give you better grip, assuming the kinematics are half reasonable.

But, the big no-no is running out of travel, so your springs and sta bar have to be soft, but not too soft. On mountain bikes we used check the suspension travel we were actually using on a given track, and soften off until we just touched the bumpers twice per lap.

Now if you do that you won't have a 'balanced' ride (subjectively), all you'll have is happy tires. Which is actually the main thing. I haven't talked about shocks, I've never set shocks up for a race car. I do help set them up for road cars.

[strad]

I'll give it consideration but it's not much of an answer.
With the subframe connectors and the three point tower brace and the IRS subframe braces and bushings what I have done is create one hell of a great track car and a poor road car.
It is indeed too stiff and I'll concede I question Eibachs choice or roll bar stiffness, there is next to no body roll at all.

[gruntguru]

Having lowered a car substantially, you are almost forced to increase wheel rates, if only to avoid running out of travel in bump.

Another consideration - for many suspension types, lowering the car will lower the roll centre by more than the CG meaning - increased body roll. Another reason for stiffer springs and bars.

[Greg Locock]

It's not much help because tuning suspensions is not just rules of thumb,if you apply rules of thumb you get thumbs not cars.