While looking to find functional, and cool looking wheels for my Kelmark GT(vw type 1 irs chassis), I ran across this video on offset on the tsw site.
At the end, he state's that the wrong offset can cause bump steer and wondering. Wondering is what I have now.
Can some one explane how offset would cause wondering?
Thanks.
More likely you have a Toe Issue.
I didn't watch the video, and I'm no engineer, but it seems like moving the wheels outward from the steering pivot point would give a longer force lever acting to turn the steering when the tire encounters something. Smaller bumps or pavement variations would have an amplified effect on steering. In a perfect world, those forces would be balance side-to-side and have zero net effect, but in the real world one tire will always encounter slightly more force, so it wanders.
I guess I'm thinking of it like bump steer but caused by differentiated pavement drag that leads to a continuous pull one way or the other, and that would change as the road surface changed. The net effect would be wandering.
ultraclyde said:I didn't watch the video, and I'm no engineer, but it seems like moving the wheels outward from the steering pivot point would give a longer force lever acting to turn the steering when the tire encounters something. Smaller bumps or pavement variations would have an amplified effect on steering. In a perfect world, those forces would be balance side-to-side and have zero net effect, but in the real world one tire will always encounter slightly more force, so it wanders.
Yep, or in other words, by altering the scrub radius. Too much (or too far negative) scrub radius will cause wandering/tramlining, and something that feels like bump steer but technically isn't.
In reply to GTXVette :
I do have a toe fungus, umm, toe issue. Driving was a chore, really bad, so I pulled some strings in the garage for a quick check. Way toed out, pulled it in and it was much better, but not perfect, not even perfect for a ACVW. The side effect of my home garage adjustment was the steering wheel was now straight when the wheels where straight. Woo Hoo ! It will be going for a professional alignment.
I dont want to buy rims and make the issue worse, or have it return after the alignment (if the alignment even takes care of it), and then have to go back to the originals.
I was thinking 90s GTA rims, see coolest 89s/90s wheels thread. Its funny some one brought that thread back to life about an hour after I started my rim search, and then the rims I was looking at where posted. Any way, the fronts and rears have different offsets, trying to figure out if I should go with fronts and rears, all rears, or all fronts. I hope to have the front tires off tomorrow to see what I currently have for offset. Spread sheet started....
Not watched the video, but it sounds like he's defining bump steer differently than usual. He's talking about steering wheel kickback, bump steer is typically toe change with suspension movement and that's not affected by the wheel choice at all.
What Keith said. "Bump steer" is having the steering angle of the wheel changed by the suspension linkage as it goes up and down in bump or roll, not the concept of having it knocked around by external forces.
I hate it when people abuse established language. I'm still seething over "literally" being redefined as useful for emphasis. Literally literally no longer means literally.
I suppose the actual question the OP asked was how offset would cause wondering (or wandering, and we do seem to have wandered off a bit...), and that's down to, as folks have mentioned, creating varying distance between the steering axis where it meets the ground, and the point from which the tire's contact patch is delivering force. There's lots of fun rabbit hole here, but let's look at the wandering/kickback thing in a pic:
It's fairly easy to see that if this is the left side of a front-driver heading away from us, if we gas it, this tire's going to try to turn to the right, as it tries to use the lever arm of the scrub radius to drive itself around the steering axis (and if we brake it'll try to steer left). Of course, the right side is doing the same thing but in mirror image, and they mostly balance. But the larger that scrub radius gets, the larger the lever arm trying to turn fore/aft (accel/braking/the fore-aft component of bumps and how they tend to add to braking/accel) forces into steering forces. When things happen to only one side of the car, the balance between sides is interrupted, and all of a sudden the steering wheel's jerking in your hands.
Or maybe you drive over a big tar line with just one side and there's just a drag on one front wheel. So you get a pull toward that side. "Wondering."
I'm pretty sure we don't want zero scrub radius, but I don't recall why. I think things get really numb, even though caster/trail would be the obvious self-centering/increasing-with-cornering-load force. I want to say some front-drivers have negative scrub radius, so that when the right slips, the left tries to turn itself leftward in an attempt to balance the offset torque of trying to accelerate the left side of the car.
I have half a mind to apologize for my answer, even though I think it's fairly accurate. 
Ransom said:What Keith said. "Bump steer" is having the steering angle of the wheel changed by the suspension linkage as it goes up and down in bump or roll, not the concept of having it knocked around by external forces.
I hate it when people abuse established language. I'm still seething over "literally" being redefined as useful for emphasis. Literally literally no longer means literally.
I suppose the actual question the OP asked was how offset would cause wondering (or wandering, and we do seem to have wandered off a bit...), and that's down to, as folks have mentioned, creating varying distance between the steering axis where it meets the ground, and the point from which the tire's contact patch is delivering force. There's lots of fun rabbit hole here, but let's look at the wandering/kickback thing in a pic:
It's fairly easy to see that if this is the left side of a front-driver heading away from us, if we gas it, this tire's going to try to turn to the right, as it tries to use the lever arm of the scrub radius to drive itself around the steering axis (and if we brake it'll try to steer left). Of course, the right side is doing the same thing but in mirror image, and they mostly balance. But the larger that scrub radius gets, the larger the lever arm trying to turn fore/aft (accel/braking/the fore-aft component of bumps and how they tend to add to braking/accel) forces into steering forces. When things happen to only one side of the car, the balance between sides is interrupted, and all of a sudden the steering wheel's jerking in your hands.
Or maybe you drive over a big tar line with just one side and there's just a drag on one front wheel. So you get a pull toward that side. "Wondering."
I'm pretty sure we don't want zero scrub radius, but I don't recall why. I think things get really numb, even though caster/trail would be the obvious self-centering/increasing-with-cornering-load force. I want to say some front-drivers have negative scrub radius, so that when the right slips, the left tries to turn itself leftward in an attempt to balance the offset torque of trying to accelerate the left side of the car.
I have half a mind to apologize for my answer, even though I think it's fairly accurate.
On point post right there. Most cars will have positive scrub radius where the steering axis hits the ground maybe an inch or two inboard of the contact patch center,except the front drivers as mentioned. I personally haven't tried zero scrub radius but I'm led to believe that it can work well without funky steering feel with the right amount of caster and trail. FWIW, trail is the primary driver of steering feel and stability, while scrub radius has a much smaller effect. Just try throwing some wheel spacers on and see for yourself. Conceptually, I think zero scrub would work better with little or no KPI (and really everything works better with less KPI, but for practical purposes it usually isn't optional).
When I start building my FSAE styled project I'm definitely planning on using zero KPI and as close to zero scrub as I can get. It's all in the packaging.
KPI=bad, Caster (the right amount of)=good
Have you ever noticed that a "caster wheel" actually has ZERO degrees of caster? They are totally reliant on TRAIL to self center. In fact they should actually be called "trailer wheels"!
You guys have got this thing totally covered, so the only thing i can still contribute is that 'we have a lot of knowledgeable people on these forums'. 
I have never had a truly objectionable side effect from huge scrub radius on my cars, EXCEPT when i had to disable one of my front brakes on the way to $2017 Challenge. The ~1.5" of additional leverage on the steering from my wheel spacers on the tow vehicle made any brake application a truly white knuckle experience. Luckily i was only driving as far as the parts store!
Related question: How does one easily check the scrub radius without breaking out the faro arm and modeling my whole suspension in cad?
Chalk the ground and turn the wheel back and forth? Is there some trickery I can do with toe plates?
ProDarwin said:Related question: How does one easily check the scrub radius without breaking out the faro arm and modeling my whole suspension in cad?
Chalk the ground and turn the wheel back and forth? Is there some trickery I can do with toe plates?
I think it can be done reasonably accurately using a square, tape measure and some chalk. Remove a wheel, but load the suspension at ride height. Maybe a jack under the lower ball joint at ride height? Align a square between the two balljoints and mark where that line hits the ground. Put the square on the wheel hub and mark that ground point. Last measurement you need is wheel offset - can get that with tape measure, or likely by just reading the wheel :)
The maths: Scrub = Our zero point is the kingpin axis point (first point). Add the distance to the hub point (positive is outboard, negative means the hub point is closer to the car center) and the wheel offset (again, sign direction matters. Negative offset means that this needs to be a negative number, positive is positive).
The main reason steering axis inclination is a problem is because it's one of the things that is measured in an axial view (caster is just like SAI, but in side view). So, as you turn the wheel, the scrub radius actually has less to do with the point on the ground defined by the SAI, and more to do with the point on the ground defined by the caster (a.k.a. trail). So, scrub radius is one of those things that is hard to set for all conditions, because it is such a dynamic thing, as the caster, camber, and SAI are constantly changing as the car drives, moving the point that defines the scrub radius around.
As with all things, moderation is the key. A modest amount of offset change won't even be noticed. When does it become a problem ?
Since we are discussing the front suspension and steering lets add instant center location and ackerman steering. Just to add to the list.
I know, they are really not part of the discussion.
I think it's also important to point out explicitly why KPI/SAI (they are the same thing for those that are new to this) should be avoided whenever possible. I'm only going to be talking about the front wheels here. For best cornering potential we know that our wheels and tires must be presented to the road surface with a certain camber angle. Ideally our car would have 0* of camber when running straight down the road so that we have the largest contact patch available for braking and accelerating. But we also know that 0* camber doesn't work that well while cornering. So this is where either your car's steering geometry can become riddled with compromise, or some clever geometry can give you the best of both worlds.
Let's start with KPI. The inclined steering axis causes some funny wheel cambers while turning. Imagine your car with a few degrees of KPI and no caster turning into a turn. Your inside tire will start gaining positive camber, which is ok, this isn't detrimental to overall grip. However, the outside tire ALSO gains positive camber forcing the tires contact patch far over to the outside edge of the tire. This causes a very large reduction in grip and makes your tires not very happy. Basically your front wheels will look like this \\=// as you turn.
Now let's imagine that same car without that KPI but with a few degrees of caster. As you turn the inside wheel still gains positive camber, again this is ok. But now the really cool stuff starts to happen! Your outside wheel will be gaining negative camber! So your car wheels will now be leaning into the turn and hooking the tires into the road like this //=//, (imagine this car as being viewed from behind and making a right hand turn). This is very good for improving grip!
So you can see that by minimizing the KPI and selecting a proper amount of caster you can make your camber change with respect to steering geometry work in your favor. Special note: this idea works really well with a nice rigid beam axle, independent front suspension have other compromises that must be taken into consideration.
Let's now talk about the downside to having 0 or very little KPI and why you don't see this philosophy in normal road cars. The only real problem is packaging. Mainly you are forced to use wheels with very high positive offsets. Essentially you end up with ugly flat faced wheels (that enthusiasts would discard asap). This also forces you to cram your brakes as far out and as close to the spokes as possible. You then must force your kingpin pivots (or ball joints on modern cars) right into the centerline of the wheel, you must put the upright's steering arm inside the wheel as well. As you can imagine this gets crowded in a hurry and it might force compromise with your desired suspension kinematics. One way around this is the really super tall uprights as found on many cars these days like the 350z pictured below where the upper ball joint is typically above the top of the tire. This lets you have more authority of where your KPI axis lives while not cluttering up the wheel as much. In summary, the above design suggestions aren't found to be practical for most OEMs. The result of this is the kludge fix of adopting some degree of KPI in an attempt to get the scrub radius back down to some reasonable dimension.
Well I hope that covered everything and I hope I didn't leave anything out.
In reply to freetors :
You didn't write a 48 page dissertation, so *of course* you left stuff out. :)
Ouch, my freaking brain! I need a video to dumb it down for me. Lol.
KPI helps with directional stability and aids the return of the wheel to straight ahead.
As has been said, positive caster will help with the camber problem. This is why we have both.
There actually were some cars in the dim past that had negative caster. Not for long.
iceracer said:KPI helps with directional stability and aids the return of the wheel to straight ahead.
As has been said, positive caster will help with the camber problem. This is why we have both.
There actually were some cars in the dim past that had negative caster. Not for long.
My dad had a '37 Toro tractor that had a buggy spring on the front swing axle, with about a 3" slot on the swing axle's pivot bolt. The uneven pressure of the spring had opened up some holes over the years, causing negative caster. It was very, very challenging to drive at high speed (>4 mph) as it always wanted to tear the steering wheel out of your hands and go to full lock, but it was handy creeping along on a side hill, since it always wanted to turn away from the abyss. It was like it was scared of ravines.
The thing is though, you can have negative caster and still have perfectly functional steering! All you need is to offset the spindle rearward with respect to the kingpin axis to create more trail! Now would I ever do this? Heck no, for all the virtues positive caster has I wouldn't want to be without it. But it can be made to work.
You'll need to log in to post.
