MG Sports Cars

OK this is intended to be simple but technical.
On the MGB, the upper control arm is about 8" long. (shock arm) The LCA is about 12" long, and the outer steering link (adjustable) is about 10" long. For these dimensions to play nice, theoretically all three would be parallel and equally spaced but they are not. The UCA and LCA are at different angles relative to each other, and the OSL is roughly parallel with the LCA and close to centered. (On wednesday I can look at Steve's car to see how close.

In the past, when lowering the steering rack we have generally tried moving the tie rod ends down by the same distance in order to maintain steering geometry and for small distances this seems to work fairly well, or at least we've not heard a lot of complaints about it. However, this thinking is somewhat incorrect. Because the OSL remains roughly the same length while moving down, the relationship between it, the LCA and the UCA changes, meaning toe-in and bump steer has to be effected.

I'm giving this a new look because, one of the benefits of the air bags is that by releasing air pressure I can easily move the suspension through it's range and look at it's behavior without having to resort to removing the springs. So it's pretty easy to lay a 2x4 against the sidewall and against the rear tire and watch to see what happens. The amount of bump steer I'm seeing is pretty surprising, and it makes me wonder just how much bump steer is present in the unmodified stock suspension. (If I could persuade Steve to install a set of air bags while he's here we might find out!)

So at this point I'm wondering what is the minimum amount of bump steer anyone has been able to achieve by moving the tie rod end? Is it centered in the range of suspension travel, or biased one way or the other? What I'm seeing is toe-in on compression, but I've got a lowered rack and lowered inner LCA pivots so it will be different from stock.

JB

Jim

To get zero bumpsteer you have to have the outer link of the rack point to the instant center and the end points of that link have to lie in the planes of the control arm inner pivots and the plane of the ball joints (or kingpin). This link can be shifted in or out so long as it still points to the instant center and the length remains the same.



This is going back quite a while but when I did my turbo motor installation I had to lower the engine to keep everything under the hood. This required lowering the rack as you know. The stock rack was too long to start with even with the steering arms angled out for ackerman. Lowering the rack made the situation even worse because of the angle of the plane of the inner control arm pivots. I had to narrow the rack about 3" as I recall to have the ends of the rack lie in the plane of the control arm pivots. I then bent the steering arm down so that the link pointed to the suspension instant center. I got lucky and the thing had zero bumpsteer when I put it back together. The steering shaft then needs to have 2 universals to hook back up to the column.

Getting rid of the bumpsteer made a noticible difference in the handling. I also did some experimenting with reducing and eventually eliminating the ackerman which also helped the grip at the front a lot but made the front tires scrub quite a bit on tight radius turns especially if there was any sand on the road surface.



Edited 1 time(s). Last edit at 04/25/2011 03:44PM by Jim Stabe.

I'll get into ackerman a bit more after I have a better handle on the bump steer. I understood that well enough at one point I guess but it goes away if you don't make concrete use of it.

Right then, what happens if the intersect point is outside of the vehicle? (instant center) Obviously I'm seeing that the instant center moves around a good bit and I'm guessing this can make locating the outer steering links interesting to say the least. Great fun to play with actual linkages though. I'm going to have to take the wheels off now and see if I can find some light, stiff rods for extensions. This ought to be a hoot. Maybe fishing poles would be a good choice! I don't now where the IC falls with the stock suspension, again, I'll have to look at Steve's. But mine is definitely to the outside rather than the inside.

JB

Unless you have some extreme angles on the control arms the IC will probably be outside the vehicle. The best way is to plot it on paper by taking some measurements of the front suspension and plotting them to scale like the diagram above. You can then establish the planes of the pickup points (easier if the axes of the inner points are parallel to each other and I think they are - they become a single line when viewed from the front) and see if the steering link is the correct length and if it pointed at the IC. If the arms are parallel to each other, the steering link should also be parallel to them. If you have the steering arms set up to provide ackerman ( a line projected from the outer tie rod centerline through the axis of the ball joints [kingpin] will intersect the center of the rear axle), I think you will find that the link is much too short and will cause toe in on bump and rebound. You will also probably find that to get the correct length for the link that you will have to narrow the rack by some amount.

As I remember with the stock setup that the rack is too long causing some toe in on bump and rebound and that only gets worse the more you lower the rack.

Can "anti" be a good thing ? I've had some , self-proclaimed suspension experts, tell me reducing anti-dive is a good thing. They claim this makes the car "less darty". I was taught positive caster makes for less "darty". Akerman can help the inside wheel "turn-in", entering a turn. It can also help the inside tire, past it's slip angle limit,(not desireable). Seems to me, anti-dive, helps create roll castor-gain, and enhanced negative camber(dynamic), in a turn,(desireable).Things were "simpler", with Morgans-sliding pillar. Cheers, roverman.

All good points to consider, but it's a bit much to have running all around your brain at one time, hence the desire to keep it as simple as I can at this point. Sort of like changing the fuel curve and timing curve at the same time, it's easy to get lost.

I've got a pretty busy day, and then Steve will be here and this has to go on the back burner until after the weekend. But next week I think I'll pull a front wheel and see how far out I have to go to get to the IC. If it's workable I'll go that way, and it might be, with the blocks below the LCA inner pivots. Nothing like actual physical parts to give a real feel for something. The LCA is inclined downward in the front something like 3/8 or 1/2" IIRC and the ackerman is stock. Moving the rack down would aggravate the excess length of the rack of course. (Can't see narrowing that as being an easy thing.) If just moving the ends of the steering arms will get the bump steer reasonably close I can live with that. I always thought it was just something you allowed for anyway, like body roll.

JB

Jim

I wish I could remember more detail about what I did but that was 35 years ago and I sold all the parts 15 years ago so I can't even measure them for you. You are probably wise to put it together and see how it behaves, the car will certainly be drivable and I'm sure you have missed driving it. At some point when the steering geometry rises to the top of the list you may want to do something about it. The rack is fairly easy to narrow, you just cut it off and thread the end as I recall. The housing gets the same amount removed from the tube and welded back together on a fixture to hold it straight. I used some aircraft U joints in the shaft to get it up to the column.

Hey Jim,

To me the best plan of attack is to first get the rack length right. I think that it's excessive length is the major player here.
Jim S clearly knows what he's talking about and plotting it on paper has many advantages. Doing the math however can be a pain so if you're more tactile like me. Then just build a simple 2D model. I use a piece of plywood with the inner control arm pivot points laid out on it. A couple of 1x1 wooden control arms are screwed on to the pivot points. Cut the kingpin/spindle out of plywood, mark your control arm lengths and screw that on. Cut a 1x1 tie rod and screw that on to your spindle and lay down a line for your steering rack plane. Build it full scale and you can measure the changes without converting.
Now its just a matter of operating your "suspension" to find the right steering rack and tie rod lengths.
You can also map out your camber change with the new blocks. The only fly in the ointment is the altered LCA angle from the uneven blocks. But the change in arc for your LCA is probably minor and shouldn't affect toe change much. My concern with the LCA angle is the caster gain under compression and loss under extension. This will make the car want to turn in harder as the car leans into a corner. It will also make the car a little darty over bumps due to the caster change. Depending upon the wheel and tire combo this could be quite severe.
If you really want to see whats going on a 3D full scale model is pretty easy and lets you make and record changes quickly.
I know that there are some very good simulation programs out there but I work better when I'm actually able to touch and feel it.
At the end of the day, as Jim S said, drive it first and see how it behaves. It might be just fine as is.
Cheers
Fred



Edited 1 time(s). Last edit at 04/26/2011 02:05PM by DiDueColpi.

Thanks guys, I really appreciate the support. I think I'll go with the full scale 3D model since I have it sitting right there on the lift, at least to get the initial measurements. Then maybe mock something up so I can move things around. With the air off I can use my floor jack with a vertical extension (a chunk of old axle shaft and a length of large conduit) to raise and lower the spindle, and it should be easy to clamp a piece of tubing or something to each of the arms and see where they intersect. Set a step ladder out there with some cardboard to mark points, lines, arcs, and arrows, and a paragraph on the back of each one to be used as evidence against us.... no wait, what?

Time to go again. Later!

JB

Barrister coming out again....

Right then,
Rear steering it is!

I'm not much of a CAD/CAM kind of guy either as witnessed by the way I figured out where the frame members needed to go at the start of the build. Lot to be said for making wood models.

>>Barrister coming out again....

No, Alice's Restaurant. 'You can get Anything You Want, at Alice's restaurant...'

Hey Dan guess what? The shop is nice and dry! No water at all. It's starting to feel clean too. Gonna be a good weekend.

The roadster is good to come off the lift, soon as we push the truck outa the way. Man, I gotta get some motors in these things.

JB

Yep, thank God for wood Jim S. I did the same when designing the front end kit.

Jim B one way to limit the Bump Steer is to set the ride height that is going to be in between the lowest point and the highest point.

This will give you the location of the steering arm on the rack. If need to lower the steering arm, just swap the steering arms from left to right. This will allow you to install the tie rod ends from the bottom thus lowering the steering arm.
As per Jim S find your instant center first, then proceed to make adjustments.