LocostUSA.com

OK, it seems I have a basic error problem?
Read several books and have tried to follow discussions here.
Have a suspension analysis program on an old PC.

Now that I have determined which ball-joints I will be using I wanted to get to work on A arms and mounting points to the chassis.
As others have posted a lot of the points are pre-determined by the spindle/upright chosen and the chassis*.

It seems that no matter what I try I cannot get negative camber in roll without bump going crazy.
I've been assured that negative camber in roll is not difficult, so I must be making a basic error of some sort.
3" compression and 2" droop travel.
Despite all the reading I have done it seems there are no "Ideal" numbers to shoot for, lots of theory but no numbers.
The assumption I've been working with is that the theoretical ideal would be zero camber at all times.
"Best?" camber I've gotten is around -4º at full bump and about +2º to +3º in roll.
*In my case the inner mount points can still be located wherever I want, within a range of several inches laterally and vertically.

No drawing yet but here are the basic dimensions.
For a level lower arm I need to be 8" from the ground at both ends.
Using -1º camber as the static ride height number.
Lower arm 10" - 11" (Max) long. Upper arm 2/3's of lower, angled less than 10º.
Spindle BJ mounts 10" apart.
Lower BJ 8" from ground.
Upper BJ 1.0" off-set to the inside.

Lower arms are to be the classic A type, BJ outer and rod-end inner.
Upper is a heavy arm from the BJ strait to the inner pivot rod-end and a leading arm going forward to a rod-end.
This is very similar to the original Alfa setup and gives me the most room for the engine and accessories.

I know I need to work up some more exact wheel and spindle drawings but thought I should be able to get close for basic suspension numbers.

_________________
If I must be a one-man PC free zone, so be it!

Richard, you'll never get perfect camber in both bump and roll. They actually work against each other, so that improving one hurts the other. It's a compromise. Generally you want to reduce the amount of positive camber increase in roll by about half. So if for example the car rolls three degrees in a turn, you want around one and a half degrees positive camber. Then in bump you'll only get a degree and a half negative camber, for the same amount of travel. It'll never be perfect but you can tweak things later by decreasing static camber.

So what is a good to excellent number to shoot for, camber in bump at 3" to 4" travel?
Can't go more as I only have 5" ground clearance at ride height.
50" track, 90" wheelbase.
With this I can at least work on laying out A arm drawings.

I'm about to try to improve on my rough spindle measurements.

_________________
If I must be a one-man PC free zone, so be it!

There isn't a good number. It all depends on how much the springs and bars let the chassis move and how much caster and KPI you're running (caster will add negative camber to the outside wheel and KPI does the opposite).

_________________
-Andrew
Build Log
Youtube

Picture a triangle. At one corner is "perfect camber under acceleration"; in the second it's "perfect camber under braking", and in the third, "perfect camber in roll." Pick your point between them.

_________________
Midlana book: Build this mid-engine Locost!, http://www.midlana.com/
Kimini book: Designing mid-engine cars using FWD drivetrains, http://www.kimini.com/book_info/

If the steering effects aren't being factored in with the roll, then the numbers are pretty much meaningless.

_________________
-Justin

Also follow my build on blogspot, tumblr, or instagram and twitter (GarageOdyssey)

I think that if I can put together a list of the fixed values I should be able to find a Pro to work out the rest.
Not "Getting it" on my own despite many hours trying.
Would rather put my time into fabbing parts!
Anyone here qualified to take this on?

_________________
If I must be a one-man PC free zone, so be it!

Some cars use more static camber with less gain in bump.....can cause other issues if you go too far (just like everything else) but it is another tool. By observation, German cars seem to do this on the rear more than others.

Off setting the kingpin axis can help camber on both wheels when turning. My solstice spindles do this, as well as a bunch of others. I am sure too much can cause problems too.

I'm not sure what software you've been using, but the 'Wishbone' suspension program will perform the 3D analysis that allows you to factor steering inputs in conjunction with roll.

http://locostusa.com/forums/viewtopic.php?f=26&t=11985

_________________
-Justin

Also follow my build on blogspot, tumblr, or instagram and twitter (GarageOdyssey)

As my head doesnt do quadratic equations and I dont have the patience to model everything, I worked it out old school by tacking / clamping my suspension to the frame and moving the arms up and down over the expected travel area. My main focus was limiting the amount of bump steer in the travel range which I think is what everyone is describing. Worked out in the end despite a ridiculous amount of bump steer at full travel. To get to full travel, Id be upside down or in the air so it didnt become a concern.

A 2x2 piece of plywood screwed to a 2x4 at 90deg angles makes a good bump steer gauge.

Consider hitting a 3" tall bump. Do you *care* what the camber change is? Unless it's outright crazy, it's going to be lost among the commotion.

Many of my old suspension design books insisted on the importance of the shape of the camber curve at the 3 to 4 inch bump range. For a '39 Ford on an unpaved road, yeah, it was be important. (remember 'washboard' roads?) But modern roads are (often...) smoother, and modern cars are sprung *much* more stiffly than in the old days, when an important design factor was getting the softest possible spring rate to help the ride on rough roads.

Richard, when you collect the necessary dims and put them into vsusp, post a link to the page so anybody can try to improve it.

Try a ratio of near 1.5/1 for lca/uca, 70-100 vsa, camber loss on the outside wheel of .75 per degree of roll with zero static neg camber, rc that is below the frame, doesn't vertically and doesn't move far from side to side in roll.

Any increase in neg static makes bump worse and roll better by the same amount.

_________________
MV8

https://www.youtube.com/watch?v=Om3C1Ep ... D3E18BB447