Suspension design for Locost with Vsusp

[horizenjob]

Because I am trying to appear as "original" as possible on #2, I am bound by packaging.

Yup. Spitfire uprights with the heim joint mod would be a good addition for looks. Then you might need to mount under the frame. I don't know which would appear more original. With a Ford xflow engine then you could lower the frame 3 inches easy. Not pushing you any direction though.

I've never heard of trying to run 3/4" toe out. It is not really the same as Ackermann. What happens is that as you do the very first initial steering input, the inboard wheel develops steering angle very quickly and the outboard wheel goes to straight ahead. At this point the inner wheel is generating turning force, but also drag on the inside of the car. It's pretty noticeable on a road course.

In your parking lot events the turns are very tight and the rear wheels are cutting inside the path of the front wheels. I don't think you spend enough time at the small steering angle for you to notice the toe out effect. So 100% Ackermann is the way to go. The toe out won't hurt though, well be careful on the highway.

Thanks for explaining your thoughts here. I really want to get the suspension right on Car9. I have been opting for adjustable roll centers and this is bearing that out as a good idea. It seems the way this conversation is going that there will be setups for autocross that are different then road racing.

Did your care also tend to push in the turn after being slow to turn in? Did you try not using your front anti-roll bar? Do you have a rear anti-roll bar? If you angle your trailing links for the rear axle we could get some roll oversteer. It might be worth an extra set of holes on the trailing rods mounts.

I'll do some numbers with Vsusp, but waiting until the fixes in progress now are available.

[SkinnyG]

I'm not after 100% "original," I'm after "original appearing." I'm not interested in using a crossflow or Trunion uprights, I'm not even interesting in saving original geometry. 99% of the "Locost" appearance annoys me - I'm changing that.

I got up to 3/4" toe out just trying things. That's when I figured toe wasn't the issue.

Initially I ran no sway bars at all, but had horrific (and fun) oversteer. Despite a very low rear roll center.

First, I changed the rear bushings from UHMWPE to rubber, which helped a lot - can't allow the rear suspension to bind at all.

Then I added a 1/2" sway bar up front, which wasn't a huge improvement.

I went up to a 5/8" front bar, softened slightly by milling the middle section until it felt right. There wasn't enough adjustment in the end link location.

Then I installed a limited slip (which helped tame oversteer considerably), but once we tamed it, it wouldn't turn in.

I started dropping the front more and more which felt good, but it still needed something more.

Somewhere around this time I switched the brakes to dual masters and a balance bar which was perhaps the single most significant improvement in making the car turn. But that's beyond the scope of this thread.

The final tweak was adding a 1/2" rear sway bar (I actually sat down and did calculations this time) and the end result was fantastic - I haven't changed it since. I think the toe is at about +1/8" or so. It's fine on the highway - just pay attention to the twitchyness.

In reflection of getting this car to handle, I believe that (for me) the front roll center is the most critical. Next being camber curve. I had a good camber curve on Lethal#1, but dropping that roll center made a bigger improvement. The camber curve on Lethal#2 hasn't found its happy place yet.

(Am I going beyond the point of this thread? Am I venturing too far down a rabbit trail? Does any of this add to this thread?)

The usual BS disclaimer - this is all just my opinion; your mileage may vary; blah blah blah.

[cheapracer]

Does any of this add to this thread?

Trouble is if someone doesn't mention that you have a ridiculously low rear roll center that's causing all your grief then others reading your post will think that a very low front roll center is a good thing, it's not and that's relative to this thread.

Having stated roughly yourself that "many Locost have oversteer problems with a 10" rear RC" and then stating "I have no turn in and I have a low rear RC (4")", can't you see it?

I urge you to go and lift your Panhard rod to a point where you have a rear RC of around 6" to 7" and experiment from there. Fitting a rear anti roll bar has compensated somewhat by allowing some weight to be transfered diagonally forward in roll and that's why you're now getting some outside front wheel bite but a higher rear RC will do this better - note I did not say high RC, I said "higher".

Relative to this thread, it is unusual that a street orientated, RWD performance car will stray very far from 2"-4" and 5"-7" RC's front and rear respectively (Honda S2000: 2.0" and 7.5", Miata: 2.5" and 5.0" for example, but like everything, there are quotable exceptions) Even a Ferrari F355 that has a very low CoG, very wide track has a 0" front but 5" high rear RC - better to understand the roll axis and it's effects, not the individual RC.

Grab a broom and push it along the floor with your hand close to the floor and turn the handle simulating chassis roll, little effect other than the 'inside' corner of the broom will lift, now lift the handle up above your knee and try again - feel the 'outside' corner now trying to dig in - roll axis at work.

[SkinnyG]

Having stated roughly yourself that "many Locost have oversteer problems with a 10" rear RC" and then stating "I have no turn in and I have a low rear RC (4")", can't you see it?

I did say that. And I see what you are saying.

I also did a considerable amount of reading, research and consultation regarding roll centers, spring rates, roll resistance, etc etc which led me to the direction I headed.

I also initially designed the front suspension around shortened struts, but changed (and did not optimize) to Chevette spindles to get through inspection. The shorter Chevette spindles raised the front roll center higher from what I had intended. << This is probably significant to note.

Considering all Solid Axle Lotus 7 and all Solid Axle and DeDion Caterhams have a very low rear roll center and they have not changed that, I don't believe my goals are out to lunch.

Is my way the best way? It doesn't matter - it's just what ~I~ did, the way that made sense to ~me~. There are many ways to skin a cat (only one way to cook it).

Perhaps we should get back to the software at hand....

..... which I think is very cool, by the way.

[cheapracer]

Considering all Solid Axle Lotus 7 and all Solid Axle and DeDion Caterhams have a very low rear roll center and they have not changed that, I don't believe my goals are out to lunch.

What they do and what they achieve isn't relevant to your car or your driving style or both. I am genuinely trying to help - every thing you have mentioned points at your rear RC being too low and then when you said adding the rear roll bar helped besides key bits such as no turn in and very confident/stable at speed, it's "most likely" (*see below), ... you are lucky it will be as easy as simply changing some brackets unlike this forum member who has a welded in IRS .. maybe this handling looks familiar to you .. http://www.youtube.com/watch?v=iaTV3FHw99I - I meant to suggest a couple of days ago in his thread he may want to try a rear roll bar to combat his apparent low rear RC issue.

Go here; viewtopic.php?f=5&t=13921&p=147386&hilit=Discussing+SkinnG%27s+understeer%2Fno+turn+in+dramas#p147386

Perhaps we should get back to the software at hand....

We are, there's no point in people blindly going in before they understand what effect RC's have

[horizenjob]

The shorter Chevette spindles raised the front roll center higher from what I had intended. << This is probably significant to note.

That's why I mentioned the Spitfire parts. I don't mean for you to include the trunions. They are commonly cut off and a little piece is welded on to take a spherical bearing mounted in the wishbone. Graham Early just recently did this and put pictures in a thread somewhere. It's what my Formula Ford uses and you can see these in lots of pictures of kits from England in our build logs. I saw them yesterday in a Sylvia Riot build log here.

Not trying to be pushy by the way, just helpful. Perhaps with the frame as high as yours is you cannot afford to lower the wishbone.

This thread is for discussing our suspensions and we will be use Vsusp to pass info bak and forth. Right now Rob is putting in an important fix and that is why I haven't provided any pointers.

So I do have some more comments about the process you went thru but first want to ask wether you were having an understeer problem in general or just a corner entry problem. As you noticed some of this relates to having limited slip, that can make the car resist turning. It seems your experience there was mixed though.

When you run a Higher rear RC and a front anti-roll bar you are reducing the rear traction at the start of turn in before the car rolls, then as the car rolls you are reducing the front traction to match. So you can affect the transitory handling that way.

[SkinnyG]

I don't mean to come across as snarky or anything, and I do appreciate your help. I have always found your posts and insights helpful and informative. I apologize if I am offending you.

I haven't spent any "real" time refining the 7 for a couple years. What I did do, is go back through some of the threads at http://www.vcmc.ca where I documented a lot of my development. This will be fresher than my dusty memory:

Roll Centers

Selecting Sway Bars

Limited Slip

Lethal Development <<< Brake bias, sway bars, spring rates

Still more Lethal modifications <<< sway bars, ride height, tire diameters & roll centers

Updates to the Lethal Locost <<< Brake bias

I have saved some other discussions between Joe and myself regarding Roll Centers in a file somewhere....

[SkinnyG]

This was also a persuading article by Mark Ortiz on roll centers and solid axles:

Chassis-Newsletter-2007.5.pdf

[SkinnyG]

================== New Rabbit Trail ==================

What are people's thoughts and/or goals on camber gain?

Degrees camber per degrees roll?

Keep the tire vertical?

????

[horizenjob]

The Mark Ortiz articles mention not going for more then %50 camber correction in roll. That was about what I was going to aim for.

[cheapracer]

I apologize if I am offending you.

Not in the slightest, I love the fact that you're a hands on guy (don't confuse that with handsome guy) and your chat is something I miss back in Oz workshops after work beer benchracing'!

[OptimusGlen]

Alright, here's my first shot at this. Admittedly, suspension engineering-speak is like Latin to me. I haven't quite picked it up yet.

I am using Volvo 240 struts that will be cut down and modified to accept an upper ball joint. Additionally, the lower ball joint will be upgraded to be able to sustain the increased duty it will see in this configuration.

http://tinyurl.com/GlenGC7v01

I haven't done anything with the rear yet, my plan is using a solid axle/4-link setup.

I understand that my setup as-it-sits is probably no good, I'm hoping to learn what I can. Am I right to say that my RC is too high? How does one get a lower RC?

Edit: playing around with VSusp more, found that if I lower the lover control arm mounting points slightly, it makes a big difference on the RC.

http://tinyurl.com/GlenGC7v02


Edit again: round 3
http://tinyurl.com/GlenGC7v03

[horizenjob]

I just tried OptimusGlen's configs and am very happy with the updates that Rob has made to Vsusp. It seems to work better in Safari, and the improved resolution of the numbers is helping.

Glen, I have a couple of simple requests. First would you mind editing your post above to mention what spindles you are using? Then people can just copy your numbers or use your pointer and edit the rest in to suite them. I'd like you to edit the post to try to keep this thread a bit shorter and make it easier for people to follow. The other request is to provide a "Project name" in the first item of the frame dimensions. You picked good names, but once I open both your links the tabs just say "default" and I can't tell which one is which... The more configs or links we get posted the harder this is going to get.

So you have excellent control of your roll center height in both roll and ride. You have strongly favored camber correction in roll as opposed to ride. This is where I started too, but I am getting to accept to trade off this with camber in ride. In your first config you get 4 degrees of negative camber with 2" of bump, which would be heavy braking. What I am doing now is checking the 2" bump case. Then I set the roll to 2 degrees and adjust the bump from -1" to +1" to simulate braking / entering and then exiting accelerating. I forget the numbers I saw but you go from considerable negative camber in braking to something like 1 degree positive on exit.

So I would draw another case that does less of that and then compare and see what you like. The second one looks better at this, I think I did lose track of which one I was looking at while I typed this.

Rob is looking at adding an item to chart the jacking effect and/or the angle of force from the contact patch thru the suspension. Those would show you directly why you don't want a high roll center with this type of suspension. A good chunk of the trouble you're having is because the spindle's lower ball joint is so high off the ground. It makes the roll center and camber correction tradeoffs more difficult.

[OptimusGlen]

I've found that people have used lower ball joint spacers successfully on Volvo 240's, I will be able to design and build my own and thus updated my VSusp design accordingly.

Again, this setup is using Volvo 240 front struts, cut down with an adapter welded on top to mate to an upper ball joint, the lower ball joint receiver will be spaced down 1-2 inches.

Here is a spacer to give you an idea of what I mean.


http://tinyurl.com/GlenGC7v04
and
http://tinyurl.com/GlenGC7v05

Edit: Had more time to tweak the design in 3D and I think I found a happy medium in design and packaging.

http://tinyurl.com/GlenGC7v07

[cheapracer]

Here is a Lotus Elan's geometry scaled to 53" to aproximate a Locost's track.

I know it's a big picture, I left it to scale for anyone to use so I'll go to the next post