Tigers
|
| posted on 1/3/04 at 01:21 PM |
|
|
Roll Center
Hi!
Maybe someone has calculated how high is front RC for locost "book" suspension?
And how high (low) is resonable (optimal)?
And also - just stupid crazy idea - what happens if I construct suspension with RC higher than CM? Would car lean towards inside of curve?
|
|
|
|
|
pbura
|
| posted on 1/3/04 at 02:33 PM |
|
|
I did, for a car with a ride height of 6", book chassis and wishbones, and Cortina uprights, and came up with about 3.5" for a RC
height.
Answers may vary  , but I think it's in the ball park. Seems pretty reasonable, as it would be about 1/3 of the rear RC height using
13" wheels and a Panhard rod.
Somewhere I read that a RC above the Cg would cause the car to lean in, but if this were a great idea it would be standard practice, I suppose. High
RCs cause lateral thrust and wheel jacking. Don't ask me explain this last point; I understood it at one time, but the knowledge has fallen out
of my ear.
Pete
Pete
|
|
|
Tigers
|
| posted on 2/3/04 at 10:52 AM |
|
|
Thanks a lot for your answer!
Only problem is that my rear RC is quite low (102mm, when ride height is 140 mm). And I don't know shoul I try to lower it or no. Also there is
no much room for that. Maybe I just leave it and put rear anti rollbar...
Recently found usefull advice - You should design your suspension BEFORE you design a chasis.
Wish I knew it couple months earlier. 
|
|
|
MikeP
|
| posted on 2/3/04 at 01:19 PM |
|
|
I've got an analysis of the book chassis on my site at http://www.7builder.com/SuspensionGeometry/ based on some measurements of the book
uprights I got off the web.
Close to Pete's, the RC I get is 3.3" on the front. From what I've read, you want the front RC lower than the rear. I've no
practical experience to validate that.
Suspension program is my own (so ymmv), the measurements that I used are from the drawing that's available there too.
HTH!
|
|
|
pbura
|
| posted on 2/3/04 at 02:11 PM |
|
|
It's not too late to check your design and to make sure that your front and rear camber curves and RC movements are in control and compatible
with each other. You may have done this already, but if you haven't, I mentioned some good free tools on this thread:
http://www.locostbuilders.co.uk/viewthread.php?tid=11057&page=2
You must have your suspension geometry worked out at this stage of your build, but springing and anti-roll bars can wait until the car is otherwise
finished, and you can weigh the corners and determine Cg location. At that point, you can determine if the springing needed to control roll is too
stiff for comfort, in which case you will need ARBs. It would probably be good to assume that you are going to use them.
A great example of an analysis that would do the job is Smithees Weight Transfer
Worksheet. Notice how they are able to predict the roll angle of the chassis at 1G lateral force, and relative roll stiffness given by the
front and rear suspensions (which affects oversteer/understeer).
I also suggest that you contact Tudor Miron, who has at times generously offered to use his suspension design software to help other builders.
BTW, I really like your Russian drill! Got a hammer and sickle to go with that?
Pete
Pete
|
|
|
Tigers
|
| posted on 3/3/04 at 10:27 AM |
|
|
Thanxs!
It seems that there is lot of research to do ...
Ok, gettin' back to CADs
|
|
|
cymtriks
|
| posted on 7/3/04 at 03:20 PM |
|
|
Caterham Roll centres
As stated in Autocar story about an experimental independent rear suspension 7.
Original Deon chassis
rear RC 120mm
front RC 60mm
New Indy design
rear RC 65mm
front RC 30mm
And for interest this is the original Elise
rear RC 75mm
front RC 30mm
My own analysis suggests that low is good down to about 1 to 1.5 inches. The new Caterham and the Elise front RC values fall right in the middle of
this range. Below this RC height road bumps can cause some strange effects on suspension geometry. F1 cars go down to about 3/4 inch at the front but
these have near zero suspension movement and run on near perfectly flat race tracks.
|
|
|
pbura
|
| posted on 8/3/04 at 01:15 AM |
|
|
quote:
Original Deon chassis
rear RC 120mm
I wonder, how do they get it so low?
Thanks for the stats, Cymtriks, very interesting. Were there any other clues in the article about camber change, etc?
Pete
Pete
|
|
|
TheGecko
|
| posted on 8/3/04 at 01:55 AM |
|
|
quote:
I wonder, how do they get it so low?
With any solid axle, including de Dion, the roll centre position is set by the type of lateral locating linkage used. A panhard rod could be set up
to cross the centreline of the car 120mm off the ground (although I wouldn't do it!). A Watts linkage could also be set up this way (I
wouldn't do that either ). A WOBlink or Mumford link would be the best way to get a low RC on a solid axle, particularly the Mumford as the
RC point is virtual rather than a physical point on the linkage, helping greatly with ground clearance issues.
Hope this helps,
Dominic
(who is currently sketching endless variations of his rear suspension in a desperate but probably futile attempt to balance camber control vs roll
centre position vs roll centre movement vs locating points for arms vs frustration factor!  )
[Edited on 8/3/2004 by TheGecko]
|
|
|
pbura
|
| posted on 8/3/04 at 03:49 AM |
|
|
Can't imagine Caterham using a WOBlink or a Mumford link, hence my confusion.
Pete
Pete
|
|
|
Kitlooney1000
|
| posted on 10/3/04 at 09:00 PM |
|
|
scuse my ignorance, but what the hell is roll centre
|
|
|
JamJah
|
| posted on 10/3/04 at 11:24 PM |
|
|
I still cant work out what the engine types all mean. DOHC SOHC WTF etc. Ok not the last one!
I also cant work out how any of this matters. You dont like it, then you goto the scrappy and change it!
|
|
|
gerr80
|
| posted on 11/3/04 at 12:05 AM |
|
|
Suspension design
Dominic,
You need to read Alan Staniforth's book 'The Race and Rally Car Source Book' which details the string computer. This unbelievably
simple tool allows an almost infinite number of suspension measurements to be visualised and measured in less time than you would believe possible.
|
|
|
pbura
|
| posted on 11/3/04 at 01:12 AM |
|
|
quote:
Originally posted by Kitlooney1000
scuse my ignorance, but what the hell is roll centre
NOW you're asking this???
Just kidding, Lew If you built by the book, you're in good shape.
Each axle has a roll center, and when the two points are connected they form a roll axis, about which the sprung mass of the car rotates in roll, due
to centrigugal force. Here's a diagram of how a front roll center is determined:
The lower the RC, the more the car will roll. The higher the RC, the more sideways thrust is generated that will upset traction and cause the inside
wheel to jack up. So, your RC works with your springs to determine how much the car will roll in a turn.
Another thing about RC is that if you keep it in the same position relative to the chassis as the car is bobbing and weaving, handling will be more
predictable and the car will feel more secure.
There's a ton of info in suspension books and on the web if you're interested. Here's a good recent thread about books:
http://www.locostbuilders.co.uk/viewthread.php?tid=10975
I like this website for explaining suspension stuff:
http://members.aol.com/sccacuda/cars/1SmthTa.html
I have tons of respect for the Locost suspension; it's very well-balanced as far as roll centers, springing, etc., and there's not much to
add to it unless you just like to experiment. Most of the guys debating suspension design are building one-offs or very non-book cars.
I, for one, am (ahem) not presently building anything at all! Gosh, this keyboard is exhausting!!
Pete
Pete
|
|
|
TheGecko
|
| posted on 11/3/04 at 04:49 AM |
|
|
Pete,
I'm in agreement with all of that except forquote:
The lower the RC, the more the car will roll.
I would qualify the word
"lower" there to be "further away from the centre of gravity". The distance between the CG and RC is called the roll couple.
The roll couple is the lever arm that cornering forces acts on to roll the chassis around the roll centre. You already know that of course, this
re-iteration is just for kitlooney's benefit.
The longer the roll couple, the more the chassis will want to roll and the stronger the springs need to be (incluing anti-roll bars) to the detriment
of straight line ride and handling.
In theory, if the RC was above the CG the car would roll inwards on corners like a motorbike! As Pete correctly states however, high RC's are
evil for other reasons. The Trebron DRC suspension that was invented in the early '70s got around that by using a clever collection of links.
Effectively there were two RC's - one (low) for the suspension movement; and a second one (high) for the body. The end result was a car that
banked slightly inwards on corners whilst maintaining sensible camber angles. The complexity level isn't too high - might be an interesting
project for some one to build a DRC Locost - not me! I have some schematics and descriptive text if anyone is interested.
gerr80, I have Staniforth's book(s) and know about the String Computer. However, making it work well for a strut suspension is trickier than
double wishbones, because of the sliding rather than rotating element. That said, I tripped over a web page a while ago that showed a possible
solution. I suppose I should try again. Has anyone else successfully built a String Computer for struts?
Dominic
|
|
|
pbura
|
| posted on 11/3/04 at 07:14 AM |
|
|
Dominic,
I forgot you have struts! So I just did a Google and found a schematic showing how an RC is determined with struts; here's a link for those who
are interested:
http://www.miracerros.com/mustang/t_roll_center.htm
I've been staring at this thing for the past 20 minutes.
I think a 3-position analysis (similar to what's on my website) would work, solving for the bottom chassis pivot point.
Considering given data:
1. The outboard control arm position and strut angle are as per the donor uprights.
2. I'm wondering, is there such a thing as an optimal height for a strut, or do you just make them as long as possible, for the least change in
IC length with chassis movements? There's got to be some reason they make them so long! Or maybe you've got a particular strut you are
using, in which case the upper strut mounting position is now a known factor.
So, armed with these three points and a preliminary assumption for RC height, you could plot your chassis in three positions, with the wheel at the
desired camber, and see if there is an inboard control arm pivot that will satisfy the conditions.
Maybe you'll remember some time ago that you and I were discussing 4-bar linkages (on Locost Theory, I think) and how to solve them. After
that, I read an engineering mechanics book and learned about 3-position analysis for solving linkages, on which my drawing exercise is based. As I
recall, the technique works on sliding joints (such as a strut, before the innuendo starts) as well.
The way this went for me with the wishbones was that I didn't quite get the camber I wanted. However, my expectations were unrealistic, as I
later learned, and what I had was about as good as could be expected. RC control was perfect.
Hope this helps, really!!
Pete
Pete
|
|
|
Tigers
|
| posted on 11/3/04 at 11:29 AM |
|
|
RC data
I did calculations of my front and rear suspension, and got results:
Front
Rc height 3.2 inches
At 2degree right roll, left wheel camber -1.33, right 1.27, rc lateral movement -0.013 inches
At 1 inch bump wheel camber 0.48 degrees, at 1 inch drop -0.67.
Rear
RC height 6.59, static camber 0.3 degrees
At 2degree right roll, left wheel camber -1.09, right 0.4, rc lateral movement -0.26 inches
At 1 inch bump wheel camber 1.15 degrees, at 1 inch drop -1.85
May be you can share your numbers just to compare or give some judgement how bad is it?
|
|
|
Bob C
|
| posted on 11/3/04 at 12:54 PM |
|
|
strut length
Struts are made long because the braking torque is trying to put a bend in them. This makes them stick under braking - so your going into a corner
with the brakes on & the suspension stops moving - not good! Of course if you have inboard brakes it's OK.........
The longer the strut the less the brake induced stiction.
Bob C
|
|
|
pbura
|
| posted on 11/3/04 at 05:14 PM |
|
|
quote:
Originally posted by Tigers
I did calculations of my front and rear suspension, and got results:
Front
Rc height 3.2 inches
At 2degree right roll, left wheel camber -1.33, right 1.27, rc lateral movement -0.013 inches
At 1 inch bump wheel camber 0.48 degrees, at 1 inch drop -0.67.
Rear
RC height 6.59, static camber 0.3 degrees
At 2degree right roll, left wheel camber -1.09, right 0.4, rc lateral movement -0.26 inches
At 1 inch bump wheel camber 1.15 degrees, at 1 inch drop -1.85
May be you can share your numbers just to compare or give some judgement how bad is it?
Tigers,
Here's what I have (nothing for the rear, sorry). Book figures were hacked out from the book and 2nd hand web info, so may not be 100%
accurate. "My" figures were a design exercise, just trying to figure out how to approach the problem. I'd take them, though! I
plugged in 0 static camber for comparability.
BOOK
Static: camber 0, RC height 3.47
2 degrees roll: unloaded wheel camber -1.34, loaded 1.22, RC lateral movement 1.86 inches
1 inch bump: camber -0.77 degrees (slight difference in terminology I think; no matter)
1 inch droop: camber 0.66 degrees
"MINE"
Static: camber 0, RC height 2.49
2 degrees roll: unloaded wheel camber -1.42, loaded 1.31, RC lateral movement 0.25 inches
1 inch bump: camber -0.78 degrees
1 inch droop: camber 0.65 degrees
I'd take your numbers, too, sir (though I'd prefer a real car)! Very nice!
Syd, you're both right and wrong, IMO. If you're building to the book, fine; the book is reliable. But what if you're not building
to the book because of chassis size, IRS, different uprights, etc? What do you do then except get out your string and pencil or hire a designer?
Pete
P.S. Thanks, Bob, for the explanation re: strut length
[Edited on 11/3/04 by pbura]
Pete
|
|
|
pbura
|
| posted on 11/3/04 at 06:18 PM |
|
|
Syd,
Won't disagree that there aren't a few ways to skin this cat I thought that perhaps we had an issue as far as whether the cat required
skinning at all 
Mr. Bura to you (apparently)
Pete
|
|
|
pbura
|
| posted on 11/3/04 at 07:26 PM |
|
|
quote:
Originally posted by Syd Bridge
What's all this talk about cats, and removing the coverings thereof??(Pussy fetish/obsession??)
LOL to that! I enjoy slightly fractured metaphors, btw, and use them frequently.
Are you serious about the boat/composites/middy? If so, that would be a blast
Well, Syd, I'm not trying to get into any pissing match about methods here. It's all good. I wonder how Tigers did his?
Pete
Pete
|
|
|
Kitlooney1000
|
| posted on 11/3/04 at 07:41 PM |
|
|
thanks for the explanation, i now have a head ache so i think i will stick to the books design for now.
|
|
|
MikeP
|
| posted on 11/3/04 at 11:18 PM |
|
|
If read quite a bit about this stuff myself out of interest - but no good real world experience. I *think* Pete's in the same boat - sorry if
I've assumed too much Pete. How about you Syd? I'd love to hear some war stories like: "We were having a real problem with the car,
till we realized the roll center was too high/low (or moving too much, or whatever). We fixed it by ... and it handled great after that".
I might have stories of my own like this once I get my own car on the road - but I hope not .
|
|
|
pbura
|
| posted on 11/3/04 at 11:35 PM |
|
|
Good call, Mike, indeed I am a tyro
I think most people will agree, though, that you can learn a lot by watching, listening, and reading. It gives you a bit of a head start on the
doing. This MO has worked for me for car repairs, messing with computers, gardening, carpentry, and many other tasks, so why not car-building?
Again, studying is a head start, not expert knowledge. To a certain extent, this is physics, so I feel OK about commenting in a limited way. You
won't catch me saying boo-wah about welding, though
Agreed that it would be great to hear some war stories
Pete
[Edited on 11/3/04 by pbura]
Pete
|
|
|
Tigers
|
| posted on 12/3/04 at 02:57 PM |
|
|
Cat thigies...
What me? I skinn no cats. Me love cats )))))
As for me, then in my desing there are lot of constraints - things I can't change. And I build as I can. But also I would like to know how to
change things I can change to make it handle normal. And the way people (who understand something) don't laugh when I show them my car.
Ps. Excuse my french ))
|
|
|
