C2,C3 Corvette Roll Center

Shocks, Springs, Brakes, Frame, Body Work, etc

Moderator: Team

66Vette

Re: C2,C3 Corvette Roll Center

Post by 66Vette »

clshore:

Thanks for the procedure. Interesting process. I think I can get through the math, though been a long time since any ME courses (I'm a practicing EE). I do not have scales, but have some contacts where maybe I can borrow them.

Cris
clshore
Expert
Expert
Posts: 703
Joined: Sat Dec 12, 2009 4:48 pm
Location: Beverly Hills, FL

Re: C2,C3 Corvette Roll Center

Post by clshore »

An alternative brute force method:
Securely attach a chain or sling off center to roll bar or similar robust lifting point.
Measure the point of attachment WRT to 3 hard points on chassis.
Raise the car entirely off the ground by chain/sling, allow to stabilize.
Once airborne and stable, centerline of the sling/chain will pass directly through the attachment point and also the CG.
Measure the chassis angle WRT level (front/rear and left/right).
Use Trig to calculate the x,y,z position of the CG relative to the attachment point and hard points based on angles and distances.

Or, you could use one of those laser levels to paint a vertical line on the sling/chain, and where the line paints the body, mark with grease pencil or tape and pencil.
Now reposition the laser (or spin the car) 30 degrees or so and repeat body marks. Do this 2-3 times on each side of the car.
The body lines represent a plane that 'slices through the body and contains the CG
The slices all intersect at the same line that also passes through the CG and also through the point of attachment.

Not sure that the math will be any easier though. ;(

Some folks have started using the XBox Kinect and such as a low resolution (and low cost) 3D scanner system:

http://grassrootsmotorsports.com/forum/ ... 379/page1/

Handy for scanning things like an engine bay for motor swaps, a car interior for building cages, suspension setups for making flares,
underbody for making exhausts, etc.

The big deal is the open source software that can handle the scans as CAD files, so that you can pull measurements, do modeling for changes, etc.

I'll bet a setup like that could also calculate the CG based on a scan of a car suspended over a level surface.

OK, I'll set the crack pipe down now, and go quietly.
User avatar
MadBill
Guru
Guru
Posts: 15024
Joined: Tue Nov 15, 2005 10:41 am
Location: The Great White North

Re: C2,C3 Corvette Roll Center

Post by MadBill »

A couple more thoughts:
1. Toe-out is always bad. I'd use enough static toe to ensure that the car retains some toe-in at full bump. The added tire drag will be undetectable in such a high powered car.

2. I have seen many early Corvettes with loose-fitting (or even missing entirely!) half shaft inner end C-clips. Since as mentioned the shaft acts as the upper control arm, this slop causes rear camber to increase positively, often by a couple of degrees, when the longitudinal forces in the outboard shaft change from compressive to tensile (which I believe happens at ~0.5 g).

3. Off-the-wall suggestion*: Since roll stiffness can of course be increased with stiffer springs, stiffer bars or a combination thereof, how about moderately stiff springs with big enough bars to achieve the needed track characteristics, but use extra long links with stiff coil (valve?) springs fitted in series with the Delrin bushings? On the street, moderate preload would allow these springs to effectively lower the roll stiffness and they could be wound down to coil bind to achieve the needed roll rates at the track. (*I have a pretty well unbroken record for reinventing the wheel, so someone may well already offer such a component.)

4. My gut feel is your CG is more like 17".

5. Some scales claim the ability to calculate CG, but I haven't seen them in action. (I will in the next few weeks however.)

6. All else being equal, lowering the roll center at either end will reduce the roll stiffness there, changing over/understeer balance.
Felix, qui potuit rerum cognscere causas.

Happy is he who can discover the cause of things.
clshore
Expert
Expert
Posts: 703
Joined: Sat Dec 12, 2009 4:48 pm
Location: Beverly Hills, FL

Re: C2,C3 Corvette Roll Center

Post by clshore »

MadBill wrote:A couple more thoughts:
1. Toe-out is always bad. I'd use enough static toe to ensure that the car retains some toe-in at full bump. The added tire drag will be undetectable in such a high powered car.
...
Actually, in AutoX, not always so.
The goal is to negotiate what are usually very tight corners, and a bit of 'on purpose" instability can often help to rotate the car, if the driver is prepared to use it.
Depends on the car, the course, and the driver.
Brian P
Guru
Guru
Posts: 1611
Joined: Wed Jan 02, 2008 7:35 pm
Location: Toronto, Ontario, Canada

Re: C2,C3 Corvette Roll Center

Post by Brian P »

Fine, but you're complaining that the car is twitchy, and toe-out that comes at the wrong time will do that. You say that on sweepers the car takes a set - that's because it has gone to a certain amount of body roll with the associating amount of rear steer (slight roll oversteer) and has stabilized. It's when you're going from side to side and the rear steer effects are moving around that it is going to do that.

The issue isn't necessarily so much that you have toe-out, but rather that you are having rear steering effects (toe-out on compression) and those will change around with body roll as you go from side to side.
clshore
Expert
Expert
Posts: 703
Joined: Sat Dec 12, 2009 4:48 pm
Location: Beverly Hills, FL

Re: C2,C3 Corvette Roll Center

Post by clshore »

Brian P wrote:Fine, but you're complaining that the car is twitchy, and toe-out that comes at the wrong time will do that. You say that on sweepers the car takes a set - that's because it has gone to a certain amount of body roll with the associating amount of rear steer (slight roll oversteer) and has stabilized. It's when you're going from side to side and the rear steer effects are moving around that it is going to do that.

The issue isn't necessarily so much that you have toe-out, but rather that you are having rear steering effects (toe-out on compression) and those will change around with body roll as you go from side to side.
True, but so what?
My statement simply addresses 'Toe-out is always bad'.
Brian P
Guru
Guru
Posts: 1611
Joined: Wed Jan 02, 2008 7:35 pm
Location: Toronto, Ontario, Canada

Re: C2,C3 Corvette Roll Center

Post by Brian P »

I'm trying to return your focus to the limitations of your suspension design and how the issues and set-up relate to the symptoms that you describe.

The issue is not so much "toe-out" as it is "rear steering" i.e. bump / roll steer.

With the way you have it right now, when you enter a corner and the weight shifts, the outside wheel will toe-out a lot on compression because the drive shaft (acts as the toe control link with that suspension design) is pulling the back of the trailing arm inward. The inside wheel will toe-in a little because the drive shaft is letting the back of the trailing arm on that side outward a little (but the drive shaft is closer to horizontal so this effect is less). The net effect is rear-steering towards the outside of the corner plus a bit of toe-out bias - "roll oversteer".

You say that it takes a "set" in high speed sweepers ... that's because once the amount of body roll stabilizes, so does the rear-steering effect, and once everything finds an equilibrium of some sort, that is your "set". The rear will be angled out a little because of the rear steering, but it's a constant amount, so it's not a problem.

You also say that the car is twitchy in transitions. The above rear-steering behaviour explains that.

And you state that in autocross, a certain amount of what you call toe-out (and I will call roll oversteer) can be beneficial. Well ... that's what "twitchy" feels like.

If you want the car to have more stable and predictable steering response - less "twitchiness" - then you need to dial out the amount of roll oversteer.

If you want the rear of the car to steer towards the outside of the corner, one way to do that is "roll oversteer", another way (if you don't have a rear suspension design with roll oversteer) is to set it up with static toe-out. Either way, it makes the car feel twitchy, and it's what you already have!
clshore
Expert
Expert
Posts: 703
Joined: Sat Dec 12, 2009 4:48 pm
Location: Beverly Hills, FL

Re: C2,C3 Corvette Roll Center

Post by clshore »

Brian P wrote:I'm trying to return your focus to the limitations of your suspension design and how the issues and set-up relate to the symptoms that you describe.

The issue is not so much "toe-out" as it is "rear steering" i.e. bump / roll steer.

With the way you have it right now, when you enter a corner and the weight shifts, the outside wheel will toe-out a lot on compression because the drive shaft (acts as the toe control link with that suspension design) is pulling the back of the trailing arm inward. The inside wheel will toe-in a little because the drive shaft is letting the back of the trailing arm on that side outward a little (but the drive shaft is closer to horizontal so this effect is less). The net effect is rear-steering towards the outside of the corner plus a bit of toe-out bias - "roll oversteer".

You say that it takes a "set" in high speed sweepers ... that's because once the amount of body roll stabilizes, so does the rear-steering effect, and once everything finds an equilibrium of some sort, that is your "set". The rear will be angled out a little because of the rear steering, but it's a constant amount, so it's not a problem.

You also say that the car is twitchy in transitions. The above rear-steering behaviour explains that.

And you state that in autocross, a certain amount of what you call toe-out (and I will call roll oversteer) can be beneficial. Well ... that's what "twitchy" feels like.

If you want the car to have more stable and predictable steering response - less "twitchiness" - then you need to dial out the amount of roll oversteer.

If you want the rear of the car to steer towards the outside of the corner, one way to do that is "roll oversteer", another way (if you don't have a rear suspension design with roll oversteer) is to set it up with static toe-out. Either way, it makes the car feel twitchy, and it's what you already have!
Just to be clear, I'm not the OP (66Vette) and it's not my car.
That said, I'm well aware of rear-steering effects and behavior (if you think the Corvette rear-steers, try a swing axle suspension).

I still believe that the major effect the OP was describing is from compliance, not geometry per-se.
But it also could be roll damping.
Brian P
Guru
Guru
Posts: 1611
Joined: Wed Jan 02, 2008 7:35 pm
Location: Toronto, Ontario, Canada

Re: C2,C3 Corvette Roll Center

Post by Brian P »

Ah heck, you're right, I mixed up who is posting what.

There could certainly be compliance issues on top of the rear-steering issues. It all adds up ...
66Vette

Re: C2,C3 Corvette Roll Center

Post by 66Vette »

Guys, sorry. I stepped away from this post figuring it was dead.

MadBill, keep me posted on your experience of measuring CG with the scales. I would assume the car has to be tipped a little to do the measurement.

Good points on the toe out under compression. If I really wanted to cut and hack, I could reverse this by changing the half shaft angles.....it could be reversed to toe in under compression. The half shafts are parallel right now and I don't have that much body roll, so I'm at least at the point where the toe (out) gain is the least. It seems I could also change the lower strut location and alter the toe gain, of course, all this stuff alters the camber curves as well....sheesh....

I'm going to take more measurements and see if a suspension guru can stuff this into his program just to look at dynamic effects.
mbrooks
Expert
Expert
Posts: 851
Joined: Fri Nov 16, 2007 8:41 am
Location: Kansas, USA

Re: C2,C3 Corvette Roll Center

Post by mbrooks »

I have an IRS on the JBL, the rear toe adjusters are linked to the A arm so there is no bump steer. You might take a gander at the pics and get some ideas.

http://www.jblmotor.com/chassis.htm
PackardV8
Guru
Guru
Posts: 7619
Joined: Sun Jul 30, 2006 2:03 pm
Location: Spokane, WA

Re: C2,C3 Corvette Roll Center

Post by PackardV8 »

I forgot where this quote came from, but "you can make any bad suspension design work if you don't let it move!"
You over-complicated the quote:

"Any suspension can be made to work if you don't let it."
Jack Vines
Studebaker-Packard V8 Limited
Obsolete Engineering
User avatar
MadBill
Guru
Guru
Posts: 15024
Joined: Tue Nov 15, 2005 10:41 am
Location: The Great White North

Re: C2,C3 Corvette Roll Center

Post by MadBill »

66Vette wrote:Guys, sorry. I stepped away from this post figuring it was dead.

MadBill, keep me posted on your experience of measuring CG with the scales. I would assume the car has to be tipped a little to do the measurement.....
Correct. Intercomp scales instructions, e.g. this model: http://www.intercomp-racing.com/Product ... LE_900.cfm just require you to baseline the corner weights then raise one end of the car, preferably by at least 1" per ft. of wheel base, enter the WB and the height raised and Presto: CG is reported.
Felix, qui potuit rerum cognscere causas.

Happy is he who can discover the cause of things.
Dan Timberlake
Guru
Guru
Posts: 1737
Joined: Tue Jan 15, 2013 8:10 pm
Location:

Re: C2,C3 Corvette Roll Center

Post by Dan Timberlake »

MadBill wrote:A couple more thoughts:
snipped

4. My gut feel is your CG is more like 17".
Looks like stock 63 at standard ride height was around 16.5.
http://www.web-cars.com/corvette/1963_S ... php?page=7

lowering the ride height ought to lower the CG. lightening efforts below CG might tend to raise it.
clshore
Expert
Expert
Posts: 703
Joined: Sat Dec 12, 2009 4:48 pm
Location: Beverly Hills, FL

Re: C2,C3 Corvette Roll Center

Post by clshore »

MadBill wrote:
66Vette wrote:Guys, sorry. I stepped away from this post figuring it was dead.

MadBill, keep me posted on your experience of measuring CG with the scales. I would assume the car has to be tipped a little to do the measurement.....
Correct. Intercomp scales instructions, e.g. this model: http://www.intercomp-racing.com/Product ... LE_900.cfm just require you to baseline the corner weights then raise one end of the car, preferably by at least 1" per ft. of wheel base, enter the WB and the height raised and Presto: CG is reported.

Cool, MUCH easier than the procedure I posted.
Post Reply