Roll Center

[BARCRacing10]

I'm building a new car and I've been told that it's very crucial to find the roll center. Every thing I can find online either tells me why it's important to know or tries to sell me some software for figureing it. Does anyone know if it's possible to simply take measurements and figure? If so how do I do it? Thanks!

[NickHolt]

I'm building a new car and I've been told that it's very crucial to find the roll center. Every thing I can find online either tells me why it's important to know or tries to sell me some software for figureing it. Does anyone know if it's possible to simply take measurements and figure? If so how do I do it? Thanks!

 

We teach how to do that in the Suspension Seminar. In fact, it's one of the first homework assignments we do - figure out the front and rear roll center heights on your racecar. Very eye-opening assignment for some... LOL

 

It's a bit detailed to exlain in a short forum post, but basically you plot out the right front suspension at ride height on some graph paper, draw lines through the A-arms to an "instant center" then draw a line back to the centerline of the opposite tire contact patch. The distance (in inches) between where that line crosses the center of your car and the ground is the approximate roll center height. I have left out a lot of detail, but that's the main idea.

 

During the seminar I pass out notes on how to do it complete with a sample graph.

 

Nick

[Truck99]

I'm building a new car and I've been told that it's very crucial to find the roll center. Every thing I can find online either tells me why it's important to know or tries to sell me some software for figureing it. Does anyone know if it's possible to simply take measurements and figure? If so how do I do it? Thanks!

 

We teach how to do that in the Suspension Seminar. In fact, it's one of the first homework assignments we do - figure out the front and rear roll center heights on your racecar. Very eye-opening assignment for some... LOL

 

It's a bit detailed to exlain in a short forum post, but basically you plot out the right front suspension at ride height on some graph paper, draw lines through the A-arms to an "instant center" then draw a line back to the centerline of the opposite tire contact patch. The distance (in inches) between where that line crosses the center of your car and the ground is the approximate roll center height. I have left out a lot of detail, but that's the main idea.

 

During the seminar I pass out notes on how to do it complete with a sample graph.

 

Nick

 

But what about the lateral location and RC migration Nick?

[NickHolt]

But what about the lateral location and RC migration Nick?

 

A couple of those "details" that I left out.. LOL.. really, the best bet is to get some suspension software and let the puter sort it out. But be warned, the software is only as good as the measurements you feed into the system. And there is some things to be gained by knowing how your RC moves around the chassis, but trying to get it perfect at the expense of say, the left side camber curve is a bad trade off... RC height is the key. The RC is going to move around for sure, but if you find a spot where the static RCH and it's associated moment arm is working for you, it's better to play with other, more important parameters than lateral RC location and migration. Now, I take it all back if you're messing with a BB/ss setup and running on stops.

 

Nick

[sm44]

I'm building a new car and I've been told that it's very crucial to find the roll center. Every thing I can find online either tells me why it's important to know or tries to sell me some software for figureing it. Does anyone know if it's possible to simply take measurements and figure? If so how do I do it? Thanks!

 

We teach how to do that in the Suspension Seminar. In fact, it's one of the first homework assignments we do - figure out the front and rear roll center heights on your racecar. Very eye-opening assignment for some... LOL

 

It's a bit detailed to exlain in a short forum post, but basically you plot out the right front suspension at ride height on some graph paper, draw lines through the A-arms to an "instant center" then draw a line back to the centerline of the opposite tire contact patch. The distance (in inches) between where that line crosses the center of your car and the ground is the approximate roll center height. I have left out a lot of detail, but that's the main idea.

 

During the seminar I pass out notes on how to do it complete with a sample graph.

 

Nick

Dang Nick, I thought you just got in pressed the loud pedal and turned left, this stuff is complicated.

[NickHolt]

Not really all that complicated. It's just that it's hard to find a reliable source of information that is not all physics jargon and complicated math.

 

That's why I teach suspension engineering principles in plain English in my seminars. When I designed the course I imagined that my Mom was sitting in the front row and made it as understandable as possible. I also pass out lots of class note because, as you say, there is a LOT to learn and it's very easy to forget all the principles. Many of my students take the seminar more than once and get more out of it each time. One student took the course five times!

 

I think the hardest theoretical concepts to really understand are 1) roll centers, 2) moment arms, and 3) the relationship between roll centers and moment arms, 4) roll couple distribution, and 5) caster.

 

But the goal of the seminar is to present a complete picture of how all the various parameters function together as a package. Lots of racers have very detailed knowledge of how this or that works, but a very fuzzy idea of how this or that should work together. They know a lot about a lot of branches, but they really don't know how they fit on the whole tree!

 

Anyway, I got very busy this off season and didn't conduct a seminar this time. I feel badly about that because there were several teams that really want to attend.

 

Nick

[Wayne Walker]

Hey Nick,

 

Don't forget about the dirt 4-Link stuff. Want to talk about "migration"

[abrungot]

Here a poor dude comes in trying to get help.. & ya'll in a matter of a few post's confuse the hell out of him...

Nice..

 

http://www.circletrack.com/howto/1816_fron...tion/index.html

[NickHolt]

Hey Nick,

 

Don't forget about the dirt 4-Link stuff. Want to talk about "migration"

You got that right! Although the obscene amount of rear roll steer induced by the typical dirt four-link setup plays a bigger role than whatever roll center changes occur in one of those bad boys.

 

Nick

[NickHolt]

Here a poor dude comes in trying to get help.. & ya'll in a matter of a few post's confuse the hell out of him...

Nice..

 

http://www.circletrack.com/howto/1816_fron...tion/index.html

 

Anyone who has taken my suspension seminar should knows that the roll center height relative to the Center of Gravity Height (CGH) forms a moment arm. The distance between the roll axis (formed by a combination of both the front and rear vertical roll center heights) and the CGH produces a moment arm that is the key moment arm in any conventional setup.

 

I place little importance on lateral roll center location because quite often camber curves are neglected in order to obtain the "correct" lateral roll center locations. To my way of thinking, achieving good camber curves is more important than moving things around to get "correct" lateral roll center locations. This is where Mr. Bolles and I tend to disagree, although I see that in this current Circle Track article he has adjusted his thinking more towards the way I teach it.

 

All roll centers migrate if the suspension moves at all as in a conventional setup. The key is to come up with a static roll center height that works well in the dynamic conditions that are present in that car and at that track. I have found a sweet spot that seems to work well with a wide variety of cars that use a conventional setup, but this is highly classified information. I do share that info during my suspension seminars.

 

Nick

[C_Malin]

After you figure out where your roll center is located. send me a message, and I'll tell you how to find out if its in the right location.

 

I'm building a new car and I've been told that it's very crucial to find the roll center. Every thing I can find online either tells me why it's important to know or tries to sell me some software for figureing it. Does anyone know if it's possible to simply take measurements and figure? If so how do I do it? Thanks!

[BARCRacing10]

Abrugnot - Thanks for the website! That is the most helpful one that I have seen yet. It actually tells you what to measure! Thanks!

 

C. Malin - Thanks for the offer... I may take you up on that before too long

 

Nick - Thats for regurgitating exactly what every website I had been to before had said! "It's really important... I teach a class on it so I'm not going to try to give you any free information that is "highly classified" but here are all the terms you need to know even though they will do you absolutly no good because without taking my suspension class I'm not going to tell you how to locate or apply those theories"

But let me know when you give another one of those suspension classes

 

I guess a better question would have been:

Is there a "good ole boy" way of measuring roll center - something that could be measured and adjusted at the track if you are unhappy with the handleing of the car?

[Truck99]

Is there a "good ole boy" way of measuring roll center - something that could be measured and adjusted at the track if you are unhappy with the handleing of the car?

 

No. To get the best and most accurate measurements, you need a flat and level surface, a plumb bob or two, a couple of tape measures, and LOTS of time.

 

Depending on the car, many find taking the measurements very difficult to do with the motor in the car.

 

This will open a whole different can of worms, but IMO, there is no perfect roll center. Any roll center within reason can be made to work well with the right springs and suspension components. Like Nick said, you may be better served focusing on things like caster and camber - both static and the camber curve in bump and rebound, and bump steer.

[Wayne Walker]

88, Check your pm's

[CuttingEdgeRacing]

Anyway, I got very busy this off season and didn't conduct a seminar this time. I feel badly about that because there were several teams that really want to attend.

 

Nick

 

With that being said, when ARE we getting another chance at it??

[jwmbishop]

I'm building a new car and I've been told that it's very crucial to find the roll center. Every thing I can find online either tells me why it's important to know or tries to sell me some software for figureing it. Does anyone know if it's possible to simply take measurements and figure? If so how do I do it? Thanks!

The software is the best bet. Static positions are only the first step - what the car is doing from up in front to down in front during entry braking and then adding total roll at apex is the most important measurement you can have and the software eliminates (or more precisely does for you) many calculation steps between the two. You can even download and install the demos for free and the visual will take you a long way toward understanding what and how best to measure and change!

 

Now if you are using just "simple" roll ctr as a starting point for trial and error tuning it is fairly easy to visualise and measure.

 

Sit on the floor looking at the front of your car. Project the plane that the right front control arms create inward (to left of car - your right) to the point in which they intersect now named A (this point is typically 1-3 feet to the left of the car). Project an imaginary line from A to the center of the right tire contact patch - line R.

Repeat (in opposite) for the left side to create point B and line L. Where R and L intersect is your roll center. To put on paper simply meaure the heights of the inner bushings (use an average between front and rear bushing) and ball joints from the floor at ride height and graph these measurements. As you will see using wheel spacers, ride heights, lowering or raising spindles etc this static (and the resulting dynamic) is quite tunable even in classes where stock parts and their mountings are required! The easiest parts to effect this is the ball joints and bushings. Just how off center can you make a stock bushing and just how far can you bend a ball joint before it breaks are lessons you will learn doing it this way! Taller upper ball joints give you more angle and shorten lines R or L - which then raises or lowers the RC. Wheel spacers and ride height have more left to right effect.

 

Here is a link to a good picture demonstrating kinematic roll center. In this example the RC is below the ground. This will create a very tight car and would thus require softer springs to get it to turn without creating push loose. Of course thats with the assumption that the center of gravity is well above ground. If the CG is only 2-3 inches higher than the RC the CG weight (acting as a lever) will have less impact than if CG is 5 inches above ground (longer lever has more effect). If the CG is BELOW roll center the car does not ROLL it BANKS (watch an indy car turn left in slow mo - you will see the weight pass under the rc and the car will actually bank left at the first moment input of left steering - before the spring rates "settle" it in to where the engineer wanted the weight to land). The lower (and further left) the RC the less weight transfers off of the inside (Left) tire - seeing a car three wheel out of the corner (winged cars excepted) is distressful to a guy who knows what is happening! Four tires will ALWAYS produce more overall traction than three!

 

As much as quoting Circle Track Magizine pains me - they have a rather extensive series of articles in the past 4 years covering just what you are wanting to learn. Free info online is always better than buying the textbooks!

[jwmbishop]

Here a poor dude comes in trying to get help.. & ya'll in a matter of a few post's confuse the hell out of him...

Nice..

 

http://www.circletrack.com/howto/1816_fron...tion/index.html

 

Anyone who has taken my suspension seminar should knows that the roll center height relative to the Center of Gravity Height (CGH) forms a moment arm. The distance between the roll axis (formed by a combination of both the front and rear vertical roll center heights) and the CGH produces a moment arm that is the key moment arm in any conventional setup.

 

I place little importance on lateral roll center location because quite often camber curves are neglected in order to obtain the "correct" lateral roll center locations. To my way of thinking, achieving good camber curves is more important than moving things around to get "correct" lateral roll center locations. This is where Mr. Bolles and I tend to disagree, although I see that in this current Circle Track article he has adjusted his thinking more towards the way I teach it.

 

All roll centers migrate if the suspension moves at all as in a conventional setup. The key is to come up with a static roll center height that works well in the dynamic conditions that are present in that car and at that track. I have found a sweet spot that seems to work well with a wide variety of cars that use a conventional setup, but this is highly classified information. I do share that info during my suspension seminars.

 

Nick

I agree with ya wholeheartedly. Mr Bolles tends to irritate me everytime he brags on being innovative and claims to have coined the prase "roll Moment center" in the late 80's.

 

I read "Concepts in Advanced Automotive Suspension Design" copyrighted 1972, while studying at San Jose in 87. This book had a wonderful explanation and several workpage examples and detailed roll center, roll moment, roll momentum and roll momentum centerline (axial comparison between front and rear). So I guess being innovative includes just dropping an "um" and omitting a "line".