Negative Rollcenter Height

[wellis77]

As I wade through the mysterious world of suspension design I'm curious about the effects of negative rollcenter height (meaning below ground) in dive. I've got a lot to learn but one thing I've read regarding rollcenter height is that static height is good around 1-3 inches. One scenario I'm playing with has a roll center just above 1" but under 2" of dive, it goes negative. What are the effects of such a change?

Also, I have a theoretical understanding of anti-dive but is there a target that suspension designers typically shoot for on a car built for handling? I read that many race cars have 0% while passenger vehicles have as much as 30%+, is there a general happy medium for a performance oriented car?

Thanks for all the feedback. I'm learning a lot going through this and am enjoying the process but there sure is a lot of theory involved.

Will

[Bryce]

The roll couple is larger therefore a higher roll stiffness is needed, larger sway bar.

Anitidive is going to depend on CG height and brake bias. Race cars usually have a really low CG so hihg antidive percentage would not do much.

[exwestracer]

The lower the roll center, the better theoretical traction at that end of the car. Lowering the roll center reduces the weight transfer up off the inside tire as the body rolls outward in a turn.

Imagine the car had a solid suspension. As you lift one side of the car, that tire will immediately come off the ground. A high roll center has much the same effect.

I say better "theoretical" traction because a lot of body roll leads to a lot of camber change, loss of contact patch, and a very unsettling feel for the driver. It also screws around with any front aero we might have aiding traction. So, as Bryce pointed out above, we usually artificially raise the roll center with bars anyway. Lowering the C/G will allow a low roll center without a lot of roll moment and allow us to run a smaller bar, giving the suspension a little more freedom to work.

[wellis77]

Thanks for the explanations guys, very helpful.

[JRouche]

And Im not sure, but Ill comment.

With street cars the front, (you are talking about the front right) roll center is usually very low. Close to ground level or like you said an inch or two above statically.

Now when you say "dive" I think of complete front end dive. Like during braking on a straight. Both sides compress at the same amount. Thats not so much of a problem.

Now if you mean suspension compression in a turn. Yes the roll center will usually dive, and it will migrate. It goes off center and to one side. Thats the biggy to think about. Because in a turn the roll center will shift sideways. It (the roll center) will dive down (below ground surface) and shift sideways also. And for the most part you cant control that with a stock front end, its due to the control arms. BUT!!! You can take advantage of it if you know how its gonna react in the turn.

Roll couple. The line thats drawn from the rear roll to the front roll. Thats very important.

Think of it like driving a solid rod through both roll centers and thats how your car will "roll" as it goes through the turns.

The rear is usually higher than the front and so the front tends to roll more because the weight of the car is ABOVE the roll center. And on front engine cars its even worse. Much more weight above the RC at the front. So we put heavy roll bars in place to counteract that.

But sway bars are linear compared to the moving front RC height and side to side location. Its like a bandaid for a dramatically moving RC on the front.

And yes, thats where the control arms come into play. A properly setup front suspension will limit the side to side migration of the RC up front. So you are working on keeping the roll couple from the front to back as steady as possible.

And solly for going all off the subject.

YES, try to keep the roll center from diving too far. But really!! Look at the migration from side to side as well. Thats gonna give more strange reactions during driving than how much it (the roll center) dives below ground. JR

[wellis77]

Thanks for the extended explanation JR. I am mostly thinking of dive in my question but your reply brings up an obviously important topic. I've been looking at RC migration recently as well but in my research haven't been able to find a reasonable amount of movement, outside of "limit it as much as possible". Like you and others have already mentioned, the anti-roll bar will help but it's a band-aid so if I can set the suspension up and limit the size of the bar, obviously that's idea. If I angle the LCA's up towards the frame (away from the ball joint), that will increase my roll center height, reduce the need as big of a bar (when considering dive only). I'll have to look closer at what that will do to the RC migration. you mentioned the LCA's are used to control RC migration. What might be some other ways the LCA's can be set up in order to reduce migration? Each build is different but and full of compromises, but it will give me some ideas to look at and see where it goes. Thanks again JR.

[ArtosDracon]

Also don't forget the leverage effect of the distance between the roll center and the center of gravity, the greater the distance between the two, the more body roll you're going to experience. When designing my front suspension, I actually found it was better to bring the roll center up almost an inch to about 4.25" in order to decrease roll center migration and decrease body roll.

[wellis77]

Good thought Robert. The best way to raise the RC is via the LCA frame mounts correct? Granted, other things will have to be considered in doing this, but is this also the best way to limit RC migration? The problem I see in raising the LCA frame mounts is the arm is then no longer centered and will experience positive camber gain at the start of dive. So in order to maintain only negative camber gain throughout dive, where is the RC adjustment made, shifting the UCA mount up? This seems like it would have an additional side effect of both slowing camber gain or as well introducing positive camber gain before going negative.

[ArtosDracon]

The best way to decrease roll center migration is to make the upper and lower control arms as parallel as possible. That can make it difficult to get proper camber gain though, which is more important. Do you have measurements for pivot locations and angles of the UCA mounts and ball joing locations?

[ArtosDracon]

I should really expand on that.

Control arms closer to parallel moves the Instant Center for a specific side further away from the tire itself, that it turn makes the lever about which the tire is moving, much longer, so a smaller distance will be covered at the instant center for the same distance in vertical movement at the tire. Which will decrease Roll Center migration, because the roll center is based on the intersection of lines drawn between the Instant Centers and the center of the tire patch. However, in order to properly gain camber under compression most suspensions are SLA designs, and with a shorter upper arm, the Instant center will move both vertically and horizontally. The easiest way to minimize the vertical movement of the IC, which is what effects the height of the Roll Center is to keep the lower control arm as level as possible at the broadest range of motion possible.

This is all overly simplified of course because despite all the math physically possible, you can only set your car up "ideally" for one corner, the rest is a balancing act between all the different corners you have to take, and very few of us even bother to look at tire deflection, bushing deflection and chassis deflection when calculating the suspension geometry. If your frame deflects a quarter inch during roll, every calculation you've done may as well be thrown out the window, the same goes for tires and bushings. It can be maddening.

One good source I've found is here:
http://www.neohio-scca.org/comp_clinic/hand_out_reprints/Vehicle%20Dynamics2007.pdf

[wellis77]

Thanks Robert. I've actually read through that paper a couple times and it has helped alot. I've got a few different setups I'm playing with and tweaking and I'm satisfied with most everything except RC migration. I can get pretty much everything in a respectable range but that. It's driving me nuts!

[ArtosDracon]

I just read your sig(might have been a good place to start huh?) And noticed you're using the Gen 3 Viper IFS and I'm willing to bet part of the problem with RCM is due to the track width difference. Control arm lenghts are optimized to track widths, so modifying the track width will mess with the suspension in a few ways, none larger than RCM though. This is likely going to be a situation where you can only do so much. Have you played with Suspension Analyzer? It has a few day trial, so once you think you're really close, you can put all your info into it and see where you're at, it will visually and numerically track things that are REALLY annoying to have to calculate through roll, like RCM and ackerman.

[wellis77]

Hey Robert,
I have been playing with suspension analyzer, my trial runs out tomorrow. BUT, I think I found a combination that works. I wish there was an easy way to post the reports to get some feedback. You are right about the track width being a problem. I had to get shorter UCA's just to fit my engine between them. In doing so, I need to adjust the length of the LCA's as well and that has been a big ? of how much. I just worked a combination that might be a good compromise of many things. On the RC migration I was able to reduce it so the max lateral movement was 1.689" at 2" dive and 3* roll, and at 2" of dive it goes from 3.115" to 1.028". Camber gain is .9* from 1"-2" and .7* from 0-1". I am not 100% certain I'm reading the ackerman report correctly but if I am, at the most I'm off by 2.1* at max steer. If I can find a USB cord I'll scan the reports and post them to get some feedback.

[wellis77]

Found a cable and got my junk scanned. I have another thread where I've asked some to help me understand the stuff the S.A. is spewing out so I'll redirect this thread over to that one, in order not to further muddy questions about negative rollcenter with analyzing my stuff. Here is the thread:

https://www.pro-touring.com/showthread.php?75816-Help-me-understand...&p=763228#post763228

[High Plains Mopars]

Roll couple. The line thats drawn from the rear roll to the front roll. Thats very important.



Actually, that is the roll axis. Roll couple is the difference between the front and rear end of the car's combined spring and sway bar rates.

[Kenova]

Control arms closer to parallel moves the Instant Center for a specific side further away from the tire itself, that it turn makes the lever about which the tire is moving, much longer, so a smaller distance will be covered at the instant center for the same distance in vertical movement at the tire. Which will decrease Roll Center migration, because the roll center is based on the intersection of lines drawn between the Instant Centers and the center of the tire patch. However, in order to properly gain camber under compression most suspensions are SLA designs, and with a shorter upper arm, the Instant center will move both vertically and horizontally.

If someone was looking to increase camber gain by using taller spindles and/or longer ball joints, at what point does the taller spindle assembly do more harm (too much RC movement) than good (camber gain) ?

Ken

[ArtosDracon]

If someone was looking to increase camber gain by using taller spindles and/or longer ball joints, at what point does the taller spindle assembly do more harm (too much RC movement) than good (camber gain) ?

Ken

I hate saying this, but there just isn't enough information to say accurately. It would depend on the lenghts of the arms, both individually and as a ratio as well as the KPI of the spindle and the vertical distance between the UCA and LCA mounting points. If you have some specific measurements, I'll be happy to try and get you in the right direction, in your own new thread. :)

[JRouche]

If someone was looking to increase camber gain by using taller spindles and/or longer ball joints, at what point does the taller spindle assembly do more harm (too much RC movement) than good (camber gain) ?

Ken


And YES. Thats the all elusive part of the front A arm suspension. It really is a give and take dance.

You look for a certain camber gain number for what you are doing then work in the RCM and see its all over the place. So it becomes what is more important and what nasty aspects of what you decide on can you fix.

Personally? I like to work on what parts of the system that CANT be fixed or helped out with any other means. So I think camber gain is a basic suspension quality. It can be achieved through many means, but cant be brought back with any other fixes.

So I look at it as one of the primal suspension needs (or sub-primal). There are others that rank higher on the pole. Like suspending needs, what it takes to suspend the body.

There are primal suspension needs that rank at the top. The need to support the cars weight, the need to steer the car, the need to brake the car. Those to me are the primal needs of a suspension. Everything after that is tuning. But you cant remove the primal needs, you can only add to them and adjust them.

So the sub-primal needs vary from car to car and its use, but the primals are still there. A street car may need different braking and a complient suspension than a nascar vehicle will. A drag car may not need much braking or steering up front, but the primals are still there. Same with F1 cars.

So in the end to tune your suspension it depends on what the need is. And thats the dance. No gain without a loss when talking suspensions.

I personally would work on the sub-primal needs first and try to fix the lower needs with tuning aids.

For an example. I think a decent camber gain is worth the cost of a low and migrating RC. Cause I think I can help the suspension out with fixes (such as a sway bar up front). But I wont ever get to go backwards and fix my lack of camber gain with fixes. Cause if you try to, for example putting tall ball joints in, or changing the spindle you are going backwards. Now you are still dealing with a low and traveling RC.

My thinking is work from the large (the primals) and whittle yer way down to the tuning parts.

Unfortunately there is no one great setup. But I think if you tackle the major suspension issues first and work downward you can get to a descent performance in the ending.

Dont overlook the tuning aids, the bottom of the suspension pole. Sway bars, anti bump steer components, and actually many other tuning aids that help to fix the must be in place suspension components and adjustments. JR

[wellis77]

Great explanation JR!