FWD Oval Cars – Beam Axle vs. Independent Rear Suspension

[Crazy Bored]

The story of my situation is I used to race Honda Civics on a ¼ mile oval, and I learned how to make one corner rather well after a couple years. The division I was running in consisted entirely of Civics, although other cars were allowed. This year some guys are running a VW Golf against the Civics, and while it has a 2.0L engine compared to a 1.5L, they are struggling to compete. It is quite literally beam axle versus independent rear suspension. I have been helping them for a short time and began playing with basic things, changing cross weight, spring rates, and ride heights (within the rules) to at least make the car more balanced. This thread is partially about how to improve the setup, and partially about the potential of a beam axle in general.

A basic outline of the rules and situation:
• 2250 lbs minimum
• 56% left side weight maximum
• 195/60–15 tires (200 treadwear minimum)
• 5” minimum ride height
• No adjustment of brake bias

Static camber rules:
LF: -2.5° RF: -5°
LR: - 1° RR: -3°

Track banking:
Turns 1-2: ~10°
Turns 3-4: ~15°

A large amount of time building my Honda Civic in the past was spent on these primary goals:
1. Lower CG: weight was reduced to 2013 lbs – 250 lbs for ballast under the floor.
2. Rear weight: car went from 61% front weight to 55% with various methods.
3. Increase dynamic camber: Caster was added to front with UCAs from an older Civic, rear camber gain increased by moving three suspension points on each side of the rear.

The first two points here are difficult in the current situation, with the Golf having a heavy cage and some unnecessary OEM material remaining throughout. The car has no ballast, and is roughly 50 lbs over weight, with 63% of the weight on the front. I am obviously trying to find ways to reduce weight and the CG. The 56% left side weight is not an issue.

Anything I did to increase camber in the Civic always ended up with faster lap times, after appropriate cross weight adjustment for balance. I have come to believe that somewhere around 7-8 degrees of negative camber on the right side would suit the given tire size. My simple trick to get more camber at the front in the Civic does not work with the Golf as there is no UCA to replace/modify (we are looking into ball joint extenders for the LCA). I have no easy way to change the camber gain on the right rear tire. Nevermind the camber on the left rear tire, in the Golf we carry it in the air for the entire corner (well, I technically mean two corners, since this is an oval).

This brings me to the main idea of the thread. I was able to achieve significant camber gain in the rear by changing the angles of the rear control arms on my Civic. We have the Golf set at the maximum static camber allowed, and while it is now reasonably balanced, the rear is clearly struggling and wearing on the right rear sidewall about half an inch down. This is with 38 psi in the tire, while I could run 28 psi in the same size tire on my Civic and have none of the sidewall touch the track (that’s also with much more weight on the rear of my Civic).

One of the key points here is a lot of the adjustments to a setup have to be free, or close to it. I used a front spring in the right rear of my civic, modified rear suspension points with scrap metal, etc. Therefore things like very stiff springs, anti-roll bars, and dampers aren’t very feasible. Springs can be cut to increase stiffness, and the ride height shimmed back up, but only so far (extreme things may cause them to get suspicious in the tech shed). We don’t have stiff enough springs for the RR and LF to make the car rotate well, and the RF tire is touching the wheel well so it cannot have a softer spring. As far as I’m concerned, the LF spring rate is irrelevant as long as the tire is in the air. Raising the RR enough to make the car turn ended up having the entire rear of the car way too high off the ground and the lap times were terrible.

With these practical limitations on spring rates and cross weight, I eventually gave into using rear toe out. My Civic ran well with ¼” out at the front and basically 0 rear toe. At this point we are over 1.5” of rear toe out on the Golf, but it was amazingly helpful. The car is very balanced, and is impressively stable and predictable.

So, the car is now well balanced and clearly faster that it used to be in the turns, and has more power than the Civics. We still can’t touch the faster Civics in the turns though, and I know from my own Civic that they can carry much more speed through the turns. We have a best finish of 5th out of 13 cars, but that was with some luck, and we were far behind the front ones (which is where I used to be with my Civic).

The left rear tire is dead weight, and the right rear tire is partially running on the sidewall. I feel like at some point, shifting weight rearwards won’t be very useful for overall speed as only one tire is supporting all the rear weight. How can I make a rear beam axle compete with independent rear suspension cars? Or can I?

[Greg Locock]

I'll just concentrate on one bit - increasing the camber in roll on the twistbeam. The original SAE paper 810420 by Terry Satchell is indispensable. Bascially to change the camber gain in roll you need to move the twistbeam longitudinally. It's not a huge effect even with large changes in X.

There's also this link which autogyro found http://www.auto-ware.com/ortiz/ChassisN ... er2006.htm

[Greg Locock]

Moving the twistbeam rearwards by 100mm on a car that may respond differently to yours reduces the camber in roll by about 5%

[Crazy Bored]

Thank you for the information and links. From what I understand after some basic FBDs and approximating deflection of some parts, I have come to the same conclusion about the camber gain in roll. I assume "by reducing the camber" you mean reducing the negative camber. Since I can only move the twistbeam roughly 100mm forwards, and I need a relatively large amount of camber gain, it doesn't seem like it would be worth the effort.

I feel like in my situation the only real way to move forward is to design an independent rear suspension using the current mounting points and integrating the rest of it with the rear section of the roll cage in some way. Unfortunately we don't have the time nor budget required to do that somewhat acceptably.

For the sake of interest, the car is currently rather well balanced with very close to 1" of rear toe out, and 33.2% cross weight. The RF, LF, and RR springs are all the stiffest ones we have available. The RF is stiff to prevent the tire hitting the wheel well, and the LF and RR are stiff for dynamic cross weight.

I am wondering if it would be worth trying a soft LF spring to increase roll resistance, and reduce body roll and thus camber loss. The static cross weight would then be lowered even more to fix the balance.

[flynfrog]

are you taking any data at all? Travel movements, tire temps, are a good start.

[Crazy Bored]

With no real budget the only thing I can do to measure travel is to measure the maximum amount by putting a plastic tie on the right side damper shafts.

Tire temperatures can be done if I find an IR gun to borrow, although the tires will have dirt/dust on them as the entire pit area is gravel. I don't know anyone with a temperature probe. I am certain the inside of the right tires will not be very hot, as rubber from the outside portion of the tread builds up on the inner portion of the tread, never wearing off. I believe if a fifth of your tread is not even contacting the track in the turns, you need more camber.