The Mark Ortiz Automotive
CHASSIS NEWSLETTER
Presented free of charge as a service
to the Motorsports Community
November 2009
Reproduction for sale subject to restrictions. Please inquire for details.
Mark Ortiz Automotive is a chassis consulting service primarily serving oval track and road racers. This newsletter is a free service intended to benefit racers and enthusiasts by offering useful insights into chassis engineering and answers to questions. Readers may mail questions to: 155 Wankel Dr., Kannapolis, NC 28083-8200; submit questions by phone at 704-933-8876; or submit questions by e-mail to: markortizauto@windstream.net. Readers are invited to subscribe to this newsletter by e-mail. Just e-mail me and request to be added to the list.
LAUNCHING A DRAG CAR ON A HOT, SLICK TRACK
We have a drag car that has trouble hooking up when the track is hot and slippery. The car is a late '60's muscle car with stock engine location, but it has slicks and an aftermarket 4-bar rear end, with coilovers.
When the track is reasonably cool and traction is good, the car hooks up well, and launches with both front wheels just clear of the track. But when the temperature goes up and the track gets slick, it just spins the wheels. We've tried adding ballast behind the rear axle. If we add enough, the car will pick up the front wheels, but the rears still spin, and the times aren't better.
We've tried lowering tire pressures. We find that 8 psi works better than 9, but we are reluctant to go lower. People tell us that these pressures are too low. Our tires have somewhat stiffer sidewalls than what they're running, and stiffer than what we used to run, so we're unsure what to do on this.
We also get conflicting advice on what to do with the 4-bar setup, specifically where the side-view instant center should be.
What are your thoughts?
First of all, if there's less grip, there's probably no way to make the car as fast as it would be with better grip. The object is to re-optimize the car for the slipperier surface, and minimize the loss. If you can do that better than your competition, the track conditions will slow you by a smaller margin, and you will be more competitive.
Let's consider what we want to achieve when launching a rear-drive car at the drag strip. Basically, we have five objectives:
1.We want as much of the vehicle's weight on the drive wheels as possible.
2.We want the loads on the two drive wheels to be as equal as possible.
3.We want a straight launch. We don't want to lose control, or have to lift, or scrub off speed with steering corrections.
4.We want the highest coefficient of friction possible at the tire/road interface.
5.We want to avoid wheel hop or tire shake.
For the first objective, the questioner is correct that we would like the car to launch with the front wheels off the ground, at least briefly. We want it to just barely lift them, so that the front tires and the wheelie bar casters are both off the ground. In that condition, the rear wheels are supporting the entire weight of the car.
Of course, we can't steer with the front wheels off the ground, so we want them off the ground only as long as we can count on the car to run straight on its own. If we do have a tiny bit of load on the fronts at launch, that can also be accepted in the interest of steering control. But we want the car either in a state of wheelstand or very close to it.
The car will reach this state when the product of its mass, its forward acceleration, and its c.g. height during launch equals the product of its static front weight and its wheelbase. As an equation:
My = maxHcg = wfLx (1)
Where:
My is the overall rearward pitch moment
m is the car's mass
ax is the forward acceleration
Hcg is the c.g. height, as it exists when the car is actually launching
wf is the total front weight (front end gravitational force; essentially, static front weight)
Lx is wheelbase
If m is in pounds, ax needs to be in g's.
When the track gets slipperier, ax decreases. Therefore, to satisfy equation (1), if we don't want to increase the car's weight, and we don't have ballast we can move to reduce wf, and we can't readily alter the wheelbase, we can only do one thing: increase Hcg – raise the center of gravity.
We can do this by raising things within the car, raising the car's static ride height, or by making the ride height rise more under power: make the rear squat less, and/or the front rise more. Of these, in many cases the easiest and best is to simply adjust the suspension for more static ride height.
Note that raising either or both ends of the car will accomplish this. Raising the front and lowering the rear will not.
Having equal loading on the two rear tires is not necessarily as simple as one might think, especially with live axle rear suspension. Drive shaft torque tries to roll the rear suspension rightward. The reaction torque at the front of the driveshaft acts through the motor mounts and tries to roll the sprung mass rightward. The driveshaft acts on the axle, trying to roll it leftward. These forces act against each other through the rear springs. The resulting roll displacement can be called torque roll.
This roll moment is resisted by both the front and rear suspensions, but the rear suspension resists it within its own linkage and springs, whereas the front suspension resists it through the tires. To the extent that the torque roll moment is resisted at the front wheels, it unloads the right rear and left front tires, and loads the left rear and right front. The resulting effect on wheel load distribution can be termed torque wedge: the increase in LR/RF diagonal percentage due to driveshaft torque.
If the rear suspension cannot move at all in roll, we get no torque roll or torque wedge. If the front suspension has no elastic roll resistance, and the rear suspension has some compliance in roll, we can get torque roll without torque wedge.
It is possible to build a front suspension with no elastic roll resistance. This would involve using a Z-bar or a pivoted transverse leaf spring, or some physically equivalent arrangement. However, that is not ideal for handling as the car gathers speed, and I have never seen it done on a drag car.
It will be apparent, though, that any front end will have no effective roll resistance if the wheels are in the air. This is one of the advantages of launching with a mild wheelstand. However, we can't keep the front wheels off the ground very far down the strip, because we have to be able to steer.
With an open differential, torque wedge causes premature wheelspin on the right rear. With a spool or a locker, or with sufficient torque transfer in a limited-slip diff, that can be prevented, but then we have unequal thrusts from the two rear wheels, and the car tries to turn right.
It is possible to build an asymmetrical rear suspension that will compensate for driveshaft torque, but the questioner's car has a traditional rear suspension with four trailing links. These generally have either a Panhard bar for lateral location, or a device that drag racers call a wishbone, which is a
triangulated structure in about the same plane as the lower trailing arms, incorporating a slider so that it locates the axle laterally but not longitudinally.
The trailing links have multiple mounting holes, allowing adjustment of their angles. To give bind-free motion in roll, the links would have to be parallel, but drag racers set them so they converge toward the front, providing a side-view instant center somewhere between 40 and 60 inches ahead of the axle.
This creates a linkage that binds in roll. Essentially, it turns the rear axle housing into an anti-roll bar – a very large, stiff anti-roll bar, but also one with 40 to 60 inch long lever arms.
We will discuss some of the tradeoffs in such linkages momentarily, but first let's consider some of the other objectives we are trying to accomplish.
To get a straight launch, we not only want the rear tires equally loaded, we also want little or no bump steer or roll steer, at either the front or the rear.
To get the highest possible coefficient of friction, we want to have the tires in optimum condition and at optimum pressure for the track conditions, and we want the car as light as possible. A tire's coefficient of friction decreases with increasing load. That is, tractive friction force does increase with normal force, but less than proportionally, while inertia increases with mass, exactly proportionally. That's why the car can get slower with added ballast in the rear, even though it wheelstands more readily. (Of course, more ballast also makes the car slower once it's up to a speed where it's limited by power rather than traction.)
Tire shake and wheel hop are related phenomena, and usually occur together, but they are not quite identical. Tire shake is the whole car oscillating up and down on the tires. Wheel hop is the rear axle oscillating up and down with respect to the sprung mass.
I don't know if anybody understands everything that goes on in tire shake or wheel hop, but here's the basic theory: when the tire takes a bite, the tire and/or the suspension induce a momentary load increase that makes the tire bite even harder. However, this load increase doesn't last, and the tire breaks loose. When the traction decreases, the dynamics that created the load increase reverse and the car settles back down. The tires then take a bite again, and the cycle repeats. The key is that either the tire or the suspension lifts the car when the tires bite, and lets it down when the tire slips.
This means that we can't have too much anti-squat, or we'll get wheel hop. We also may encounter tire shake if the tire pressure is too low, and the tire wraps up excessively.
Optimal tire pressure definitely is related to sidewall stiffness. Tires with stiffer sidewalls will want lower pressures to work best, as a rule. With any tire, though, the basic principle is the same: go
softer until the car gets worse, then back up a little. Generally, a slipperier track calls for lower tire pressures. One measurement – in this case 60-foot time – is worth a thousand expert opinions.
Although a small modification to tire pressures may be needed once the rest of the chassis is optimized, the basic strategy should be to find the best tire pressure first, and then dial the rest of the car in around that.
When we have no rear anti-roll bar, optimizing the four-bar geometry involves a compromise among three conflicting requirements: roll steer, anti-squat, and bind in roll.
To minimize roll steer, we want the instant center near wheel center height. To optimize anti-squat, we want the instant center somewhere on a line passing through the contact patch center and sloping up toward the front of the car at no more than about 8 degrees. A steeper slope than this tends to invite wheel hop, as a rule of thumb, although this will vary some with track conditions and tire properties. To maximize bind in roll, and thereby minimize torque roll and torque wedge, we want the instant center as close to the axle as possible. It will be evident that there is no way to satisfy all these requirements at once.
The compromise that seems to work best is to have the instant center around the height of the lower links – about six inches off the ground – and about 45 inches ahead of the axle. That's really too close to the axle, and too low, but if we keep it at 45 inches, and raise it, we get wheel hop. If we raise it and move it forward, the roll steer improves, but we get more torque roll.
Rules permitting, there is a way out of this: instead of using the rear axle housing as an anti-roll bar, add a separate anti-roll bar. Then the 4-bar geometry can be optimized for roll steer and anti-squat, without the need to keep the side-view swing arm length short. This has in fact become a popular approach in drag racing over the last two or three years. With an anti-roll bar, you want an instant center around axle height or a bit below, and 100 or more inches ahead of the axle.
The anti-roll bars I see on drag cars are a bit short on lever arm length. The arm angles aren't too bad in pure roll, because the bars are short and attach to the axle far inboard from the wheels, but if you add some ride motion, the angles get extreme pretty fast. If I wanted to improve one of these, I'd go a bit longer on the lever arms, and use a fatter, possibly hollow, bar – packaging constraints permitting.
If you really want to steal a march on the competition, consider just omitting the left upper trailing link. Run the lower links horizontal, and have a single upper link, on the right, angled down at the front just enough to compensate for driveshaft torque and eliminate torque roll. You might still want an anti-roll bar, but not compensating for driveshaft torque, and using a lot of rear roll stiffness instead, is kind of a crude, brute-force approach. It works, because drag strips are generally pretty smooth, but I think the next improvement will come when somebody figures out how to compensate for driveshaft torque with the linkage instead.