The Mark Ortiz Automotive
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: firstname.lastname@example.org. Readers are invited to subscribe to this newsletter by e-mail. Just e-mail me and request to be added to the list.
I was recently reading the book 'Motorcycle Handling and Chassis Design' by Tony Foale, in particular his brief but concise description of the forces involved while power sliding a motorcycle on a dirt flat track.
I then proceeded to search everywhere I could think of for a good description of the dynamics involved in a four wheel vehicle power slide on a dirt or pavement surface. Very little or anything out there in the literature, SAE papers, respectable chassis books, internet, etc.
My curiosity is now piqued as to the setup implications of power sliding a racecar.
We know that while power sliding we are using a portion of the longitudinal force vector to maintain our curved path, i.e. provide a component of force to add to centripetal acceleration, and a portion of that same vector is being used to propel us around the corner.
So what if any effect is there from say stagger? If we say that a power slide necessitates rear wheel spin than we would have to conclude that stagger is really irrelevant in this situation.
What does our dynamic wedge look like under these conditions?
What setup parameters can we change to promote or increase or decrease the oversteer yaw rotation of a power slide, besides what might be obvious to make a vehicle oversteer to the extreme?
Or is this more of a driver induced condition controlled by throttle and counter steering?
And finally in dry slick conditions on a dirt track (no accumulated rubber, just hard dry dirt with accumulated fine dust on the surface) is this technique even the best tactic to better the lap times?
If you know of some literature I have overlooked I would appreciate if you would point me in the right direction.
Or, which seems to be more the case in these questions, would you mind giving us your thoughts on what you consider to be the main chassis dynamic variables in play during a power slide?
As a practical matter, we know that powersliding only works on relatively low-grip surfaces, simply because if we do it with sticky tires on a high-grip surface we scrub off too much speed and quickly destroy the tires. With enough power, and at a sufficiently low speed (as in actual drifting competition), we can powerslide a car on pavement, but it’s not normally a way to win races.
It is certainly true that we are using a portion of the car-forward thrust as cornering force and part of it as road-forward propulsion force.
I have never heard of anybody actually trying an experiment taking segment times on a relatively low-grip surface, powersliding versus trying to make the car track, and comparing the times. One thing I would predict, just based on my own experience in low-grip conditions, is that when not powersliding, the times would be less consistent, and there would be more instances of loss of control. When we try to corner at the limit of adhesion on a low-grip surface, the car will often go into dramatic understeer or oversteer on its own, in response to variations in the road surface. It can take a lot of road to recover, and we may lose a lot of speed, or even crash. We can avoid this by going slowly enough, but we don’t want to do that if we’re racing.
If we powerslide the car, we have oversteer, but the amount is under the driver’s control to a fair degree, and we at least know the car isn’t going to simply push into the wall. Cars having a wide range of understeer gradients on their own can be held in a controlled attitude, to a degree not possible without powersliding.
In particular, we can “drive a tight car loose” by powersliding.
In some cases, it can be unclear whether a car is powersliding or not. Cars that naturally oversteer generally need to corner with some power on. If available power is fairly modest, full throttle may be the setting least likely to spin the car. The “stab it and steer” technique will be familiar to Porsche and Corvair drivers. Is a full-power oversteer condition a powerslide, when the driver is actually applying that power to minimize oversteer?
The effect of rear tire stagger does diminish greatly as the rear tires reach the point of breakaway. A car with ample stagger is easier to get into a slide, and tends to stabilize somewhat once the tail is hung out. A car with little or no stagger is hard to get rotating, and tends to snap loose more abruptly as the rear tires reach breakaway. So a car with more stagger tends to be easier to drive. The downside is more drag when the rear tires are not sliding, particularly down the straights. Hence, my usual recommendation is to use as little stagger as the driver can live with.
Compared to a technique of not deliberately hanging the tail out with power, does powersliding add or reduce dynamic wedge? That depends on the car. It could go either way.
Actually, the term “dynamic wedge” is a bit unclear. It’s easy to say that a particular effect adds or reduces wedge dynamically. Anything that adds an upward jacking force at the inside rear, for example, adds wedge dynamically compared to a baseline without the effect. But if we wanted to move beyond conversational physics or qualitative discussion, and actually put a number on “dynamic wedge”, how would we calculate that? If we were doing K&C testing, and wanted to output a “dynamic wedge” value for the client, how would we do that? Would we just calculate diagonal percentage, and call the car unwedged when that is 50%?
I am inclined to call a car unwedged when the front and rear wheel pairs exhibit identical left percentages, which also implies that the right and left wheel pairs exhibit identical rear percentages. But when the car has its wheel loads very unevenly distributed both laterally and longitudinally, an unwedged condition by that definition is a long way from 50% diagonal.
For example, suppose we have a car with equal wheel loads statically, that is cornering hard under heavy power and dynamically has 70% right and 60% rear. If the car is unwedged by my definition, the right front has 70% of 40%, or 28% of the total load. The right rear has 70% of 60%, or 42%. The left front has 30% of 40%, or 12%. The left rear has 30% of 60%, or 18%. What’s the diagonal percentage (RF + LR)? It’s 46%. If it were 50%, the rear tires would be more evenly loaded than the fronts.
Note that this has nothing to do with how sideways the car is. This could be a pavement car, operating at a fairly small yaw or attitude angle. We could also get the same wheel load distribution if the car is accelerating only laterally, and has 60% rear statically.
Since there is no agreed way of expressing “dynamic wedge”, my recommendation for either simulation or K&C testing would be to look at front and rear left percentages, and also diagonal percentage, and let it go at that. To move beyond that, we could look at the ratio of front to rear inside or outside wheel percentages, or the percentage point difference between those. However, that would be a measure adopted by a particular team or person, and would not be useful for comparison across a broad spectrum of cars or teams.
ANTI-LIFT IN FRONT-DRIVE CAR
Your recent discussion of anti-dive geometry brought up a question about other possibilities. As you know, strong acceleration in a front wheel drive race car causes the front to rise and it loses traction. My question is: Is there any possible suspension geometry that would minimize or eliminate the chassis rise in a FWD car under acceleration?
I have used a system to mechanically oppose suspension rise using a cable and hydraulic cylinder system but it is somewhat complex and adds weight, although it does help off the line traction. The car is a Classic Mini tube frame D Modified(SCCA) with a Modified Acura Integra GSR drivetrain.
Rearward load transfer under power does reduce drive traction with front wheel drive, and that is one reason it is not the first choice for a high-powered car.
Front end lift is not the cause of rearward load transfer; it is a result. For a given forward acceleration of a given mass, the total amount of rearward load transfer depends purely on the c.g. height and the wheelbase. To improve matters, we need to lengthen the wheelbase and/or lower the c.g. However, front end rise does have a small effect on dynamic c.g. height. We can have a slightly lower dynamic c.g. height by reducing front end rise and also by permitting rear end drop. Note that the idea is not to prevent the car from pitching, but to keep the car down overall.
It is possible to have some anti-lift in a front-drive front suspension. It is not possible to have an anti-squat or pro-squat effect at the rear under power, because there is no ground plane force at the rear contact patches.
Because the front suspension is generally independent, with no gearing in the uprights, there is no torque reaction through the suspension linkage. All anti-lift has to be thrust anti-lift: the geometry has to make the wheel center move forward as the suspension compresses and rearward as the suspension extends. The side-view projected instant center has to be above hub height and behind the wheel, or below hub height and ahead of the wheel.
Such geometry adversely affects impact harshness, but that can be tolerated in a race car. The real limitation on anti-lift is wheel hop. The more anti-lift we use, the more prone the car becomes to wheel hop at the point of front wheelspin. Exactly how much anti-lift will make this intolerable, or make the car slower because of it, is hard to predict. It depends a great deal on the tires and the road surface. Undamped compliance in the suspension makes it worse, so making the control arms and their mountings as stiff as possible is helpful.
Using stiff low-speed rebound or extension damping at the front can momentarily slow front end rise, but will not keep the front down in steady-state acceleration. Similarly, using soft low-speed compression damping at the rear will reduce overall ride height momentarily, but not steady-state.
Using high wheel rate in ride at the front helps reduce front lift. Using low wheel rate at the rear helps increase rear squat. Since we generally want enough rear roll stiffness to make the car lift the inside rear wheel, this means using very little anti-roll bar at the front, and a lot at the rear.