The Mark Ortiz Automotive
Presented free of charge as a service
to the Motorsports Community
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.
SMALL LOAD TRANSFER WITH LOW ROLL RESISTANCE?
Below is a quote from Ben Bowlby about the Nissan delta wing.
“Because the vast majority of the roll stiffness was on the rear, the front tyres saw virtually no change in vertical load.”
Could you explain to a virtual newbie like me how this happens?
Having very little lateral load transfer in cornering is actually possible with any car – for one end of the car only, and only up to the point where the opposite end lifts a wheel. This is true whether the car has an extremely narrow track at one end or not. The only difference when one end has a very narrow track, and also a small percentage of the static weight, is that very small load transfer cannot be obtained at the wide end, except for very small lateral accelerations. But that is definitely possible at the narrow end.
When the wide end has a very wide track and well over 50% of the static weight, it is possible to have all the load transfer at the wide end and none at all at the narrow end, all the way up to maximum lateral acceleration. It might even be possible to have negative load transfer (inside tire gains load in cornering) at the soft end of a car.
Any wheel pair can only resist roll by exerting force against the ground. Regardless of how the system generates its roll resistance, it has to act through the tires, and therefore has to change their loadings.
For the car not to tip over, the total roll resisting moment for both wheel pairs has to equal the overall overturning moment for the entire car, which is the product of the c.g. height above ground and the total ground plane lateral force at the tires.
Or, neglecting the unsprung mass load transfer, which is not affected by relative suspension roll resistance, it is the product of the sprung mass c.g. height above ground and the portion of the
ground plane force at each end that reacts the sprung mass inertia. It simplifies explanation if we ignore the unsprung load transfer.
Treating things in this simplified manner, the roll resisting moment at each end of the car is the load transfer at that end times the track at that end. This is also the change in wheel load difference times half the track. (Adding 100 pounds to one tire and taking 100 off the other changes the difference by 200.)
The sum of the front and rear roll resisting moments has to equal the overall roll moment. Therefore, we can reduce load transfer at one end of the car by making it softer relative to the other. However, this will correspondingly increase the load transfer at the other end of the car.
It will be apparent that when either factor (the track or the load transfer) is small, the product (the roll resisting moment for the wheel pair) cannot be large. However, if the product can be small, then neither factor has to be large.
Thus, it is entirely possible to have small load transfer at the narrow end of a near-tricycle four-wheel car. We can even arrange for it to have no load transfer at all, provided that the wide end can resist the roll moment by itself without lifting a wheel.
We can even arrange for the narrow end to have negative load transfer, provided the wide end can react more than 100% of the roll moment without lifting a wheel.
This requires the suspension system to have negative net roll resistance: when it experiences ground plane force to the left, it tries to roll the car to the right, not resist rightward roll. This can be arranged by having zero or very small elastic anti-roll (for example, all springing by a z-bar), and geometric pro-roll (roll center below ground). I do not know of any actual car that has net negative roll resistance at either end, nor do I expect to see one, but this is a theoretical possibility.
With a very narrow front track, it is possible to greatly affect the load transfer at the front with by adjustment of the front suspension. This allows us to tune the oversteer/understeer balance of the car. However, it is not possible to greatly affect load transfer at the rear by adjusting the front suspension. With very little track to work with, the front can’t generate a lot of anti-roll moment even with a lot of load transfer. In a tail-heavy car with a front track similar to the rear, if we add front roll resistance to the point where the inside front is very light or airborne, we get not only an oversteer reduction/understeer increase, but also good loading on the inside rear, helping the car put power down with that tire. When the front track is extremely narrow, the inside rear will unload dramatically in limit cornering, even if the front end is stiff enough to pick up a wheel.
EFFECTIVE UPPER CONTROL ARM PLANE IN STRUT SUSPENSION
A question has been
rattling around in my head for a while which you may be able to shed light on.
From all of the books I have read regarding finding the swing axle length on strut suspension they say that you project a line from the top of the strut, square to the centreline of the strut until it intersects with the line of the lower control arm.
I understand this 'in the old days' when struts were in line with the lower ball joint. However I am confused with modern suspension where the strut is bolted to the side of the steering knuckle. Should the upper line project square to the strut? Or should it be square to the 'virtual' strut (line between strut top pivot and ball joint)?
For purposes of analyzing suspension geometry, the axis of the strut itself is what matters. The effective upper control arm plane is the plane perpendicular to that axis, containing the strut pivot center of rotation.
For purposes of analyzing steering geometry, the steering axis is the line containing the ball joint center of rotation and the strut pivot center of rotation.
UNDERSTEER IN TIGHT TURNS, OVERSTEER IN HIGH SPEED TURNS
Why does the same car understeer in a low speed tight corner but oversteer in a high speed bend? Aero aside, is this just a product of yaw center location or is it due to other effects (Ackermann, etc)?
It isn’t strictly true that all cars do this, but when they don’t it’s usually because of aerodynamic effects – those being significantly greater downforce and/or less lift at the rear than at the front.
I will confess that I don’t fully understand the reasons for this, but I understand some of them.
Ackermann does enter into it. A car needs more Ackermann in a low speed turn than it does in a high speed one, and we have to compromise.
The rear wheels always track more to the inside of the turn, relative to the front wheels, in tight turns than in sweepers, for the same amount of understeer. With rear drive, this “off-tracking” puts the center of front wheel drag further to the outside of the turn and the center of rear wheel propulsive thrust further to the inside, relative to the track of the c.g., in a tight turn than in a sweeper. This creates a yaw moment out of the turn, adding understeer.
This effect should reverse with front-wheel drive. However, even most front-drive cars tend to understeer more in tight turns than in sweepers.
If the car has a limited-slip differential, a locker, or a spool, the understeer-inducing effect from this tends to increase as the speed difference between the drive wheels increases. That difference increases as the turn gets tighter.
With rear drive, the amount
of rear tire grip available for cornering diminishes as we demand more
longitudinal force for propulsion. The faster the car is going, the more of
the total rear tire force capability we have to use for propulsion, just to
maintain constant speed. This is a big effect. In a reasonably powerful car,
if we are applying close to full power in a tight turn, we’re throttle-steering
the car. But the same throttle opening in a sweeper is merely what we need to
keep from slowing down. Of course we’ll be in a higher gear for the sweeper,
so the rear wheel torque won’t be the same at identical throttle opening, but
the actual longitudinal force needed at the drive wheels will go up roughly
with the square of speed.