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.
CONTROLLING AWD FRONT/REAR TORQUE SPLIT ON A BUDGET
The last couple of months you have been discussing differentials and so I thought I would throw in my question(s) on the subject.
I am building an AWD car for road racing. Rules require production unit-body with modifications and the rules allow an alternate transmission but orientation must be as from the factory. The engine is transverse and in front of the front axle, will have around 475 hp, weigh about 3100 lb with driver and run the same size tire all around. So far, my research suggests that I should run a LSD in the front and rear positions and a diff (as opposed to a spool) in the middle. Research also brings up "torque split" to reduce on-throttle understeer which, in a passive system, is a final drive ratio difference front-to-rear that is usually achieved through a biased open center diff with a viscous clutch to control front-to-rear slip.
Sounds great, except I don't have access to that kind of center diff and viscous clutch arrangement. Can I achieve the above dynamics by running a different FD in the front and rear diffs, to achieve the "torque split", and run a Torsen or a Weismann diff in the center position? I considered running an open center diff in this case but it seems I would be converting the car to two wheel drive thereby losing the traction advantages of AWD. Or am I stuck with a 50/50 torque split?
The driveline is designed to take a Salisbury-type diff, a spool, a Weismann locker, and I presume an open diff. Any of these can be used in any of the positions. From this you can surmise that Weismann is building the driveline.
They claim their diff is "magic", without saying how it works, and have suggested running it at the front and the rear and a spool in the center. I suspect they may have run this configuration in off-road truck racing. I think this will exacerbate understeer and consequently benefit from a center diff. I have tried to research "Weismann Diff" and "Weismann Locker" but haven't found a good enough explanation of how it works to understand it. Some media say it is locker and some put it in
the same category as a Torsen, i.e. not an overrunning locker but a diff that is "torque-sensing" and biases the torque to the wheel with traction before any overrunning condition sets in.
The overall problem I am grappling with is how to get a driveline that has the least throttle-on understeer while still maintaining the benefits of AWD traction. The torque split idea seems popular and makes some sense as the car would drive more like RWD but still have part of its power used at the front and in really slippery conditions it would drive like AWD.
What about the locker in the center position? It seems to me that the locker in the center position will act like a spool as a locker only differentiates when one of the outputs is caused to overrun the speed of the diff and the other output (or in this case, one axle overrunning the other axle). The only situation that causes one axle to overrun the other is when turning as the front axle will scribe a larger arc than the rear. But in a race car, where the tires are operating at some degree of slip to generate a cornering force, this speed differential may be insignificant. An alternate scenario is to run different FD ratios front-to-rear in which case the locker will drive the taller gear, as the shorter gear functions as overrun, until the tires at the taller end slip enough to exceed the speed on the other side of the diff. The effect will be a driveline switching between 100% RWD and 100% FWD. AWD drive could be achieved if the slip at one end matches the output speed of the other side
of the diff but then you are back to 50/50 torque split and the effect is probably transitory on the way between RWD and FWD. Am I overlooking something here?
I tried a search for Weismann differential too, and likewise came up dry. Any readers who know more about this, including any working with the company, are invited to fill us in. I am familiar with the Weismann locker, which is a Detroit Locker style design, made to fit road racing dog-ring transaxles. It's been around since the late 1960's. There is no law of nature that says the company has to make only that design forever, but I can at least speak to the characteristics of this device.
For high-speed pavement use, I don’t like a spool in the middle. It’s not so much a matter of drag as a matter of being able to control the car with the throttle in a normal manner. Ideally, you’d like the car to be able to throttle-steer more like a rear-drive car, but have the front drive pick up as the rears begin to slip. A nose-heavy car with 50/50 torque split will sort of do that. When cornering hard, you can only get a little forward acceleration – hence little rearward load transfer – before the rears start to slip, so the rears should spin first. Trouble is, the car understeers steady-state unless you have it roll-stiff in the rear, making the inside rear wheel very lightly loaded in cornering – much like a similar layout with front-wheel drive. That means the rear diff must either be a locker or have enough preload to lock when one wheel is airborne or nearly so. Otherwise, when you try to throttle-steer, you just spin the inside rear and the outside rear stays stuck rather than sliding.
As I see it, if you can’t get a planetary center diff, you have three options: open center diff, with equal final drives or taller final drive at the front; worm gear or clutch pack center diff, with equal final drives; or locker in the middle with taller final drive at the rear.
If you try unequal final drives with a limited-slip in the middle, you’ll have lots of drag, friction, heat, and wear, so you’re pretty much limited to equal final drives in that case.
If you have an open center diff, and taller final drive in front, the gears in the center diff are working constantly, but at least you’re not wearing out any locking mechanism. The mean front/mean rear torque split (at the respective ring gears or diff carriers) has the same proportionality as the respective final drives. For example, if we use a 4:1 at the rear and a 3:1 at the front, the rear wheels get 4/7 (57.1%) of the torque and the fronts get 3/7 (42.9%). The torque split is always in this ratio, and the speeds can be in any ratio.
With a locker in the middle, and taller final drive in the rear, the rear wheels do drive, and the fronts do overrun, most of the time, as the questioner notes. But when the rear wheels spin, and the locker locks, we don’t get front-wheel drive. We now have all wheels driven. There is no controlled torque distribution. The torque goes where the resistance is. What is controlled is the ratio of front and rear mean wheel speeds. These will be inversely proportional to the final drive ratios.
Suppose, for example, that we have a 3.00:1 rear final drive and a 3.30:1 front final drive. When the locker is unlocked, we know all the torque is going to the rear. When the locker is locked, we don’t know what the torque distribution is, but we know that the mean rear wheel rpm is 10% greater than the mean front wheel rpm. We know that we have to exceed 10% slip at the rear to get lockup. What torque it takes to get this slip will vary, but the 10% threshold is a constant, which we can adjust by our choice of final drive ratios.
This is probably a good set of properties. The one problem is that when lockup comes, it is abrupt. This is an inescapable characteristic of lockers. Either one output shaft is overrunning, or the whole thing is locked. When cornering hard, the lockup will come when the car is at the limit of adhesion and the power is on, and it may upset the car. I would expect you will need some seat time to get used to it.
The center locker does give us front-wheel drive in one situation: when we’re backing up. In reverse, a locker drives the faster shaft, and it will not lock. This could be a drawback for off-road use. For racing on pavement, it shouldn’t be any worry.
For the front diff, I’d try either a worm gear or clutch pack unit. Clutch pack with preload is best for minimizing inside front wheelspin, but adds understeer and “fight” at the steering wheel. Worm gear is smoother but prone to inside wheelspin. This is less of a problem than it would be in a front-drive car, because the rear wheels have to both spin for either of the fronts to spin.
My best guess for what to try first with your constraints would be: lockers middle and rear; worm gear in front; front final drive about 10% greater numerically than rear final drive; battery, oil tank, and any ballast at the rear.
IS THERE SUCH A THING AS TOO BIG (IN TIRES, THAT IS)?
My question this time is regarding slicks and the width thereof.
I have an MG Midget set up for road racing and at present run 8in slicks. The front is the usual double wishbone and the rear is live axle. All are converted to telescopic shocks.
Given the same suspension system, is there a point where tyres can be too wide? Another midget runs 7in slicks and the comment was made that 7in was as wide as was needed but the previous owner went to 8in as the tyres were easier to get at the time. Would there be any advantage in going to 10in wide? Would the setup need to be changed a lot to get any advantage? I was going to offset the rims so that the extra width added to the overall track.
If a road course were a skidpad, the answer would be fairly simple. In general, wider tires give greater lateral acceleration, within any practical limits, on practically any car, provided that setup and inflation pressure are optimized. However, most road courses have significant straights. In many cases, road races are won on the straightaways. In most of these cases, the straights are at least partially won in the turns: faster cornering lets us start each straightaway at a higher speed, and brake later at the end.
It may be that there are cases where a narrower tire actually will give more cornering power due to higher temperature. Some tread compounds – most, in fact – have a threshold temperature, below which they do not make good grip. However, hotter is not better without limit. Above a certain point, higher temperature hurts grip. It may also cause more severe heat cycling, which makes the tire go off more as a run progresses. If we do have a case where a narrower tire grips better due to temperature, we can say with great confidence that the effect will be highly weather-sensitive: the colder the day, the narrower the tire should be. That makes it difficult to state categorically what tire size we should put on a particular car.
In general, we can say that bigger is better for grip within any practical limits, especially if we have some freedom in choosing compounds. The main drawbacks to big tires are that they weigh more, have more rolling resistance, and make the car wider. Greater car width compromises the line we can take, and it adds aerodynamic drag. The line issue is more important for autocross or hillclimbs on narrow roads than on higher-speed road courses. In general, for road course work, the issue mainly comes down to grip versus drag.
I mentioned the possibility of winning the straightaway in the turn. Perhaps an example will help illustrate. Suppose you are on wide tires and are racing another car on narrow tires, and consequently you have more grip than your competitor, but he has less weight and drag, and therefore has better top speed and top-end acceleration. Suppose your grip advantage is good for a 1mph advantage in a particular turn. Suppose that this competitor is right on your tail approaching the turn.
You will be able to brake later, partly because your car can slow at a greater rate (not only because of greater grip; drag actually helps you here), and partly because you are able to take the turn faster. So you will start pulling out some distance, or at least some time, as soon as your competitor begins braking. You will continue to build your lead through the turn.
At exit, you are still 1mph faster, and still building your lead. Now your competitor can build speed faster than you can, at least if the turn was fast enough so that both cars are now power-limited rather than grip-limited. Is your competitor gaining time on you? Not yet. His drag and weight advantage has to get him up to your speed before he even starts cutting your lead. Once he starts cutting your lead, he has to catch you. Once he catches you, he has to pass you. He has to at least get up even with you before the next braking zone to be able to even have his nose alongside your tail entering the next turn. Remember, he has to brake earlier than you.
If the straight is long enough, he can get by. But the length required is greater than might be imagined.
It will be apparent that in at least some cases, the choice will depend on the course. Long straights and high speeds favor the narrow tires more. A track where most of the lap is spent cornering will favor wide tires more.
The amount of engine power influences the choice as well. Ample power favors wide tires. Meager power favors narrower ones.