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.
OFFSET FRONT ENGINE FOR ROAD RACING?
Regarding engine offsets and any related configuring of a-arms, here is a related question.
I know of one dominant, high powered road race front-engined, live/solid axle GT race car (configuration dictated by rules as ďproduction basedĒ) built and raced years ago by a major auto manufacturer that used slight engine offset to the right (driver on left) with maximum rearward engine/transmission setback, with driver as far to rear as possible. If there are relative design merits of this besides the rearward engine/transmission (balancing/offsetting weight of the 180-190 lb driver, and getting him some room with max engine setback), what would be the a-arm and other suspension geometry design and tuning considerations of such a configuration?
A related obvious question for a solid axle car would be the effects of greater asymmetry at rear with differing axle/housing lengths as a result of right engine offset?
And of course the big question then is do you think all the related and required changes related to the engine side offset would be worth the bother and make the chassis work all that much better?
If the car is in the design phase, thereís probably no more bother to have the engine offset than not. Small offsets are actually quite common in production cars. If weíre talking about an existing car with the engine centered and weíre considering modifying it, thatís a lot of bother. It would involve a fairly comprehensive rebuild of the whole car, so itís the sort of thing that would only make sense if weíre already committed to that.
When the car has to turn both ways, we generally want the front suspension symmetrical. However, there may in some cases be an argument for tolerating small asymmetries due to packaging constraints. At the rear, with a beam axle, itís desirable to have asymmetries that cancel torque roll, as discussed in previous issues, but no other asymmetries that affect wheel loads.
However, that doesnít mean the center section canít be offset. That doesnít affect torque roll, or at least not in and of itself. What counts for that is the members that transmit axle torque and thrust
from the housing to the frame. Where the pinion shaft transmits its forces to the housing doesnít matter. Dramatically offset center sections are the norm on front axles in four wheel drive vehicles, and those offsets donít have any noticeable effect on suspension behavior.
In some cases, elements that do transmit torque and/or thrust from the axle to the frame necessarily are offset along with the center section Ė for example, in the case of a torque tube. But ordinarily, packaging constraints permitting, we can have any linkage design we want, with any center section offset we want.
Unless the front of the engine is entirely behind the front wheels, generally the front wheel housings and front suspension will limit how far we can offset the engine. Having no front wheel housings helps some, but the wheels still have to steer. Having a narrow engine helps. Having a wide track helps.
REVERSED LOWER A FRAME ON A LOTUS 7 STYLE REAR SUSPENSION?
I'm in the process of building a car that uses the A-H Sprite rear axle (no laughing it's what I have). Back in the day the Lotus 7 used a similar axle with a lower A frame attached to the center of the axle to replace the lower trailing links and a Panhard bar. I always thought this was pretty slick as it allowed transverse location without hanging structure out behind the axle. Of course it didn't work as the axle didn't like the load and, as a result, leaked its fluid.
My question: if I flipped the A arm around so its apex mounted at the center rear of the chassis floor and with the other ends mounting where lower trailing arms would normally mount on the axle would I end up with a workable set-up similar to a dual trailing link and Panhard bar?
Yes, it is possible to turn the layout around like that, and this has the additional advantage of making the roll center move up and down with the sprung mass rather than the axle. This makes it a more harmonious partner to an independent front suspension.
The most common obstacle to this idea is finding room for it. If the A frame is made short, the driveshaft tends to hit it in droop. If the A frame is made long, it intrudes on seat or footwell space.
These problems are lessened greatly if we donít need a lot of ground clearance or suspension travel, as in a pavement race car. It may be possible to integrate the mount for the front of the A frame into a driveshaft containment structure. That point will see very large longitudinal loads, so the structure needs to be built to take those.
In the Lotus design, those large longitudinal loads are applied at the middle of the axle housing, and the resulting bending loads are what cause the problems. The A frame is about 6 inches above the
ground. The upper links are about 15 inches above ground. When each tire exerts a propulsion force of 100 pounds, the A frame is loaded in compression and applies 333 pounds rearward to the center of the housing. The axles each apply 100 pounds forward at the outer bearings. The upper links each apply 67 pounds forward. This combination of forces tries to bow the housing back in the middle and forward at the ends. There are also torsional loadings, but itís the bending thatís the real problem and causes leaks at the gasket where the diff snout goes in. Other axles, including the Ford 9 inch, can have the same problem if they are loaded sufficiently in this manner.
If we reverse the layout, the 333 pounds rearward is applied as 167 pounds rearward at each end of the axle. There are still big torsional loads on the tubes Ė bigger in fact Ė but the bending loads are now very small. And in braking, only the ends of the tubes are loaded in torsion.
It is possible to reinforce the housing so it will take the bending loads. One method is to add a channel-section brace across the rear of the housing. This adds torsional stiffness as well as bending. Alternatively, it is possible to use angled tubular members on the back of the housing, anchored to a bracket protruding rearward a little behind the diff. This creates a triangulated structure. It is possible to have opposite-hand threaded clevises or Heims at the ends of the tubes, and get limited toe adjustment. Itís possible to add another set above or below the axle, and get limited camber adjustment. All these ideas can be used with any linkage layout. The Lotus design just makes them unusually desirable.
COPING WITH TIGHTENING TURNS IN AUTOCROSS
Would you advise on autocross? I see that your specialty is oval-track and roadracing, but I get the sense that a few cursory minutes of your thoughts on the matter might be as good as some experts in this smaller field of interest.
Do you have any thought on whats and hows for shock absorbers when the driving is to keep control during all quick large steering movements and sharp transitions between off-throttle and/or trail-braking into tightening slaloms and other no-rest decreasing radius turns, and powering out of tight corners?
Iíll actually advise on any kind of vehicle. Right now Iím mentoring a team at UNC Charlotte who are designing a human-powered vehicle. US-style autocross is basically road racing (asphalt surface; car turns both ways) in a parking lot, one car at a time, around traffic cones. Usually the turns are tight and the straights are very short. Well maintained parking lots tend not to be very bumpy, but they can be quite wavy, since they cover a lot of area and are generally intended to be used at very low speeds.
Iíve driven a few autocrosses, and Iíd say ďno-restĒ pretty well describes the event. The turns and gates come at you so fast that the steering wheel is in nearly constant motion and the car spends very little time in steady state.
The event also rewards aggressiveness. You get a time penalty for every cone you knock down, and if you miss a gate your run is not counted, but on the other hand there is usually not much to hit except cones, and you are scored only by your best run.
Because most of the turns are tight and of short duration, the car is generally undergoing large yaw accelerations for a substantial portion of the run. The usual challenge in very tight turns or slaloms is to overcome understeer when initiating the turn and oversteer when concluding the turn. My usual recommendation for this is to use more low-speed damping, both compression and rebound, at the rear than at the front. That tends to de-wedge the car (unload the inside rear and outside front and load the other two wheels) when itís rolling outward and add wedge when itís de-rolling or rolling inward.
This is not without some disadvantages. The rear wheels will tend to unload more than the fronts over crests. Using compression damping, not just rebound, at the rear can help here. This may call for more than a simple adjustment on a single-adjustable shock, however.
The questioner asks about decreasing-radius turns and slaloms. Courses Iíve driven havenít had these, but I can easily imagine course designers including them to add a driving and setup challenge. When weíre trying to slow down during sustained cornering, we are most often worried about oversteer rather than understeer. If weíve adopted setup tricks that are geared to tossing the car entering tight turns, especially extra rear brake, a steady turn of decreasing radius can catch us out.
Barring elaborate active suspension and braking systems, we canít really expect the car to know whether weíre trying to toss it or slow it without spinning it. However, if we use rear low-speed damping to free the car up on entry rather than using rear brake, that will have relatively great pro-oversteer effect entering abrupt turns, and relatively little effect in sustained braking and turning.