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: email@example.com. Readers are invited to subscribe to this newsletter by e-mail. Just e-mail me and request to be added to the list.
HIGHLY MODIFIED FRONT-DRIVE CAR ON STICKY TIRES
I have had quite a bit of correspondence with paying clients who are racing front-drive sedans all over the world. While I have a policy of keeping specifics of clientsí issues and cars confidential, I do address the generalities in my newsletters and columns, and also specifics of a particular personís car, when that person wants the correspondence to be published. I have recently had such correspondence with a person in Colombia, and I will present it here.
I engineer a Nissan Tiida/Versa. Itīs highly modified but now Iīm having issues with the rear suspension. The rear is an H beam and it does everything I donīt want it to do (toe-in in droop).
Iīve been trying with stiffer and stiffer springs (front 600lb/in; rear 900 lb/in) but this solution isnīt what I feel would be best. Lately Iīve been considering a 1100 or more rear spring rate.
So my questions revolve around the following:
-Adjustment of an LSD with adjustable preload for less understeer in FWD.
-How to make the car oversteer in slow corners and understeer a little in fast corners.
-Additional links or rear H beam redesign for racing.
-Any tips in regards to suspension geometry to make the front end more planted.
I feel there is very little info in regards to FWD cars and itís something a lot of people race. Maybe you could do a column in regards to FWD setup and tricks.
I have looked at pictures of the carís suspension on-line, and the rear suspension appears to be a conventional twist beam layout, as pioneered on the VW Rabbit/Golf. Readers can view the system via the following links:
I donít see how there could be toe change in ride. The whole thing pivots about the two bushings as a unit.
I donít see how it would have roll steer either. Such suspensions have had compliance steer, historically. Because the bushings provide lateral location and are well forward of the tire contact patches, there is an inherent tendency for cornering force to cause the rear wheels to aim out of the turn, creating compliance oversteer. The original VW design had such compliance oversteer, but
VW came up with some ingenious bushings that compensated for that, and such designs are now common in this type of suspension. I believe the original patent on the bushings has expired, allowing any manufacturer to use them without paying VW. Actually, it appears Nissan has achieved similar effects with ordinary bushings, simply by angling them in plan view.
What are you using the car for? What sort of tires does it run on? What is the maximum lateral acceleration you get?
I imagine youíve lowered the car some. How much?
What modifications to the suspension are allowed? Can you provide your rules, or a link to them?
Springs, bushings, uprights, bearings are free.
Suspension arms must be modified from the original material but are free in shape (longer, shorter, etc).
You can add additional links to the suspension.
All parts where the suspension arms and links mount must be original but can be reinforced.
From reading some of your past articles and some various ideas I think the way to go to make the car handle better would be the following:
∑ Shift some weight back (we currently have 62% front distribution). I was thinking of putting battery, oil cooler and others in the back.
∑ Start using a front bar and reduce spring in the front. Would help with camber gain in roll.
∑ Optimize rear wing to provide more downforce at required speeds. Also try to find some more downforce at the front.
∑ Use a modified watts linkage for the rear H beam. Would need to modify bushings.
Illustration of modified rear suspension
I think that the bushes in front due to their angle tend to try to bend outwards the side that is compressing. If you looked at that wheel from behind the car I think it tries to describe a circle with a center outwards.
I might do a basic solidworks model to test the twist and confirm this. I think VW and Nissan did this to counteract the tendency for toe out. Since the beam tends to toe in with the angle of the bushings it cancels the toe out caused by the lateral compliance of the bushings.
The car is used for a 2 liter super touring class. We currently pull around 1.42 lat gīs in the corners. Our tracks are usually twisty and depend quite a bit on cornering for good lap times.
We donīt have much power but we have planned to step to a turbo motor later this season. Currently we have a 160 whp motor (we are very high here, 8000 ft above sea level, and having a NA motor is very difficult). Some of the turbo cars are at around 240 whp.
We use Continental Extreme Contact slicks (225/45 R15) and I think we might be able to pull around 1.6 to 1.7 gīs max. We donīt have much banking and speeds are not high enough for extreme aero.
This is a picture of our car:
I donít think there would be any harm in adding a Watt linkage to the rear, but before going to the trouble Iíd want to do an actual bump steer test, and then a compliance steer test, to see what the system is actually doing. My guess is that in ride and in roll or one-wheel bump (no lateral force), you wonít see much toe change. I expect that the steel beam is so much more rigid than the bushings that it will hold the wheels in alignment and the bushings will flex as needed to let the beam move. If the bushings were rigid, the beam would have to flex for the system to move, but hopefully you havenít tried to substitute solid cylindrical bushings for the rubber ones. You would not only get some bump steer, youíd get serious binding. Sphericals would probably work, if the angle of the mounts is within their misalignment capacity. It looks like the system in your picture, with the Watt linkage added, has the bushings repositioned so the axes are aligned with each other, rather than having axes that converge to a point well forward of the bushings as they do in the stock configuration.
The question would be how much change you get under lateral load, without the Watt linkage. The plan-view angularity of the bushings is intended to make axial compliance (along the bushing axes) point the rear wheels into the turn, compensating for the unavoidable tendency for radial compliance
to let them point out of the turn. Ideally youíd like to get the car on a kinematics and compliance rig to see what the rear wheels actually do under a side load. In fact, I would expect that any major manufacturer has a K&C rig and has done that, but checking their work yourself might be a bit difficult in the highlands of Colombia. As a poor manís way of testing, you might try taping a bump steer plate or other nice flat piece of plate to the side of one of the rear tires, and positioning a bump steer gauge to read toe change from that. Then use a come-along anchored to something solid to pull laterally on the roll cage at about the side view c.g. location to simulate lateral inertia force, and see what reading you get on the bump steer gauge.
I would also note whether the car appears to wiggle at the rear in hard cornering, and whether the driver feels that the rear wiggles. If thatís the case, Iíd proceed with adding the Watt linkage but Iíd still want to do at least a crude compliance test before and after, out of curiosity if nothing else. If cost is no object at all, Iíd plan on using the Watt linkage, because thereís no way it can hurt. It would probably work fine with the stock rubber bushings at the front pivots, or with sphericals. Youíd only need to reposition the bushings if you want to use solid cylindrical ones.
I see that you can use different or modified uprights. That could have real possibilities for the front. I expect you lower the car some. That always presents problems with strut suspensions, because the control arms end up with the bushings too low with respect to the ball joint.
You could make new uprights with lowered pickup points for the ball joints. That would give you better camber recovery in roll, and also more geometric anti-roll (higher roll center). It might also be desirable to lower the attachment points for the strut. That might allow more suspension travel or lower ride height, provided nothing else runs out of travel. Probably the most desirable approach would be to machine entirely new uprights from solid. It might also be possible to build up the lower portion where the ball joint attaches by welding, and then remachine that and re-heat treat the part. Lowered billet uprights might be a part you could sell to others. You want the roll center somewhere around three inches above the ground if possible. That probably will require the ball joint center of rotation to be at least as low as the bushings, or a bit lower.
You would want to lower the outer tie rod ends by a similar amount to the ball joints, to prevent bump steer. It would also be desirable to make the height of the tie rod end adjustable, so bump steer could be adjusted to suit various caster settings.
Using more front anti-roll bar and less spring will not help camber recovery, but it may make the car ride bumps better, provided nothing bottoms or runs out of travel with the softer springs. Using more spring and/or more bar, for more total front roll stiffness, will not give more camber recovery per unit of roll, but it will improve camber control by reducing roll. This will come at some cost in terms of how the car rides bumps. Other factors permitting, a reasonable rule of thumb is to try to get about half of the angular roll resistance from the bar and half from the springs.
Moving masses rearward will reduce understeer and probably help braking a bit, but will hurt forward traction. There is no escape from this dilemma with front drive. This means the optimal
front percentage depends on the track. For a track with sweeping turns and no straights, more rear helps. For a track with slow turns separated by drag strips, you might want more front.
Differentials for any car that pulls over 1.3g lateral acceleration are a problem. Any diff that is torque-sensitive works poorly when the inside tire is very lightly loaded. Preloading the diff helps, but adds understeer. Making ramp angles more aggressive helps, but not as much, and adds understeer as well.
The only thing that works halfway decently is a locker, such as a Detroit locker. Most people donít want to put up with those for a front axle, because they create a kick in the steering when they lock. However, a front drive car with a locker can be fast, while the driverís hands hold out. Power steering offers some help, and is more common than it used to be. Probably the biggest obstacle to putting a locker in a front-drive race car nowadays is simply finding one for such a car.
This is not nearly as much of a problem when the tires arenít so sticky. Front-drive sedans only pull around 0.85g on the tires theyíre designed for. If the car has a track to c.g. height ratio of about 3.5, which is typical, at 0.85g only about a fourth of the carís weight transfers to the outside tires. The outside wheel pair then has about 75% of the load.
Suppose that the car has 60% of its weight on the front and 40% on the rear. Then if the rear suspension has enough roll resistance to pick up the inside rear wheel, we have 40% of the carís weight on the outside rear. Percent of total on the outside front is then 75% Ė 40% = 35%, and percent of total on the inside front is 60% Ė 35% = 25%. Front tire loading is distributed 25/35, or 41.7/58.3 Ė fairly equal; equal enough to kill a lot of understeer and allow even an open diff to put a fair amount of power down, or allow a torque sensitive limited slip to work well.
However, if the tires can generate as much as twice that lateral acceleration (1.7g), close to 50% of the carís weight transfers. Itís at the point of bicycling, or flipping. Even if the inside rear is completely unloaded, the inside front is pretty much unloaded too. At 1.5g, about 44% of the weight transfers. The outside wheel pair now carries 94% of the total. Assuming the car picks up the inside rear, the outside rear still has the same 40% of total, but the outside front now has 54% of total and the inside front only has 6% of total. Front tire loadings are now distributed 6/54, or 10/90. At this lateral acceleration, it is very difficult to kill understeer with rear roll stiffness. It is also very difficult to get enough load on the inside front to put power down with anything but a spool, a locker, or a clutch pack diff with a lot of preload.
Rear drive offers no real escape either. However, with rear drive at least longitudinal load transfer aids traction rather than hurting it. And a locker makes the car twitch toward oversteer when it locks, but at least it doesnít beat up the driverís hands and arms.