Landracing Forum
Tech Information => Technical Discussion => Topic started by: Count on November 12, 2009, 10:21:29 PM
-
Starting to set up the rearend for my corvette, and I am wondering if anybody has any thoughts on camber and toe in the rearend. I'm working up a ford 9" full floater housing and I need to decide if I'm going to add anything or run 0 camber and toe.
-
Just an odd question here. On the Breedlove/Fossett/Ahlstrom car the original designers set it up to have (supposedly) 1 degree of camber and toe in on the rear wheels. We actually measured it across the range of suspension travel and it varied quite a bit. What is typical in LSR and 300+ mph drag racing?
Granted we are comparing wheel-driven to thrust-driven; I'm just curious.
-
You camber and toe the rear end for cornering. For LSR you want it straight and flat.
-
On the rear engine roadster I am working on we designed it with ½ degree of camber and zero toe in the rear. We tried to get a balance of the least amount of drag with stability. A lot also have to do with the type of tire you are running and its construction. Our rear suspension is built around the 28” Good Year LSR tire. Tony
-
Thanks guys, I'll be running the goodyear 28's also. I've been considering .5 deg. camber, just not sure if that will help or hurt things.
-
Tony! Does camber on a non-independant rear housing cause drag? The concept for LSR doesn't sound right but what works is more important.............JD
-
JD, Both the independent and non independent rears with camber causes some drag. But the benefits of stability gained by being on the borderline of drag outweigh the negatives. In the case of our roadster ½” of camber was the boarder line. As I stated in my earlier post a lot also has to do with the tire your running and the construction of the tire. What works with a Good Year may or may not work with a M/T and visa versa. With 30 plus years of suspension design from Indy cars to prototypes under his belt, Steve Conover designed the suspension on our car along with consultation with the Good Year engineers and the Koni engineers. It took about three months to finalize everything. Even though with the many-ply construction of the Good Year LSR tire doesn’t contribute as much to pneumatic trail as a standard racing tire, it was taken into consideration (Before everyone jumps on me saying this only maters in a car turning, it dose mater going straight not as much, but still maters. Just ask a tire engineer). It’s all built around the 28” Good Year LSR tire in the rear and the 21” Good Year LSR tire in the front and there unique characteristics and will not work with the M/T tires. This wont work on every car as there is no cookie cutter approach to all of this. Tony
-
Thanks Tony for the explanation. Putting an axle without u-joints in a calculated bind is a step way above my knowledge. I believe the first any layman knew about camber in a passenger car style rear end was from listening to NACSAR on Sunday and having competitors go out of the race with problems. I can see lees of a "strain" with axle width of 55 to 60"; narrow ones may cause more problems.
For a guy like me who uses 4 different rear tire combinations (brands too) for diameter this would be impossible. I don't want to give up that Pontiac pinion angle and with only so many ratios available for them I need the tire diameter choice.
Thanks you've enlightned me again...............JD
-
JD, After reading my post it may sound like only the tire characteristics were taken into account. But other things that are also taken into consideration, suspension design, weight of car, speed objective, length of car and many more other factors. It’s the whole package, and sometimes compromises have to be taken. Tony
-
The NASCAR guys can run camber becasue they have crowned splines on their axles and run full floating style hubs and axles. The crowned splines allow them to bend some camber into their rear end. Even with crowned splines you still have friction when you add camber. Friction equals horse power. If you are running a tubo 500 inch+ engine maybe it won't matter. Tony's setup uses CV joints which have considerably less friction when angled. If the tire can run with some small degree of camber I don't think that there is any doubt that it will add a certain amount of stability to the car, the real question is "is it worth all of the hassel on a straight axle"?? On an independent design like Tony is using it is just a matter of screwing some rod ends until you have the camber you want.
Rex
-
In maybe '65 Pete Robinson had normal rear tread width with his top fueler but had a wider than normal rear and turned his Halibrands backwards for negative offset to put the wheel bearings at the center of the tread . He lost several final rounds due to having to lift because of handling problems . It then became obvious that the standard combination with the wheel bearing farther inward was bowing the axle and toeing the wheels in under acceleration and improving handling . My friend said with their Formula 5000 Lola they always ran rear toe in . Trailers sway less with the wheels toed out .
-
(edited). . . The crowned splines allow them to bend some camber into their rear end. Even with crowned splines you still have friction when you add camber. . . . If the tire can run with some small degree of camber I don't think that there is any doubt that it will add a certain amount of stability to the car,
Rex I'm trying to understand your post . . . sorry for all the questions. :cry: If you bend the housing to get camber, but use crowned splines, then the tire could run flat (zero camber). Or are you saying they bend the housing and add even more camber which the crown splines will allow. If wheel camber is equal to the bent housing why would you need to crown the splines? The axel centerline is exactly perpendicular to the wheel plane and there would be no relative motion between them. Or are you indicating the crown splines are in the third member where the misalignment motion takes place?
Also, we are talking very small values here (one degree)! How do you manufacture that into the housing with any accuracy? Is the housing bent, the bearing-end flanges machined that way, or the ends welded on with 1 degree misalignment?
Another thought. Could you shim the lower part of the bearing flanges to "tilt" the flange to get camber?
Very interesting thread and thanks to all.
-
I have one 1” cambered and one ½ “ cambered 9” floater left over from my IMSA car. Both housings came from Speedway Engineering and are both spot on in accuracy. For LSR the floater 9" seems like a lot of bearing drag. Though I have heard some guys running ceramic bearings in floater hubs along with axle seals in the tubes to cut down on the drag. Tony
-
Saltfever,
You have to remember that the NASCAR guys are running "full floating axle" rear ends which is means the wheel is not mounted to an axle flange like your street car. The wheel is actually mounted to a hub that has a set of typically Timken style bearings that run on a stub that is welded to the rearend housing. Part of this hub is a drive flange that is splined and the actual drive axle is a straight shaft that is splined on both ends, one end runs in the differential and the other end engages the splines in the drive flange. If you bend the axle housing on a floater you will angle the wheel mounting hub and drive flange the amount of the bend now the straight splines on the drive axle will not fit into the drive flange without binding so they "crown" the splines on the drive axle outer end to allow it to engage the drive flange with out bind. There is still friction generated by the crown spline rotating in the drive hub but there is not binding.
If you bend a standard style housing to get camber you will actually bend the drive axle and put large stresses in the outer wheel bearing and cause bearing failure. Bearings are extremely "stiff" so any miss alignment even what we may consider small, 1/2 degree?, will cause very high stresses on the bearing and cause failure.
Rex
-
Thanks, Rex.
I have seen full floating hubs and they seem to protrude out quite far. I think the recommended rim width for the 28" Goodyear is either 4.5' or 5". That is not a very wide rim. Isn't the floater hub likely to stick out past the rim? Can you get shorter hubs so they can be covered up?
Both you and Tony mentioned the drag caused by camber. What values are we talking about here? Assuming the pumpkin is removed (to eliminate it's drag) can the axles be turned over by hand? Are we talking inch pounds or foot lbs of torque here?
TIA :-)
-
On the housings that we use at the shop I work at, the camber is machined into the snout that is welded on to the axle tube. The axle tube itself is still straight.
So what I am doing is having the housing built to fit my car with the 4 1/2" wheels. The engine is also offset an inch and a half to the right, which complicates things further. Luckily we have a guy who does nothing but build rearends and he has been willing to work with me on this project. I'm finding out that there is a whole lot that goes into building your own rearend.
-
Thanks, Count. How much camber are you going to use? Will the wheel mounting face be in the middle of the rim or offset?
-
I'm thinking 1/4 deg. camber, and mabybe some toe. I wish there was a way that I could test this before doing it. But I've really only got one shot at this. As for the offset of the rims I had Marsh do the wheels for me and had them get the hubs as close to center as they could. I'll try to check that out for you tomorrow. Seems like they were able to do 1 1/2" maybe 2".
-
i dont know about lsr but ive straightened a lot of drag race rearends that were bent or made crooked and the cars always ran faster in mph and lower in et with a rearend that was straight willie buchta
-
I was just thinking...
No matter what, the traction (thrust) on the two rear tires is never equal. The tire with the most traction will want to steer the car to the opposite side. To keep the car straight, the steering is adjusted, so to at least a minor degree, the car body is travelling a little "sideways".
If you look at the attached caveman wall painting (MSPaint), the red arrow is straight ahead, zero toe. This will still cause a little steering offset to the right when that is the "dominant" tire. Green would be the absolute worse case, and require huge steering offset to the right. The Blue arrow (exaggerated) is toe-in. At some point between the red and blue arrows, there is a point that will require no steering offset to go straight, hence keep the body perfectly aero.
I don't think it could actually be calculated exactly, but it seems there is a toe-in number that should actually increase your speed. The more "perfect" your weighting, tire dia and pressure is at full engine torque, the less toe it will take to make the body straight. CG location probably comes into it.
-
If you are running a full-floater with that narrow rim width; aren't the hubs going to stick out past the rim or through the hubcab? Thanks for this thread . . . very interesting. :-)
-
The 9'' ford rear end on the 222 camaro has a drag race style housing. It looks like they used a 2'' wide by 3 or 4'' piece deep of channel steel and welded- fitted and tapered it across the middle of the housing from side to side. Between the brackets for the 4 link there is a 1 1/2 tube welded from side to side and bottom of bar is just even with the bottom of the back of housing. This insures that the housing does not flex up or down or foward or back.
JL222
-
Saltfever:
The race car floaters aren't like a 3/4 ton floater. They're much more compact and don't stick out very much at all. The only amount that sticks out is the cap and that's probably only a 1/4" or so.
Pete
-
Thanks, Pete. I forgot the NASCAR type were more compact.
Mcrat I wish I was as handy with MsPaint. It’s a great drawing that displays perfectly what we are talking about. I’m not sure green is the worst case. Why wouldn’t the blue line be as bad? Any thrust on either side of the red line would want to move about the CG. If we are discounting toe-out on the left for the green line then toe-in on the left side should be discarded for the blue line no? :|
-
Thanks, Pete. I forgot the NASCAR type were more compact.
Mcrat I wish I was as handy with MsPaint. It’s a great drawing that displays perfectly what we are talking about. I’m not sure green is the worst case. Why wouldn’t the blue line be as bad? Any thrust on either side of the red line would want to move about the CG. If we are discounting toe-out on the left for the green line then toe-in on the left side should be discarded for the blue line no? :|
The same angle of toe-out and toe-in, the toe-out will have a much more pronounced force applied trying to steer the car from the back end. It acts on the CG, and has far more "leverage" against it. And a certain amount of toe in will make it so no steering input is required to go "straight" even when the rear tire traction is very unequal.
-
By the way, I'm not sure I'm correct, or whether it's significant. It just makes sense to me.
-
McRat, I tend to agree with you, now if I can just figure out that magic number. It seems to me that if you can make the car more stable, with a minimal amount of drag, and less driver input of countersteer, you should go faster.
-
I know that with the sports-racing and formula cars I've worked on over the years that rear toe out led to immediate, often drastic handling problems and rear toe in was usually totally stable. My tendency would be to run only a touch of toe-in if any. As for camber I wouldn't run any in this application but if a person wanted to I'd again run only a touch. The main thing is to ensure that both sides are identical.
Bye the way a touch is a very precise measurement related very closely to a smidgen!
Pete
-
Only slightly related, we had serious handling issues when first started 4x4 dragracing. Toe-in in the FRONT, will make the truck change directions violently if traction shifts from left to right. Since we had an open front axle, when one tire broke loose the truck headed with great drama one direction, then as it caught, violently in the other. We fixed it, so now in the worst case no matter how many tires spin, the most it will do will be having the body point straight down the track, no steering, but it's path curves like when you're bowling. I figure that's torque reaction, and as long as you don't try to correct it, you're fine.
-
0 rear camber, 0 rear toe, works good for us.
-
Bye the way a touch is a very precise measurement related very closely to a smidgen!
Pete
a smidgen is ten poofteenths, whereas a bees-dick is three fifths of five eighths of F all. Either way these measurements are below the usual tolerances of the human eye multiplied by the tolerance of the calipers you just stepped on. :roll:
-
LOL :-D
Ok, so now all I have to do is figure out just how much that is.
-
Ask your wife. If she does any home cooking at all, she can tell you!
Lynda
-
i asked sheri she said when she was younger she used to have a lot of toe out but as she gets older gravity has taken its toll willie buchta
-
Saltfever, This is one of my floater hubs. When we ran in IMSA we made up spacers for the inside to keep the bearings from riding up on the axle snout to cut down on drag. Also I built a dry sump for my 9” in my gas roadster. It is very similar to the one that the Corvette team used at Lemans on there 9”. I also tested that setup on a test rig I built against others and saw the gains. I posted some pictures on another thread of the dry sump system. Tony
(http://i273.photobucket.com/albums/jj236/maguromic/hubjpg.jpg)
-
The same angle of toe-out and toe-in, the toe-out will have a much more pronounced force applied trying to steer the car from the back end. It acts on the CG, and has far more "leverage" against it. And a certain amount of toe in will make it so no steering input is required to go "straight" even when the rear tire traction is very unequal.
Thanks, McRat. I can't argue about your personal experience I'm just having trouble understanding it. Maybe there were other things influencing toe and they affect "out" more than "in". But in my mind if out or in toe angles are equal then the force should be equal but in different directions. To keep it simple think of a motorcycle. If the rear wheel is misaligned and pointing to the right the bike will go that way. If the wheel points left then left you go. To me a car wheel acts the same way. The CG is the location that everything pivots around. Whether the toe is out or in only affects the direction of spin. But as you mention there is uneven traction, and the other tire, so maybe I'm missing something. At any rate this is a thought provoking thread and has gotten me thinking, and focusing, on how critical alignment is. I would have not considered checking the straightness of my rear end until I read this thread. Alignment to the frame yes, but straightness I would have missed. Many thanks to all.
Tony thanks for posting the picture. It alleviates a concern I had.
-
The simple way to measure if a rear end is bent when the axle is out of the car is to mount a set of wheels. Block the wheels so they can't roll, then roll the axle through 360 degrees. A tape measure between the two wheels during the rolling process will show if the rear is bent and how much. If the amount is at all significant you will notice a wobble in the wheels.
Pete
-
Thanks for the idea, Pete. I was going to mount a bar on each side with a cheap mag base laser projecting forward only to get toe and missed the idea of "swinging" the housing. I'm assuming you are talking about a floater otherwise the wheels would have to be mounted to the axles. If it is a floater is there bearing slop or other built-in clearance that I should be aware off when swinging the housing? TIA :-)
-
It will work with any rear axle. The wheels are mounted on the axle and held in place by the blocks while you swing the axle housing. Its commonly used in oval track racing where there's lots of contact. The laser will work just fine to detect any camber or toe errors. Just note where the housing is in relation to the pointer.
The next trick is using a rosebud torch to straighten a bent housing.
Pete
-
Hell Yeah! Fire up the oxyacetylene! I love those old fabrication techniques.
-
I always used the welder method to straighten a housing. Find out were the most bend is and then lay a bead on the concave side and quench it with a wet rag. It will pull the bow right out of a nine inch housing. I did it to a nine inch that was in a Mazda RX7 during the Daytona 24 hours once.
Saltfever, a car is not even close to a bike from the affects of wheel missalignment. If every thing on a car when moving did not change, no weight transfer, no flexing, no bumps, no outside disturbances at all, then toe in or toe out might not affect the cars handling. But since we don't live in a perfect world, thank God, toe out will be much more destablizing than toe in on a race car. In the venacular of NASCAR, toe out make you "lose" and toe in makes you "tight" and lose is not a stable condition.
Ever driven a fork lift? There is a real case of rear toe steer, and unstability.
Rex
-
As a matter of fact in my days as a starving student I drove a forklift for Golden Grain Marconi and became quite proficient at it. With time and practice your subconscious will sense subtle nuances of rear steering that gives you an enhanced capability you could not have if you tried to process all feedback consciously. But I digress . . . :-)
Sorry for trying to simplify with the motorcycle example because you are right about all the disturbances that affect a 4 wheel vehicle. So here is the deal. There is a vast difference on whether the steering axis is before the CG or after the CG. The former is an inherently stable condition. If the steering input is behind the CG then the CG adds to the lateral acceleration . . . an unstable condition! As you mentioned there are many disturbances. Because of the disturbances we are assuming that one rear wheel has enough influence to start a deviation from the longitudinal track. I don’t care if it is toed in or out. That wheel will move the car in the direction the wheel is pointed and since it is behind the CG (and with a shorter couple) the force to correct or stop the course deviation is far greater than front-steer. I understand the “feel” of loose or tight and how toe causes either effect. However, I think we are confusing front-steer “feel” with rear-end physics. Intuitively, we assume toe-out is bad in back because that is the effect up front. However, if you are traveling in an absolutely straight line and the car’s rear end darts to the left (because disturbances to the wheel that takes command has toe-in) the deviation from track is going to be quick and difficult to stop because the lateral acceleration of the CG is in the direction of travel!
I'm not trying to be argumentative here. I hope you guys consider this a discussion. I'm just trying to understand why there is some preference to toe-in. As I think about this I'm starting to think any rear end toe is a very bad situation. Either case is unstable due to its position behind the CG. And building in any propensity to go left or right of track is precarious.
"Loose" or "tight" are simply the same as understear or overstear. Compared to "neutral" all those are unstable conditions. LSR travels in a straight line. I suspect "neutral" is the best condition.
-
Thinking more on it, it's possible that rear toe out might induce rear end oscillation. I was just looking at it like a flat board, but it's 3D. The CG is above the thrust point on the tires. If you try to "pull" the CG to the right, it would apply downforce to the opposite tire, making it the new dominant tire, then swap back and forth. With toe in, the dominant tire tries to lift the opposite tire, so it maintains it's dominance.
-
Hmmm, interesting and good point. The tire with toe-in will remain dominant but it is still pointed to the left. Me thinks it wants to drive the CG to the left. If that tire remained dominant but was pointed forward (zero toe) the tendenancy would be to track straight.
-
I know this isn't my post but thanks for all your helpful insight. I has been very informative.
Later
Robbie
-
Jump on in Robbie. The waters fine.
Ok, lets call the left tire A and the right tire B. going off the centerline of the car to a point ahead of the car we'll call X. If A and B are toed out then the intersection of the centerline will be behind the car. Going forward the imaginary lines go away from the centerline of the car. It seems to me that this would want to pull the car in two different directions. With A and B toed in the imaginary lines cross the centerline X ahead of the car at the same point pushing the car in the same direction. Which seems to me why toe in would be more stable. Am I heading in the right direction with this?
-
Great description Count and nice observation. However, I don’t think it is applicable to the issue. If I could only draw with Paint as well as McRat, well a picture is worth . . . :-)
So let’s try an example. If you are an overhead crane operator you will instantly understand. Picture a 3 ton weight hanging from a crane. As you start to move the overhead crane, the mass will lag behind the overhead carriage but eventually start to move forward. When the crane stops the mass will continue on for a bit. This starts a swinging action like a pendulum. To stop the swinging the crane operator watches the swinging mass and as it reaches its maximum swing point and starts back he moves the overhead carriage in the same direction. This dampens the swing and the load will hang vertical. (for illustration only. Swinging a load is dangerous and could lead to termination). Don’t ask me to explain the physics – I can’t.
We are talking about a car being disturbed for any reason and starting to move the CG from its intended path. The only forces to stop that movement are, wind, CG mass (one of Newton's laws :wink:), and tire-to-surface friction. (Would that be called the scrub angle?). Since we are already misaligned with the course we can assume all three forces are not sufficient to arrest the problem.
I will use the assumptions already posted in this thread. One rear tire is dominant. I don’t care if it is pointed left or right (toe in or out) but because it is dominant it will drive the rear-end the direction it is pointed. One degree at Bonneville means you are pointing either left or right of Floating Mountain. Just kidding . . . don’t know exactly. :-D
When a (zero toe) rear-end starts its deviation the CG also moves in that direction. (Think of the 3 ton mass above) The driver steers into the direction of rear-travel and by moving the front of the car in that direction it dampens out the “swing” of the CG. This allows the tire’s scrub angle to return to normal and assist in recovery through the tire-to-surface interface (friction). Two forces are in your favor.
Now picture either left or right toe. As the rear swings in the direction of the toe (remember it is dominant) the driver will turn the front end in that direction. The toe has already assisted to start the “swing” of the CG and is actually driving the CG away from the center-line. As the driver turns toward the direction of the toe he actually increases the steering angle (the left toe) away from the center line further accelerating the CG away from centerline. The “dampened-pendulum-effect” is all that you have for help because you have worsened the scrub angle at the tire surface. The more you steer into the spin the more you drive the CG away from the cenetr-line. Very bad IMHO.
Sorry for the dissertation. McRat’s skill could have saved 1,000 words.
-
(edited . . ) Ok, lets call the left tire A and the right tire B. going off the centerline of the car to a point ahead of the car we'll call X. . . With A and B toed in the imaginary lines cross the centerline X ahead of the car at the same point pushing the car in the same direction.
In a perfect world, you are right. You have perfectly divided thrust at each wheel, with perfect adhesion, and no crosswind. Its a nice triangle with the CG in the middle being driven in a straight line. I think engineers call that something like a force vector analysis to show everything balances and thrust is through your imaginary point. However, it ain't perfect out there. To simplify take out the spool and use an open diff. (A lot of cars run Bonneville without a spool). Now you have applied almost all thrust through only one side of your triangle. Also, that one-sided thrust is constantly increasing and decreasing due to every condition on the course! To make matters worse, with an open diff, your thrust can be changing from side to side, again depending on traction! So as line A thrust overwhelms line B thrust, what will be the tendency! Doesn't sound stable to me. If you install a spool you have perfectly divided torque to each side but not equally divided thrust. Thrust is the application of torque to the ground interface. For too many reasons to list here, traction will not be exactly equal on both sides of the car! So even with a spool you have the problem of unequal thrust along your A or B vectors. Even with no toe-in you still have the problem of unequal thrust. That is why dragsters narrowed their rear ends. By bringing the wheels closer to the centerline you decrease (but don't eliminate) the offset thrust angle. YMMV and IMHO :-D