A Test of Correct Rear Wheel Alignment

Secrets, tips, tools, design considerations, materials, the "science" behind it all, and other topics related to building the cars and semi-trucks.
Post Reply
User avatar
Stan Pope
Pine Head Legend
Pine Head Legend
Posts: 6856
Joined: Sat Jul 05, 2003 7:01 pm
Location: Morton, Illinois
Contact:

A Test of Correct Rear Wheel Alignment

Post by Stan Pope »

Is this a valid test of a process of Bent-axle Rear Wheel Alignment?

Apply the following steps to each rear wheel in turn:
1. Apply the alignment process
2. Run on a track, noting time.
3. Nudge the alignment of a rear axle.
4. Run on the same track, noting time.
If the time degraded, then the change to that axle degraded alignment.

Suppose that the above is repeated several times, producing equal "aligned" heat times and each time the car runs slower after nudging the alignment. Question 1: Can we then say that the alignment process produced best alignment for the given DFW Toe, Rear Axle Camber and Rear Wheel Clearance?

Question 2: What should the magnitude of the step 3 "nudges" be? My guess is that some should be smaller than the last adjustment during alignment and some should be at various somewhat larger magnitudes. Your thoughts?

If the answer to Question 1 is "Yes," then we have a way of evaluating Rear Axle Camber alternatives accurately. And evaluating Rear Wheel Clearance accurately.

Why do this? While this would be rather time consuming if it were required every time we aligned a car, it really only needs to be done once in order to prove the alignment process and one's skill at applying it! Then for subsequent builds, we can apply the alignment process with confidence.
Stan
"If it's not for the boys, it's for the birds!"
User avatar
Vitamin K
Pine Head Legend
Pine Head Legend
Posts: 1249
Joined: Sat Apr 20, 2013 7:26 pm
Location: Spotsylvania, VA

Re: A Test of Correct Rear Wheel Alignment

Post by Vitamin K »

It seems like one could argue that you could only "prove" that the aligned axle was better than any of the non-tweaked axles. Some detractors of bent-axle alignment seem to claim that it will be inherently slower than straight axles no matter what. I'm not certain how you could prove or disprove this.
User avatar
Stan Pope
Pine Head Legend
Pine Head Legend
Posts: 6856
Joined: Sat Jul 05, 2003 7:01 pm
Location: Morton, Illinois
Contact:

Re: A Test of Correct Rear Wheel Alignment

Post by Stan Pope »

Vitamin K wrote:It seems like one could argue that you could only "prove" that the aligned axle was better than any of the non-tweaked axles. Some detractors of bent-axle alignment seem to claim that it will be inherently slower than straight axles no matter what. I'm not certain how you could prove or disprove this.
Yes, if there is no variability in other approaches, and testing those involves introducing uncontrolled variation in other factors, then they have exempted themselves from such focused comparison.

I left out a question above: How many iterations of the 4-step test sequence are required to give 95% confidence that the proposition is true?
Stan
"If it's not for the boys, it's for the birds!"
User avatar
davet
Master Pine Head
Master Pine Head
Posts: 539
Joined: Wed Jan 16, 2013 1:33 am
Location: MN

Re: A Test of Correct Rear Wheel Alignment

Post by davet »

I can't see this as being even remotely viable. There are too many variables between runs, even on the same track.
You could adjust one axle incrementally to achieve best time but that is based on how that one axle relates to the axle opposing it. You are finding the best time with most likely a poorly aligned opposing axle. It also is being skewed by the front wheels and how they are aligned.
You might then start incrementally adjusting the opposing axle looking for the best time however that is also in relation to the alignment of the first axle which is probably still off because you were adjusting it in relation to this one.

Our first year in Scouts we thought we needed a straight-runner. We aligned just as you describe above except we only had a 6' board and a tape line judge by. There was no way to know if any changes were helping or hurting. For all I knew we had all 4 wheels with toe-out yet it ran straight.

I've never tried your weighting system, Stan for aligning rears. It just seemed too complicated and more like an art form in reading the reaction of weight shifting.

I've found that aligning the rears without any front wheels at all is the best for us. We use Lightninboy's method with a square and a board. It is absolutely foolproof and very quick. You can easily identify a bad drill job. It is such a good method that we may run bent axles again if we race next year. That way I don't have to strive so hard to drill perfect holes then to find perfectly straight axles.

Any scout can drill the holes any way he wants and with a slight bend it can be aligned absolutely perfectly very quickly.
User avatar
Stan Pope
Pine Head Legend
Pine Head Legend
Posts: 6856
Joined: Sat Jul 05, 2003 7:01 pm
Location: Morton, Illinois
Contact:

Re: A Test of Correct Rear Wheel Alignment

Post by Stan Pope »

davet wrote:I can't see this as being even remotely viable. There are too many variables between runs, even on the same track.
You could adjust one axle incrementally to achieve best time but that is based on how that one axle relates to the axle opposing it. You are finding the best time with most likely a poorly aligned opposing axle. It also is being skewed by the front wheels and how they are aligned.
You might then start incrementally adjusting the opposing axle looking for the best time however that is also in relation to the alignment of the first axle which is probably still off because you were adjusting it in relation to this one.

...
I wish that I were able to describe these procedures so that their simplicity could be more easily grasped. :(

The test that I described could be applied to Lightninboy's method "with a square and a board."

The method of the test is to assume that an axle is poorly aligned with respect to the opposing axle. Then show that performance is not improved by a sampling of alternative alignments of that axle. This is a statistical application of a few proofs that your teacher showed in geometry class. (To prove proposition A, assume A is false and show that this leads to a contradiction.) While I know that the quality of this "proof" is not on a par with the quality of the instructor's geometry proofs, it should be statistically reliable if enough trials are used. And, remember ... this is not intended to prove the alignment of a specific car. Rather it is to prove a rear wheel alignment method and one's mastery of that method!

BTW, I think that there is little "art" in the weight biased rear wheel alignment procedure. By, in effect, isolating each rear wheel, it can tackle each rear wheel independently and uses a common indicator (the wheel behind the DFW rides the planned line or not) to judge individual rear wheel alignment. It is fast to apply and produces its results based on as nearly real running environment as is possible on a short (3 ft.) test track. It has the "plus" that it can be applied to test that alignment has not been altered as a result of racing and/or mishandling.
Stan
"If it's not for the boys, it's for the birds!"
User avatar
Vitamin K
Pine Head Legend
Pine Head Legend
Posts: 1249
Joined: Sat Apr 20, 2013 7:26 pm
Location: Spotsylvania, VA

Re: A Test of Correct Rear Wheel Alignment

Post by Vitamin K »

Stan Pope wrote: I wish that I were able to describe these procedures so that their simplicity could be more easily grasped. :(

The test that I described could be applied to Lightninboy's method "with a square and a board."

The method of the test is to assume that an axle is poorly aligned with respect to the opposing axle. Then show that performance is not improved by a sampling of alternative alignments of that axle. This is a statistical application of a few proofs that your teacher showed in geometry class. (To prove proposition A, assume A is false and show that this leads to a contradiction.) While I know that the quality of this "proof" is not on a par with the quality of the instructor's geometry proofs, it should be statistically reliable if enough trials are used. And, remember ... this is not intended to prove the alignment of a specific car. Rather it is to prove a rear wheel alignment method and one's mastery of that method!

BTW, I think that there is little "art" in the weight biased rear wheel alignment procedure. By, in effect, isolating each rear wheel, it can tackle each rear wheel independently and uses a common indicator (the wheel behind the DFW rides the planned line or not) to judge individual rear wheel alignment. It is fast to apply and produces its results based on as nearly real running environment as is possible on a short (3 ft.) test track. It has the "plus" that it can be applied to test that alignment has not been altered as a result of racing and/or mishandling.
A little bit off topic here, but maybe relevant. I do like the bias-weight alignment method, but I actually resisted investigating it for a few years because, well, it seemed overly complicated. It wasn't until after my frustrations with drilling cant (and dealing with the inconsistencies of BSA axles) that I decided that I wanted an iterative approach to alignment, and that brought me back to actually investigating Stan's method. Once I did that, I discovered that the complexity was really just my perception, and that the actual method was straightforward and simple.

For what it's worth, I have gotten absolutely torn up on some other boards for suggesting that there might be merits to using bent axle alignment in certain situations, provided that one has a workable method of aligning wheels. Of course, most of the people ridiculing me have not actually tried, or even investigated, Stan's method, But that's just human nature, right?

I guess my point is, maybe the technique needs some better 'PR', to attempt to dispel some misconceptions people might have in their minds about what it entails. I think there are perfectly valid reasons to go with drilled cant and straight axles, but I see bent axle alignment dismissed out of hand for all of the wrong reasons.
Post Reply