Moving a rear axle forward or aft

[FatSebastian]

Conventional construction has the rear axles installed co-axially (e.g., in the same slot, or aligned holes, with the nail tips pointed toward each other).

What if rear axles are not kept coaxial? That is to say, what if one of the rear axles was moved forward or aft relative to the other rear axle? What are the potential consequences, and how would these consequences have a positive or negative outcome on elapsed time?

[A1nogoslo]

Conventional construction has the rear axles installed co-axially (e.g., in the same slot, or aligned holes, with the nail tips pointed toward each other).

What if rear axles are not kept coaxial? That is to say, what if one of the rear axles was moved forward or aft relative to the other rear axle? What are the potential consequences, and how would these consequences have a positive or negative outcome on elapsed time?

So many ways to build a car.
Unbalance of the rear weight placement due to the moving of axle placement forward on one side could be a potential consequence.
A good track and timer would be the only way to find out the anwser you are looking for if you want to know the outcome on elapsed time.
Have the car made with extra axle holes or slots in the rear so you change one thing at a time so you don't change other accpects to the original cars setup.
We have made cars with the rear axles very slightly forward to back in location
we haven't seen much or difference in times...
It would probably have to move them more like a 1/16" + or - offset to really see if that does anything or not


So many things to try and test on these little cars... thats what makes it fun.

[Stan Pope]

Conventional construction has the rear axles installed co-axially (e.g., in the same slot, or aligned holes, with the nail tips pointed toward each other).

What if rear axles are not kept coaxial? That is to say, what if one of the rear axles was moved forward or aft relative to the other rear axle? What are the potential consequences, and how would these consequences have a positive or negative outcome on elapsed time?

Interesting question!

Factors involved:
1. Moving one axle forward shifts weight distribution slightly to the car's rear wheels, and the moved wheel bears slightly more weight. So, if you are depending on a certain amount of weight on the front to maintain stability, then you might lose it.
2. If the track is "perfect" (no joints, etc., to cause bumps), then there should be no effect other than caused by #1. If the track has poor joints with "bumps" on both side of the rail at each joint, then your car's CM moves less over symmetrical bumps. I think that this results in slightly less loss from such bumps.

I think that the effect is small enough that you must construct tests very carefully to avoid masking the effect with unintended influences. To make the effect large enough to measure on-track (without being swamped), you may have to construct exaggerated "joint bumps" on the flat near the bottom of the slope. Or, maybe a series of smaller bumps spaced about twice the difference between the rear axles.

[FatSebastian]

Interesting question! [...] 2. If the track is "perfect" (no joints, etc., to cause bumps),

So let's clarify the question a bit more by assuming we are dealing with a 3-wheeler and there are bumps. Suppose we are able move ballast side to side to keep the COM in the same place (or where necessary) to maintain front traction identical to the coaxial situation. Consider the following geometry questions:

Because COM is usually referenced relative to "the" rear axle, what reference line would become appropriate, or should one spit the difference between rear axles?

Suppose the front hits a bump such that it wants to pitch up. The pivot axis is no longer perpendicular to the direction of travel. Is this a good thing or a bad thing?

Suppose the front hits a rail defect such that the car wants to yaw. Same question.

[pwrd by tungsten]

Interesting question! [...] 2. If the track is "perfect" (no joints, etc., to cause bumps),

So let's clarify the question a bit more by assuming we are dealing with a 3-wheeler and there are bumps. Suppose we are able move ballast side to side to keep the COM in the same place (or where necessary) to maintain front traction identical to the coaxial situation. Consider the following geometry questions:

Because COM is usually referenced relative to "the" rear axle, what reference line would become appropriate, or should one spit the difference between rear axles?

Suppose the front hits a bump such that it wants to pitch up. The pivot axis is no longer perpendicular to the direction of travel. Is this a good thing or a bad thing?

Suppose the front hits a rail defect such that the car wants to yaw. Same question.

You would have to split the diffrence between the rear axles is my thought.

It is possible to build a very fast car without doing this. I agree with previous posters about this being a benifit mostly on tracks with bad joints where both rears may hit at the same time...

I would not risk it without a track to test. Even then I would test something else before this

[Stan Pope]

... Consider the following geometry questions:

Because COM is usually referenced relative to "the" rear axle, what reference line would become appropriate, or should one spit the difference between rear axles?

This is another reason that I prefer to use a weigh scale rather than a ruler to locate the CM!

Suppose the front hits a bump such that it wants to pitch up. The pivot axis is no longer perpendicular to the direction of travel. Is this a good thing or a bad thing?

Suppose the front hits a rail defect such that the car wants to yaw. Same question.

All bets are off if the track is that bad and you chose the wrong wheel to make dominant!

[FatSebastian]

All bets are off if the track is that bad and you chose the wrong wheel to make dominant!

Obviously I phrased my question poorly.

Previous responses suggested a potentially small benefit for the back wheels negotiating a bump. Considering small disturbances to the front wheel, either from the bottom or from the side (every racing environment will potentially introduce many small disturbances), how might the fact that the back wheels are not coaxial affect the dynamical stability of forward motion, if at all?

[Stan Pope]

... Considering small disturbances to the front wheel, either from the bottom or from the side (every racing environment will potentially introduce many small disturbances), how might the fact that the back wheels are not coaxial affect the dynamical stability of forward motion, if at all?

Noting my prior sidestepping of the more detailed question, I'll simply say that the details of this case are too subtle for my experience/intuition to analyze to a reliable conclusion! I suspect that it is inconsequential, but as we know, every 0.0001 second helps/hurts at the top end of competition! Again, tests must be extremely sensitive because the effect is easily masked by "noise".

[FatSebastian]

Noting my prior sidestepping of the more detailed question...

Fair enough.

I'll simply say that the details of this case are too subtle for my experience/intuition to analyze to a reliable conclusion!

Since the question may be largely academic - I'd find an unreliable conclusion just as interesting!

[FinePine]

During a curved section of the track, the forward rear wheel will tend to lose contact with the track. With a rear weighted car, I expect that the car would diagonally tilt to maintain/restore contact on both rear wheels. If your car is a 4-wheeler, it will become a 3-wheeler. If it is a 3-wheeler, then the result depends some on whether the dominant front wheel is on the same side as the forward rear wheel or not. In any case, I don't think the consequences would be beneficial. Well, if the COM was unusually placed, a 3-wheeler could become a 2-wheeler for the duration of the curve, so if your track is a continuous curve, you might be able to delay wheel spin-up for the third wheel until you reach the flat, but then you have an abrupt change to deal with when the wheel does make contact. Same effect for a 4-wheeler, too, so if you are required to have 4 wheels touching (on a flat surface), then you could make the 4th wheel so that it barely touches, but would be raised for most of the curve. Of course, if your rules are restrictive enough to require 4 touching, then you probably also have to use the stock slots anyway.

[Stan Pope]

Since the question may be largely academic - I'd find an unreliable conclusion just as interesting!

Okay. Note that if the two rear axles work as a hinge during front wheel impacts, that the initial motion of the front axle tries to be almost exactly vertical, even if the rear axles are offset as much as an inch. This is almost exactly the same as when the rear axles are coaxial! From this, I would expect no significant difference between the two configurations on a track with only small defects.

Another concern is the steering effect of bumps encountered by only one rear wheel. Each bump can be viewed as a transcient increase in friction which would cause an impulse to rotate the car about it's vertical axis (yaw). For instance, a bump encountered by the right rear wheel would cause the car to try to steer to the right. If the track is really good and the section join lines are the most severe bumps, then the car with offset axles will encounter two such steering torques at each join. The car with aligned axles would encounter one steering torque which is almost totally nulled by opposing impulses. For bumps which appear on only one side of the lane, the rear axle offset should not be a factor.

I think that we don't need to be concerned with the transcient imbalance issues on 3-wheel cars, since they would apply only if the car's CM were near the side-to-side limits of stability. 3-wheel car builders don't go there!

[FatSebastian]

During a curved section of the track, the forward rear wheel will tend to lose contact with the track.

Great observations, FinePine. It would then seem desirable that the rear wheel on the non-dominant side should be the forward one.

Another concern is the steering effect of bumps encountered by only one rear wheel. Each bump can be viewed as a transcient increase in friction which would cause an impulse to rotate the car about it's vertical axis (yaw). For instance, a bump encountered by the right rear wheel would cause the car to try to steer to the right.

Can one precisely predict (or measure) the location of the vertical (yaw) axis, and does moving one of the rear axles change the axis location?

I think that we don't need to be concerned with the transcient imbalance issues on 3-wheel cars, since they would apply only if the car's CM were near the side-to-side limits of stability. 3-wheel car builders don't go there!

You have written what I was hesitating to put into words. Hitting a seam might cause the car to yaw slightly left, then right, so does this tend to cause instability or unstable feedback? I take it that your guess seems to be "probably not" except for extreme configurations. Great stuff!

[Stan Pope]

... Can one precisely predict (or measure) the location of the vertical (yaw) axis, and does moving one of the rear axles change the axis location?

This is a great point! Absent friction (i.e. car rotating in free space), the axis would be through the car's CM. However, when wheels are touching the track with some normal gravity induced force, they will try to become the center of rotation ... unless they break free. Then, I think, the issue gets very cloudy. If all three (or four) wheels are touching and sliding with the same CF, then the yaw axis is still through the CM (I think). However, if one of the wheels jumps up (probably the lightly loaded DFW), then (I think) that the yaw axis shifts toward the remaining working (probably the rear) wheels. Predictability may depend on our ability to assess the CF's and weight distributions!

Now, what we need is a real physicist to delve into the specific case and tell us of the real motions involved!

[Stan Pope]

... If all three (or four) wheels are touching and sliding with the same CF, then the yaw axis is still through the CM (I think)...

Just realized that the friction of the sliding wheels gets divided into two vectors ... one is sliding friction (parallel to the axle) and the other causes the wheel to rotate differently than when the car was not sliding. I think that the wheels closest to the CM have more involvement with "rotating" than wheels farther away. Having to factor wheel moment into the analysis really mucks it up!