Wheel balancing bemusement

[RobertRO]

A few days ago I went to a garage to balance my wheels. The worker did his job properly, presetting the balancing machine according to the specs of the wheel. Then he applied the weights, again correctly, and gave the wheel the final check spin. The machine displayed all zero, confirming a correct balancing.

Then - don't ask me why - I asked the worker to rotate the wheel about 90 degrees on the machine spindle, and recheck the balancing. To everyone's bemusement, the machine showed that the wheel was unbalanced, and that other weights were needed (of considerable size, 10g+). Of course, no other weights were applied.
We tested again, first with the wheel rotated by 180 deg., then by 270 deg. from initial balancing position. The machine reported the same result: the wheel is not balanced, other weights must be applied. Only when the wheel was rotated back to the initial position it had during balancing (relative to the spindle) did the machine zeroed again.

I assumed that the garage had a defective / uncalibrated balancing machine, and went to another one. And another one. Until now, I tested with 10 (ten!) different balancing machines, from various manufacturers, some of the machines almost new, others used for years, and every time the result was the same: a wheel balanced on any machine is reported unbalanced by the respective machine when rotated at any angle on the machine spindle, other than the initial balancing position (0 / 360 deg. rotation).

I lost two full days with this and I can confirm that the machine operators did set them up properly, and applied the weighs correctly. I have even used a precision scale to apply weights trimmed to the gram. By default, the balancing machines round up the weights to a multiple of 5, but they can be turned into a "precision" mode and can display the required weights in 1 gram precision.

I think that it is very unlikely (although not impossible...) that all these 10 machines are defective, and/or all the 10 operators are clueless about their job.
I wonder if anyone here has observed something similar. If not, I ask the first member on this message board who will balance a wheel, please do the test described above (= rotate the wheel on the machine spindle after balancing) and report back.

 

[andysat]

Bloody ell-that's interesting!

 

[woolybear]

In an ideal world, one would hope that wheel balancing machines, were independently Calibrated and Serviced, by a thorough and totally Independant company to the one, that's carrying out the service, that you require - i.e: digital wheel balancing, in this instance.. A minimum of every, 3 to 4 months, on a revolving cycle, as it were..
My machines are and it's a basic requirement, to ensure the accuracy of the equipment, to be honest..
It's a cost and one which is far from a cheap one. However, I cannot recall the last time, I had a client back to me, with an issue..
Cheers..

 

[RobertRO]

My machines are and it's a basic requirement, to ensure the accuracy of the equipment, to be honest.

Woolybear, thanks a lot for your feedback.
Considering that your machines are properly maintained, I would highly appreciate if you can perform the test described below at your earliest convenience.
After you balance a wheel, please rotate it on the spindle of the same machine with a certain angle (e.g. any angle between 90 and 270 deg.) and recheck the balance. Let us know if the machine zeroes, or reports a certain unbalance.

 

[Oakandacorns]

This is interesting, would be interested what the reason for your experience has been RobertRO. Will follow this with interest.

 

[RobertRO]

I have submitted the issue described above to Mr. Mike Miller, the technical editor of Roundel Magazine (BMW CCA). I guess he wouldn't mind quoting here his feedback:

You have described normal wheel balancer function.
Once you’ve achieved 000, you’re done. There’s no reason to rotate the tire at all after that. Instead, you remove the wheel and tire assembly from the balancer and bolt it onto the car.
BMW wheel/tire assemblies should be balanced “to the gram,” and if that requires a precision mode on the balancer then that’s what is required.

On the same subject, here's an excerpt from the user manual of a noname wheel balancing machine:

INCONSISTENT UNBALANCE READINGS
It could happen that after balancing a wheel and removing it from the balancing machine, then again mounting it on the balancing machine, the wheel is found not to be balanced.
This does not depend on incorrect indication of the machine, but only on a faulty mounting of the wheel on the adapter; i.e. between the two mountings, the wheel has assumed a different position with respect to the balancing machine shaft center line.
If the wheel is mounted on the adapter with screws, it could be possible that the screws have not been correctly tightened: they should be tightened one in crosswise fashion: or else (as is frequently the case ) the wheel has been drilled with too wide tolerances.
Small errors, up to 10 grams (0.4 oz) are be considered normal in wheels locked by a cone: the error is normally greater for wheels locked with screws or studs.

To conclude, it seems that - for reasons which I still do not understand - it is normal for a wheel to be reported unbalanced by the very machine that balanced it, when the respective wheel is remounted in a different position relative to the machine spindle.
I would like to find somewhere a solid explanation of this odd phenomenon. I guess it lays in the very principle and technological limits of the balancing machines.
Although makes some sense, I am not buying too much into the explanation that the mounting adapter is the source of errors. If that is true, then it means that most wheels are incorrectly centered on the machine, thus incorrectly balanced.
The common sense tells that a balanced wheel should be reported balanced (within very small margins of error) no matter how I retest it on any calibrated machine.

It would be great if woolybear, who posted upstream in this message thread, could perform the test and report back, considering that his wheel balancing machines are properly maintained and calibrated.

 

[dell]

Think there are out of balance forces in any machine, these would be zeroed on initial setting. Just a thought! Dell.

 

[Anonymous]

So basically, the machines arent consistant! If you balance the wheel...then for example want to double check, it will give a different reading.

No consistency means you will never know its correct.

 

[RobertRO]

I've got another opinion from Mr. Woody Rogers , Product Information Specialist at Tire Rack USA.
I am posting hereafter his feedback (again, hoping that we wouldn't mind quoting him).

Obviously the mass of the tire and wheel assembly hasn't changed, so if the same balancer is giving a different reading then there is an outside influence. Most balancers in good running order are very precise, and also very accurate.

Tire Rack uses Hunter balancing equipment, and our area rep has given a number of demos and instructional seminars to our team. One of his (and my) favorite demonstrations is to balance a tire and wheel assembly, recheck it and get a zero weight needed reading, loosen the mounting spindle and insert a folded over business card behind the wheel and retighten the spindle chuck. He then checks the balance and the machine says more weight is needed.
I think the crux of the issue relates to the illustration of our balancer rep's demo. When you change the orientation of the mass the balancer sees it as an imbalance.[...] The folded business card effectively tilts the assembly and therefore moves the plane of the mass of the wheel, causing the change in reading.

Turing the wheel on the balancer can effectively be the same thing, depending on how the assembly is held on the balancer spindle. There are several common combinations of flat back plate, tapered cone, finger plate front and/or rear to hold the wheel on the spindle. As shown by the business card example it takes very little change in the orientation of the mass to register on the balancer.

I also had a discussion with the engineers at one of the tire companies some years ago. They had done a study on repeatability based on how the tire/wheel assembly is secured to the balancer. Depending on the method used the repeatability in measuring the balance of an assembly went from the high 40's% to somewhere near 85-90%

We've found the best practice is a tapered cone (with the most shallow taper angle available) at the rear and something like the Haweka FixPlate on the front as the best way to secure the assembly precisely, and repeatably. This method is the closest to using the lug hardware to secure the assembly to your vehicle, and was the one that scored around 85% repeatability in the tire manufacturer's study I mentioned below.

The machines you are balancing on may be very well calibrated. But if the assembly isn't attached accurately, the small change will be greater than the sensitivity of the machine and show up as imbalance.

My investigation ends here. Although the topic is not fully exhausted, I hope that the information posted in this thread puts some light into this intriguing subject.