The magnetic balancer is defective or has a defective component. For some as-yet-undetermined reason, the empty balancer shaft finds the same location as top each time it's spun up. When I balance a prop on the magnetic one, it looks good, but if the shaft turns during the process, the balance goes awry. I'm certain that's not how it's supposed to be.

 I've been using the Dubro roller-balancer in the interim and the results are quite good. I'm up to 14 of 16 props balanced and expect to complete the other two tonight.

I'd like to spin the motors prior to assembly, because my assembly plans are days in the future and I can use the "slack time" to do the tiny focus type work of balancing stuff.

I've also found an inexpensive ESC that should work with my radio and receiver, and I'll swing by the local hobby shop to see if they have anything similar and equally effective.

An interesting idea. From the placement of the magnets, one can determine that one flat face is north and one flat face is south. Per the instructions, one north face points to the balancing shaft, while the other magnet has the south face pointing to the balancing shaft. I had considered that the polarity was 90 degrees to the face, but soon realized that was not accurate. I've rotated the magnets in the holders in 45 degree increments, and the same spot on the balancing shaft points upward each time.

The shaft I'm using is the one that came with the balancer and it is too large for the diameter of the motor bores. I do want to test the motor bells in the magnetic balancer, since the motor bells contain magnets and I would expect some attraction and cogging. I'll find some rod locally to fit the motor.

Yes, I did read that thread. Your suggestion to get a shaft from the local hobby shop was appreciated, although I think the shaft on my motors is smaller than 5mm. I'm having trouble getting the bell off the first one, but I've seen that it's an aluminum bell on a steel shaft, so I'm trying to freeze-release the bell from the shaft. That method has worked for me in other projects. The motor is in a sealed bag in the freezer at this moment, so I don't know if it will release more easily once it's removed. I don't have a gear puller or bearing puller as small as this requires, but I'll be adding that to my hobby shop list of things to check and/or purchase.

I've checked the specs on the motors I have and they have a max figure of 20 ampere draw, which is the primary specification that's needed for an ESC, yes? I'm still aiming for a single motor operation test rig, for vibration/balancing and other purposes.

I can't get the bell free of the motor shaft and I don't want to use excessive force and damage something. Both setscrews have been removed and the entire motor dropped to nearly 4 degrees F (-15 degrees C) but still the bell is well attached.

I'll probably be using the trial and error method of running the motor, attaching balance tape to an arbitrary point and checking the results.

There should be no vibration in the Z axis, and if there is, little can be done about it. Any vibration in line with the boom/outrigger will also show up at 90 degrees to the boom, so as the X axis vibration is reduced (or increased), the Y axis should follow.

The motors are the MK3638 units that came with the okto kit.

Thanks to your message, I've discovered that there's a circlip holding the shaft to the bearing on the non-rotating end. Removing that clip, carefully, enables me to remove the bell, with the shaft still attached to the bell. Perhaps now that it's removed in this manner, I can again release the set screws and push the shaft out of the bell. I no longer have to worry about damaging the bearings or other parts of the motor.

I have a drill press and will probably try the method you described to push out the shaft. It seems a good bit more controlled. I've measured the shaft and it comes out to 0.157 inches, which google says is 3.9878 mm or one might think, 4 mm in diameter. Perhaps I can find a 3mm or 3.5mm drill bit and perform the task you suggest. I'm hopeful that I'll be able to re-insert the shaft, so as to not have to purchase a complete motor replacement. Since the shaft has set screws holding it in, one would think it is designed for replacement. I've also found a parts replacement kit here on the shop site:
 http://www.mikrokopter.us/shop/sp-set-mk3638-spare-parts-mk3638

I discovered that I have a 0.157 drill bit, but I used a 0.151 bit to push out the shaft, quite easily. I also learned, after the fact, to enclose the motor in a plastic bag before placing it on the drill press! Duct tape, patience and a magnifying glass enabled me to remove the metal particles that the magnets in the bell attracted.

The balancing shaft for the Top Flight magnetic balancer fits the bell and centers well with the cones. The shaft just barely fits on the Dubro balancer, but I can turn the plates around and get another five mm on each end, but I don't think I need to do this. I may also be able to use the Dubro shaft, since the cones are metal, and a smaller diameter which will fit more securely in the center bore of the bell.

I think the most difficult part of this project is to remove the e-clip or c-clip from the shaft. I used two tiny screwdrivers and a bit of modeling clay to capture the clip but it's time consuming and prone to error or to damage the clip. I've found a specialty tool:
http://www3.towerhobbies.com/cgi-bin/wti0001p?&I=LXVWC5&P=FR
and hope to find that my local hobby shop has one in stock. Faster to pick it up locally.

This is great news and back to the main focus, I have picked up a low power ESC with which to run the motor for test purposes.

I have yet to build the iPod touch bracket, so that's tomorrow's project.