[QuinnSjoblom]

Getting ready to start putting together my servo driven bt30 spindle setup and balance has been on my mind. The spindle cartridge itself is rated for 14k rpm and i assume its probably already pretty well balanced, but i will be adding other parts to the shaft of the spindle for locking its position, basically a disk with tapers milled into it at 180 degrees apart. I can hold very tight tolerances for the turning and milling so it will be very close to balanced to begin with, but im sure it would benefit from being balanced, especially if running up to 14k. I will also want to balance my servo pulley which is 5mgt 80t, pretty big hunk of steel. Ive already mounted that up and spun it to max rpm, no noticeable vibrations that i can tell, but im sure any slight amount will affect surface finish when milling.

So the question is, how can i balance these assemblies without buying very expensive equipment? All rotating assemblies will have 8 tapped holes around outer edge to use various set screws as balance. Back when i built my mill turn spindle, i was able to successfully balance that with a vibration app on my phone with the phone resting on the headstock. process is pretty simple, put a set screw in each of the 8 holes one at a time and measure vibration. See which hole gives best improvement then increase or decrease set screw size until vibration is at minimum. The vibration app worked, but the readings were not always clear and consistent when getting down to finer adjustments. The end result was i think good enough, but that spindle only turns 3k rpm. For 14k rpm i dont think the phone app is gonna get it as balanced as i would like. Ive seen a few vibration meters for a couple hundred bucks. If i knew confidently that i could get precise vibration readings for that much i would buy one, but for all i know the accelerometers are no better than whats in a smart phone.

Another method ive seen used before is with a laser. The example i saw was for balancing props on a multirotor. A mirror is mounted to the vibrating object (the arm of the multirotor in the example i saw) and a laser is mounted to a stationary object with it pointing at the mirror. The laser is reflected off the mirror onto another stationary surface. As the mirror vibrates, the reflected beam widens proportionally to the amount of mirror displacement. Im thinking it should be possible to get pretty fine resolution by reflecting the laser a very long distance, 50 feet or so along the length of my shop. Could even use a second mirror to bring the beam back over to where im working and double the resolution. A guage can be drawn on whatever the laser is shining at with various tick marks to observe and record the relative vibration amount for each trial. Ive only seen one example of this so i dont know how well it will work compared to other vibration meters.

Does anyone have any experience, known methods, or advice for precise balancing? i realize another option is assembling it and sending it off to get balanced, but i would really like to have the ability to do this myself especially since the locking feature of my spindle is experimental at this point and may change later.

[Jim Dawson]

That laser idea seems like it would be very accurate, and very inexpensive to do.