- Joined
- Jan 27, 2005
- Messages
- 13,039
- Likes
- 5,481
- Location
- Dalworthington Gardens, TX
- Website
- pbase.com
Bill, if you are checking radial runout at the threads of the spindle then that is not the correct procedure. Threads can't and shouldn't be depended on as part of the alignment system -- they are primarily fasteners. Your spindle should have one or two register surfaces that provide precision alignment. Almost all woodturning lathes have a flat register surface near the base of the threads which is perpendicular to the spindle axis. A few lathes also have a cylindrical register surface between the flat surface and the threaded part of the spindle. My Delta 1440 has this configuration and I love it (one of the few things that I love about this lathe) because it provides both angular and radial precision alignment. However, it also requires the chuck insert to be counterbored to take advantage of this feature. I sent an engineering drawing to Oneway to modify a couple inserts and now I am as happy as a clam. (Note to those who stick a piece of milk carton plastic over the register -- the plastic is not a precision part so don't whine if the chuck wobbles.)
One other thing is to measure the runout with the spindle unloaded. This means removing the drive belt. All bearing assemblies deflect under load.
If money were no object, you could consider going to high precision bearings built to specification from Fafnir or Torrington, but such bearings are unbelievably expensive. Woodturning lathe manufacturers have to decide where to draw the line between cost and performance benefit and that means using off the shelf parts. It sounds like OW has recently answered you question about the change in noise level and given you their rationale for their design change. Having worked in industry, I understand OW's point about time and cost of adjusting preload. A huge part of manufacturing costs go into things that need precision alignment especially if it is time consuming. Companies don't make design changes on a whim since the recovery of NRE costs is slow. It appears that the previous design may have been overly problematic to them. and going to their current design looked like the better choice of two evils.
One other thing is to measure the runout with the spindle unloaded. This means removing the drive belt. All bearing assemblies deflect under load.
If money were no object, you could consider going to high precision bearings built to specification from Fafnir or Torrington, but such bearings are unbelievably expensive. Woodturning lathe manufacturers have to decide where to draw the line between cost and performance benefit and that means using off the shelf parts. It sounds like OW has recently answered you question about the change in noise level and given you their rationale for their design change. Having worked in industry, I understand OW's point about time and cost of adjusting preload. A huge part of manufacturing costs go into things that need precision alignment especially if it is time consuming. Companies don't make design changes on a whim since the recovery of NRE costs is slow. It appears that the previous design may have been overly problematic to them. and going to their current design looked like the better choice of two evils.