When I acquired the grinder in 2013 the spindle was mounted on rubber sealed Korean FAG 6204 bearings. I replaced the bearings in 2019 with rubber sealed Japanese Nacchi 6204s due to excessive heat being generated by the rear pair of FAGs. Hercus supplied the machine new with the spindle mounted on oil lubricated SKF 6204s.

IMG_20240130_170831425.jpg

The grinder had always been capable of achieving a reasonable finish especially when pressed into service as a cylindrical grinder. The spindle had some axial play which I naively thought was addressed by the cluster of preload springs and being axial didn't seem to have too much of a detrimental effect on the surface finish. There was also radial play.

DSC_8509 a (Large).jpg DSC_6546.jpg

A month or so ago I removed the rear cap from the grinding spindle headstock in an attempt to discover the cause of the approximately 0.010" axial and 0.001" radial play in the spindle. To my dismay the outer race of the visible bearing was rotating at around maybe a 100 RPM with the 3600 RPM spindle under power. It turned out to be a result of the fit of the bearings in the bore of the headstock.

IMG_20240130_171216700.jpg 6596766907_97bec24179_o (1).jpg

A combination of seal friction and insufficient belt tension were probably the main culprits. The rear bearing pair should be a reasonably firm sliding fit. I had discussion with one of Statewide Bearing's engineers. Adam Rex and Adam said that H7 tolerance is typical for a sliding fit. For a bearing diameter between 30 to 50mm (6204s are 47od ) the tolerance range is 0 to 25 micron (0.001") oversize. The bearings themselves were within tolerance when measured in comparison to gauge blocks. They were roughly 6 microns/ 0.0002" undersize.



IMG_20240131_101817392.jpg IMG_20240131_103459346.jpg

Measurement of the bores was iffy. I only have telescopic gauges. The bores were both 0.001" plus or minus a few tenths oversize. The bores are not ground so any bearing race rotation results in bore wear and additional clearance. Adam thought that with increased belt tension and unsealed bearings, race rotation would cease. With the bearings located in the worn portion of the bore and with a broomstick acting as a lever to increase belt tension I managed to reduce the rotation to about 1 RPM. The belt would have snapped and the spindle and motor bearings would have let go if I kept it up!!

I'm concerned about timing out so I will continue this later. Part of the solution was repositioning the rear bearing pair in the unworn portion of the bore.

BT
Attached Images




Read the full thread at metalworkforums.com...