Atlas 618 - Timken Bearings

[JRT]

Are these Timken bearings worn out & how do I know? Here is a brief video demonstrating the "play" in the chuck side bearing. The rear one is similar. Is this normal or do I need to order new bearings and races?

Also, here is the spindle inserted into the head with the bearings in place to show the play as well.

Thanks for any assistance! -John

 

[rabler]

Are these Timken bearings worn out & how do I know? Here is a brief video demonstrating the "play" in the chuck side bearing. The rear one is similar. Is this normal or do I need to order new bearings and races?

Also, here is the spindle inserted into the head with the bearings in place to show the play as well.

Thanks for any assistance! -John

Those bearings are meant to be run paired under preload. So play in the individual races doesn’t mean anything.

Generally you tighten them until the shaft runout is small, around a few tenths. You want the preload to be just enough to minimize the runout. If they are gritty, or run hot when pre-loaded, they’re bad. Others may have better advice.

If it were me I’d price replacing them while it is apart.

 

[pontiac428]

What Randall said on preload. Unless there are flat spots (brinelling) or tracks worn into the races, the bearings are probably good. When setting preload, there needs to be room for thermal expansion, so follow the manual (MOLO) on that one. It typically involves setting a cold preload torque with a spring scale (fish scale) at a prescribed radius. When treated well, those bearings can last for life on a hobbyist's duty cycle.

 

[Ischgl99]

It’s hard to tell from a video, but it didn’t look like there was any wear on the rolling elements. The play you showed of the cage is completely normal for an unloaded bearing, a home shop is unlikely to have the equipment to properly check an unmounted bearing for runout, so do as mentioned above and test runout while properly preloaded in the machine. If the races show wear patterns, brinelling/false brinelling, discoloring from overheating, or spalling (small “potholes” in the surface) then it would be a good idea to replace them.

It’s not a good idea to spin a bearing like you did in the video, that causes skidding of the rolling elements against the races and could cause some damage to the rolling elements and/or races.

 

[JRT]

Thanks guys, appreciate the info. Did not realize the bearings had to be under preload to tighten up but that does make sense.

I have no idea what a Spring Scale is but will look it up. Is the little threaded collar on the gear end of the spindle the only mechanism for tighten the preload on the shaft through the bearings?

 

[rabler]

Thanks guys, appreciate the info. Did not realize the bearings had to be under preload to tighten up but that does make sense.

I have no idea what a Spring Scale is but will look it up. Is the little threaded collar on the gear end of the spindle the only mechanism for tighten the preload on the shaft through the bearings?

Spring scale = any simple scale based on a spring. Like a scale used for weighing fish.
Generally yes the threaded collar is how the bearings are preloaded, which is why the collar usually has fairly fine threads. There needs to be a way to lock that collar into place at the proper preload. Typically a second collar tightened against the first.

 

[JRT]

It’s hard to tell from a video, but it didn’t look like there was any wear on the rolling elements. The play you showed of the cage is completely normal for an unloaded bearing, a home shop is unlikely to have the equipment to properly check an unmounted bearing for runout, so do as mentioned above and test runout while properly preloaded in the machine. If the races show wear patterns, brinelling/false brinelling, discoloring from overheating, or spalling (small “potholes” in the surface) then it would be a good idea to replace them.

It’s not a good idea to spin a bearing like you did in the video, that causes skidding of the rolling elements against the races and could cause some damage to the rolling elements and/or races.

I didn’t realize that about spinning the bearings, I’ll not do that again. The races have no wear at all, seemingly. Good to know that the cages are loose like that until under preload. It’s easy enough to to disassemble and reassemble that I may run it as is for now as I think all is okay and order new races and bearings for later in case they do wear out.

I have a dial caliper that I can use to check the runout as soon as I reassemble everything. If the runout is a few 10ths, I’ll be happy.

 

[JRT]

Spring scale = any simple scale based on a spring. Like a scale used for weighing fish.
Generally yes the threaded collar is how the bearings are preloaded, which is why the collar usually has fairly fine threads. There needs to be a way to lock that collar into place at the proper preload. Typically a second collar tightened against the first.

Oh, interesting about the “spring scale”.

Mine doesn’t have an extra collar but I can get one at my local hardware store. I need some of these #4 lead shots to keep from marring the spindle threads.

 

[rabler]

If the runout is a few 10ths, I’ll be happy

Just to correct myself from earlier: what the bearing preload really controls is how much deflection occurs under applied load, (how much radial and axial play the spindle has). So it isn't the true runout (deviation from round when turning). You put a DTI on the spindle and apply radial pressure to see how much the DTI deflects. Actual runout is a function of several other things, although insufficient preload on the bearings will add to that. Too much preload wears out the bearings, pretty quickly/dramatically if the preload is really excessive. Which is why getting the right preload is important.

 

[markba633csi]

The collar on Atlas usually has a small setscrew to lock the collar in place- and there is a small brass or copper slug under the setscrew to protect the spindle threads. You can sub a small chunk of thick copper wire if the original slug is missing...