Lateral pillow block tolerances for a lathe spindle? 1

[BurningKrome]

I'm very much an amateur in this subject.

I've recently purchased a lathe spindle rod threaded for use with a K01-50 lathe chuck. Dimensions 153mm long, 14.94 mm diameter. The rod is not tapered. Link below. I'll be mounting it in in two SKF SY15FM pillow block bearings.

To check the fit, I inserted the rod through both bearings - without the pillow blocks mounted to a surface, and simply hand-turned the rod. If I held one of the pillow blocks solidly stationary, the other block would noticeably "wiggle"- as if the rod was slightly out-of-straight. I have contacted the rod manufacturer to inquire about the tolerances.

However, my questions here are:
- Is the rod actually out of straight, or are the bearings mounted slightly askew in one or both of the blocks?

- If it's the bearing, will they "straighten themselves out", after things are solidly mounted and the spindle is turned with force?

- If it's the rod, is this normal for a "precision" lathe rod? (What is the acceptable tolerance?)

- Pillow blocks (as I understand it) are designed to absorb some amount of lateral distortion by twisting in the block, to remove the distortion pressure off the internal bearings themselves. How much can they/should they reasonably "absorb" this way ?

- How might I measure the "straightness" of the rod in a meaningful way?

Thanks!

NOTE: The pillow blocks are NOT the pillow blocks in the link's picture. I purchased only the rod. The pillow blocks were purchased separately.

Link

 

[Compositepro]

You must clamp the pillow blocks down so that the spherical bushings can slide into the correct alignment. The sliding has a lot of friction.

 

[Tmoose]

What is the shaft retention method of the pillow blocks you bought? Are they cylindrical bore, or do they use tapered adapters? I think they are 15 mm bore, so cylindrical bores are pretty much a certainty.
I anticipate you will have .002" or more shaft runout when things are tightened and secured.
And the bearings will have .001" or so internal clearance, so the "precision" of this device will be limited.

.
I'd be checking a bunch of things with a dial indicator before, during, and after assembly.
Shaft "straightness" would be just one.

 

[BurningKrome]

@Tmoose,

Thanks for the post. The shaft is 14.97 mm cylindrical (not tapered.) The bearing bore is cylindrical.

The shaft is held in place by collets, which have an off-center oblong opening that matches an off-center, oblong protrusion on the bearing. I assemble the unit by inserting the shaft (quite easy, because the bearings are 15mm - leaving about half a mm gap), mating the collets to the front of the each bearing, turning the collet clockwise until finger-tight, and tightening the collet's grub screw. I'm not sure if there's any secret to ideally mounting those collets.

If I insert the shaft, and don't tighten the collets, I can turn it (inside the the bearing's borehole but not actually turning the bearing) and the shaft seems to be straight (nothing "wiggles".) So, I assume the shaft is actually straight, and (as @Compositepro mentioned) it may just be the bearings needing to settle in.

When I do the permanent assembly, I'll be using some lock-tight semi-permanent sealant to close the gap between the bearing bore hole and the shaft. I'm also not sure if there's an ideally recommended method for doing that.

 

[BurningKrome]

Hello @Compositepro,

I wanted to follow up on your post about allowing the bushings to slide into place. How long does this normally take? I'm assuming the sliding into place happens while the spindle is turning?

I have clamped down the pillow blocks, and ran them at about 1/4 normal use speed for a short time (perhaps 10-15 minutes.) Heat was indeed developed, but the bearing alignment did not seem to change (if I unclamped one pillow block, and manually turned the spindle, the unclamped pillow block still "waggled.")

An additional related question. When clamping down the blocks, the spindle seemed to bind a small amount. I.e. ...
[ul]
[li]When both pillows are are unclamped from the mounting surface (with the collets also unclamped) the spindle moves very freely through the bore holes (literally no resistance.)[/li]
[li]When one pillow block is clamped down and the other is unclamped (and no collets), the spindle moves freely, but i can feel some drag/resistance.[/li]
[li]And when both pillow blocks are clamped down (without collet) the spindle can be moved in and out of the bore hole, but with some effort.[/li]
[li][/li]
[/ul]

Q: Is the latter too much binding?

Q: Will this bind alleviate after the bushings have slid into place?

Q: Should I attempt to remount the pillow blocks? The technique I used was to mount one pillow block completely, and then adjust the position/shim of the second pillow block based on the free movement of the spindle through the borehole. The best I've achieved is the last bullet above. Do I need to try again?

Thanks again for any help!

 

[Compositepro]

Pillow blocks are a robust way to mount ball bearings to almost any structure. The spherical bushings allow the ball bearing axis to self align to the shaft axis even though the base structure surface is not precise or aligned.
The self alignment occurs within one revolution of the shaft. The pillow block cannot compensate for a bent shaft, or one that is off-center with its bearing due a loose fit between shaft and bearing. These issues will cause the spherical bearing to constantly shift during each a every revolution of the shaft, which is not what it is for. You need a new shaft that is straight and of the correct diameter.