Wobble with Collett Chuck

[PenHog]

Hi All,

I'm asking for help. When I put a new Beall collett chuck on my lathe (PSI starter lathe), I can see wobble. I have done the following:

(1) Checked that my eyes are not deceiving me. I placed a thin parting tool at the base of collett chuck, there is no noticeable bouncing, it's smooth. Placing the same tool near the end of the collett chuck produces noticeable up and down movement.

(2) Measured runout at the point closest to the lathe's headstock (on the base of the collett chuck), it is 0.0025 inches.

(3) Measured runout at the point farthest from the lathe's headstock (at the end of the collett chuck, before its grooves), it is 0.005 inches.

(4) Cleaned the 1"x8 tpi threads on the headstock.

(5) Used a leveler to check that the lathe is level across and in depth.

(6) Taken out the screws of the headstock to check they are not bent, and they look fine, and tightened them down. Repositioned the headstock ever so slightly to try to cutdown on the wobble, with no improvement.

(7) Repeated with the older PSI collett chuck I have and found essentially the same runout (it's a little larger), and the wobble persists.

I don't know what to do now. This is the first time I've taken a close look at this issue, since I used to turn in a jaw instead of a collett chuck. But I noticed that there was a noticeable (to me) deviation from true when attaching the cap to the body of a recent pen.

I suspect this has been a problem all along, and I'm only noticing it now since I'm turning thinner pens, and perhaps my standards are getting higher.

Any ideas on what I could try to do to fix this wobble? I would prefer to avoid purchasing a new lathe, but I'll do so if there's no other good option.

Thanks in advance.

 

[howsitwork]

Is it possible to check the bearing preload in the headstock eg tighten down the bearings ?? Assuming taper roller bearings? Check the manual and see if they are adjustable

 

[Curly]

You really can't use the outside of the chuck as it doesn't really matter. The critical part is the collets gripping a known round and straight bar. First is to pull the collet nut off and put an indicator on the inside of the cone that holds the collet and take a reading. Then tip a collet into the nut to properly hold it and then put a piece of steel rod like a steel dowel, close tolerance bolt, length of ground drill rod, that sort of thing. Tighten the collet normally and then take a reading on bar by the collet and a few inches away. The numbers should all be within 0.001" or 0.002", a little more at the end of the rod. If they are still out a bunch you have to determine if the shaft threads are not concentric to the axis of the shaft, the seat the chuck sits against is out, the chuck is out, or the collets are of poor quality. The tolerances of all parts can stack up to so don't get too hung up on perfection as woodworking lathes have looser tolerances that metal lathes.

 

[Curly]

Do you have the same issues when you check the inside of the shaft morse taper concentricity?

 

[PenHog]

Hi Ian,

Thanks for the reply. I'm sorry, but I don't know what you mean by bearings (treat me like a child) My lathe is a Turncrafter Commander from PSI 10", if that helps with whether your solution is possible to pursue.

 

[PenHog]

Hi Peter,

Okay, that makes sense, thanks for the advice! I don't think I have a steel rod or anything like that. I could try a mandrel, or a live center maybe?

 

[Curly]

Live centre and mandrel can be used to verify the morse taper but not the chuck. You can't hold the taper with the collet. You could use the shank, unthreaded portion, of a bolt but it may not be round or straight enough for a good reading. There needs to be enough bar the full length of the collet plus a couple inches out of the chuck.

 

[PenHog]

Hi Peter,

I measured the runout inside the collett cone, and it is essentially 0.002 inches. So, that's encouraging.

I don't have a rod that meets the requirements you specified. I searched around the house, and I simply don't have the tools to make one. I'll try to find one for purchase, and then I'll report back.

Thanks for your help.

 

[monophoto]

Hi Peter,

I measured the runout inside the collett cone, and it is essentially 0.002 inches. So, that's encouraging.

I don't have a rod that meets the requirements you specified. I searched around the house, and I simply don't have the tools to make one. I'll try to find one for purchase, and then I'll report back.

Do you have a set of drill bits? The unfluted portion of a drill bit should work fine for checking runout in a collet chuck. Select a bit that matches the diameter of a collet, and that you can confirm to be straight (roll in on a flat surface - a straight bit will roll unimpeded, while a bent bit will bump across the surface). Just grip the bit leaving about an inch or so of the unfluted portion extending beyond the end of the collet, and then measure runout on that portion. Obviously, there could be a problem measuring runout on the fluted portion of a bit!

 

[PenHog]

Do you have a set of drill bits? The unfluted portion of a drill bit should work fine for checking runout in a collet chuck. Select a bit that matches the diameter of a collet, and that you can confirm to be straight (roll in on a flat surface - a straight bit will roll unimpeded, while a bent bit will bump across the surface). Just grip the bit leaving about an inch or so of the unfluted portion extending beyond the end of the collet, and then measure runout on that portion. Obviously, there could be a problem measuring runout on the fluted portion of a bit!

Hi Louie,

Thanks for the suggestion! I contemplated this last night, but I was worried it might damage the drill. Given your advice, I would try it now, but I've got a better alternative on the way from Rick (I ordered a round metal rod). I'll update when that arrives and I can run the test.

 

[howsitwork]

Max
looking at the manual and parts diagram for your lathe it uses plain ball bearings on the spindle. These are not adjustable but equally they are unlikely to have developed eccentricity from a collet chuck so that's good. Try the round rod test and see what it shows up. On the bright side IF there is wear it will not cost you a great deal to replace the bearings with good quality ones . Over here I'd recommend SK&F but there are loads of good bearing makers over there too.

 

[PenHog]

Max
looking at the manual and parts diagram for your lathe it uses plain ball bearings on the spindle. These are not adjustable but equally they are unlikely to have developed eccentricity from a collet chuck so that's good. Try the round rod test and see what it shows up. On the bright side IF there is wear it will not cost you a great deal to replace the bearings with good quality ones . Over here I'd recommend SK&F but there are loads of good bearing makers over there too.

Hi Ian,

Wow, thanks for looking into that! So, you embarrassed me enough to go look at the manual too (which is what I should have done in the first place), and I think I see the part that you're referring to (part 17 in the schematic, which is a round piece of metal with balls embedded in it). As you suggest, I'll run the rod test first and then we'll see if I need to venture into that part of the lathe.

 

[howsitwork]

Max

don't disturb the bearings unless you have to!! Reseating them properly can be real challenge. Yes it looks like either part 17 or part 15 depending upon the model - both are listed as ball bearings with a reference number. IF you do need to remove the bearings take the old ones along with you and ask for high quality ones cos the cost difference between average and high quality is usually not too great .

 

[PenHog]

Max

don't disturb the bearings unless you have to!! Reseating them properly can be real challenge. Yes it looks like either part 17 or part 15 depending upon the model - both are listed as ball bearings with a reference number. IF you do need to remove the bearings take the old ones along with you and ask for high quality ones cos the cost difference between average and high quality is usually not too great .

Good advice, thanks Ian! I do hope I don't have to go that far and peek at the bearings, but we'll see...

 

[PenHog]

Hi All,

I used the rod that I received from Rick, set it in the collett chuck and measured.

I see about 0.009" runout about 0.5 inch from the start of the rod as it comes out the chuck.

I see a little over 0.0145" runout about 2.5 inches along the rod.

And, yeah, I can see noticeable wobble in the rod when I turn on the lathe at low speed and just eyeball it.

So, that's no good. I suppose I could go further and try to figure out if it's the ball bearings. However, I bought this lathe on sale from PSI with the idea that I would upgrade if pen turning became a steady hobby. And since that seems to be the case, this may be an opportune time to be looking for a new/better wood lathe...

 

[Curly]

Your bearings won't cause that out of round unless they are totally shot. You would hear and feel the grinding and noise if that was the case.

If you are not putting the collet into the chuck correctly, that could do it. They should be tipped and snapped under that half arc ridge just above the short taper that engages the nose of the collet in the nut. Never mind if you already do that.

The back of the thread may not be hard up against the shoulder in the headstock shaft. You may need a washer to bear against.

The threads of the shaft may not be concentric to the shaft. Nothing you can do to fix that.

The lathe shaft may be bent. Also nothing you can do about that.

The chuck thread could be out. I doubt that due to the reputation Beall have for making a good product.

The collet you have, especially if an import could be out but even they are usually okay. Trying another (borrowed from another turner) would rule that out.

There could be a few things off that when added together give you what you have.

 

[PenHog]

Your bearings won't cause that out of round unless they are totally shot. You would hear and feel the grinding and noise if that was the case.

If you are not putting the collet into the chuck correctly, that could do it. They should be tipped and snapped under that half arc ridge just above the short taper that engages the nose of the collet in the nut. Never mind if you already do that.

The back of the thread may not be hard up against the shoulder in the headstock shaft. You may need a washer to bear against.

The threads of the shaft may not be concentric to the shaft. Nothing you can do to fix that.

The lathe shaft may be bent. Also nothing you can do about that.

The chuck thread could be out. I doubt that due to the reputation Beall have for making a good product.

The collet you have, especially if an import could be out but even they are usually okay. Trying another (borrowed from another turner) would rule that out.

There could be a few things off that when added together give you what you have.

Thanks, Louie. I did double check that I had the collett chuck set up properly.

 

[PenHog]

Weird, I'm pretty sure I didn't mean to post just that sentence, sorry if that sounded abrupt. Anyway, I'll take some photos and post to confirm.

I guess I wouldn't mind trying to take a look at the bearings. There is some noise going on when I activate the lathe; I usually don't notice it because I have my dust collection on (and previously, I've had a fan going to keep cool). I'll see if I can figure out how to access those, it looks a bit tricky.

 

[Curly]

It didn't sound abrupt but you surprised me with the new name. I'll get used to it in time.

 

[Penultimate]

I have a collet chuck on my metal lathe. I improved my runout by changing the collet nut. I got a better one at Shars. Also, I try removing your chuck and remount it by turning it by one bolt hole. It's a PITA to do it 3 or four times but it's worth it. If you find a better runout mark the position and always mount the same way. If you have a screw on then disregard. Good luck.

 

[PenHog]

It didn't sound abrupt but you surprised me with the new name. I'll get used to it in time.

Argh, I'm sorry, apparently I'm not quite awake these days. I feel fuzzy headed ever since I had my covid shot.

 

[PenHog]

I have a collet chuck on my metal lathe. I improved my runout by changing the collet nut. I got a better one at Shars. Also, I try removing your chuck and remount it by turning it by one bolt hole. It's a PITA to do it 3 or four times but it's worth it. If you find a better runout mark the position and always mount the same way. If you have a screw on then disregard. Good luck.

Hi Mike, thanks for the advice. Unfortunately, my chuck screws into the 1x8 headstock.

 

[PenHog]

Hi Peter, Ian, Mike, Louie,

I'm sorry for going silent on this; work and kids and sleep were priorities lately. I took some photos yesterday, in part because a new lathe arrived. If you get a chance to look and notice anything wrong with how I'm measuring runout, please let me know.

To review, the Turncrafter was giving me:
0.002" inside the open lip of the chuck,
0.009" about 0.5 inch along the test bar in the collett, and
0.0145 about 2.5 inches along the test bar in the collett

This is roughly the same with both the Beall and PSI collett chucks I have.

The new Rikon lathe is giving me
< 0.002" inside the open lip of the chuck,
0.003" about 1 inch along the test bar in the collett chuck, and
0.0055 about 3 inches along the test bar in the collett chuck

Again, roughly the same with both chucks. I do make sure to have the collett click into the place; I think I'm doing that correctly. Runout doesn't seem to change regardless of how much I tighten the chuck.

The Rikon is obviously much better, but I'm still a little disappointed... perhaps my expectations are too high.

 

[Curly]

Your measurement method is fine. You are expecting a little too much for an inexpensive wood lathe. I would expect a high end lathe, Vicmarc, Oneway, Robust etc would be better but never having measured one I couldn't say with certainty. Even metal lathes have some runout and there are chucks made that can be adjusted to take out those kind of errors but they are not found on hobby sized lathes and they carry a price premium to match.

 

[PenHog]

Thanks, Peter, I appreciate the info, and I'm glad I'm not measuring incorrectly. Given that you mentioned them, and I didn't know about these brands, I looked at the price tags on Oneway and Robust lathes...wow, certainly outside my price range at this point.

Assuming that I don't want to attempt any major modifications to the new lathe, are there ways of reducing the runout? Will bringing up the tailstock to hold the piece from both ends reduce this? I would try to this with the calibration bar I have, but it's not hollowed out, and I'm worried about dulling my live center.

 

[Curly]

There is nothing you can really do to reduce the runout. It is the manufacturing tolerances at work. Keep in mind that a sheet of copy paper is about 0.004". If the thread of the shaft was cut on the loose side of the tolerance when the chuck is tightened it may be slightly to one side. If a shoulder is off the tiniest amount the chuck would kick off to the side when snugged. Both are examples where the tolerance is within specification and for normal use is more than fine. That is why when making things you try to complete as much as possible without moving the part. If you have to turn it around to complete you then use an indicator to dial it back to round with the insertion of shims. Rolling papers or thin brass shim stock forexample. Most things pen related don't need that kind of perfection.

 

[PenHog]

Thanks, Peter!