Spacer Between Chuck and Spindle?

mcload · Feb 7, 2020

Sorry for the lame question. I did a thread search and I think I found the answer,
but not definitive. I do know that I have a couple of chucks that for one reason or another,
do not tighten against (or "meet") the shoulder of the spindle.

Is it correct to assume that all chucks should tighten on the spindle shoulder, and if not,
need an appropriate width spacer ring to fill the gap? (I seem to remember one on a heavy 10 my Dad
had, but that was a long time ago.)

Would aluminum be okay?

Thanks in advance.

PMc

wheels17 · Feb 7, 2020

I would cheat and find an inexpensive bearing where the outer race width is proper and it just fits over the spindle when the bearing bits are removed.. It will be ground flatter and squarer than anything you could make. I have a few chucks with the opposite problem, very low numbers of threads engaged when the chuck is on the spindle.

McMaster also sells ring shims, but I don't know what their tolerances are.

DrHook · Feb 7, 2020

On my bud's Logan I was helping him with, we had the same issue, until we found the tiny burr in the new appearing chuck and removed it...

iwananew10K · Feb 7, 2020

assuming clean and burr free, bore out the chuck back a little so it will fit.
very common issue.

SLK001 · Feb 7, 2020

iwananew10K said:
assuming clean and burr free, bore out the chuck back a little so it will fit.
very common issue.

Exactly.

Take the back mounting plate off the chuck, then thread it on backwards. Bore out the plate to the diameter of the clear space between the threads and the shoulder +0.020". Once completed, you shouldn't have any more problems with the chuck not properly registering against the shoulder.

If the plate goes on too far, make up a washer about 0.25" thick that will fit over the threads and spin the plate on with the washer in place.

Derek Smalls · Feb 7, 2020

SLK001 said:
Bore out the plate to the diameter of the clear space between the threads and the shoulder +0.020".

The unthreaded portion of the spindle is supposed to be a close fit to the c-bore, or relief, you are producing in the backplate. This register is what aligns the chuck radially to the spindle. I would be shooting for about .0005"-.001" clearance between the two diameters.
You can bore this oversize and let only the thread do the aligning, but there is a fair amount of clearance in the thread and you may see this as runout when the chuck is mounted.

mcload · Feb 7, 2020

SLK001 said:
Exactly.

Take the back mounting plate off the chuck, then thread it on backwards. Bore out the plate to the diameter of the clear space between the threads and the shoulder +0.020". Once completed, you shouldn't have any more problems with the chuck not properly registering against the shoulder.

If the plate goes on too far, make up a washer about 0.25" thick that will fit over the threads and spin the plate on with the washer in place.

Okay, I see what you're doing. Reducing the number of threads in the backside of the chuck to make up for the extra space, minus 20 thou.
And a C-bore within 4 thou of spindle diameter. Makes sense.

Thanks; I'll look into that.

iwananew10K · Feb 7, 2020

you wont have to touch the threads in the chuck....the untreaded portion of a SB9 spindle is only about 1/4"....certainly the counterbore in the chuck is deeper than that as is....just open it up a bit.

mcload · Feb 7, 2020

iwananew10K said:
you wont have to touch the threads in the chuck....the untreaded portion of a SB9 spindle is only about 1/4"....certainly the counterbore in the chuck is deeper than that as is....just open it up a bit.

Okay. Let me get the spindle and test in the chucks to see if any come short of shoulder (I haven't put this lathe back together yet).
Thanks for the responses.

PMc

mcload · Feb 7, 2020

Well, first let me say that I apologize to everyone for wasting their time in either reading
or posting to this question. I just checked all of my chucks and they all land on the shoulder
of the spindle. I must have been thinking of another machine. I should have checked them first
before posting. I will delete this post tonight.
Sorry again!

PMc

View attachment 277989

SLK001 · Feb 7, 2020

Derek Smalls said:
The unthreaded portion of the spindle is supposed to be a close fit to the c-bore, or relief, you are producing in the backplate. This register is what aligns the chuck radially to the spindle. I would be shooting for about .0005"-.001" clearance between the two diameters.
You can bore this oversize and let only the thread do the aligning, but there is a fair amount of clearance in the thread and you may see this as runout when the chuck is mounted.

Not true. The register is the shoulder of the spindle. The unthreaded portion is just that - an unthreaded portion. South Bend lathes generally have 0.020" clearance between the unthreaded part and the bore in the back of the plate.

thermite · Feb 7, 2020

Derek Smalls said:
The unthreaded portion of the spindle is supposed to be a close fit to the c-bore, or relief, you are producing in the backplate. This register is what aligns the chuck radially to the spindle. I would be shooting for about .0005"-.001" clearance between the two diameters.
You can bore this oversize and let only the thread do the aligning, but there is a fair amount of clearance in the thread and you may see this as runout when the chuck is mounted.

+1 That fit is VERY important to the "working goodness" - most especially once under load - of any threaded-on goods, be it chuck, faceplate - wotever.

SLK001 · Feb 8, 2020

thermite said:
+1 That fit is VERY important to the "working goodness" - most especially once under load - of any threaded-on goods, be it chuck, faceplate - wotever.

Sorry, but even THIS is untrue. It the two surfaces were tapered, then a taper to taper intersection will act as a proper register. Two cylinders will only register if the clearances are such that they would require a press-fit - not something you would want on a removable object such as a chuck.

If you really think your statement is true, measure the OD of your spindle clear spot and the ID of your chuck counterbore. When you find that there is 0.010-0.020" clearance, please explain how two cylinders that don't touch (thus don't register) can be important to the "working goodness" of a chuck.

You can bore this oversize and let only the thread do the aligning, but there is a fair amount of clearance in the thread and you may see this as runout when the chuck is mounted.

The shoulder AND the threads are what do the aligning/registering. Yes, there is variablility in the threads between spindle to spindle. This is why EVERY chuck has its mounting plate machined DIRECTLY on the lathe it is meant to be mounted on.

mcload · Feb 8, 2020

Hmmm...interesting discussion.
So the shoulder provides the chuck to be square; and the spindle threads provide longitudinal thrust against
the shoulder, but also provides concentricity between chuck and spindle?
Perhaps in the end, that's the difference between using a 3-jaw and 4-jaw, the latter can be "zeroed" in to remove
any deflection or misalignment artifact.
I would also agree that a chucks C-bore would have to nearly be a sliding/press fit to align concentricity every time.
The statement about mounting plates being machined on their specific lathes makes perfect sense too. So in practice (and theory),
every 3-jaw universal needs to be matched and trued for a specific spindle on a specific lathe?

SLK001 · Feb 8, 2020

mcload said:
So in practice (and theory), every 3-jaw universal needs to be matched and trued for a specific spindle on a specific lathe?

Exactly. That's why EVERY chuck mounting plate is machined on the VERY LATHE it is meant to be used on. This machining MINIMIZES (counters) any error that can pop up on the spindle manufacturing.

thermite · Feb 8, 2020

SLK001 said:
Sorry, but even THIS is untrue. It the two surfaces were tapered, then a taper to taper intersection will act as a proper register. Two cylinders will only register if the clearances are such that they would require a press-fit - not something you would want on a removable object such as a chuck.

If you really think your statement is true, measure the OD of your spindle clear spot and the ID of your chuck counterbore. When you find that there is 0.010-0.020" clearance, please explain how two cylinders that don't touch (thus don't register) can be important to the "working goodness" of a chuck.

The shoulder AND the threads are what do the aligning/registering. Yes, there is variablility in the threads between spindle to spindle. This is why EVERY chuck has its mounting plate machined DIRECTLY on the lathe it is meant to be mounted on.

You meant to say "I didn't take the time to understand"?

Chucks (faceplates, etc..) ARE meant to have an interference fit and DO.

All of them. It is NOT a true cylinder. ANY of them.

On threaded goods it is done at the "register" shoulder. Not just the flat. A short taper, not a straight cylinder. Otherwise it is useless for serious work. Why are we not surprised?

Tapered THREADS have also existed.

On L series and Hardinge taper it's more obvious. American Standard short taper (A series and D1) or Cazeneuve-proprietary it is less obvious, but still crucial that the final skosh is drawn up tight to an interference fit.

AND NOT a "sliding" fit. Which would be useless, any sort of precision at all.

And, no, not every chuck needs to be fitted ON the machine.

Integral back and OEM-fitted chucks in Hardinge taper, A, L, D1 series, and Cazeneuve are more often shipped "ready to go" than not.

So are plenty of threaded goods. Some of THOSE need to be turned about for fitting to either of a SB or Logan, same thread, but slightly different registers. BTDTGTTS.

Big World. Long history. Multiple millions of spindles. More to that world than just South Bends, after all.

thermite · Feb 8, 2020

SLK001 said:
Exactly. That's why EVERY chuck mounting plate is machined on the VERY LATHE it is meant to be used on. This machining MINIMIZES (counters) any error that can pop up on the spindle manufacturing.

No they are not. Nor do they need to be. Plenty of such goods as HAVE NO separate plate.

Stop blowing smoke and go do some research, wilyah?

"Errors in spindle manufacturing"?

That needs a solution of its own. Sounds like Harbour Freight?

South Bend spindles - "Iron bearing" ones even from early days - were always rather GOOD, even for light, cheap lathes.

And then there were Hardinge, Monarch, Hendey, ATW, L&S...Axelson, Sidney, Nebel, Rivett, LeBlond Heavy Duty with 50, 35, even 25 millionths TIR, and on medium and large lathes, not "WW" collet-drivers.

jim rozen · Feb 8, 2020

thermite said:
South Bend spindles - "Iron bearing" ones even from early days - were always rather GOOD, even for light, cheap lathes.

Not just southbend, but any lathe with a threaded spindle.

Current wisdom is threaded spindles have two and only two alignment features. One is the plane formed by the flat on the end of the
chuck, and the flat on the spindle. The other is the cylinder formed by the spindle register diameter OD and the chuck's counterbore
ID. The first pair come together mostly naturally unless there are chips in there. The second alignment feature is best when it has
a clearance of around 0.001 diameteral clearance. Ie fucking close.

The threads are only a device to be sure the first alignment features come into actual solid contact. Anyone who decides to rely
on the spindle/chuck threads as an alignment feature will be doomed to frustration and disapointment.

Sure if you overbore the register diameter on the chuck counterbore, it will fit any lathe at all. But the chuck won't repeat on any of them.

A good chuck properly mounted can have the chuck, with workpiece installed, removed from the spindle and then put back, and have the
part still running true.

promacjoe · Feb 8, 2020

Although South Bend was very meticulous about their spindles, The area we call the register was not as exact as you might think. There were slight variations in this area.

Several years ago, a member had problems mounting a Chuck. He said the threads were too tight. Since I had a tap specifically for this, I offered to help. He sent me the Chuck, And after inspection, I tried it on my machine. And yes the threads were too tight. I clean the threads, using the tap, by hand only. Basically cleaning the dirt or chips out of it. Now it threaded on my machine perfectly. When he got it back, it would now threat on his machine but it stopped at the counterbore. After taking a very small cut on the counterbore, he said it fit perfectly.

This tells me that South Bend paid more attention to the threads then they did to the counterbore area. The threads are basically a taper, Which will center the Chuck. If there is any misalignment In these two surfaces, you will have problems. Only one of these elements should be used to center the chuck. Not both. You can only eliminate one, The counterbore. you cannot eliminate the threads. If the threads are made in a precise manner, they will be concentric to the spindle.

Stay safe and have fun. Joe.

Spacer Between Chuck and Spindle?

Hot Rolled

Stainless

Hot Rolled

Diamond

Stainless

Stainless

Hot Rolled

Diamond

Hot Rolled

Hot Rolled

Stainless

Diamond

Stainless

Hot Rolled

Stainless

Diamond

Diamond

Diamond

Stainless

Similar threads