ER32 Collet Thread Problem

[Nigel Graham 2Participant @nigelgraham2]

My ER32 collet chuck has become extraordinarily tight – in fact unuseable.

This is the closing-nut and its mating thread, not that of the spindle thread.

Obviously I’ve examined and soon discounted the obvious, a swarf crumb in the thread.

Testing without a collet shows the nut binds on the body only a couple of turns or so down. I cannot see any sign of damage but this is quite a fine thread so might be distorted… perhaps?

I don’t know if I’d been a bit too heavy-handed at some point and distorted the thread – is that possible? Perhaps if trying to clamp a work-piece whose diameter was low on the collet’s nominal range. It has however been developing this problem for some while even though I do not use it very often, and then only on the Myford lathe and a Warco dividing-head.

Any ideas why this has happened, or if there is any cure short of buying a new chuck?

…..

It also has a very annoyingly poor match between its own ‘C’-spanner and the nut slots – necessitating trying it in two or three positions before the stamped spanner actually fits the milled nut!

[Robert Atkinson 2Participant @robertatkinson2]

There are a number of possible causes but as the chuck has problems and may have been poor quality as indicated by the spanner issues I’d suggest you cut your losses, bin it and buy a new one rather than waste you time investigating it. further.

Robert.

[BazyleParticipant @bazyle]

It is almost certainly cross threaded – forcing a new thread groove to start a little way round from teh real thread where the initial thread wall is thin. To concentrate your examination run a scriber point (or pin) along the thread from say 4 grooves from the end and focus only on where the point is. You are looking for any breaks in the smooth wall. Repeat for the nut. One may be perfect but the other damaged, not necessarily both.

[Robert Atkinson 2]

On Robert Atkinson 2 Said:

… …cut your losses, bin it and buy a new one… ..

Robert.

Sound advice

 

[Nigel Graham 2]

On 15 September 2025 at 21:33 Nigel Graham 2 Said:

..if there is any cure short of buying a new chuck?

 

Two items are involved: the chuck body and the closing nut. A new nut for the existing body is a possibility.

Borrow a nut from someone and carefully use it to determine the location of your own problem.

I think the thread pitch is 1.5mm. Use a thread file to finesse it. Set up in the lathe and chase the thread, turning the spindle by hand. If your time is worth less than your money, do the same for the internal thread on the nut.

[JasonBModerator @jasonb]

Are you snapping the collets into the nut before putting the nut onto the holder? Failure to do that means the nut does not engage with as many turns of the thread and therfore can overload the first few that it engages with when tightened.

[JasonB]

On JasonB Said:

Are you snapping the collets into the nut before putting the nut onto the holder? Failure to do that means the nut does not engage with as many turns of the thread and therfore can overload the first few that it engages with when tightened.

Been there done that.  Surprising how many people do not know this vital point.

[Howard LewisParticipant @howardlewis46836]

The thread for an ER32 is 40 x 1.5 mm

It will be worth examining the internal and external threads to see which one has been damaged by being cross threaded. Possibly both!

If you have a 1.5 mm pitch thread file, it might be possible to clean up the threads.

If not, you are in the market for a new collet chuck and clamp nut.

At some time you must have cross threaded, or there is a LOT of dirt in the threads, and use of excessive force has done the damage.

Howard

Howard

[howardbParticipant @howardb]

Before you scrap the closing nut, try this.

Set the collet chuck body up in whatever you have, lathe, mill etc.

Select about 60 -80 rpm.

Find a decent triangular needle file – the angle of the section will be at or near to 60 degs – close or true to the thread angle.

Set the chuck body rotating and dress the thread gently with the file at first until you get the hang of it, and then a little harder to clean out the thread. Then use the same file to dress and round over the peaks of the thread to remove any trace of likely damage or swarf adherence.

Try the collet nut on as you go to see if it improves the fit.

To clean out the closing nut female thread is more difficult, if you  have a 60 degree internal screw cutting tool use it to do the same cleaning out operation as before, like a thread chaser, set up true in a chuck, but in the female thread.

[Martin ConnellyParticipant @martinconnelly55370]

Have you only got one collet holder and one nut? If so do as already suggested and get another, then at least you can check the new nut on the old holder and the old nut on the new holder. This will allow you to identify the source of the problem before you throw anything out.

Martin C

[renardiere7Participant @renardiere7]

And if you going to buy a new nut go the extra mile and buy one with a bearing, it makes a huge difference.

[Nigel Graham 2Participant @nigelgraham2]

Thankyou chaps.

I’ve just brought the offending items in for a close look….

Even with a lens I could not see any sign of cross-threading – no distortion or tearing. There did seem a very tiny chip or bruise in the nut, and using an M10 tap as a chaser (by hand) I think I reduced that slightly, but it made no difference.

I tried running a piece of kitchen towel round the threads, pushing it in with a thumb-nail, but that brought out only oil residue.

I am very careful about cleanliness of chuck threads, and there was no swarf in these.

Carefully screwing the two together without a collet, the nut rotates barely a turn and a half before binding, so obviously something is seriously wrong somewhere.

 

Then I remembered I have two Stevenson’s Collet Blocks – barely used because they are right awkward so-and-sos to set up. Not only that, but they are accompanied by a ball-bearing collet-nut. I’d forgotten all about that!

Right:

Original, plain nut runs onto both the SC Blocks perfectly well (without a collet).

Ball-bearing nut, never used, engages the collet-chuck thread but binds within one turn. (Hand only!)

 

So it’s the chuck thread that is damaged. It might be repairable, if I have a suitable triangular file as Howard suggests; or better, thread chaser. I’d be reluctant to try a single-point threading tool, but it might come to that, rotating the lathe very carefully by hand only.

I will try it.

 

It does look as if buying a complete new chuck may be the only option….

.

Jason, Oldiron  –

I am normally careful about fitting the collet properly but testing it on the front room table showed it is more of a controlled wobble into place than any sort of positive clip action. It falls out at the slightest sneeze. So perhaps I’d tightened it one day not properly engaged. Though that should show by the collet face not being flush with the chuck face.

I don’t use the thing very often partly because I have never been convinced it properly grips even the high-quality material I keep it for, so it may be too easy to over-strain even with the nominally correct collet. It appears to tighten further in use though, without necessarily increasing its grip on the work, and releasing it can be an almighty struggle.

 

Option two…..

Don’t bother to replace it!

.

The spanner, I discovered, will fit properly in all twelve possible positions (6 each way-uppedness of the spanner); but with the chuck sitting upright on the computer table, not on a lathe spindle level with my chest so making it very awkward to manipulate the spanner.

.

Why little use for the Stevenson’s Blocks?

They are awkward to use due to their short bodies and A/F sizes much smaller than the collet-nut diameter.

A machine-vice grips them only by the end portion of the jaws: not good even if the collet nut, clear of the jaw-ends, is not fouled by the vice flanges and Tee-slot bolts. Used against T-slot fences, as for much of my milling, they need clamping down on parallels to lift the nut from the table, and their small bodies leave precious little area for clamps.

Really they need their own, dedicated “vices” making….  as if I don’t have already too much to do.

[JasonBModerator @jasonb]

If your collet is just dropping out there is something wrong with the nut though try it with another size collet as you may have mucked up some collets as well, it should stay in if properly clicked into place. I’ll go and fit one later and video how much you should be able to shake it about.

Not had a problem using the Collet blocks, just need to be careful they sit down level if you have a long piece of work sticking out the back. Protruding nut is used by many as a “stop” against the side of a chuck jaw so you can repeatably position it for 4 or six cuts.

[DiogenesParticipant @diogenes]

Not all nuts ‘snap’ the collets in – the Bison ones are a ‘drop-in’ fit – you can easily swap collets in and out with the tools ‘pre-loaded’..

..would suit your pace of work Jason.. 😊

[JasonBModerator @jasonb]

How do you get the collets out if they don’t have the ring to pull them out of the holders long taper?

Though the Bison website shows a lip for the collet to snap into https://rdbarrett.co.uk/product/morse-taper-er-collet-chucks-by-bison/

Nigel. a photo of the inside of your nut may help. Or try snapping a collet into the nut that came with your block and see if that holds the collet.

[Taf_PembsParticipant @taf_pembs]

Hi Nigel,

Dumb question time as I couldn’t see if you said but have you been using the chuck for a while, I take it it’s not new?

Very unlikely but run the chuck (with no nut) in a spindle with an indicator to measure the run out / roundness of it very near the mouth as you say it catches around a thread or so in.

With all the checks you have already done, you have proved it’s the chuck body for what ever reason and there is nothing visible then just replace it.

I replaced mine recently with the Byson ER32 and it is perfect, it costs a little more but not much and it’s more than worth the extra money. The very one that Jason has linked to above

 

As a side line, why not use the ball bearing nut’s? –  they tighten up far easier, require less torque. Since I started using them I only do a short sharp snatch or 2 on the collet spanner with the mill in low gear (no spanner on the chuck flats) and never had anything slip, even swinging largish diameter stuff where as with the std nut you had to use a spanner on the chuck flats to tighten the nut suitably.

[JohnFParticipant @johnf59703]

Nigel said

So it’s the chuck thread that is damaged. It might be repairable, if I have a suitable triangular file as Howard suggests; or better, thread chaser. I’d be reluctant to try a single-point threading tool, but it might come to that, rotating the lathe very carefully by hand only.

I would use a single point tool either by hand or under power with the correct tip,  I did a similar type of job for a chap last year i think.  It was an internet trapezoidal thread [12mm?] from a woodworking machine and the internet thread had galled or ??  I did this by hand turning !

John

[MacolmParticipant @macolm]

For curiosity, I would check across the body thread with a micrometer (if you have one of that size) to see if it is oval, or with a dial gauge inside on the taper. Unfortunately if it is oval, there little chance of recovery to a usable standard.

[FulmenParticipant @fulmen]

I see no real risk in chasing the threads using a single point cutter. As it is now it’s useless, so either you’ll fix it or make sure it’s scrap 😉

[Nicholas FarrParticipant @nicholasfarr14254]

Hi, I agree with Fulmen, that is, if it’s no good as it is, you can’t make it any worse (other words were used), which is something that was pointed out to me when I first started work. Even if you have to turn the first two or three threads off down to the root, I doubt it would make a lot of difference to hold your tooling.

Regards Nick.

[DiogenesParticipant @diogenes]

Jason, you just drop a collet in, kind of give the thing a shake and it slips neatly down the taper and into engagement with the ring – if the collet tooling is on size you never need to remove it at all – so stuff like my favourite spotting drill and edge-finder sit in their own collets and I can just grab them and drop them into the nut with one hand and stuff it straight into the chuck.

[Nigel Graham 2Participant @nigelgraham2]

Tra-La!

I fitted the chuck to the lathe, loosened the secondary dive-belt then very carefully “scraped” the thread using a single-point, 60º insert tool, manually.

Left hand rotating the chuck, right hand holding the tool against the thread, with the index finger inside the taper as a guide and steady. The action was a little like peeling a potato, except that I was pulling the thread against a stationary (ish) cutter.

The damage seemed in the middle half of the length of the thread, not at the start at all. Even with a magnifying-glass, and white paper background, I could not see what or where though, apart from what seem chatter marks in the root.

Little by little the nut, without a collet, would screw a tiny bit further a time, with a lot of testing over invisible high spots.

Eventually the nut would pass the tight section and screw up the last few turns to the chuck face almost as easily as it should.

A cautious touch of fine grinding-paste mixed with ‘Brasso’ eased a few further spots, (with a paper cover on the lathe bed) then I washed both parts thoroughly in white spirit, and lightly oiled them.

 

There is still an irregularly, slightly tight region but I was able to wind the empty nut on by hand, fully.

Testing with a collet picked at random, to grip the shank of a milling-cutter of the right size, succeeded.

 

So I still have no idea what was actually wrong. If I had cross-threaded it as alleged, the damage would be right from the start and fairly obvious at least under magnification.

 

.

Taf-Pembs:

No, the collet set is not “new” in years, but I have not used it very much, and never as a milling-cutter chuck although it has held turned work for milling flats. All the collets are still in their “presentation box” and more than half of them are still in cling-film.

.

Jason:

No I have not “mucked up” any collets!

They’ve only ever held good-quality rods of their appropriate diameters, and not very often at that!

The fault was in the chuck thread, though I was unable to identify exactly what. I’ve a feeling it was always iffy so perhaps trying to tighten it over tight areas of thread distorted the thread slightly. The thread tightness would have been disguised by the normal tightening torque, but it always seemed far harder to undo than to load.

 

 

 

[JasonBModerator @jasonb]

Just trying to get to the possible causes of the thread damage. Your mention of the collets dropping out sounded odd, as generally they should snap in and stay there. If not being snapped in or not seating correctly in the nut that may cause damage to the threads and or collet. Does not matter what they are gripping if they are in wrong then they risk being damaged.

This is how a collet usually snaps in and won’t drop out. Bearing nut first then a standard one. Green mark shows the widest part of the lip, tilt collet towards that and then bring inline as you snap it in. Not the best camera position but you get the idea

 

[Nigel Graham 2]

On Nigel Graham 2 Said:

[…]

 

apart from what seem chatter marks in the root

[…]

Well done, Nigel !

I would say that you have identified the underlying problem, which is poor ‘process control’ by the manufacturer.

Occasional use of Go/NoGo gauges is intended to prevent automatic machines from mindlessly producing defective components … but there is too much greed/laziness in the modern world.

MichaelG.

[Bazyle]

On Bazyle Said:

It is almost certainly cross threaded – forcing a new thread groove to start a little way round from teh real thread where the initial thread wall is thin. To concentrate your examination run a scriber point (or pin) along the thread from say 4 grooves from the end and focus only on where the point is. You are looking for any breaks in the smooth wall. Repeat for the nut. One may be perfect but the other damaged, not necessarily both.

Definitely not cross threaded!

Tony

[Author]

Posts