Reducing collet nut play by electroplating thread?

[Ian P]

On Ian P Said:

Hard to imagine what caused the crack in the first place […]

That’s strange ^^^

I thought Adam had explained it quite succinctly in the opening sentence.

MichaelG.

[MacolmParticipant @macolm]

As already pointed out, the problem will be that the taper expands the cracked section of the sleeve, but not the intact diametrically opposite section. Assuming the thread is a close running clearance, the resultant eccentricity may be one to two thou.

One plan might be to grind a diametrically opposite slot so that the expansion is symmetrical – or  better a total of three or four slots (including the crack). The tightening ring will now take the force of the taper closing, assuming the deflection does not cause breakage. Unfortunately, the collet is likely to become a bit bell mouthed when tightened, though it should otherwise run more true than it will at present.

Sleeving and a recut thread would appear best, but as with gluing things, if it broke in the first place, an inevitably weaker repair has a low chance of success.

[Michael Gilligan]

On Michael Gilligan Said:

On Ian P Said:

Hard to imagine what caused the crack in the first place […]

That’s strange ^^^

I thought Adam had explained it quite succinctly in the opening sentence.

MichaelG.

I must have read that explanation originally but then dipped in and out of the the thread, much as I do nowadays with this new improved (not!) forum, so forgot that salient info.

Not keen on Malcolm’s idea of putting more weakness into the spindle taper by adding slots.

Overall I think the best chance of success is to shrink on (not ‘glue’ on) a sleeve and make an oversize nut. For the highest precision the spindle bore taper could be reground in-situ, as it were.

Ian P

[HopperParticipant @hopper]

What RPM will the spindle be doing? Is it one of their 20,000 RPM grinding speeders, or one of their 150 per centers attached to a milling spindle?

[Michael GilliganParticipant @michaelgilligan61133]

For anyone still pondering this

[ I have had my say already ]

Here is the collet system in question:

https://www.lyndexnikken.com/images/files/SKColletChuck_CAT2011-TC.pdf

Adam has already confirmed that his ‘Speeder’ takes SK10 … but also that a reduction to SK6 would suit his intended use-case.

MichaelG.

[MacolmParticipant @macolm]

“Not keen on Malcolm’s idea of putting more weakness into the spindle taper by adding slots”.

 

You are right of course, this was just an attempt to indicate “best bodge”. Incidentally though, it is fairly easy to create circa 1mm wide slots in hardened steel using a thin grinder disk in a circular saw, with improvised wooden guides (avoids abrasive on your lathe or mill). Proper rectification will require tricky precision work, with success dependant on available resources and skills.

 

So a my suggestion is to check if a “next size up” ER collet nut has a suitable thread diameter to allow for an intermediate sleeve. ER nuts are readily available and of suitable quality. Modification or an adapter would be necessary to provide the flat collet clamping surface, but otherwise much of the precision spadework would be already done, and what’s left is easier. Now either shrink on a sleeve as suggested, or keep the existing thread and make a sleeve that fits it.

 

The shrink sleeve has the advantage that the split will be tightly closed, and thus directly maintain the hoop strength of the original configuration. However, given the split, there is considerable risk in turning off the thread, though less if it can be done by grinding. Also, any mistake in the interference fit can allow the sleeve to pull off with the tightening force

 

The alternative is to make a sleeve with an internal thread having as close a fit as possible to the original thread to contain the split. Then it would be fitted with locktite. Think of locktite as cast in place Perspex, with micro keying but weak adhesion. Therefore the fit of the thread needs to ensure good metal to metal contact to provide the predominantly path for the high tightening forces. The locktite can only stabilise things, it is not a dependable structural element for the level of forces necessary for collet closing.

 

Now, either way, the spindle with the fitted sleeve can be set up with the taper true, and the sleeve threaded to fit the ER nut. This will need to be modified for flat contact with the collet, or perhaps just a turned adapter could do the trick without having to machine the nut. Also worth investigating would be the versions of the ER nuts with integral torque reducing ball race. If dismantling this by removing the balls proved possible, an alternative custom piece could be turned up. Dismantling would entail drilling out the plug closing the hole used to feed in the balls, then blow them out with compressed air, doing it all inside a deep poly bag to prevent escapes. Then reassemble with the new part, which would ideally be correctly hardened.

 

With some luck, either of these modifications would provide near original concentricity, and near enough original strength.

[MacolmParticipant @macolm]

If you did not follow that, here is a noddy drawing. Yellow = existing spindle and collet, grey = ER ball race nut, red = threaded adapter, green = ball race end plate of ER nut, mauve = flat plate adapter. Plugged ball insertion hole shown at top.

[old martParticipant @oldmart]

I would endorse Malcom’s repair and modification as being best from an engineering point of view. However, there is the added complication of breaking the taper when removing the workpiece or tool and removing the collet. I don’t understand how the Nikken nut attaches to the groove at the front of the collet.

If you have the ability to singlepoint threads, why not make a replica Nikken nut with snug fit threads that keep the crack from opening? Made from en24t, or similar machinable and stronger than mild steel.

[old mart]

On old mart Said:

[…] I don’t understand how the Nikken nut attaches to the groove at the front of the collet.

It doesn’t … the SKs are not that sort of collet.

MichaelG.

.

Ref.__ from the PDF that I linked earlier:

[Michael Gilligan]

On Michael Gilligan Said:

For anyone still pondering this

[ I have had my say already ]

Here is the collet system in question:

https://www.lyndexnikken.com/images/files/SKColletChuck_CAT2011-TC.pdf

Adam has already confirmed that his ‘Speeder’ takes SK10 … but also that a reduction to SK6 would suit his intended use-case.

MichaelG.

^^^

[MacolmParticipant @macolm]

Yes indeed, that needs to be addressed. What I have suggested is a generalised scheme, with the scope to copy in the necessary features from the original.

[Michael Gilligan]

On Michael Gilligan Said:

On old mart Said:

[…] I don’t understand how the Nikken nut attaches to the groove at the front of the collet.

It doesn’t … the SKs are not that sort of collet.

MichaelG.

.

Ref.__ from the PDF that I linked earlier:

Sticking my neck out here, What sort of collets are they then?

The SK’s have a groove that the nut appears to engage with to extract it, similar to ER.

Ian P

[Michael GilliganParticipant @michaelgilligan61133]

They are described, and illustrated, as “Flat Shoulder Pushing”

MichaelG.

.

The first few seconds show assembly:

https://youtu.be/TlZncSCa2mw?feature=shared

… Probably best without the audio !!

[MacolmParticipant @macolm]

Updated configuration – details not claimed!

[Michael GilliganParticipant @michaelgilligan61133]

https://youtu.be/IZ6svVyl3QA?feature=shared

No, I am not on commission … just very impressed 🙂

MichaelG.

[Michael Gilligan]

On Michael Gilligan Said:

On old mart Said:

[…] I don’t understand how the Nikken nut attaches to the groove at the front of the collet.

It doesn’t … the SKs are not that sort of collet.

MichaelG.

.

Here is another illustation that might help:

 

Ref. __

https://samchully.com/_data/kor/manual/BT_SLIM_SK_COLLET_CHUCK.pdf

 

 

[MacolmParticipant @macolm]

Though obviously superior in guaranteed concentricity relative to the common systems, I am sure the costs are many times those of mass produced ER chucks and collets. For a start, each collet only covers a range of 0.5mm.

Thus the (empty) collet is captivated on the nut by fully compressing the large diameter using the extraction tool, which allows the nut to engage. Once fitted with a diameter within its range it cannot escape, allowing the taper to be subsequently released by unscrewing the nut. Demounting is then the reverse procedure.

[Michael GilliganParticipant @michaelgilligan61133]

Nicely put, Malcolm

Hopefully everyone is ‘on the same page’ now.

MichaelG.

[BazParticipant @baz89810]

So basically these SK collets are like ER collets but don’t have the front angle, is that right?

[duncan webster 1Participant @duncanwebster1]

Similar to Myford collects I’d say

[Adam HarrisParticipant @adamharris13683]

[Adam HarrisParticipant @adamharris13683]

The SK collets do not pull out of the collet nut at all easily! There is no eccentric extractor ring like the ER system, just a concentric lip, indeed like the Myford system but much more difficult to pull out than Myford collets as there is no extraction tool albeit one could make one.

[old martParticipant @oldmart]

With the very simple way of holding the collet in the nut which has been explained fully now, I would be even more keen on making a replica nut with threads which are tighter on the cracked part. Singlepointing can easily make any grade of fit.

[Author]

Posts