Thread: ER Collet Chuck Spindle Nose Blanks.

In principle yes, but......

Some further details would be required before I would be able to give a definite answer.

Such as,Hercus only or other threads as well, what material, OD, price etc.

Would these be backplate mounted or direct on the spindle, ie would I have to cut the M39x4mm internal rear thread for my 9x20.

I have an ER25 chuck that screws direct on to my M30x3.5mm wood lathe spindle thread and a similar chuck for the M39x4mm metal lathe spindle thread would be nice.

Material would be 4140,OD Aprox 62mm,length 70mm.
I would have to make a Dummy spindle for the 39x4.
They would all be Spindle mounted as mentioned above.
Price not sure yet,I would have to see whats available and work a price from that.

Can't say I would buy as described ...

The fine thread up to an edge for the nut would be the harder bit for me (limited successful fine thread cutting experience - this would be a metric one on my imperial machine so no undoing the clasp nuts)

I'd possibly consider a back plate mounted collet chuck that was finished. (happy to supply nut) The chuck would have a step like a chuck has on the back. -- (then again that is what the UK sites sell?)

I'd supply (or buy) a partially made back plate, face on my lathe and fit to the step on the pre-made collet chuck. -- 3 m6/1 quarter inch threaded holes on the periphery of the collet chuck would aid the process of making the backplate and the chuck stay together.

Er-32 is where I am at the moment.

Would that give an acceptable result from a run-out perspective?

There would be No shoulder to cut the Metric thread to, you would have an undercut.

To mount a Finished Collet chuck to a Backplate may be a little hard to keep concentricity.

There is the chance that these Chucks could be supplied with the Thread for the nut allready machined and then all you would need to do is cut the Taper to suit your collet.

Thanks for adding the pictures -- if they were there before something went wrong at my end. My visualisation did not match reality

more room than I thought for run-out of the thread cutting exercise.

Hmm......

I am with havinago on this one .

I would be more likely to buy if the external metric thread was there ready to go .

Its one thing making a few stuffups on your on project , in BMS for example. But 4010 is another thing altogether , if you stuffup the fine threading , then you have done your money . Seems to me that if somebody is competent enough to do the internal thread, then , the external one will be a breeze for them. Having said that. Its a great idea and good on you for asking. i hope there is enough interest to go ahead . And the Price is right !

Mike

The most difficult to make part is the nut, but standard ER nuts can be bought ready made on eBay for peanuts.

The taper is best made by setting the compound to the required angle of 8 degrees as good as possible. Then put a collet into the nut and insert some round bar of the collet's maximum clamping diameter, smear a little engineer's blue paste (as used for scraping) onto the taper, insert and lightly clamp the collet into the collet chuck. This will show in which direction to lightly tweak the compound angle. Take a very light cut, and repeat until the contact is perfect.

Even the cheapest Chinese eBay collets have a runout of under 0.005mm. But to take advantage of this accuracy, it is crucial that the collet chuck mounts perfectly concentrically to the spindle nose, and that it repeats accurately every time it is mounted. This is not possible by the spindle threads alone. It can only be achieved by a perfect "sucking" fit to the short cylindrical register behind the threads on the spindle nose. Sch a registration fit must be made on purpose for the particular spindle nose the collet chuck will be mounted to. Any slop whatever small in this registration will invariably lead to a runout error that will be different in magnitude and direction everytime the collet chuck is mounted.

That said, it depends what a collet chuck is used for. A collet chuck is not always sought after only for accuracy, it will also allow to work safely very close to the chuck, without fear of emery cloth or freehand tools or fingers (or shirt sleeves or long hairs or ties) being caught in chuck jaws. Another advantage over conventional chucks is the very low inertia a collet chuck has - it accellerates and stops faster, reducing the time taken for taking control measurement or checking fits (though this advantage is somewhat lost on Hercus type lathes because of the very heavy cast iron step pulley around the spindle that adds greatly to inertia). And yet another advantage of collets is that it causes less out of balace vibrations at very high speeds as can be achieved when fitting a VFD variable speed motor to the lathe.

Chris

Thanks chris

The external thread can be done .
The reason I thought not to do it was most people could use the experience in thread cutting,they also have the oppertunity to finish the OD to what ever shape/appearence they like.
I am still not convinced about the tight fit/snug fit on the spindle,maybe some of the members on this forum who have purchased screw on collet chucks from some of the tooling suppliers may be able to comment on theres.

Why 4140?

One reason is the surface finish you can get.
It dosent attract rusting as quick.
It is harder wearing than mild.
They could be made from anything.
Most but not all tooling I make is usually something more than mild.
All arbors I make and other items associated with rotated cutting are from Hg Ten.

Originally Posted by pipeclay

I am still not convinced about the tight fit/snug fit on the spindle,maybe some of the members on this forum who have purchased screw on collet chucks from some of the tooling suppliers may be able to comment on theres.

Bought collet chucks for threaded spindle noses have necessarily always a slightly oversize register. This is so, because every lathe maker has a slightly different register length and diameter. If the register in the collet cuck was too small, the chuck cannot mount and would be returned and the business is lost. Hence they make the chuck register oversize so it will fit on any lathe with the same thread diameter and pitch.

With a 4-jaw chuck the registration does not need to be a perfect fit.

A good quality 3-jaw chuck ($400 class for a 5" chuck) has a repetition accuracy of 2 to 3 hundrest of a millimeter when new. If the spindle nose has a ground thread and the backplate thread is very well finished too, it is possible to get away without a perfect registration.

But from a collet chuck one would expect better than 1 hundretst of a millimeter repetition. This can in my opinion only be achieved with an interference fit of the registration.

There are set-tru collet chucks that are adjusted for optimal runout after each mounting to the spindle and thus can overcome the registration problem of threaded spindle noses.

This is maybe the reason that lathes with threaded spindle noses had traditionally collets that fit directly into the spindle taper, either with a hollow drawbar or with a front mounted lever closer. These are more accurate because they do not rely on the threads and register at all.
Unfortunately the Hercus 260 used C4 size collets that are near impossible to find at a reasonable price.

More modern lathes have precision ground tapered registers on spindle and chuck (like the popular camlock) and do not have this problem. I think collet chucks really only became popular with these modern type of spindle mounts.

Then again, as I said earlier, accuracy is not always the main reason to use a collet chuck on Hercus. But if it is.... then the registration better be a perfect fit.

Chris

Can you if you dont mind measure the fit on your chucks,face plate etc to give an example of clearance please.
Not to disagree with what your saying as I am no expert in the correct registration methods employed with screw on spindle fittings,I am having trouble trying to understand how you would fit a collet chuck to a spindle if this Fit on the Regerstration was Interference.