Pros and Cons of the ER collet system

[Michael GilliganParticipant @michaelgilligan61133]

As a slight aside:

In a previous thread, Raymond linked to the ISO standard … from which I quote:

________________________________________

INTERNATIONAL STANDARD ISO 15488:2003(E)

© ISO 2003 — All rights reserved 1

Collets with 8° setting angle for tool shanks — Collets, nuts and fitting dimensions

1 Scope

This International Standard specifies the dimensions, materials and manufacturing requirements, and designation of collets (of form A or of form B) for tools with cylindrical shanks and their corresponding holders and nuts. For non-standardized clamping devices, such as clamping devices specified in drawings, these holders can be agreed upon between customer and supplier.

Form A applies to milling and any other application where a hard collet bore is required, provided that the clamping range of h10 be sufficient.

Form B applies for general purposes where an extended clamping range is required.

__________________________________________

.

With all the interest in ER collets, it is surprising how little interest has been expressed in the difference between the two forms.

MichaelG.

[just stirring the pot]

Edited By Michael Gilligan on 05/09/2016 18:57:36

[Raymond AndersonParticipant @raymondanderson34407]

Ahh, So Mr Guhring and Mr Albrecht are only talking chaff Well in the words of Johnny Nash "I can see clearly now" . and here's me thinking they surely must know what they are talking about. well well, we live and learn .

cheers.

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Michael Horner on 05/09/2016 17:59:23:

We need M.G. to dig out the patent application that would stipulate what the original intention was.

.

I'm guessing that I would be M.G. …

In which case; what you actually need is to learn how to 'dig things out' for yourself.

MichaelG.

[Raymond AndersonParticipant @raymondanderson34407]

Hi Michael, I clean forgot about that Document.. Thanks,. maybe folks never realised there were 2 types.. I certainly didn't until I read the document.. The pot is going "still " now

cheers

[JasonBModerator @jasonb]

May be of interest, seems the hydralic chucks slip first, the SX single taper is better and there BC best, no ER in that test

This one is the same guy testing a ER40 which outperforms the hydralic ones in teh video above

[Raymond AndersonParticipant @raymondanderson34407]

Jason, thats an interesting video. It was tests similar to those that Albrecht did, although I got it in a brochure not on a video. I am going to try find the brochure and also more about the SCHUNK Sino R Milling chuck.. both of which outperform the ER's, although us hobbyists will never use them. or need them. It is interesting.

cheers

[JasonBModerator @jasonb]

The makers of those chucks here

[Raymond AndersonParticipant @raymondanderson34407]

Some good looking gear there. I have never heard of that make.. Another query, why do most manufacturers that put videos up on you tube nearly always show their Vmc's / Hmc's using milling chucks of various types but very little using ER. I wonder how that can be. just wondering.

cheers.

[blowlampParticipant @blowlamp]

I would imagine the ER collets are just about as rigid as most other systems when used at their largest holding capacity. However, with the collet pulled further into its holder, as when gripping smaller tools or workpieces, the collets will have only line contact with the collet holder and the item being held,

The first picture is of a simplified holder/collet/tool setup and the second picture is my attempt to show where the gaps would be between the holder, collet and tool.

How much this affects accuracy or rigidity I don't know.

Martin.

[Ian PParticipant @ianp]

Martin

As you show, when the tool (or work) is smaller than the nominal size of the collet it will only make line (hopefully broad lines) contact with whatever it is gripping. The OD of the collet though should still make full contact as it is a cone within a cone.

Ian

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Ian Phillips on 05/09/2016 20:12:24:

… it will only make line (hopefully broad lines) contact with whatever it is gripping.

.

The breadth of the lines will be determined by the compressibility of the materials

… If both collet and 'whatever' are very stiff materials, then contact will closely approximate a true line.

MichaelG.

[AjohnwParticipant @ajohnw51620]

Exact size on size would be needed to avoid point contact Martin. A condition that is highly unlikely to be achieved in practice.

The plus of ER against the negative aspect is that one collet can hold a small range of sizes. That isn't the case with typical traditional collets. Stuff that is to be fitted into them needs to be pretty close to the correct size.

I suspect that the radial myth comes about because traditional collets are keyed to prevent rotation. They need to be because the taper part is only designed to close the collet onto the work. If used on R8 for milling sizes need to be pretty precise as well.

Comments about Berhard don't fit in at all with my training. They were referred to a bar chucks and for use turning bar where the od isn't going to be touched and accuracy isn't that important. I'd expect up to a couple of thou or so error. Their grip length is also pretty long. For real work I was told to use the ones that came with the lathe I was using. On things like CVA, Langs, DSG's there are rather a lot of them, For the size range they cover there is rather a lot with my Pultra as well.

Crawford specify accuracy some where in their catalogue. My recollection is that it's not a lot different to standard ER that have been made correctly. Like ER there are probably higher spec ones available.

I'm as bad on work holding collets as are some others. I've had my Boxford for some years now and it came with collets that may well have been bought with it. Still in preservative. Lathe dates from 1977, probably 3 owners. I've had the size of Burnerd multi grip that was supplied to go with Myford 7's for a long long time along with a boxed set of the collets. I've never bothered or had the need to mount it. Worse still I recently bought a Jacobs version. There is a reason though, chuck type rather than lever and I tend to worry about not having larger sizes of collet in case I need them. Crazy really as correctly machined soft jaws will hold work pretty precisely. Some think that the scroll in a 3 jaw must be miles out and will wobble all over the place. They don't. Maybe cheap ones do.

I have ER16 and E25. I spent some time wondering about that. I don't really want ER32 or ER40 dangling on the end of my millers spindle. I was sort of tempted to ER32 but there isn't much of a gain over 25 really in cutter size as far as machining goes so I will adapt a 25mm indexable end mill and past that I feel that a face mill make more sense.

Some one passed a comment on here about 5C collets on here. I was thinking of buying a set and a chuck. The comment was don't. I did and have scarcely used any of them so if you do buy as needed.

Really the best all round answer is to think how to make things in one setting as then they aren't needed. Next comes ok can't do that entirely but I can still …………. what ever solution crops up. I find there usually is one.

John

–

[JasonBModerator @jasonb]

Interesting comment about 5C sets vs. individuals. I would say I have used 75% of the ones in my 1/32" increments set in the couple of years since I have had them. Yet it was something like 40% cheaper to buy a set than individuals so by my reconing if you only use 60% of teh ones in the set you are quids in.

 

Set of 29 = £109, 29 individuals = £174

 

J

Edited By JasonB on 05/09/2016 21:06:21

[Anonymous]

Posted by Michael Gilligan on 05/09/2016 20:38:01:

… If both collet and 'whatever' are very stiff materials, then contact will closely approximate a true line.

Always assuming that the clamped item is a true cylinder and not a cone.

Andrew

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Michael Gilligan on 05/09/2016 18:33:04:

Posted by Michael Horner on 05/09/2016 17:59:23:

We need M.G. to dig out the patent application that would stipulate what the original intention was.

.

I'm guessing that I would be M.G. …

In which case; what you actually need is to learn how to 'dig things out' for yourself.

MichaelG.

.

Michael Horner,

What I believe to be the original Patent states the following Claims:

.

As you will see; the claims do not include any reference to the 'intention' [in the sense of what it's intended to clamp] … This Patent concentrates on the benefits of the circumfrential groove.

Now: For your 'homework' … 

See if you can find the Patent, and the ones that build upon it.

MichaelG.

Edited By Michael Gilligan on 05/09/2016 21:34:06

[blowlampParticipant @blowlamp]

Posted by Ian Phillips on 05/09/2016 20:12:24:

Martin

As you show, when the tool (or work) is smaller than the nominal size of the collet it will only make line (hopefully broad lines) contact with whatever it is gripping. The OD of the collet though should still make full contact as it is a cone within a cone.

Ian

I don't think it will Ian. As the collet closes in on itself, the radii of the segments stay the same, but contact at a smaller diameter of the cone.

Martin.

[Ian PParticipant @ianp]

Posted by blowlamp on 05/09/2016 21:36:21:

Posted by Ian Phillips on 05/09/2016 20:12:24:

Martin

As you show, when the tool (or work) is smaller than the nominal size of the collet it will only make line (hopefully broad lines) contact with whatever it is gripping. The OD of the collet though should still make full contact as it is a cone within a cone.

Ian

I don't think it will Ian. As the collet closes in on itself, the radii of the segments stay the same, but contact at a smaller diameter of the cone.

Martin.

Yes I see your point.

The more we look into this whole collet business the more one realises how imperfect they ALL are.

The cone within a cone only make full metal to metal contact in one axial position, for the ID of the collet to reduce the collet has to move axially so immediately compromising the fit.

I suppose we could liken collets to the bumble bee which carries on flying even though the laws of aerodynamics say that it too heavy (or something)

I am happy with all the collets I have and use, 5c, ER16, ER25, MT2, Bosch router, Dremel, (actually Dremel no!)

Ian P

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Andrew Johnston on 05/09/2016 21:21:42:

Posted by Michael Gilligan on 05/09/2016 20:38:01:

… If both collet and 'whatever' are very stiff materials, then contact will closely approximate a true line.

Always assuming that the clamped item is a true cylinder and not a cone.

Andrew

.

Very true, Andrew

… Or indeed one of many other mismatches of shape.

MichaelG.

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Ajohnw on 05/09/2016 20:52:35:

(a) I suspect that the radial myth comes about because traditional collets are keyed to prevent rotation.

(b) Really the best all round answer is to think how to make things in one setting as then they aren't needed.

.

John,

Could you explain (a) please ? … [do we have a different definition of the word radial ?]

I totally agree with (b) … to the extent that it is possible

.

MichaelG.

Edited By Michael Gilligan on 05/09/2016 22:32:25

[AjohnwParticipant @ajohnw51620]

It initially started off as a tangential force but got repeated by so many people that it turned into radial as each person whispered it to another.

Myths always change like that.

John

–

[Michael GilliganParticipant @michaelgilligan61133]

Aha … A 'factoid'

[Using the word as coined by Norman Mailer, rather than as recently abused by some radio DJ whose name I don't recall.]

If that's the case, it might explain some of the differences of opinion that are evident in this thread.

MichaelG.

.

Deep breath everybody … When some people say 'radial' they might mean 'tangential'.

… Is there any hope ?

[John HaineParticipant @johnhaine32865]

The claims of a patent would not state the intention. They only describe the features that are protected. The intention might be described in the specification, but that is only descriptive text.

[Martin KyteParticipant @martinkyte99762]

Totally confused now.

Radial force should be side acting along the radius through the centre.

Axial should be end acting coaxial with the collet.

Torsional should be rotational about the cylindrical axis.

For all the uses I put the things to they work admirably so I cannot see what all the fuss is about.

regards Martin

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Martin Kyte on 06/09/2016 09:09:53:

Radial force should be side acting along the radius through the centre.

Axial should be end acting coaxial with the collet.

Torsional should be rotational about the cylindrical axis.

.

Quite so, Martin

… That's why I raised the point.

The mis-use of these terms might well be what has caused so much difficulty.

MichaelG.

[Michael GilliganParticipant @michaelgilligan61133]

Posted by John Haine on 06/09/2016 08:28:33:

The claims of a patent would not state the intention. They only describe the features that are protected. The intention might be described in the specification, but that is only descriptive text.

.

The whole document is properly referred to as a 'Patent Specification', and, in this specific case, there is no more description of what the collets are intended to hold.

… I quoted the 'Claims' only because they are concise.

MichaelG.

[Author]

Posts