Has anyone tried the 0um ER Adjusters?

[Sub MandrelParticipant @submandrel]

[Sub MandrelParticipant @submandrel]

The Gloster site has a flier for micro-adjustable ER closing nuts. They must be pretty good – if you follow the 'practical advice' you can 'produce in complete tranquillity'.

LINK

Neil

[John Stevenson 1Participant @johnstevenson1]

You don't even have to buy one.

Just sit there cross legged with your pink cloak on and go Uummm.

OK I'll get me coat…………….

[Ian PParticipant @ianp]

I almost fell asleep watching the 'video'!

Presumably its not a cheap device but they have skimped on the brochure (spelling errors) and the video is appallingly bad.

I dont quite see how it works but at best it can only adjust for zero runout at one point so potentially even a perfectly parallel slot drill might cut a dovetail shaped groove.

I'm not a potential purchaser.

Ian

[BazyleParticipant @bazyle]

Interesting concept. Maybe I'm a bit out of touch but I hadn't seen this before. Bit like a Griptrue in effect but simpler. I guess it would work for a chuck too, especially a camlock.

[Clive FosterParticipant @clivefoster55965]

Ahem! ER collets don't hit the specified run out until they are torqued up properly. Which is a lot. 70 ft lb or so I think for ER-25. Nipping up by hand as shown won't do it. Nor will it hold the tool or whatever properly. That skinny ring and adjuster screws will do sweet FA if the thing is tightened up. As shown all its going to do is push the cap around so that the slack collet is square in the taper before you lean on the spanner. Very theoretically it might encourage things to settle squarer than they would if left to their own devices but in a practical world just pulling up tight does just fine unless there is a fault in collet or holder.

ER collets are, by design, aligned by the rear taper so there is sufficient clearance in the cap threads et al to allow front to follow back. Which it will do just fine provided the component being held is rigid enough not to crush and inserted far enough to fully support the fingers.

Clive

[Sub MandrelParticipant @submandrel]

My worry was that the rear taper, not the front, is the critical one, and that in any case as Ian says will it just make it run true at one point? My guess is that it works, but perhaps not as well as they suggest.

My home made chuck (with commercial nut) runs true to 2 tenths, and even with a home-made collet it was 4 tenths, so I'm not syure micrometre accuracy would give me any practical benefit.

An intreresting concept though,

Neil

[John McNamaraParticipant @johnmcnamara74883]

Hi All

I think I can see how it works. From the video you can see the first part is a screwed ring being attached to a standard chuck body I will call this the backing ring for want of a better name, Then the actual collet closing ring is attached. It appears to be made from two parts. First what appears to be a standard collet closing nut that may have had a shoulder turned on and then a press fitted ring bedded firmly on the shoulder. This ring is drilled and tapped then fitted with radially disposed hex grub screws. These screws bear against the previously attached "backing ring".

On first look you would have to wonder how this actually makes a difference; clearly the back taper not the nose taper is what aligns the tool, it should not make a difference?

I think the answer lies on the inherent flexibility of an ER collet. Let us first consider if the collet was composed of separate wedge segments that you placed around a tool in the holder and drove them in to clamp the tool. You would have to drive them in evenly or the tool would not be cantered in the holder.

ER collets are not separate pieces but with so many saw cuts they are quite flexible. by using this device attached to the nut working against the backing ring (Which would need a slight Clarence fit between the collet closing nut and the holder to allow it to rock slightly) The nut could be canted forming a micro swash plate that forces the wedge segments of the collet in unevenly and the tool slightly out of alignment. thus the device is able to correct small errors. correcting a few tenths or even a thou or two would appear to be quite feasible.

It is also a tool that would be fairly easy to make on a small lathe on the small lathe, ideally starting with a commercial nut.

Cheers

John

Edited By John McNamara on 24/03/2013 13:39:32

[John Stevenson 1Participant @johnstevenson1]

I think John has it but unless you are on R8, INT / BT 30 / 40 with drive doogs you have no way of getting the collet holder back into the machine to guarantee repeatability with the spindle so every time you do a tool change you need to adjust.

Might be Ok for the big boys with thier 40,000 rev spindles at 25 horsepower to get the last nano out of a job but can't see any use in a home shop with, lets be honest hobby type machines and spindles.

True story.

Some while ago I had a job to take the edge down on a piece of 1.6mm stainless to 0.5mm for a width of 3mm on long strips pre cut. So basially just a sliver off.

Set these up on the CNC in rows with a 6mm gap inbetween so I could do two at one pass with a 12mm solid carbide cutter. Started running the job and after a while I noticed the edge had deteriated on the cutter. Swapped cutters and carried on but second cutter did the same.

As I only had two 12mm carbide cutters I had to swap to a HSS cutter which did the rest of the job, over a 1,000 pieces on the one cutter.

I saved a sample piece and sent it and the cutters back to the supplier for a reason why they didn't last.

The answer was I was not cutting deep or agressively enough and I didn't have a rigid enough machine. The CNC only weighs 2 1/2 tonnes and has a 7 HP motor

[Clive FosterParticipant @clivefoster55965]

Johns theory is interesting but improbable give the forces involved. The micro misalignment if the back taper isn't entered dead true probably does occur but doing the cap nut up properly removes any such effect. Probably more likely with cheaper sets with lesser attention to taper and finish tolerances. Many years back I did some exploratory tests with my Vertex branded set after being told how important it was to do things up properly. Hand or light nip tight would sometimes leave a small runout, memory says thou or two, which would always disappear once something approaching book tightness was achieved. The effect didn't seem to be especially repeatable but appeared to be most likely towards the bottom end of the holding range. Assuming my judgement of hand / light nip torque was reasonably reliable its defensible to assume that greater closure uses up a greater portion of the applied torque in collapsing the collet leaving less for alignment and grip so errors were more likely.

Despite the slits an ER collet isn't all that flexible and there are mutual alignment forces from the interaction curved walls on the male and female tapers. Closing onto the tool also helps drive towards accuracy for similar reasons. Might be interesting to do a proper analysis of the forces and deflections involved with special attention to how the errors purported to be fixed by this device could arise and whether or not its actually strong enough to work as advertised.

Clive

[John McNamaraParticipant @johnmcnamara74883]

Hi John

Every time I have to machine stainless steel particularly sheet I feel grumpy… What little tricks will it pay on me this time! The setup you describe had an interrupted cut, the 6mm gap between the two pieces. My guess is little pieces of (after being cut work hardened stainless) floated around in the gap and got caught as the tool rotated into the next cut or so. Carbide does not particularly like interrupted cuts anyway and the hard bits were to much for the hard but brittle edge. HSS being tougher held up.

Hi Clive

A simple test to prove this theory will be to insert a piece of paper under say a third of the circumference of the face taper (the taper that touches the nut) then tighten up on a piece of silver steel or ground stock and checking the runout with and without the paper. I will try it over the weekend.

It would be best to not remove the holder from the spindle to eliminate any error caused by reinsertion.

I chose paper to avoid the risk of damaging the tool.

Cheers

John

Edited By John McNamara on 24/03/2013 22:28:21

[Joseph RamonParticipant @josephramon28170]

I can't find them in their ebay shop, but they sell some esoteric stuff although THIS seems to bear a strange resemblance to a deburring tool!

Joey

[Clive FosterParticipant @clivefoster55965]

Hi John

An ounce of experiment beats a ton of theory so I shall be interested to see your results. If you have time it might be illuminating to verify the validity of my reported experiences that excess run out is primarily a property of insufficient tightening torque and doing things up to book torque brings things into line. It was around 30 years back that I played with things so "Just do 'em up tight." is about the only thing that stuck.

Link here for book torques in case anyone doesn't have them :-

http://www.techniksusa.com/metal/torque_chart.htm

Clive

[Sub MandrelParticipant @submandrel]

77 ftlbs for an ER25 slotted nut or 36 for a 'mini nut'. The former sounds high to me, but using a 4" tommy bar in the holder and an 8" c-spanner on the nut I just tightened it up till it all locked – there is a marked and very sudden change as the collet grips. I can't for the life of me see why a hex nut needs a different torque to a slotted nut – they are both the same inside.

There's no figure for the ball bearing nuts, but I would imagine it needs to be a lot less – I don't have to pull as hard to get the collet to 'lock'.

Neil

[Author]

Posts