PDA

View Full Version : Collets and Collets



PeteF
12-26-2010, 06:27 PM
A recent thread had a few points on collets, but rather than going OT there I wanted to start a new thread with the questions.

I use collets quite often, indeed prefer them to a conventional 3 jaw chuck for many reasons. My setup has an MT3 to ER32 adaptor and generally runs with extremely small runout.

On the other thread Jerry made the comment that 5C collets are susceptible to contamination causing runout. I had an ER32 collet yesterday that wasn't running true and I figured it was some contamination. Reading the other thread I began to wonder out of say, MT, 5C and the ER series, which collets are least prone to contamination causing inaccuracy? Also, which of those, again just in the basic design engineering not how they're made, is considered the most accurate? Off the top of my head, I'm guessing an ER collet is least prone to contamination causing runout, while a 3MT collet in a 3MT spindle taper is going to be the most accurate as there are no chucks or adaptors involved?

Finally, while on the topic of adaptors I like the ER32 series but a few things coming up for me are 5C. I've seen adaptors like this but wonder if anyone has used them?

http://cgi.ebay.com.au/ws/eBayISAPI.dll?ViewItem&item=270656271491&ssPageName=STRK:MEWAX:IT#ht_2786wt_907

philbur
12-26-2010, 06:59 PM
Why not just use 5C collets.

Phil:)


Finally, while on the topic of adaptors I like the ER32 series but a few things coming up for me are 5C. I've seen adaptors like this but wonder if anyone has used them?

http://cgi.ebay.com.au/ws/eBayISAPI.dll?ViewItem&item=270656271491&ssPageName=STRK:MEWAX:IT#ht_2786wt_907

.RC.
12-26-2010, 07:04 PM
Why not just use 5C collets.

Phil:)

5C collets are very expensive compared to ER32 or ER40... Freight is the killer because a full set is so heavy.

philbur
12-26-2010, 08:05 PM
Just an example:

http://cgi.ebay.co.uk/Set-30-5C-collets-1-8-1-1-32x1-32-inch-suit-Hardinge-/370467744396?pt=LH_DefaultDomain_3&hash=item5641992a8c

with free shipping.

Phil:)


5C collets are very expensive compared to ER32 or ER40... Freight is the killer because a full set is so heavy.

SGW
12-26-2010, 08:17 PM
Why do you think you need a "full set" of 5C collets? I've never had the slightest need for, say, a 13/64" collet. I'd buy:
1/16, 3/32, 1/8, 5/32, 3/16, 7/32, 1/4, 5/16, 3/8, 7/16, 1/2, 9/16, 5/8, 3/4, 7/8, 1.
If I had an occasion where I needed anything else I'd buy it when I needed it.

.RC.
12-26-2010, 08:35 PM
Just an example:

http://cgi.ebay.co.uk/Set-30-5C-collets-1-8-1-1-32x1-32-inch-suit-Hardinge-/370467744396?pt=LH_DefaultDomain_3&hash=item5641992a8c

with free shipping.

Phil:)
Like I said... expensive, compared to ER collets which have a 1mm holding range...


Why do you think you need a "full set" of 5C collets?
We need two sets here, as we are a metric country but imperial sized material is just as easy to get...


I often have odd sized off cuts I put in the collet chuck..

Pete probably has a set of ER32's already, so why drop a few hundred $ on a metric and imperial 5C set that do nothing better then the ER32 set... I am in the same boat as him... I have a full ER32 set, but I bought a TC grinder that uses 5C collets but did not come with a full set and as such I cannot sharpen all my blunt tools....

Spin Doctor
12-26-2010, 08:49 PM
Yes, 5C's have their faults. But the one big strength they have over ERs is the complete adaptability of the whole collet system. From the Step Collets to the whole gammit of expansion collets not to mention the emergency collets, both internal and external. Add in the sqaure and hex collets not to mention the fixture plates. As I understand it the ER system was designed as a Tool Holding System for rotating tools similiar to other double angle collets along with the collets similiar to the Kwik-Switch System.

oldtiffie
12-26-2010, 09:04 PM
I have 5C spindles on my T&C grinders as well as my Spindexer and square/hex holding blocks. I have a good range of 5C collets from 3>25mm but I limit them to my grinders.

I have several of John Stevenson's 5C>ER-32 adaptors (from ArcEurotrade - UK) so that I can use my ER-32 (metric) collets in the 5C spindles. They are not cheap but they are excellent:

http://i200.photobucket.com/albums/aa294/oldtiffie/Collets/ER32-C5collets1.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/Collets/ER32-C5collets2.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/ER-32_C5/ER32-C5-18.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/ER-32_C5/ER32-C5-19.jpg

PeteF
12-26-2010, 09:34 PM
Yes RC has pretty much nailed the situation, I have ER32 collets and don't necessarily want to go out and buy another set of collets just to do the same job! Tiff, yes my application is pretty much identical to yours, I saw that Spindex of yours with the adaptor, but was thinking of buying the adaptor separately from the ebay source I quoted.

Many of the suggested advantages of 3C collets really don't apply to the ER32 series, such as an emergency collet for example. Generally it's used because an appropriate size collet isn't held; whereas the ER32 can simply cover the range with its 1mm closure in each size.

oldtiffie
12-26-2010, 09:57 PM
Pete.

I hear a lot of the 5C collets that are made for specific cross-sections and some that are machinable, but with a solid tube with two rows of 12 equi-spaced tapped holes (30* spacings) you can use it as a "spider". Each screw needs a lock-nut. But if you only use as many screws as are necessary you only need to loosen one or two screws to remove/insert the job, re-tighten the one or two crews and your alignment on the new part should be very close to the old. It will handle rounds, ovals, squares, rectangles and some irregular sections.

Its an idea that's as old as bu**ery. It needs care - but it works.

I use my 5C>ER-32 adaptors i my C5 bores on my grinders so I can cover everything from 2mm (~0.08") to 20mm (~0.80") and everything in between - "inch" stuff included. That should solve .RC. (aka Ringer)'s problem with sharpening his "inch" cutters. I use both inch and mm cutters and I have no problem. I think that Ringer's latest T&C grinder is the same as one of mine.

I prefer ER collets and they are my collet of choice.

Mcgyver
12-26-2010, 10:02 PM
Why do you think you need a "full set" of 5C collets? I've never had the slightest need for, say, a 13/64" collet. I'd buy:
1/16, 3/32, 1/8, 5/32, 3/16, 7/32, 1/4, 5/16, 3/8, 7/16, 1/2, 9/16, 5/8, 3/4, 7/8, 1.
If I had an occasion where I needed anything else I'd buy it when I needed it.

+1, just get the main sizes, even by 1/8 if 16th are too much.... + mm......but you know you need different kinds of collets for different work...it doesn't just stop at one style. Handy to have the 5c or 2J as well as the flexible collets

AllThumbz
12-26-2010, 11:16 PM
But the one big strength they have over ERs is the complete adaptability of the whole collet system. From the Step Collets to the whole gammit of expansion collets not to mention the emergency collets, both internal and external.


Forgive a stupid newbie question, but what is means by this statement- the complete adaptibility of the whole collet system?

I bought a small set of 5c collets for my South Bend 10L which has a handlev collet closer system, but am clueless about ER collets, collet chucks and other devices for holding round and square pieces accurately.

Your assistance in clearing this up is greatly appreciated.



Nelson

moldmonkey
12-27-2010, 07:46 AM
Forgive a stupid newbie question, but what is means by this statement- the complete adaptibility of the whole collet system

5Cs have so many variations and are used in many accessories:

http://www.shophardinge.com/categories.aspx?catid=34

http://www.shophardinge.com/categories.aspx?catid=6746

For the contamination in ER collets, a neat trick from Metalworking, Sink or Swim is to clean the collets & nut in an ultrasonic cleaner. It's amazing all of the fine crud that will come out of the nooks & crannies.

John Stevenson
12-27-2010, 08:06 AM
Yes, 5C's have their faults. But the one big strength they have over ERs is the complete adaptability of the whole collet system. From the Step Collets to the whole gammit of expansion collets not to mention the emergency collets, both internal and external. Add in the sqaure and hex collets not to mention the fixture plates. As I understand it the ER system was designed as a Tool Holding System for rotating tools similiar to other double angle collets along with the collets similiar to the Kwik-Switch System.

Not wanting to take side but this is a rather sweeping statement and a little one sided in that the number of collets isn't mentioned.

If we are talking general turning collets i.e. just to hold diameters then in 5C for a range equal to the ER32 system you have from 1/16" to 25/32" in 1/64" steps that's 50 collets as against 19 ER32's

Those nineteen ER32's will hold any size between 1/16" and 25/32" with no gaps. The fifty 5C will still have some gaps or not close correctly on certain sizes. The ER's can also be used as work holding AND tool holding so halving the expense even more.

The 5C's will excel at sizes over 25/32" that the 32's can't reach, they will also do internal holding and special shapes.

All this does is prove what I have said for a long while.

There is no one size fits all when it comes to collets.

It's a personal choice based on many factors like does the machine take 5C's as native in the spindle, what collets do you already own, do you need collets for mill and lathe, are they readily available ? 5C's are quite rare in OZ but 10 a penny in the US.

Everyone's answer is going to be different

philbur
12-27-2010, 11:19 AM
Where do you buy ER collets for significantly less than GBP 2.67 each.

Phil:)



Like I said... expensive, compared to ER collets ...

Arthur.Marks
12-27-2010, 11:41 AM
To try and swing this discussion back toward the OP's question... (are you still here? :p) I'm not sure there is a "going rate" for used collets simply because of the variations on people's idea of wear. That said, Hardinge collets really do hold up very, very well. This keeps their used prices slightly higher than other brands. Personally, I still find a new Hardinge collet price to be a bargain for the quality you receive. On average, they are between $26-30 US dollars ordered direct. When you look at the "set" prices, you have to realize a good amount is really in the very small bore collets. Get down to 1/64th inch and the price goes up. The three smallest in a 64th's set eat up a full $155. In a used purchase, you often need to weigh the balance of wear in the set. Likely some will never have been used and others worn visibly. This makes it hard to generalize, but by the time you're talking 1/3rd of retail new---you should make out fine for your purposes.

Myself, I have a set of Hardinge's in the major sizes and a few odd 32nd's. The rest I filled in with no-name imports. I find that when I need the odd 64th or, say, 32nd over 1/2" it is usually to fit something which does not require the greatest of accuracy anyway. Generally, in those instances, I am using the collet out of convenience rather than a 3-jaw. Interestingly, it seems to work out good for me. The odd size I might need is likely for work that prefers a Hi-Accuracy collet. So I've ended up with a very small number of those from jobs that required them. I would have had to get them anyway---even if I had the "full" regular 64th's set. So I wouldn't sweat it.

What I have clearly learned is that no matter what system you use, you will never have a "full" set. There will always be something difficult to hold or needs a special workholding method at some point. So the drive for "fullness" :) is really more of a completist thing than a decision of guaranteed utility. Admittedly, that is my opinion, but I'm sticking to it.

John Stevenson
12-27-2010, 11:42 AM
Where do you buy ER collets for significantly less than GBP 2.67 each.

Phil:)

Won't hold metric sizes and still has big gaps in the set so the 2.67 isn't relevant.:)
What milling machine uses 5C's :D :)

John Stevenson
12-27-2010, 11:44 AM
What I have clearly learned is that no matter what system you use, you will never have a "full" set. There will always be something difficult to hold or needs a special workholding method at some point. So the drive for "fullness" :) is really more of a completist thing than a useful thing. Admittedly, that is my opinion at least.

Could not agree more.

philbur
12-27-2010, 11:49 AM
I think the OP's question is getting lost in the discussion.

He implies that he doesn't have any collets but is looking at buying a 5C to ER32 adapter in order to be able to use ER collets. If you start with a 5C spindle taper why would you convert it to ER. Collets are about concentricity and repeatability. Stacking components potentially reduces concentricity due to stacking of tolerances, especially with the cheap stuff.

The seemless range tha ER collets give has to be balanced against lose of accuracy when used at other than their nominal diameter.

If you are looking at a 5C spindle why would you buy anything but 5C collets.

Phil:)



Finally, while on the topic of adaptors I like the ER32 series but a few things coming up for me are 5C. I've seen adaptors like this but wonder if anyone has used them?

http://cgi.ebay.com.au/ws/eBayISAPI.dll?ViewItem&item=270656271491&ssPageName=STRK:MEWAX:IT#ht_2786wt_907

lazlo
12-27-2010, 11:58 AM
What I have clearly learned is that no matter what system you use, you will never have a "full" set. There will always be something difficult to hold or needs a special workholding method at some point. So the drive for "fullness" is really more of a completist thing than a useful thing.

Could not agree more.

+1.

People here get all worked up that whatever collet set they bought {5C, ER, TG, Morse, ...} is the collet system to have. But there's no One Size Fits All collet system -- they all have their strengths and weaknesses.

I bought an ER-40 system, and wished I had gone with ER-32 -- the damn chuck is the size of a softball, and you have to eat your Wheaties to get the 140 Ft/lbs of closing torque. But it's very accurate (ETM-brand), and if you're doing a lot of tool changes, including drills, very flexible.

But on a lathe, 5C is so nice -- through-hole system, so you can chuck long stock, collet stops, emergency collets, hex and square collets, pie-chuck collets...

Then you have collets that everyone loves to hate: R8 on my mill, Morse on my drill press, Brown & Sharpe on my dividing head. I've probably got a Jarno somewhere in the shop :)

lazlo
12-27-2010, 12:01 PM
What milling machine uses 5C's :D :)

The Hardinge TM/UM "Toolroom Mill" does :p

John Stevenson
12-27-2010, 12:15 PM
The Hardinge TM/UM "Toolroom Mill" does :p

Is that better than a Bridgy ? :confused:

lazlo
12-27-2010, 12:17 PM
The Hardinge TM/UM "Toolroom Mill" does

Is that better than a Bridgy ? :confused:

Don sure thinks so! :p

Arthur.Marks
12-27-2010, 12:25 PM
If we're talking 5C mills here... Thru-bore systems are really beneficial when it comes to reamers. Who needs an 8" straight shank to ream a .5" hole? 5C or ER/5C shank.. problem solved! :D ...and no need for the cut-off grinder to come out.

PeteF
12-27-2010, 04:52 PM
Yes the OP is still here, but none the wiser ;)

As mentioned I DO have ER32 collets and as mentioned use them in preference to a 3 jaw for small work. But my initial questions weren't a discussion on which system I should be using, the questions were actually; which system is considered least susceptible to contamination causing runout errors? And, all things being the same, which system (out of MT, ER, and 5C) is considered the most accurate if being used in a morse taper?

Again, simply repeating, but my gut feeling is that the ER series seem to be fairly tolerant of swarf etc in practice, and I'm assuming MT would be the most accurate in this situation, but what about the accuracy difference between ER32 and 5C?

Pete

lazlo
12-27-2010, 04:59 PM
which system is considered least susceptible to contamination causing runout errors? And, all things being the same, which system (out of MT, ER, and 5C) is considered the most accurate if being used in a morse taper?

Wow, that's a loaded question. I think the quality of the collet and the collet chuck (in the case of ER's) will predominate any inherent accuracy advantage of the various collet systems.
Why would you think Morse is more accurate than 5C or ER?

ER collets have minimum runout at the nominal clamp size. I've posted the RegoFix chart showing the amount of runout as you approach the max clamping range (increases by several tenths). So I presume you're talking about comparing the accuracy of the nominal bore sizes?

As far as swarf tolerance, I've had problems catching swarf in the ER slots. You can also gank the collet nose if you catch a piece of swarf in it before clamping it down (which is surprisingly easy to do, since you often have the collet nose sitting on the mill table).

But like the doctor says, don't do that :p

PeteF
12-27-2010, 05:26 PM
Wow, that's a loaded question. I think the quality of the collet and the collet chuck (in the case of ER's) will predominate any inherent accuracy advantage of the various collet systems.
Why would you think Morse is more accurate than 5C or ER?

As I said, all things being the same which would be more accurate (in the situation I suggested). Again, simply repeating, but since a morse taper collet goes directly in the taper there are no adapters involved. On the other hand since the MT seats along a long surface I'd guess it would be much more susceptible to contamination affecting it compared to the ERs. I'll always clean the taper and spindle before mating a morse taper, but while I don't exactly roll my ER collets in the chip pan before shoving them in, in my experience it seems they don't need quite the same level of care in their use. I have never used 5C but imagine they're between the two in this regard?

Arthur.Marks
12-27-2010, 05:51 PM
Even though the MT is a longer surface contact, it is perhaps the easiest to clean reliably. That is a large bonus! I use one of those spindle wipers---it is a stick with a male MT shaft with soft wipers on it. The male collet connection is of course pretty easy to clean with reliability on any 3-finger closing system. ER's, once contaminated, are arguably the most difficult to just "wipe off" / clean, but even a 5C you must reach in the spindle with a wipe and clean by hand. What if you get a minuscule particle that tears off your wipe on the collet pin and then fouls your rear mounting interface?

I agree that the systems themselves are largely based on 1) accuracy and precision tolerance in the manufacturing process for both the collet and spindle of your machine 2) the least cumulative error... i.e. whatever your spindle is directly made for will always (theoretically) be the most accurate 3) cleanliness of the system each time you use it.

If you look at the tolerances, it is really all comparable. Even if one system trumps another by a "tenth"---you likely can get a special accuracy collet to equal or better that difference in the event it is necessary. This is all assuming you are using the nominal diameter of the collet itself... such as was mentioned with the ER's and the differing runout based on varied closing diameter.

Arthur.Marks
12-27-2010, 06:02 PM
Thinking about this more, I would expect shrink-fit tooling to be the most accurate as there is no bending or spring action of the toolholder as in the other systems. I would imagine shrink-fit holders would thus produce the most reliable long term accuracy of any of the systems mentioned so far?

PeteF
12-27-2010, 08:43 PM
Even though the MT is a longer surface contact, it is perhaps the easiest to clean reliably. That is a large bonus! I use one of those spindle wipers---it is a stick with a male MT shaft with soft wipers on it. The male collet connection is of course pretty easy to clean with reliability on any 3-finger closing system. ER's, once contaminated, are arguably the most difficult to just "wipe off" / clean, but even a 5C you must reach in the spindle with a wipe and clean by hand. What if you get a minuscule particle that tears off your wipe on the collet pin and then fouls your rear mounting interface?

Arthur you make good points and you're clearly a much more civilized man than I, with your spindle wider. My "spindle wiper" is typically one of my digits!

lazlo
12-27-2010, 08:59 PM
Even if one system trumps another by a "tenth"---you likely can get a special accuracy collet to equal or better that difference in the event it is necessary.

That's a good point -- there are special accuracy 5C collets, and I ended up with a couple of Rego-Fix specially accuracy ER-40's in a mixed lot I bought on Ebay. I'm afraid to use the Rego-Fix's, after I looked up the list price!

By the way, the same company that makes those taper wipers has 5C wipers too.

Arthur.Marks
12-27-2010, 09:35 PM
By the way, the same company that makes those taper wipers has 5C wipers too.
Who's that? I looked for them once to no avail... even asked around at IMTS this year too and received blank stares. I could definitely use some. Got a link?

PeteF
12-27-2010, 11:22 PM
I was figuring it wouldn't be the accuracy of the collets per se, more the way they were secured within their respective tapers. Apart from the fact that the morse taper collet seats directly in the spindle taper, being a very long taper I wondered if that would in itself make it more accurate? I figured that what looks like a relatively smoothly ground surface is actually a series of waves and inaccuracies, albeit the better it's made the smaller they are. Over a long taper would they have more opportunity to cancel themselves out such that the average is very consistent when compared to a short but steep taper?

Rich Carlstedt
12-28-2010, 12:38 AM
As a general , yet probable controversial comment here, the more accurate collets are those with a single angle. ( MT etc.)
This is because you can only have one boss. Double angles inheirently have both manufacturing and holder issues.
Single tapers allow total control by a single surface. While being retained, you do not want deflection from the nut or drawbar. Having them retained with a square thread makes them almost perfect as the collet can then conform to the taper bore. The Universal Engineering collets called Kwik-Switch or Accura-Grip fits this criteria and is a very accurate/high precision system.
They guarantee .0004 " (.001mm) runout at 8 times the diameter from the collet. This is only exceeded in shrink fit tooling

Page 159 of this PDF starts the discription
http://www.universaldevlieg.com/UniversalCatalog2008.pdf
Universal was absorbed by Devlieg a few years ago
They have some good photos of different collets on page 159 and 160 and they describe the Accura-grip system. Note they promise 150 to 200 % of the holding of a ER collet.
Rich

luthor
12-28-2010, 01:50 AM
[QUOTE=Rich Carlstedt]They guarantee .0004 " (.001mm) runout at 8 times the diameter from the collet. This is only exceeded in shrink fit tooling

Rich, I assume that is .01mm and not .00004".

Rich Carlstedt
12-28-2010, 10:06 AM
You are correct luthor !
it is .01mm

thanks
Rich

lazlo
12-28-2010, 10:14 AM
As a general , yet probable controversial comment here, the more accurate collets are those with a single angle. ( MT etc.)

...and 5C and R8 :)


This is because you can only have one boss. Double angles inheirently have both manufacturing and holder issues.

Single tapers allow total control by a single surface.

That makes a lot of sense Rich. I have a whitepaper from Hardinge somewhere that describes the pro's and cons of the various collets, and I remember them saying something similar.

Edit: found it. Hardinge's "Basic Workholding Techniques"
http://www.hardingeus.com/usr/pdf/collet/2316.pdf


Collet Systems

There are three different types of collet systems— the Draw-In, the Push-Out, and the Stationary.

Draw-In Collet

The Draw-In collet is the most common and also the most accurate for holding concentricity. To
close the collet, it is drawn into the spindle angle. The only moving part (other than the collet) is
the draw tube.

Concentricity depends on the accuracy of the spindle and the collet. The order hole of the collet
must run concentric with its head angle and back bearing which also must be concentric with
each other. There are no other factors involved.

Push Sleeve

The collet is closed with a sleeve which pushes against the angle on the collet causing the
collet to close, but not move longitudinally.

There are even more parts in this system to affect concentricity: the collet; the cap and how it
aligns with its locating shoulder and thread; the sleeve with its allowance; as well as the clearance
between the spindle back bearing and the collet’s back bearing. This system is the least
accurate when trying to hold concentricity because of the large number of parts involved.

Arthur.Marks
12-28-2010, 10:57 AM
...and 5C and R8 :)

Yes and no. A MT has a single bearing surface which both centers and closes the collet. Most other pull-type collets have two bearing surfaces: the angular closing taper on the front and the central locating bearing on the shank.

BTW, that was a great Hardinge link!

lazlo
12-28-2010, 11:24 AM
A MT has a single bearing surface which both centers and closes the collet. Most other pull-type collets have two bearing surfaces: the angular closing taper on the front and the central locating bearing on the shank.

I agree in principle, but there's only a single angle on an R8 or 5C collet. You pull the collet back against the angle, and it closes. That's the "angular closing taper on the front" -- what's the "central locating bearing on the shank"?

By the way Arthur, my 5C wiper is "SpinLMate" model 55110.

Arthur.Marks
12-28-2010, 11:38 AM
Lazlo, look at pg. 9 in the above link. + thanks for the part #! :)

BACK BEARING:
Probably one of the least [cited? mentioned?] (but most) critical factors in close tolerance machining is the allowance between the back bearing of the collet and the spindle. When the clearance is too great, the collet will pivot or shift when tool pressure is applied to the part, causing excessive runout. Excessive runout can be caused by a worn spindle back bearing or a worn collet back bearing.

lazlo
12-28-2010, 11:57 AM
BACK BEARING:
Probably one of the least [cited? mentioned?] (but most) critical factors in close tolerance machining is the allowance between the back bearing of the collet and the spindle. When the clearance is too great, the collet will pivot or shift when tool pressure is applied to the part, causing excessive runout. Excessive runout can be caused by a worn spindle back bearing or a worn collet back bearing.

Ah, very interesting. Collets are complicated little animals! :)

PeteF
12-28-2010, 09:23 PM
Thanks guys, this is all great stuff and precisely what I was looking for. So since an MT collet is pulled in by a drawbar that would be another reason for them to be more accurate (in theory) than say the ER style that uses a closer?

I still like my ER collets but I'm interested in knowing the advantages of the other systems. One day I really should get around to making a collet chuck for the spindle nose so I can do away with the drawbar, a significant disadvantage of what I currently do, although sometimes I'll use the drawbar to seat the MT3 collet adaptor then take the drawbar back out again.

Pete

oldtiffie
12-29-2010, 01:35 AM
Pete,

going back to your OP, it seems that the MT collets are to fit into the MT in your lathe head-stock.

That may be fine where the MT collet fits straight into the spindle MT, but if the spindle MT needs to be sleeved to get to the MT collet size, you have introduced another possible error or errors.

There are special MT sleeves for head-stocks, quills etc.

Just hacking/cutting a standard MT sleeve may not be the answer either as that will be subject to error.

If it were me I'd be less than enthusiastic about turning etc. in an MT collet with the draw-bar removed as if the MT taper "came loose" or "let go" I might have quite a problem on my hands.

I have MT adaptors and sleeves on my grinders but grinding loads are way different to turning loads.

I prefer my ER-32 collets in my "tap-true" adaptor. If needs be I can "tap" it radially for parallel eccentricity and I can shim it for "conical" eccentricity if needs be.

I rarely have to do either with the work I do, but its quick and easily done if needs be.

The ultimate alignment is to bore and face the soft jaws on a 3-jaw chuck.

PeteF
12-29-2010, 02:03 AM
Yeah good point Tiff, it's more a hypothetical question to better understand the operation of the different collet systems than a nuts-n-bolts consideration of what I should do.

Regarding the drawbar, it's not something I do as a matter or routine, but certainly have, but it's just light cuts on small diameter stock and I haven't had the taper release as yet. Maybe I've just been lucky though ;)

Mcgyver
12-29-2010, 11:37 AM
So since an MT collet is pulled in by a drawbar that would be another reason for them to be more accurate (in theory) than say the ER style that uses a closer?


I would agree with that, I have both and the adapter in the spindle and drawbar is preferred imo. Less sources of error and zero overhang.

headstock adapters can be made very accurately without too much trouble. start with something tough, say a chrome moly steel. Turn the taper between centres with occasional dismounts to check the fit in the spindle using a thin layer of blue to check....adjust taper until perfect. Finish the ID in situ. Since the spindle should have been ground in place, and the adapter its turned in place in the in place turned spindle, there should be few sources of error.

In a newbs moment of weakness I've bought collets with several thou runout, busy bee, which to their credit they accepted a return and refund. But the good ones will have negligible runout....if you buy used or the common sizes as need they hopefully.

As for different types, the flexible ones have the obvious advantage of gripping any size perfectly. I've also got 2j and 5c and where they rule the day is they can hold short pieces and there's nothing proud of the front of the collet; ie on my rubber collet closer the collet end is a 1/4" back from the front the chuck which can be a really pita when you need to work right up the chuck on small parts. Like all shop things is seems, you really need one absolutely everything :D

PeteF
12-29-2010, 09:24 PM
With regarding the ER32 collet system, as mentioned, my chuck adapter uses a drawbar to secure it. Is it possible to buy these MT3- ER32 adapters with draw tubes instead of bars? Most of the work I do with them is quite small diameter and would easily fit down a tube but unfortunately the drawbar stops it dead in its tracks. The alternative is to cut the rod into smaller pieces but that is wasteful, alternatively work without the drawbar with the dangers in doing so Tiff pointed out.

Pete

franco
12-30-2010, 11:02 PM
Pete,

The easiest way to get the capability to use longer stock with the full range of ER32 collets is to make or buy a spindle nose chuck.

For a threaded spindle nose it is easy to make one which screws straight onto the spindle nose. Provided that the 16 degree included angle taper for the ER32 collet is machined while the new chuck body is screwed onto the spindle of the same lathe on which it will be used, this will center quite accurately every time. New 40 x 1.5 collet nuts are readily available, or just use the collet nut from your present chuck.

If you have a D1 series spindle nose or one where the chuck bolts directly onto a flange, make or buy a backplate mounting ER32 chuck and mount it on a spare backplate which fits the spindle. Both these cases will allow the maximum capacity of the ER32 collets to be used provided your headstock spindle will pass 20 mm stock.

If your spindle nose taper is one of the larger MT sizes and the spindle bore is over about 38 mm it is also possible to make a chuck which mounts in the spindle nose taper and has a large enough draw tube internal diameter to pass stock up to the maximum size for the ER32 collets. I saw a photo of one of these recently on another board, and will probably make one for myself for the 13" lathe eventually.

There is no reason I can see why a draw tube could not be used on a 3MT mounted chuck instead of a solid draw bar, though you might have to make it yourself, but as you said, it would restrict the diameter of the stock which could be passed through the spindle.

franco