PDA

View Full Version : making Er16 collet chuck



dovidu
03-13-2019, 09:22 PM
i needed an er16 collet chuck for my stepper motor.

purchased two from ebay: "er16 collet chuck motor extension"



i've had much luck with chinese tools, but this time... it was a terrible experience.

the specs said if i purchased 8mm ID for motor shaft, I would receive 7.98mm shaft so i could induction heat it, then put it on the shaft



however, i recieved 8.1mm oversized hole, so i couldn't use it... too much runout.



took over a month for it to arrive, wasn't going to wait again.



based on google and forums er collet chucks have 8 degree internal taper angle



my compound was not gonna have the accuracy required,

so i used a bevel protractor to get the precision within 0.1 degrees, which for ER collets is enough



even greater precision could be set, if i purchased electronic bevel protractor, but it would cost me over $600 for quality one.

i just went with vernier one, which wasn't too bad.



below is the link to the making vid of this collet holder :)



feel free to comment

https://youtu.be/bQxgMjGpD48

TGTool
03-13-2019, 10:39 PM
I haven't watched the vid yet, but you can set the 8 degree to your level of measurement by using one of the collets themselves, assuming they're good ones. Anyway, if you chuck an appropriate piece of stock in the chuck to turn down to a snug fit on your chosen collet, you can then mount the collet on the stub arbor. With an indicator set on the compound, at lathe center height, adjust the compound so it track the collet without indicator movement. I make ER-32 collet chucks and use a sine bar to set the compound and consider it good enough when I get a tenth or less deviation moving the compound. Using the collet itself means you're depending on it to be correct and you've got much less travel to watch for variation but it should be good enough.

Bob La Londe
03-13-2019, 11:08 PM
Nice video. Why do you need an ER collet chuck on a stepper motor?

dovidu
03-14-2019, 02:15 AM
Nice video. Why do you need an ER collet chuck on a stepper motor?

just for a simple cnc machine

dovidu
03-14-2019, 07:06 AM
I haven't watched the vid yet, but you can set the 8 degree to your level of measurement by using one of the collets themselves, assuming they're good ones. Anyway, if you chuck an appropriate piece of stock in the chuck to turn down to a snug fit on your chosen collet, you can then mount the collet on the stub arbor. With an indicator set on the compound, at lathe center height, adjust the compound so it track the collet without indicator movement. I make ER-32 collet chucks and use a sine bar to set the compound and consider it good enough when I get a tenth or less deviation moving the compound. Using the collet itself means you're depending on it to be correct and you've got much less travel to watch for variation but it should be good enough.

hm, i've heard you could set the agle correctly without the collet.
you need to have DRO though.
just indicate the inside taper of the er collet chuck, and use trig. to figure out the appropriate Z and X amount.
very simple indeed. need calculator though

MattiJ
03-14-2019, 07:12 AM
hm, i've heard you could set the agle correctly without the collet.
you need to have DRO though.
just indicate the inside taper of the er collet chuck, and use trig. to figure out the appropriate Z and X amount.
very simple indeed. need calculator though

Or use small sine bar to set the reference angle, check with test indicator. (edit: did I read anything what was already written?)
Or two dial indicators mounted temporarily to lathe instead of DRO. Better if one of them is long stroke (1") version.

dovidu
03-14-2019, 07:14 AM
Or use small sine bar to set the reference angle, check with test indicator.
Or two dial indicators mounted temporarily to lathe instead of DRO. Better if one of them is long stroke (1") version.oh sinebar, that should be more accurate. what i ve used is bevel protractor. i indicated the bevel protractor, which was quite easy.

Sent from my LGM-K120L using Tapatalk

kitno455
03-14-2019, 07:20 AM
you don't need a fancy protractor or DRO- just put some blue on the inside of the bore, insert the collet lightly with a piece of bar stock in it, and spin the collet. If the blue wipes evenly, your angle is good. If not, adjust compound and try again.

allan

Mcgyver
03-14-2019, 08:39 AM
one way to get tapers dead on is turn a male part, set it up on sine bar, indicate it, semi loosen the compound, tap tap tap gently with a brass hammer and try again. It's easier than it sounds and the basic way to get perfect tapers with basic measuring equipment.

old mart
03-14-2019, 12:38 PM
Seems like far too much bother, why not just bore out the 8.1mm hole and press in a bush and finish bore to 7.98mm.

Dan Dubeau
03-14-2019, 01:13 PM
Bluing against a known good taper is the best way IMO. When I made my Er 20 collet holder I made a male master along with an er16, and 32 while I was dialed in. It's all well and good to make stuff to the #'s and standards, but fit and finish are what matters in the end, so if you can measure directly, cut out the middle man and go direct.

BCRider
03-14-2019, 02:17 PM
Bluing the final taper is a great way to use for testing. But on something like an ER chuck by the time you know it's wrong it's pretty much too late unless you do the check when you've still got about 1mm of material to remove off the ID to go. So the check needs to be done while the collet is still sticking out proud of the final depth.

There is certainly "more than one road to Rome" and any number of options to set the compound angle will work. But do do so to within 0.1 by using a is one fine trick. That must be one fancy bevel protractor ! ! ! ! Looking around at google images of bevel protractors only the very best has any better a chance than half a degree.

I don't have one of those and I don't have a sine plate or gauge blocks. And besides, even with all that we still need a dial gauge. So my choice would be as mentioned earlier and make up a turned true stub and with that still held true I'd slide on one of the collets so it's a firm sprung fit and then with the dial gauge in the tool holder adjust the compound angle until there's zero runout and call it a day. In effect using one of the collets as my sine plate.

old mart
03-15-2019, 08:52 AM
If you absolutely have to reproduce the internal taper, then 0.1 degrees is not good enough. You need to setup one of the existing er holders running true in a four jaw chuck better than 0.0005" tir. Then with a lever indicator (plunger dti will not do) set up with its pivot vertical and the tip exactly on the centreline to get the compound angle perfect. Then you can finish bore the taper with the tip of the boring tool exactly on the centreline. If the indicator or the boring tool are not exactly on centre, the taper will be wrong.
I still think the sleeve in my post #10 would be easier. Also, an 8mm spindle on that er collet seems as about as stiff as wet spaghetti.

MattiJ
03-15-2019, 09:11 AM
Bluing the final taper is a great way to use for testing. But on something like an ER chuck by the time you know it's wrong it's pretty much too late unless you do the check when you've still got about 1mm of material to remove off the ID to go. So the check needs to be done while the collet is still sticking out proud of the final depth.

There is certainly "more than one road to Rome" and any number of options to set the compound angle will work. But do do so to within 0.1 by using a is one fine trick. That must be one fancy bevel protractor ! ! ! ! Looking around at google images of bevel protractors only the very best has any better a chance than half a degree.

I don't have one of those and I don't have a sine plate or gauge blocks. And besides, even with all that we still need a dial gauge. So my choice would be as mentioned earlier and make up a turned true stub and with that still held true I'd slide on one of the collets so it's a firm sprung fit and then with the dial gauge in the tool holder adjust the compound angle until there's zero runout and call it a day. In effect using one of the collets as my sine plate.

Lots of bevel protractors come with nonius scale that gives you degree split to 5 minutes. So you can read it to 60/5=20 = 0.05 degree accuracy. But even that is still kind of coarse for any taper.

Using existing ER collet to indicate is also bit risky business IMO. Lot depends on how well the collet "settles" to rod without external clamping like its normally used.
Partly same problem if you use blue.

Doozer
03-15-2019, 09:58 AM
I would venture to say that the compound slide on the lathes
that most of you home shop guys have has more slop than
the accuracy required to set and cut the angle needed for
precision collet taper setup work. I would thoroughly scrutinize
the slideways and the gib for wear and proper adjustment
before attempting a precision job like this if you want to
achieve satisfactory results.

-Doozer

old mart
03-15-2019, 10:56 AM
Dead right, Doozer, I would not try it on the Smart & Brown model A at the museum, although the compound is pretty well adjusted. I did a test of the taper turning attachment for that lathe to produce an R8 taper. Using a 12" length bar and trigonometry, I got a passable taper within a minute of arc, but not exact.
Using the collet to set the taper is a fools errand.
ER16 is capable of holding up to 13mm, 1/2", but with an 8mm spindle on the motor, 3mm, 1/8" cutters would be risky

TGTool
03-15-2019, 12:01 PM
<snip>
ER16 is capable of holding up to 13mm, 1/2", but with an 8mm spindle on the motor, 3mm, 1/8" cutters would be risky

DIN standard says ER16 goes up to 10mm so he might be a little safer there.

BCRider
03-15-2019, 01:15 PM
Reading through all the replies in this thread one would think that it would be best if the OP just gave up and took up some other sort of hobby.

Oh sure, proper care is a concern. And there's lots of good suggestions for options in the replies. But overall I find that this thread in particular has a strong current of anticipated failure unless he's got some pretty snazzy stuff to use.

Let's face it, no group of people can be more OCD at times than machinists. We constantly work in amounts that are well past invisible and require special instruments to even detect. But "we" have been making functional parts on basic machinery that works just fine for a long time. Far longer than any of us have been around.


Between the bevel protractor, the mis-sized adapter socket and a collet pushed onto a turned arbor and a good DTI you should be able to set and confirm the angle to a degree that is sufficient to produce a good working part.

old mart
03-15-2019, 01:33 PM
10MM it is, TGT, I must have misread the chart. On an 8mm spindle I would steer clear of steel and take very small cuts in aluminium. A motor spindle of 12mm would be better though at minimum.

enginuity
03-15-2019, 01:52 PM
Reading through all the replies in this thread one would think that it would be best if the OP just gave up and took up some other sort of hobby.

Oh sure, proper care is a concern. And there's lots of good suggestions for options in the replies. But overall I find that this thread in particular has a strong current of anticipated failure unless he's got some pretty snazzy

I've made a number of home shop tools out of 12L14 free cutting steel. Soft as butter. I had some guys tell me you can't make tools out of it. Won't last. All the usual stuff on why it won't work.

They work great. I'm not a production shop. It's home shop machining. Most of us aren't making stuff for CERN or something.

Same people who think Rong Fu mill drills are useless boat anchors and you can't do anything at all with them. I just refer them to George Britnell's work.