Er25 Collet Chuck For Atlas 618

[mattthemuppet2]

Thanks David, I'll be putting all this into my Atlas 618 mods thread - I don't want to clutter up the forum with lots of duplicate threads

If it helps with inspiration, I can easily cut off a suitable chunk of that stock and stick it in the post. Although looking at your location, that would probably cost more than buying it locally!

 

[mattthemuppet2]

didn't have any oil paint, but I used a thin coat of fabric paint which worked almost as well as I'd hoped - thanks Andre! Went from rubbing paint off the bottom 1/2 of the collet to rubbing it off the whole length.
then, being an inquisitive smart arse, I thought I'd try grinding the taper. Probably shouldn't have, but it looks nice. Strapped my Dremel extension to the boring bar and very carefully had at it


sparks flying! Adjusted 1/2 thou at a time and did several "spring" passes in between. Took about 2 thou to smooth out all the tool marks. I took a video but it's really not very interesting.

Grinding done. Oddly the bore seemed slightly out of round as I was grinding - you could hear the pitch change rhythmically. This might be the cause of problems I'll talk about in a bit

looks and feels damn nice though

my DTI didn't want to work in the cold (18F) so I used my DI instead. Less than 0.0005 total run out using a 0.377 gage pin (had 3 thou until I remembered to tighten the nut, doh!)


about 1 thou 1in out


this is a 14mm bar from an old dot matrix printer, which in theory should be perfectly round and straight. About 1.5thou end to end (without turning the rod) over 6in


but about 8thou total run out when I rotated it. I tried the other one I have too and got the same results


at the nose it was still about 1/2 thou.


so, not terrible but not great either. I don't have anything better to test run out with than this unfortunately - I have a couple of other bars from the printer, but these ones are more likely to be true than anything else.

Possible causes - it was late and I was excited to test it, so I didn't do a very good job of cleaning the chuck out after grinding it. So fingers crossed it's just me being an idiot.

The slightly out of roundness feeling I got while grinding - could it have been true and then something I did during grinding made it out of true? The compound set up was the same and even though the set up wasn't ideal, the "out of roundness feeling" came through on the very first light touch pass, when the grinding stone was barely kissing the metal.

Is it worth trying to indicate the taper? How? At different depths, rotating the chuck? I can't do a single front to back reading as the compound screw has a slight bend in it so the tip will wiggle in an out just because of that.

The only 2 solutions I can think of would be to recut the taper (sadness at loss of pretty ground finish) or to chuck up that printer bar in my 4 jaw, dial it true (radially and axially), mount the chuck to that and then take a skim pass off the collet chuck spindle register.

Ideas?

 

[francist]

Maybe the collet itself is at fault? I have an ER25 chuck and collet set from LMS (also on my 618) and some collets seem to run better than others. Just a thought.

-frank

 

[mattthemuppet2]

Good point Frank and not all the collets are the same, with two types of markings. the one i used came from the crappier looking lot (a couple were bad enough i got a partial refund from the eBay seller), when i tried one of the others the run out dropped. however, i found the cause of the run out - both the printer rods were out of true . only found out by chucking it up in my 4 jaw and getting exactly the same amount of run out! bah, you know what they say about assume. tried another polished rod and got less than 1/2 thou out at the chuck and about a thou 4 inches out. that's good enough for me filed down the sharp edges and stoned the sides. now I'm making a rack for the collets and I'll be done!

 

[mattthemuppet2]

all right, project is done

some more photos, just because (the paper they're sitting on are the plans)


not very flattering picture of the spindle threads - my homemade ID threading bar is a bit springy


however, it's balanced out by the simply gorgeous finish on the collet taper


rack made out of some scrap wood, scrap dowels (5/16 and 1/4), scrap 1/8 brass rod and a left over bicycle spoke


in all it's glory!


my wall of chucks - homemade collet chuck, 4" Buck 4 jaw with homemade back plate and 4" Bison 3 jaw (other jaws are in the toolbox). I think that's good for now, I can't see myself needing a 6 jaw anytime soon, but you know, if one falls into my lap..


thanks for watching!

 

[Rob]

What size of hex did you start with? I plan to make a collet chuck and hadn't thought about using hex, had planned on round but like the idea of hex.

 

[mattthemuppet2]

Hey Rob, from memory it was 38mm across the flats (I was looking to see if I had a big enough wrench), which is 1 1/2in I think. Would have been nice if it had been a wee bit bigger as the nut is slightly wider than the flats, but it was what I had to hand. If I were buying stock I'd probably look for something bigger to clear the nut or even step up to ER32. You can't fit more than 1/2in or so stock through the spindle, but the MT2 taper lets larger stock go in a ways and there's the length of the chuck too.

 

[Fabrickator]

I recently made an ER 25 adapter. To cut the taper, I set the compound to the basic "advertised" taper. I started way early to get the taper right before final sizing for correct depth. I blued the shaft and did a soft "rub" test with the collet, made a small compound adjustment and took another light cut again until I got it dialed in. After reaching the correct depth, I finished it with a quick hit of some very fine emery cloth.

 

[mattthemuppet2]

good point and thanks, I didn't describe that very well I found that the 8deg mark on my compound was dead nuts on for full contact on the length of the collet, although if I'd used Dykem I might have gotten it a hair more accurate. Still, they're complete buggers to get out if the collet is inserted fully without the nut, so I'm pretty confident that the taper is right. I might end up taking a little off the nose if I don't get the full range for the collet as the nut bottoms out on the nose before the collet is squished completely closed, but we'll see how it goes. I have to say though, I'm still absolutely amazed by the finish from grinding - making a proper mount for that dremel extension is on my list of things to do

 

[Silverbullet]

get that Dremel set up and have some 60 degree stones you can grind the threads in your next one ? nice job on this project . I think I'll be making one to.