Er Collet angle question

marchyman

marchyman

Registered
Registered
Joined
Aug 14, 2016
Messages
20
I'm (attempting) to make a collet block as practice for making a collet chuck for my Craftsman lathe. My Taig mill used ER16 collets so that is the size collet I'm trying to fit. The collet spec calls for an 8° angle. I set the compound for 8° and then put dial indicators on the Z (carriage movement) and X (the side of the compound as in moves in/out with the carriage) and looked for 0.1406" movement on X for 1" movement on Z since tan(8°) is 0.1406... It didn't take too long to get the angle dialed in.

I thought I'd double check the angle by putting a collet on a mandrel and an indicator between the compound slide and the collet. I checked for a constant indicator reading as I dialed the compound slide in and out. That did not give me warm and fuzzy feelings. An uncompressed collet measures about 8.3° (calculated). I checked a different ER16 collet from a different manufacturer. It, too, was greater than 8°.

So finally to my question. Do I cut the taper to the 8° spec or do I match the collets? Yeah, it's a test piece and I can always do it again, but I'd like to know why the collets don't match the spec.
 
My opinion.. so you cut the 8 degrees, but not from the compound setting. You need to really dial in 8 degrees. Hopefully you have some precision angles and can set them up against your faceplate and maybe use a magnet to hold it against the face plate perpendicular to the plate. Then with an indicator sweep the compound back and forth dialing in the 8 degrees. when you get zero from one end of the angle to the other, you are really at 8 degrees. Lock the compounds angle down. and go to town.
 
I'd get the angle from a collet: in the 4-jaw centre a piece of accurate stock (e.g . turned-ground-polished aka TGP) the correct (max) size for the collet, indicate to zero run out, then slip the collet over it. Wrap some soft wire in the collet's extraction groove and twist it tight so it grips the stock fully, full length of the collet, THEN set the topslide by traversing the taper and adjusting for zero movement on Best dti AT CENTRE HEIGHT.

If you have one handy, dead inkjet printers are a good source for 8 - 10 mm TGP rods that the print head runs on...

Dave H. (the other one)
 
Jeff, the compound was set such that there was 0.140x" movement in the X for exactly 1" in the Z where "exactly" is defined as the tolerance of my dial indicators. The one I used to measure the X is pretty good, the other is a run of the mill Chinese 0-1".

Dave H. Yes, I think I'll use the collet to set the angle. Except my 4 jaw is currently holding the partly machined piece that needs the taper. Well, this is a practice/text piece. I expect to screw up and do it again. Hopefully not too many times. I'm surprised I didn't screw up cutting the M22 x 1.5 threads on my well worn 65+ year old Craftsman version of the Atlas 618.
 
As the collet is a full circumference of steel at 1 end and split into segments at the other end the "spring" in the open end is probably giving you the 8.3° from the uncompressed collet, but when tightened in the holder will close down to 8° and match your collet chuck. On close inspection of an uncompressed collet would you say the slits are exactly parallel or do they appear to get wider as they get towards the open end?

Cheers Phil
 
marchyman said:
Jeff, the compound was set such that there was 0.140x" movement in the X for exactly 1" in the Z where "exactly" is defined as the tolerance of my dial indicators. The one I used to measure the X is pretty good, the other is a run of the mill Chinese 0-1".

Dave H. Yes, I think I'll use the collet to set the angle. Except my 4 jaw is currently holding the partly machined piece that needs the taper. Well, this is a practice/text piece. I expect to screw up and do it again. Hopefully not too many times. I'm surprised I didn't screw up cutting the M22 x 1.5 threads on my well worn 65+ year old Craftsman version of the Atlas 618.
Click to expand...
My bad, I hadn't carefully read what you were doing. you did the math.
 
The ER collet "springs" are alternately connected at opposite ends. I was using the shank of a 5/16 reamer that looked straight when rolled on my surface plate in a 15/16 collet. The collet had a firm fit around the shank.

It may be easier to see in pictures. Craftsman/Atlas 618 [page 17] describes today's fun and games.
 
12bolts said:
As the collet is a full circumference of steel at 1 end and split into segments at the other end the "spring" in the open end is probably giving you the 8.3° from the uncompressed collet, but when tightened in the holder will close down to 8° and match your collet chuck. On close inspection of an uncompressed collet would you say the slits are exactly parallel or do they appear to get wider as they get towards the open end?

Cheers Phil
Click to expand...

ER collets are split into segments from both ends - slit alternate from opposite ends - just to overcome the problem you mention. That way they compress/close more parallel than other collets.

When I did my collet blocks I set my compount to 8 deg and made a first cut. Then blued a collet and checked the contact, adjusted the compound, cut and repeated until I was happy i was getting good contact all along the uncompressed collet.
 
You must log in or register to reply here.
Back
Top