on collets, chucks and precision

din 64997/iso15488B sais to measure runout of er (other?) collets at 2.5-3.5 diameter. up to 6mm runout is supposed to be 0.015mm and then 0.02mm. for 3 mm diameter its 10mm from collet, for 10mm diameter its 25mm from collet. for larger it probably doesnt matter that much.

drill chucks are measured at 50mm for 3mm diam. and 100mm for 10mm diam. (according to roehm.)

no idea how lathe chucks are measured. mine has 0.02mm at chuck and 0.03 at 60mm (forkhardt 250).

i can get no spec. er collets (so supposedly 0.015/0.02), 0.01, 0.007 or 0.005. hardinge seems to spec. 0.01 on er. quality is a big problem. i bought a er25 set of "0.007" to be sure its good. it was from an idustial supplier and it was total crap, the bores even had ridges in front. were $8 a piece. sent it back. bought a 0.005 set at $15 per piece. terible, no ridges, but smaller sizes rusted inside, i have a pic. of all the shit that came out of the set after vibatory cleaning. the 16mm collet averaged 25 micron at collet and 62mu 40mm out. 6mm was 5mu and 4mu at 40mm. 3mm was 5mu and 1.5mu at 7mm. this support the theory, that spindle axis and tool axis are two lines in space that do not have to touch and the angle between them is unknown. interestingly the 6mm collet, although rusted inside has 5mu and 4 mu at 30mm (i will keep it, send rest back).

so i got myself a regofix nut and the runout dropped by 50%.

i then bought a regofix up (ultra precision) 10mm er25 collet at $40 and with the regofix nut its within spec. of 0.005mm at 2.8/5.2/6.8 mu at 3/20/40mm compared to the former collets 10/22.5/31mu with the regofix nut.

roehm spiro drill chucks are spec. at 0.05mm run out and albrecht is 0.04m. (had to sent spiro back twice.) so for 10mm its at 100mm. so roughly 3x colet runout. 4/3 x = 0.013mm. similar to "standart" er collet.

above was measured with a holder that shows no movement on a 0.002mm test indicator on top and botom of taper, and validated by a a 0.0005mm graduated hicator.

so, stating collets are much more accurate than chucks is pretty brave.

is there any u.s. standart for collets and chucks? i have seen 5c collets up to $100 per piece. is the precision up to the manufacturer? so, a set of precision 5c collets will be severat thousands bucks?

I don’t know the answer to your question, but some time ago these collet nuts were discussed here and I bookmarked them. Would this help you?

Home — easyZERO Nut

I have a pretty complete set of 5C collets and once gathered they make life pretty easy. But buying a full set brand new at full price might be rough. Almost as good as increased accuracy is that collets are great for holding things that you don’t want damaged or scuffed.

Also, there are adjustable 5C chucks made - you can dial in the lack of run out you need.

I have done a lot of close grinding and feel a good collect can run off .0002 and a good spindle or holder the same, so just throwing a part in a machine a half thow is possible.

Yes miles of posts on this subject...

so, where to get good er collets for less than $40 and are 5c better (although nobody is giving out any 5c spec.)?

dian said:

so, where to get good er collets for less than $40 and are 5c better (although nobody is giving out any 5c spec.)?

Click to expand...

Actual good quality collets from reputable companies with actual quality control are around that price range for a 10µm class ER32 for example. Certainly not much cheaper.

General purpose ER I buy YG1 5µm from Cutwel here in the UK. They might pass 5µm at the spec gauge length, I haven't tried. They are cheap and fairly well finished so you might be happy with them.

When I want quality that I know will be quality I buy 5µm from Iscar/ETM, Sandvik or WNT (Ceratizit). WNT also sell 2µm ER collets, and I have a few, it's hard to tell the difference between them and 5µm when you stack all your external errors up. And they are much more expensive.

Plus the first time you spin a part or cutter there goes the .0002 or so.

Yep, or when someone clamps down on a drill that was previously held in a drill chuck, or doesn't clean everything meticulously when assembling. It takes nothing to ruin them.

i have seen many rusty dirty collets and often tool is changed without removing collet to clean and the collet holder clamping surfaces cleaned
.
then there is long tools not straight or end mills not sharpened concentrically and of course each machine spindle taper can be dirty or have its own runout
.
long end mills i have seen .020" dia difference runout that in inch
long drills i have often seen 0.100" dia difference runout
sure short carbide drills i try to keep under .005" dia difference runout. my spec is based on over 10,000 measurements and what dull expired tool looks like after it performed its normal machining
.
sure runout is not good. but at what level do you replace your tools holders ? do you replace 95% of your tool holders cause they are over the brand new tool holder runout spec ? boss naturally not going to replace 95% of tool holders unless you got massive sudden tool failures happening every 10 minutes on 95% of you tool holders