[Frogblender]

Shown is a 1/16" ER20 collet. Note the 7 "leaves" of the collet are all dissimilar at the tool contact point. Is this normal, or is this collet junk?

[PixMan]

Certainly not a collet of "Rego-Fix" quality. You generally get what you pay for. Have you tried putting a gage pin it it and checking runout?

[John Welden]

Those rego-fix are so nice.

[PoiToi]

From what I understand on how ER collets work, the surface area of contact on each "leaf" shouldn't affect the accuracy. The forces should all still be balanced around the tool. (the pressure will change with area, but total force should stay equal.)

But I would definitely put a gage pin in it and check it with a .0001" indicator.

let us know what you find out, I am curious.

[stevespo]

I'm definitely not an expert. It does look like there are several fairly equal contact points. Test it out and tell us how it works.

As a point of comparison, here are my 2 smallest ER20 collets (1mm and 2mm). The 1mm only has 4 contact points, but the 2mm has a full 8 points, all accurately machined. I think I paid about $30+sh for a 13 piece set.

Steve

[dertsap]

are all dissimilar at the tool contact point. Is this normal

you have three points of contact , the other points are cut in a way to be for clearance only

[fatal-exception]

Would be interesting to know the runout of that, but I would say it's junk. Looks like they tried to make 8 points of contact, but EPIC FAIL.

[Frogblender]

Here is the TIR for the collet in question (numbers are in .001" (thou). Measured on a .001" indicator (I don't have a .0001" indicator))

TIR for my spindle alone: 1.0
TIR for ER20 arbor: 1.0
TIR with 1/16" collet pictured above with a 1/16" drillbit, measured on the shank of the drillbit about .15" down from where the drillbit emerges from the collet: 2.6

The collet pictured above was from an ebay set. I also have a couple of collets from novakon for comparison - here are the numbers:

TIR for 3/8" collet novakon: 1.6
TIR for 3/8" collet ebay : 2.4

TIR for 1/8" collet novakon: 1.8
TIR for 1/8" collet ebay : 0.9

TIR for 1/2" collet ebay: 1.9



Can anyone comment on these numbers? Good, bad, or ugly?

[macona]

You have .001" Runout in your spindle?

Yikes!

You dont know if your spindle is even holding the er-20 holder straight at this point.

The collet is garbage. Send it back or throw it out.

Maritool.com has good collets. And unless you are using metric cutters get the native imperial sized sets as often the imports are native metric and the dont have as good of a grip. They will still work due to the large grip range but there is not as much contact area as you squeeze them down.

[crabbass]

If you don't have a .0001" indicator, you shouldn't be able to even guage the acceptable runnout (to most shops standards). Buy a quality .0001" test indicator, then you can check that spindle out!!!!!

[Mcgyver]

runout of collet should be measured in tenths not thou's, else what's the point of using a collet?

others have noted that the size of the area of contact isn't going to materially affect accuracy which is mostly true if the three areas are close to 120 degrees apart. However having only three areas definitely will affect peformance if for no other reason that gripping power and it looks like several leaves are form a sharp v to the extent they could damage the finish of the work.

I bet its not a hardinge lol. I would have no confidence in a piece of tooling so poorly made and would consider it junk; collets are a precision piece of tooling if they can't get the slots right its not indicative that the rest weould be made well.

what's its pedigree?