[tbaker2500]

I'm getting ready to run a project which needs a 1/32" EM. I've done some work with them before, and had good results. But having just replaced my spindle, I though I'd check the runout for giggles.
With no collet installed, not even the R8 collet, I measured 0.0002" runout on the spindle. Pretty good I thought. I then installed the R8 TTS collet again, inserted a ER20 holder, 1/8" ER20 collet, and 1/32" EM (carbide with 1/8" shank). The runout on the tool shank (about 1/4" from the holder) was about 0.0011". Rotating the holder in the R8 collet didn't change anything.

So at this point I'm trying to figure out where the runout is coming from. So I removed the bit and ER20 collet, but left the holder in the spindle. Same runout. I tried an older ER20 holder (pre-TTS) and had about 0.0014" runout.

I doesn't seem to matter if I rotate the holder 180 or 90 deg in the R8 collet.

Time ran out and I left for the day, but it got me musing about where the runout is coming from. Since turning it in the R8 collet didn't really change things, that implies that it's not particularly the cumulative runout of R8 collet and tool holder. So that means the runout is either in the R8 or in the ER20 holder?

I guess I'm just cutting with one tooth on that EM, unless I track it down. :-)

[SCzEngrgGroup]

Are you sure both sides of the taper are absolutely clean?

[tbaker2500]

Well, no. I have been using anti-sieze compound on the mating surface of the R8 collet, and I remember reading a tormach paper which suggested that. You make me wonder if that was too thick.

[SCzEngrgGroup]

What I'm referring to is any debris that might have gotten in there. You should have anti-seize, but a very thin film. I would suggest removing the collet, carefully cleaning both the collet and spindle tapers, perhaps give both a quick rub with some very fine abrasive cloth to make sure there are no chips or other debris mashed into the surfaces, then apply fresh anti-seize and try again. Try measuring collet runout with just a drill or endmill shank mounted, rather than a collet chuck.

Regards,
Ray L.

[tbaker2500]

I gave it a once over, but it needs further attention.
A direct 3/4 shank tool is a good idea, I'll do that next.

[neilw20]

Any runout or dings in the face of the end of the spindle, or on the TTS holder pullup face?

[tbaker2500]

Another good thing to check. I didn't think about it until watching btu44's videos on runout.

[tbaker2500]

Whoo hoo!
I did a thorough cleaning of the R8 collet and the inside of the spindle.
It then followed the steps in btu44's video: [ame=http://www.youtube.com/watch?v=UePMwYP4BYE]TTS Runout - Holder.wmv - YouTube[/ame]
Without TTS engagement, the empty ER holder showed between 0.0004" and zero runout, depending on how it was rotated. After marking alignment, I then did the TTS engagement, and I saw about 0.0008" runout. Rotating the holder only made very slight differences. So I took the high point of the holder's TTS mating surface and honed it down slightly. After 2 tries I got the runout down to virtually imperceptible needle movement. I installed the 1/32" EM in an ER20 collet and it still read perfect on the EM. Sweet!
A second holder showed similar results.

I am very happy now, I can run that 1/32" EM with a little more confidence.

Things I learned- holders and spindle have pretty small runout, and can cancel each other out. TTS surface greatly affects runout. Slight honing can fix it.

Thanks for the help guys!

[Steve Seebold]

If you put a tool holder in your machine with no cutter in it and check the runout in the hole, I'll bet it will run at near zero. Then without moving the holder, put a tool in it only finger tight. Check the runout again. Now, whthout moving the holder, tighten your cutter like you're going to run it. The runout has no doubt increased significantly.

[tbaker2500]

You're referring to a set screw style holder, right? I've been using an ER holder to try and remove that from the equation for this testing. Or do ER collets also have that problem?
My final result tonight was on a 1/32"EM with a 1/8" shank in a ER20 holder, tightened enough to use.
The next thing I was going to look at was the set screw holders. That being said, everything works just fine normally. Not going to put more energy into it, I just wanted to make sure that 1/32"EM had a fighting chance. I don't think it will matter much on my 1/2". :-)

[Steve Seebold]

If you stick an indicator inside the ER collet holder, it should run dead nuts.

If it runs out, then you paobably have something either on the R8 collet that needs to be removed or the shank on tour tool holder if dirty. A couple of tenths runout should be OK. I have 6 ER 20 and 6 ER 16 collet holders, and the way I keep the collets clean and run my machine, there is no runout at all. And I have checkes it with both a .0001 and a .0005 Interapid indicator.

[philbur]

Standard precision ER collets have a maximum run-out of 0.02mm. Then to that you need to add tool holder, R8, spindle and bearing run-out. No run-out at all with a 0.0001" indicator, Interapid or otherwise, would be mostly wishful thinking, or luck on the same level as winning the national lottery.

Phil

[tbaker2500]

Well, with everything properly cleaned and with a little fiddling, I'm happy. Now I know to pay close attention to little bits of junk on the TTS collar.

[GLCarlson]

Way too many years ago, I read a small notice posted on the door of the analytical (chem) lab:

"Cleanliness is next to Accuracy"

Good advice for machinists, too.

[Steve Seebold]

Standard precision ER collets have a maximum run-out of 0.02mm. Then to that you need to add tool holder, R8, spindle and bearing run-out. No run-out at all with a 0.0001" indicator, Interapid or otherwise, would be mostly wishful thinking, or luck on the same level as winning the national lottery.

Phil

Then, I guess I won the lottery. BIG TIME.