[andrew2085]

I measured my spindle runout for the hell of it today. My mill (g0704) is still very much a work in progress, and I'm still getting the feel of the whole machining thing. I used a 0.0005" dial indicator mounted to a magnetic base attached to the table. I indicated off of three different surfaces.

the outside diameter of the spindle itself: 0.0002"
off of the R8 taper at the best angle I could get: 0.0005"
off the shank of a 3/8" end mill in a collet: 0.0007"

That all sound about right? I think I have OCD when it comes to this kinda stuff. Once I've converted to cnc I plan on making some TTS style tool holders. I figure something like a tool holder ought to have better tolerances than 0.001" which is gonna be more of an art than science if my tools have almost that much runout.

[hoss2006]

For the stock bearings those are perfectly acceptable numbers.
TTS tools are cheap enough especially some of the er ones on ebay that you can avoid making them yourself.
There are several videos on youtube however showing how to make your own relatively easy.
Hoss

[andrew2085]

Well that's good to know. I think I've seen the videos on youtube you're talking about where he uses a lathe tool holder in a vice. The TTS tool holders are more of a project/practice/fun type thing than a money/time thing.

[andrew2085]

I actually was able to get the runout down to about 0.0001" measured at the end mill shank. I covered the collet's taper in blue sharpie marker, then tightened it with an end mill in. I put a dial indicator on the shank and rotated the spindle by hand. I marked the side of the collet where the highest positive runout was. When I took the collet, out I could see the high spots where the collet contacts the spindle taper. I just lightly filed the high spots opposite where my max runout mark was. After that I smoothed the file marks with some sandpaper to bring the original finish back. Just doing that brought the runout from 0.0007" to what looks like about 0.0001".

The real test would be to have something perfectly straight stick out 4 or so inches and measure runout at the end. I only know that the shank is centered on the axis of rotation at the end of the collet. Its not necessarily perfectly parallel to it. That would also be a lot harder to fix. I think I can live with it.

[SCzEngrgGroup]

I actually was able to get the runout down to about 0.0001" measured at the end mill shank. I covered the collet's taper in blue sharpie marker, then tightened it with an end mill in. I put a dial indicator on the shank and rotated the spindle by hand. I marked the side of the collet where the highest positive runout was. When I took the collet, out I could see the high spots where the collet contacts the spindle taper. I just lightly filed the high spots opposite where my max runout mark was. After that I smoothed the file marks with some sandpaper to bring the original finish back. Just doing that brought the runout from 0.0007" to what looks like about 0.0001".

The real test would be to have something perfectly straight stick out 4 or so inches and measure runout at the end. I only know that the shank is centered on the axis of rotation at the end of the collet. Its not necessarily perfectly parallel to it. That would also be a lot harder to fix. I think I can live with it.

Now take the collet out, rotate it 180 degrees, put it back in, and check runout again. You might have ground the spindle off-center to compensate for runout in the collet....

Regards,
Ray L.

[andrew2085]

i only filed the collet not the spindle. an r8 collet only goes in one way so i cant rotate it.

[hoss2006]

I actually was able to get the runout down to about 0.0001" measured at the end mill shank. I covered the collet's taper in blue sharpie marker, then tightened it with an end mill in. I put a dial indicator on the shank and rotated the spindle by hand. I marked the side of the collet where the highest positive runout was. When I took the collet, out I could see the high spots where the collet contacts the spindle taper. I just lightly filed the high spots opposite where my max runout mark was. After that I smoothed the file marks with some sandpaper to bring the original finish back. Just doing that brought the runout from 0.0007" to what looks like about 0.0001".

The real test would be to have something perfectly straight stick out 4 or so inches and measure runout at the end. I only know that the shank is centered on the axis of rotation at the end of the collet. Its not necessarily perfectly parallel to it. That would also be a lot harder to fix. I think I can live with it.

Good job. Was it an economical collet?
When you go to TTS you'll only need the one 3/4 inch special, get the one from Tormach.
Hoss

[andrew2085]

Yea, it was just a cheap collet that I wasn't afraid to experiment with. My only concern is that doing this might actually make the runout at the end of a long end mill even worse. The collet is kept parallel to the axis of rotation by the contact of the tapered end with the spindle, and the slip fit of the other 0.949 ground end right? if I'm moving the tapered end to center it, then the collet is now at a slight angle. I won't be able to tell until I measure it. it could be that only the taper on the spindle is slightly offset, but the 0.950 section up inside is perfectly centered.