How much spindle runout is acceptable?

[sbalder]

I restored a Bridgeport Boss last spring that came equipped with an Erickson 30 (NMTB) spindle that was in fairly rough shape. Having a set of ER-40 collets in my poseesion, I plan to use them as my primary tool holders and leave the collet chuck in the spindle. I put a brand new ETM ER-40 collet chuck on the machine tonight and was able to get runout on the outside of the tool holder to roughly .0005". With a 1/2" drill rod chucked in a collet and the dial indicator set about an inch below the collet face, I get a little less than .001" runout. Is this acceptable for general CNC machining?

-Steven Balder

[Newby2]

Steven...Have you tried to check the runout in the bore of the collet chuck? Is the drill rod you are checking straight? (This can be checked in a V block and a dial indicator. You may want to check the concentricity of the collet chuck and its' bore with the chuck out of the spindle. Also have you checked the runout of the spinlde bore itself? From that you are indicating, the longer the tool length, the more runout you will have.
Hope this may help.
Steve

[ctate2000]

Check the spindle run out inside the spindle taper. Measuring anything mounted in the taper will add error and give a false reading of true spindle run out. However, 5 tenths does not sound bad for a tool holder mounted in an older machine. The tool holder ID to OD run out can be off by more than .0005. To check the run out of tool holder and spindle together put the indicator on the ID where collet seats.

If you plan to leave the tool holder mounted you can check run out and turn the holder in small increments and possibly get it seated so that the run out is minimal, sort of hunting for the sweet spot.

Mounting a gage pin in the collet is a good way to check collet runout and flex.

[Newby2]

Using a guage pin is always better than a piece of drill rod. Use the largest pin possible to avoid any flex and be sure the spindle bore, collet chuck OD and ID are absolutely clean and burr free. I use a green pad, (used for scrubbing kitchen pots, similar to scotchbrite) to clean all the interchangeable tool holders. After 10+ years of doing it this way, I haven't seen any change in the sizes greater than .00005". This was verified by CMM.
Also there are scotchbrite spindle cleaning tools available on the open market. (I make my own with a dowel and a piece of scotchbrite!)
Steve

[sbalder]

Thanks to everyone for taking the time to post; the information on this forum has been invaluable to me.

My runout is definitely in the spindle. That's part of the reason for the new collet chuck, and I was able to verify this by "indexing" the chuck in different positions relative to the spindle bore and checking runout. I've also used one of Dremel's new scotchbright wheels and taken down some of the roughness I can feel in the bore. Where I was measuring measuring runout was about 3.5" from the spindle face, so I figure the spindle is off by about .0002".

I don't think I am going to be able to get this any better than I have. Is this amount of runout at the tool similar to other Bridgeport-type mills? Is it worth the trouble to have the spindle reground?

-Steven Balder

[fpworks]

0.001 measured at 1" from the collet face is not too bad.

The amount of runout you can accept depends on your application. For reaming accurate hole sizes, 0.001 is too much. Try "tapping in" your reamers for better results.

If you're doing a fair amount of steel work (especially production), high runout will degrade your tool life. The "high" side of your tool will be taking a larger chipload, thus wearing that flute(s) faster...you see where this is going. This problem is hugely amplified when cutting hardened tool steels.

If you're doing aluminum, tool life won't be an issue.

In my opinion (as long as tool life isn't an issue), the biggest pain of high runout is tracking tool diameter offsets. High runout systems (being spindle, toolholder tapers, collet, collet nut) also tend to have inconsistently high runout. Since your tool will take a larger cut with more runout, and if runout is inconsistent, you won't be able to "predict" or "track" runout for specific tools, unless you don't ever take your tool out of your toolholders.

I commonly use a "roughing" toolholder that has tremendous vibration dampening characterics, but can have up to 0.0015" of runout. (on the toolholder!) We'll see 0.002" of runout on the sytem. (in spindle, carbide test pin, measured 2.5 x tool diameter from face) Tool life is unaffected in aluminum. I use it for finishing as well because the vibration dampening helps with surface finish.

Just my take on runout.

[NC Cams]

"Acceptable" is relative to the demands of the user and/or the type of work they need/want to do.

If you know how to compensate, the runout will make cut sizes a bit larger/wider than anticipated but that can be compensated for once you learn the "character" of the machine. Post #6 explains this well.

Finish is something that may have to be dealt with but, again, once you know what you're dealing with, you either live with it or fix it.

For a spindle that is off only 0.0002", you might find that a re-orientation of the bearings on the spindle could possibly improve that. Some of the OEM spindle bearings were selectively fit ABEC 1's which can have some bore to raceway runout (check the ABEC specs, it can be substantial).

By offsetting the raceway eccentricity offset to that of the spindle ID eccentricity, you can sometimes eliminate or reduce the spindle bore runout without grinding the spindle.

Bridgeport did this via selectively assembling and fitting spindles to bearings at assembly. If it got too bad and couldn't move bearings to eliminate it, then they gound the ID taper.