Post PM-833T Spindle Bore Runout

Bill Kirkley

Registered
Registered
Joined
Sep 22, 2019
Messages
89
I decided to check the runout after boring out the upper part of the R8 spindle.

The Morse taper arbor PM supplied had at most two ten-thousands runout. (See attached video)

A 1/2" end mill

shaft in the Interstate collet had about one thousands runout.

I clamped a 1X2 board in the vise. Using the Y axis handle I pressed the board hard against the end mill. There was about one thousands deflection.

I am not familiar with typical runout values. These seem pretty good to me.

I have been of the opinion that a sloppy fit between the threaded end of an R8 insert and the spindle would have no adverse affects on performance. I think wedging the insert into the R8 taper with the draw bar is what produces stability, not the upper end of the insert.



















 
I am unable to see the videos you posted, but looking and what you wrote, I'm unimpressed. I've posted the results of my PM935 spindle runout and end mill testing after a replacement head unit was provided by PM to correct for the original defective head. Collet quality does matter. Results are here.
 
If you want to vet the accuracy of your spindle, then test your spindle, not something stuck into it. First, look carefully at the R8 taper in the spindle with a good light and look for damage or other problems. If you find issues, fix them properly. If the spindle taper looks OK, put a tenths indicator against the spindle taper and turn the spindle by hand. A decent spindle should be within a couple tenths overall. If that works and you still have a lot of runout out with collets, it might be the collets, it might be trash in the spindle recess, and it might be the anti rotation screw pressing too far into the key way of the collets.

When measuring stuff, try to avoid using surrogates instead of the actual piece you want to know about. It often only makes you more confused...
 
The question arose if enlarging the upper end of the R8 spindle would affect the spindles function. So I stuck an arbor in the spindle and found it deviates less than two ten thousands.

Davidpbest stuck an end mill in a collet and had 4 ten thousands runout in the end mill if I read it correctly.

Davidpbest. For what ever reason you have to be logged in. They worked for me.
 
The question arose if enlarging the upper end of the R8 spindle would affect the spindles function. So I stuck an arbor in the spindle and found it deviates less than two ten thousands.

Davidpbest stuck an end mill in a collet and had 4 ten thousands runout in the end mill if I read it correctly.

Davidpbest. For what ever reason you have to be logged in. They worked for me.

I agree with Bob. The critical reference surface of the spindle is the taper - the upper end shouldn't matter unless it's not concentric with the spindle rotation. The TIR of my PM935 spindle as measured at the R8 taper is 0.0002". With the best collet I can buy, I'm still at 0.0002" TIR on the OD of a 1/2" carbide end mill. With lower quality collets, the TIR measured at the OD of an end mill bumps up to 0.001".
 
David, does the threaded end of an R8 insert contact the spindle wall? If not, the inside diameter of the spindle or being concentric should matter.
 
It touches the wall of the spindle for stability. Blue it up and insert to check.
 
The R8 collet touches the wall at the top and the bottom, there is typically a relief in the center of the collets/spindle. The top is precision ground to prevent lateral deflection and should be a close fit to the spindle wall. Otherwise you would get axial skew on tightening and deflection of the collet with milling. The draw bar provide vertical tension, but is not for significant axial deflection. In tightening the bar there is no centering of the rod with respect to the collet other than screwing into the top of it. The top of the collet provides the centering.
Hardinge R8 Collets.jpg


The R-8 collet clamping range is very narrow, typically within 0.002". You may not be able to measure any noticeable runout at the collet tip holding a precision rod, but 2-3" out if the collet top is not aligned at the top, then you will see more runout at the cutting tip. As David pointed out and also mentioned previously, the cheaper R8 collets often have much worse TIR and sometimes collet fitment issues. I had problems until I got a set of Lyndex and royal R8 collets. It may also have something to do with the machine specific tolerances of the spindle. To measure runout of the spindle you want to an indicator on the mating surface of the spindle.

I use two touch probes that require you align the tips to less than 0.0004" with a test indicator, I use them with a R8 ER-32 collet setup which has a TIR of ~0.0002" out to about 3" from the collet. R-8 collets had too much runout.
 
Thanks, it would appear I rendered my spindle unstable by increasing the upper diameter by four thousandths so my Interstate collets would fit.

In spite of this, it performs well for my purposes. The collets remain wedged in place after use and have to be tapped with a mallet to free them.
 
If my math is correct the tip of my half inch end mill will deflect .001 inches at an angle of .03 degrees. Fortunately, that is acceptable for the projects I do.

4F581FBB-F747-499F-91DF-9DB23AB82054.jpeg
 
Back
Top