lapping a VN spindle ID- insane? Or possible?

So along with the positioning question I have been spewing forth, I have another-

The VN spindle seems to have about .0005- .001 runout, measured on a 1/2" ground shaft inserted into a collet. The measurement is taken an inch down from the collet.
When I got this machine, the spindle had a bunch of built up material in the taper, like would result from spinning an unhardened collet or holder. So I used a steel scraper and got it out.
Any collet or end mill holder used in the spindle ends up with a ring of "fretting" marks around the taper , about an 1/8" wide and located close to the small end of the taper. This wear cannot be felt with a fingernail. Nor can any burrs etc be felt in the spindle taper.
I am thinking about making a brass lap, with the VN taper, and using a fine diamond compound to very gently lap the spindle, checking with bluing and a new collet. The object to remove any tiny high spots left on the taper ID.
Is this nuts?

It's not going to work. The lap will cut more on the larger diameters - more speed.

Stoneaxe,
Years ago we bought an Excello CNC mill with a #40 Taper that
was pretty beat up. A new spindle was out of the question so
we ground the existing taper without taking the machine apart.
Using a Dumore Tool Post Grinder mounted to the table with shaft
in a Vertical position. A bracket was custom made for the job.
With the Spindle running slow we programed the Z and X axis
movements to both dress the grinding wheel then grind the taper.
It worked great. As the #40 taper is a Standard the dimensions
were easily obtained. Bluing was used to check the fit to a new
tapered holder. If yours is not real bad I'd try lapping it like
you said as you would have nothing to lose. If you are able to
machine a accurate lap from a piece of cast iron it should work.
spaeth

Spent some time tonight playing with a bunch of different collets and end mill holders, both old VN and new Chinese. Big differences in runout. The low end was about .0005, the worst was about .005.
Best of the lot was an old 1/2" end mill holder. I am now going through all the collets and end mill holders to separate the good from the bad.
I think I will buy a new Hardinge 1/2" 5V if I can find one-
Before I start meddling with the spindle , I better have a known quantity to measure.

stoneaxe said:

Spent some time tonight playing with a bunch of different collets and end mill holders, both old VN and new Chinese. Big differences in runout. The low end was about .0005, the worst was about .005.
Best of the lot was an old 1/2" end mill holder. I am now going through all the collets and end mill holders to separate the good from the bad.
I think I will buy a new Hardinge 1/2" 5V if I can find one-
Before I start meddling with the spindle , I better have a known quantity to measure.

Click to expand...

Good thought - as you know you weren't measuring the runout of the spindle taper, but rather the runout of a system that includes the spindle, the toolholder or collet and the tool itself. Lots of chances for error to stack up there. Is there a reason you can't indicate the taper directly?

I don't know if VN C/5V tooling can be inserted in more than one orientation, but if it can, you should try different orientations of the tool in the spindle to see if you can get errors to cancel. If you find a good arrangement, you can mark the spindle nose and toolholder so you can repeat it (nail polish is good if you don't want the permanence of punch marks).

I've thought about this from time to time. One thing you might consider is making the lap with a pilot that is a close fit to the cylindrical part of the spindle, so that any pumps on the ID of the taper can't push the lap out of line.

How are you planning to cut the taper on your lap?

Cal

I will put my most accurate VN endmill holder in the lathe, and set the compound to the taper using a .0001 indicator. then I will double chh a couple others. Then I will turn the brass lap, matching both the cylindrical size and taper. Then I will blue the lap and see how close the spindle is, then go to lapping compound. This will be done by hand to start, then, if it cuts too slow, the lap will be driven by a low speed cordless drill using a short length of hose to drive the lap, so as not to introduce side forces.

This evening I put a .0001 indicator on the taper directly, of course it had a little angularity to get a reading, but there seems to be about .0003" difference from the high side to the low side. I am hoping that knowing the high side, and turning the lap to a slip fit in the cylindrical portion of the spindle, will allow careful hand work to bring it in.

I ordered a 1/2" collet today from Hardinge- $117, 18 days to make it. They were pleasant and professional people to deal with. I figured the collet, which will grip any number of other tools like a chuck , end mills etc, needed to be accurate. Absolutely no sense worrying about a spindle if the tool holders are out of whack.

The other option here is to grind the spindle, by using a die grinder in a vise on the mill table, and setting the VN head to a 15.25 degree (or whatever it is) angle and using the Y axis to feed the grinder, and the Z axis to control depth of cut. Exact centering of the stone becomes critical here, or the angle will be off. My main worry though is the fineness of the adjustment on the Z axis-it would be easy to take to deep a cut.

Runout

stoneaxe said:

So along with the positioning question I have been spewing forth, I have another-

The VN spindle seems to have about .0005- .001 runout, measured on a 1/2" ground shaft inserted into a collet. The measurement is taken an inch down from the collet.
When I got this machine, the spindle had a bunch of built up material in the taper, like would result from spinning an unhardened collet or holder. So I used a steel scraper and got it out.
Any collet or end mill holder used in the spindle ends up with a ring of "fretting" marks around the taper , about an 1/8" wide and located close to the small end of the taper. This wear cannot be felt with a fingernail. Nor can any burrs etc be felt in the spindle taper.
I am thinking about making a brass lap, with the VN taper, and using a fine diamond compound to very gently lap the spindle, checking with bluing and a new collet. The object to remove any tiny high spots left on the taper ID.
Is this nuts?

Click to expand...

If I understand your assuming the collet holding the ground shaft is running true on center. It in all likelihood is not. Turn a mandrel on centers or a spud mandrel and polish to a ring fit for your collet and re-inspect.

Roger

I made a VN #2 to R-8 sleeve for the 22L at the museum, years ago. I laid the head horizontal and adapted the compound from the SB10L to mount on the table. Dial the compound in to the taper angle and have at it. Put your toolpost grinder into the compound. I made an adapter for a Dremel and used that to grind the taper. Worked.

Make your lap so it just makes contact in the area where your toolholders show contact, or just put the lapping compound in this area. If the mouth of the taper is good don't waste your time trying to rework it. If you remove a little to much material at the small end of the taper you will never notice the performance.

Check the key.
My VN6 has runout if I don't seat the collets very carefully over the keystock in the spindle.
This was improved by fabing a new key. But, I still need to be careful about seating the collets and watching how they draw in.

Daryl
MN

Uglydog said:

Check the key.
My VN6 has runout if I don't seat the collets very carefully over the keystock in the spindle.
This was improved by fabing a new key. But, I still need to be careful about seating the collets and watching how they draw in.

Daryl
MN

Click to expand...

Yes, that makes good sense. I have been watching to make sure the key does not bind- also put a .0001 indicator directly on the taper and it is out a couple tenths.