What's new
What's new

Regrinding spindle taper on VN 12

stoneaxe

Stainless
Joined
Mar 2, 2010
Location
pacific northwest
The old VN 12 has a out of center spindle taper.

.0006" TIR, measured on the taper. This translates to about .002", two inches from the spindle nose.

If you are not familiar with the VN12, the taper is a short 15 degree (30 degree included) land about 3/8" wide.

I am a complete novice to grinding operations. To hire this done is way more than the machine is worth.
Hence-

Plan is to incline the head to 15 degrees, check exact taper parallel with the table, clamp up a die grinder and use the table Y axis to feed the grinder in.

The taper also has a slight bellmouth to it, as bluing on all the toolholders shows the same band midway up the taper. This bell mouth is so small a .0001" DI will not pick it up- but the blueing does.

I have the die grinder clamped up parallel with the table, the stone is dressed parallel to the table, and I was going to left -right center the grinder in the bore, so the taper is correct.

The plan is to blue the taper, and run the stone in stationary, till it just scratches the blue- due to the bellmouth, this should also serve to check exact parallel, as the bluing should scratch halfway up the taper.
Then to set up the grinder a few thousandths low, and bring it up by .0002" at a time, till I get a full grind on the rotating spindle.

Is a 1/4" diameter stone going to work OK to take off .001? The stone is Pferd, 60 grit, style A24. Plan is to run it at whatever speed the die grinder runs smoothest, up to 30K. I was not planning on any coolant as the stock removal is so slight. Also have a few Grier 1/2" diameter stones, maybe 60 or 80 grit, not sure. Which should I use?

I am confident I can hold the geometry, and move the table up by .0002 increments, (this has been checked using a .0001 DI and a long bar through the elevation handle) but I don't know anything about grinding, removal amounts, grits, finish, etc.

Advice? Reassurance? Instructions to report to the asylum?

Thanks, stoneaxe.
 
Advice? Reassurance? Instructions to report to the asylum?

I'd want to borrow the compound off a lathe and block it to align so I could lock all mill travel and didn't have to move the mill's table at all. Any axis.

Makes picking-up the angle easier and with a lighter-touch, too, does it not?

Otherwise, so long as you are careful, and the spindle BEARINGS are sound, seems OK to me.
 
Seems reasonable, but in application you may find that most die grinders (all?) do not have the bearings needed for this precision. Especially trying to hold tenths. You may end up with a worse spindle than you have now. Machine a taper on your lathe and try your setup if you cans hold tolerance and prove it, then take a big breath carefully try your spindle. Good luck.


Sent from my iPhone using Tapatalk
 
I'd want to borrow the compound off a lathe and block it to align so I could lock all mill travel and didn't have to move the mill's table at all. Any axis.

Makes picking-up the angle easier and with a lighter-touch, too, does it not?

Otherwise, so long as you are careful, and the spindle BEARINGS are sound, seems OK to me.

The mill table will have to do this case, the only compound I have is wimpy! I have scraped in the ways and it is smooth and easy - hopefully the mass will be helpful in damping any vibration. The head on the VN has a large and accurate scale , getting it close and tapping it in with a copper hammer has got the angle just right- I used a DI .0001 scale on the table,running in and out against the upper taper, and there is no movement at all. I fixed the dressing point to the overarm and ran the stone into it, so the stone should be exactly parallel with the table travel and taper. Since the taper does have a bell shape, if the first pass cuts the center of the taper, that will be the tell.

The bearing on the VN are 77 years old, tapered roller IIRC, when the machine was gone through I checked clearance on them- conveniently, the entire head is set up to bolt to the table to ease inspection and setting bearing end play. There is a question- There was a small clearance specified when the head was cold, probably best to run the machine for a while to get it warm?
The grinder bearing are unknown, it is an older Makita, but certainly not a toolpost grade tool. Some have had success with these, and some think disaster awaits...
 
Seems reasonable, but in application you may find that most die grinders (all?) do not have the bearings needed for this precision. Especially trying to hold tenths. You may end up with a worse spindle than you have now. Machine a taper on your lathe and try your setup if you cans hold tolerance and prove it, then take a big breath carefully try your spindle. Good luck.


Sent from my iPhone using Tapatalk

Hmmm. That is an interesting idea-
 
You could try turning it with a CBN insert, if needed.
Theres a walk through here using the same head kicked over approach. With fresh scraped ways might be nicer to use the machine over a compound etc.
Spindle taper repair

Cheers
D
 
...to run the machine for a while to get it warm?
The grinder bearing are unknown, it is an older Makita, but certainly not a toolpost grade tool. Some have had success with these, and some think disaster awaits...

My "Precise" brand grinders are the older generation, Variac powered. Both spec a 15 or 20 minute slow-ramp & warm-up for the benefit of THEIR bearings. The McGonegal J-35 isn't meant to be put straight to work, cold, either.

Makita? Pass. No exposure.

You might be ahead even with an air die grinder of the sort I use on the pantograph engraver.

Not because its bearings are worth a damn.

Because the error averages out faster at uber-rpm levels!

CBN turning could be far the better starting point?
 
As far as air grinders go the snap-on ones are the smoothest I’ve found with industrial ingersol with extended reach body’s. Spent way too many hrs porting heads with air grinders before I started cnc. Had to have both hands fixed when I was 45 ( more than a little while ago ).
This would be for normal available air die grinders not your tool and die specials.


Sent from my iPhone using Tapatalk
 
My "Precise" brand grinders are the older generation, Variac powered. Both spec a 15 or 20 minute slow-ramp & warm-up for the benefit of THEIR bearings. The McGonegal J-35 isn't meant to be put straight to work, cold, either.

Makita? Pass. No exposure.

You might be ahead even with an air die grinder of the sort I use on the pantograph engraver.

Not because its bearings are worth a damn.

Because the error averages out faster at uber-rpm levels!

CBN turning could be far the better starting point?

I considered using a boring bar, but there were some factors that discouraged me-
first, my lack of experience!, second there are two drive dog slots and a keyway interrupting the cut which concerned me, and third, I was really unsure about being able to take off such a minute amount.

Your comment about error averaging out-- now I don't know anything about probability theory, but is it possible for error to concentrate in one spot, given an unfortunate convergence of spindle and grinder rpm? Something that sets up some sort of harmonic bounce?
 
I considered using a boring bar, but there were some factors that discouraged me-
first, my lack of experience!, second there are two drive dog slots and a keyway interrupting the cut which concerned me, and third, I was really unsure about being able to take off such a minute amount.

Your comment about error averaging out-- now I don't know anything about probability theory, but is it possible for error to concentrate in one spot, given an unfortunate convergence of spindle and grinder rpm? Something that sets up some sort of harmonic bounce?

Yes, but it would not be "one spot". It would be a pattern, evenly distributed - think knurling - . Not a worry because the speeds in this case are so greatly disparate, nor are they in sync, one revolution after another.

"Averaging out" is that even if the mounted stone had a pronounced runout?

As you work to final-final, only the high spot is contacting the spindle!

:)

Given understanding, skill, and the patience of desperation, poor tools can produce better tools.

As they have done.

The lathes in use before Sumer had their little water problem were no match for a South Bend, nor a South Bend for a 10EE or a Nebel Microturn.

:)
 
"The patience of desperation"- I resemble that remark!

I am a woodworker by trade, the whole machining thing started when I needed a simple plug turned for a motorcycle crankcase, and it was going to be a three week wait for my local guy to have time to do it. So I bought a lathe, and pulled a dirt encrusted VN out of a barn, and took it apart, and scraped the ways and replaced bearings and on and on and here I am, having crawled up to the 101 freshman level of first day in class. I am in awe of real machinists. The skill set and problem solving abilities are amazing.

The thought occurs that rather than a generic "make it better", it would be useful to have a goal for runout.
There is .0006 TIR on the taper, .002" two inches beyond the spindle nose with a good collet . Bringing that to .001 is a 100% improvement. Perhaps truing the taper to improve contact and reduce fretting is just as important as runout.

Should the drawbar be tightened with a torque wrench to some specified level?


All you guys- thanks for the help!
 
I do not know why you would intentionally start .002 below the centerline. If you want to grind a true taper, you need to be on the centerline.

I would want the table elevation locked. The small .0002 you see with the indicator could be the table tilting. A sensitive level on the table should let you know if that is occuring.

I would not assume the position of the stone at rest is going tho be the same as if it was running. Ithis could be tested using the same set up you currently have but with a piece of steel in the collet. Grind a small half moon, turn off the grinder, blue the ground portion, run in the grinder and see what you find.

Yes- you want the spindle warmed up- both when making your measurements and doing the grinding.

Last- but by no means least, measure the run out with all your collets, and see if you get the same results. If the spindle has a key for the collets, try removing that, and repeat the measurements with the collet rotated 90, 180 and 270 degrees.
 
Qt J.R. :I would not assume the position of the stone at rest is going tho be the same as if it was running.
Or when it is loaded with some grinding pressure on the grinder and the holding fixture.
Or Is Y travel is good for .0002.
I have polishing rouged up a good taper holder and lapped a touch-off spindle better. Guess I might try that first.
 
I do not know why you would intentionally start .002 below the centerline. If you want to grind a true taper, you need to be on the centerline.

I would want the table elevation locked. The small .0002 you see with the indicator could be the table tilting. A sensitive level on the table should let you know if that is occuring.

I would not assume the position of the stone at rest is going tho be the same as if it was running. Ithis could be tested using the same set up you currently have but with a piece of steel in the collet. Grind a small half moon, turn off the grinder, blue the ground portion, run in the grinder and see what you find.

Yes- you want the spindle warmed up- both when making your measurements and doing the grinding.

Last- but by no means least, measure the run out with all your collets, and see if you get the same results. If the spindle has a key for the collets, try removing that, and repeat the measurements with the collet rotated 90, 180 and 270 degrees.

The.002 referred to is having the grinding stone .002 below the spindle taper contact, on a vertical centerline axis- so as to give some clearance if the stone does not run true. I will definitely lock the knee when grinding.

The .0006 runout is measured right on the taper itself, and I have tried many different collets and end mill holders, including newly made Chinese, old VN, shopmade, brand new Hardinge , etc- they pretty much all show the same trend with minor variations - ie the runout may vary from .001-.003 or so, but it is always at the same point in relationship to the spindle. I have removed the key, no difference.
 
Qt J.R. :I would not assume the position of the stone at rest is going tho be the same as if it was running.
Or when it is loaded with some grinding pressure on the grinder and the holding fixture.
Or Is Y travel is good for .0002.
I have polishing rouged up a good taper holder and lapped a touch-off spindle better. Guess I might try that first.

When I got the mill, it had deposits welded in from spun toolholders. These I was able to scrape out with a hard steel tool- the spindle is so hard the tool would slide on the surface yet cut the deposits. That was a huge improvement!
Later I did some lapping, but the off center bore is not improved, nor did I expect it to be- lapping will remove a high point, but just follow an out of center condition on an otherwise round bore- unless there are methods to do this?
 
When I got the mill, it had deposits welded in from spun toolholders. These I was able to scrape out with a hard steel tool- the spindle is so hard the tool would slide on the surface yet cut the deposits. That was a huge improvement!
Later I did some lapping, but the off center bore is not improved, nor did I expect it to be- lapping will remove a high point, but just follow an out of center condition on an otherwise round bore- unless there are methods to do this?

not if you tilt the spindle at 15 degrees and lock the lap to the table and as the lap wears you bring the table up. you could easilly remove .0oo3" of metal with a blob of steel held in a vice and grinding compound.
 
My experience with CBN is limited but I had no issues. Its available in interrupted cut grades, have cut hard jaws like butter.
Frank used a carbide boring bar that helped on the rigidity side of things im thinking. If the finish is a little squirly by the keys you could lap for final finish. Id go with diamond rolled into a lap and used with a little mineral spirits over lapping compound, the lap keeps its form better as the cut is concentrated on the work side, and youve a tool to clean up tapers.
So:-
Bore for form and concentricity.
Lap for final fit an finish.
The result should be about as good as the bearings allow.

Cheers
D
 
You might tighten up the clearance on them Timken bearings on the spindle a little before doing this operation. Just don't run the spindle over 75-100 RPM. Once done, go in and readjust your spindle bearings back to what they need to be. Really, if adjusted properly, there should be no slop in them to begin with.
 
What you're proposing has been done before with pretty good success. There is at least one youtube video online showing it being done. It's not going to be professional spindle re-grind good, but it works for touching one up.

I've considered doing it myself, but my route would be to use one of the widely, pretty affordable and available CNC spindle motors instead of a die grinder and make a nice plate mount for it to mount to the table. When I was curious about it, I chucked up a cheap die grinder I had and checked the runout. NOPE. And I don't want to have to run air for it. The spindle motors are in the ballpark of what a good quality air die grinder costs.
 
While searching on this site, I found this- Doug made a new spindle for his VN12. His was wartime vintage like mine. The part that concerned me was this-

"I took the original spindle down to the heat treat shop to get their opinion. They hardness tested it and we were both surprised at the results. Basically the hardened skin on the factory spindle was so very thin that the diamond simply broke through and never would give a reading over 25C."

If I grind through that thin skin it would be bad. Starting to wonder if I should just keep living with the .002 runout.
 








 
Back
Top