VMC Mill spindle runout

John

As an idea of the cost of the repair I had done to my Hayes Diemaster spindle that was very bent, if I recall correctly runout was about .015" four inches from the spindle, how it got bent I can't imagine as it was a chunky spindle. Spindle Services charged around £300 about six or seven years ago, which may sound a lot, but the work carried out was, Straighten the bent spindle, regrind the bearing seatings on the spindle to enable a more easily available metric bearing to be fitted, for the same reason bore out the quill housing,supply and fit new bearings,adapt to take a modern garter oil seal, regrind the 2mt internal taper running in its own bearings, quite a lot of work all done to a very high standard..

The original imperial bearings were only available from a specialist in obsolete bearings at a cost not far short of the charge for the whole job, so for me it made economic sense. In your case I'm sure the job would be much more straightforward. Fortunately mine wasn't an expensive machine but it was a shock to find the bent spindle, and the state of the bearings where they had been running out of line for years was horrendous.

I'm sure yours won't need such drastic measures.well worth checking the bearings as advised by Jason above before pulling it to bits.

Dave

It might be worth just first checking the preload on the spindle bearings, although you confirm they are smooth and quiet. If the spindle is floating about a bit then this would show up on the outside diameter as well as the taper, a slight adjustment of the preload may eliminate the excessive run out you have.

I have once machined an INT 30 milling spindle taper in situe on the mill, with the head tilted over to the correct angle, using a boring tool secured in the milling vice. Only as a last resort, and it worked well.

I can vaguely remember removing the quill and spindle assembly from my Warco VMC many years ago. I'm sure once the quill handwheel and pinion shaft and spring are removed, together with the depth stop, the quill will slide out the bottom of the head. The spindle has two castelated lock nuts for adjusting the preload, as far I can remember. If the gear rack teeth cut on the rear of the quill continue to the top edge of the quill then the pinion shaft may not have to be removed.

The very early Warco VMC mill that they offered in the 1980's was identical to the Taiwanese made Myford VMC, but in the 1990's the Warco VMC was no longer the same machine, using different castings and other detail differences, although overall it looks the same, and has similar dimensions and capacity. The Warco VMC I have was made in Taiwan in 1997 and is the version that has now been produced in China for some years.

Anyway, prompted by this thread, I have just been in my workshop to measure the run out on my Warco VMC R8 spindle. I get a reading of 0.0002" run out on the lower portion of the spindle taper, and no change further up the taper.

Although the Taiwanes made Myford VMC is generally made to a higher standard than my later differing Taiwanese made Warco VMC, I can recall a conversation I had while on a visit to the original Myford plant at Beeston in Nottingham, around the mid 1990's I think. While being shown some machines hidden away in some back rooms I spotted a new Warco VMC that they said they were examining, I think because the Myford VMC was being discontinued. They didn't really like the way the Warco VMC was made, but they did admit that they did have to put a stop to the Myford VMC manufacturer's method of "finishing" the spindle taper by inserting a piece of wood wrapped in sand paper.

Edited By Lathejack on 13/11/2022 23:31:07

Edited By Lathejack on 13/11/2022 23:37:27

John

I don't know if it will be any use to you but you haven't said whether you have a diagram of the headstock/spindle but here is the one for mine. As I said earlier mine was made in Taiwan same as the Myford so may be of some use.

John

Yes I would definitely inspect the spindle and all bearings and seatings very carefully before any regrinding etc. You will need to put the spindle between two V blocks and rotate it with a dial indicator at various locations to check for straightness.

Bearings should be checked for runout too. And check what brand of bearings it has in it. If no name or cheap Chinese fare, replace with SKF or similar quality. (FAG, Timken, Koyo, Nachi)

A useful diagram of the milling head from John Purdy, looks impressive with five bearings supporting the spindle. It does appear to be a diagram of the slightly larger turret mill, that uses a quill of the same diameter, about 86mm I think, for its full length.

The VMC mills such as the Warco have a 75mm diameter quill that increases to 100mm diameter at the bottom, and the Myford VMC quill looks similar to this although the head casting and depth stop are different.

I've just had another closer look at my VMC quill through the depth stop slot in the front of the head casting. It looks to have one castelated nut sat on a tab washer above the top bearing to secure the spindle.

 I have owned this mill for about 24 years and it is a long time ago when I removed the quill. I didn't remove the spindle but i think it uses two bearings in the bottom 100mm diameter of the quill, but I am not sure if there are two in the top or just one. They are not taper roller bearings, they are ball bearings of some type and I am not sure if they are designed to take preload or not.

Edited By Lathejack on 14/11/2022 10:39:18

Edited By Lathejack on 14/11/2022 10:46:28

Further to my last and to elaborate on Lathejack's post my quill is a constant 80 mm in diameter. The top two bearings supporting the pulley are 6208 radial ball bearings. The spindle runs in two 6206 radial ball bearings at the top and two 7207B annular contact bearings at the bottom preloaded by the ring "M108" in the diagram. Although the diagram and parts list shows three 6007 radial ball bearings at the bottom, when I disassembled the spindle I found that the design had been changed to the two annular contact bearings. You will see that I have hatched out the top of thee bearings at the bottom of the spindle and amended the parts list.

John

Edited By John Purdy on 14/11/2022 18:14:40

Thank you all for your further replies. I have now managed to strip down the quill and spindle which was indeed very straight forward. Placing the spindle between centres on the lathe I am getting in the range of .0075-.01mm runout in the centre. I am getting the same when testing with it on two vee blocks. (Supporting it on the lathe on a centre in the spline and and on its bearing on the other I was getting pretty much no run out in the taper.). I have therefore concluded that it is indeed very slightly banana shaped.

Spindle services prices were reasonable for a simple taper regrind but climbed to well over 1k if doing anything involving bearings too.

Any thoughts on straightening this fairly small amount without machining would be welcome. New bearings are on order.

It may be worth a call to Warco to enquire the availability and price of a new spindle. Our maintenance shop had some magicians at straightening shafts that had been bent, it may be worth a try if you have access to a press but probably not a job to learn on.

Mike

You need to support the appropriate bearings in vee blocks and rotate the spindle checking runout to get an idea of what is going on, be very careful how you check things.

Tony

Edited By Tony Pratt 1 on 15/11/2022 19:45:59

John D.

The quill assembly on your Myford VMC is a very similar design to the quill on my later type Warco VMC. Now you have dismantled it, just out of interest, what type of bearings does it have? Does it have two bearings in the bottom housing of the quill, and how many in the top?

The rather poor manual that came with my Warco VMC claimed that it had two bearings in the bottom, if I remember correctly, but the manuals are not always accurate.

Hi Lathejack

starting from the bottom up:

At the bottom end of the quill

1 x 6207zz (confirmed on removed bearing)

1 x 2907 thrust (from myford manual)

at the top of the quill

1x 6206zz (confirmed on removed bearing)

and then, I think, under the pulley

2x 6009zz

Edited By John D on 15/11/2022 20:57:55

I don’t have access to a press but think that may be my best option. I just need to find someone with a press near me (Herts/Bucks).

Tony, you make a good point about how I measure. How should one get bearings on and off a shaft without damaging things? I could heat the bearings or chill the shaft I guess.

Edited By John D on 15/11/2022 21:13:53

John, you are now talking about less than .01mm runout?

Tony

Not even sure if the OD in the middle of the shaft is something you can trust for any measurement. Looking at the drawing above it could be possible that the middle is not finished to the same standard as the two diameters that the bearings fit on.

Also can't see how 0.01mm deviation in the middle equates to 0.035 at the socket end if it is just a bend particularly as the bearings are placed towards the ends.

I think using a press on your spindle is about the worse thing you could do, you quote 'I am getting in the range of .0075-.01mm runout in the centre', I'm not really sure of your checking method but that is damm good on a maybe unimportant diameter as per Jason B. Just wait for new bearings and rotate the spindle in them with the installed bearings on vee blocks, you can then check the R8 taper.

Tony

The critical items on the shaft are the bearing locations, or registers and the R8 socket. The relationship of these items is what matters.

Checking between centers on the central portion between the bearing registers proves nothing.

To check this properly, place the bearing registers on vee blocks, (you will need to pack one vee block due to the difference in diameters), and clock the R8 taper off these. An alternative would be to mount in the lathe using a fixed steady on the front bearing journal. The splined end held in a 4 jaw is then adjusted to get the other bearing register running true. Once this is done the R8 taper can be clocked.

I would also be inclined to check for a cracked spindle at the cone end. Over tightening of the R8 collets can cause this to happen. Especially if the hardening process on the spindle nose has not been carried out correctly.

Regards

Gray,

A big +1 on what Graham Meek said above ^^^^^^

Only the bearing register surfaces and the R8 taper surface would usually be precision finished at the factory, probably ground. The area in the middle of the spindle could well have been left "rough machined" and not precision ground because is does not contact anything, It just sits there in space. So runout there is ok.

Graham's very handy suggestion of setting the spindle up in the four jaw and fixed steady so that both the bearing register surfaces are running dead dead true, then clock the taper is a much more stable setup than trying to rotate the spindle on V blocks that are not bolted down and clock the taper at the same time.

Then atfer this above test has shown the spindle is (hopefully) running true, when you get the new bearings and install them, I would set it up similarly with one bearing held in the chuck and the other in the fixed steady and then rotate the spindle and clock the R8 taper. It should run dead true if all is well.

Installing the new bearings is usually done using mild heat to expand them so they slip easily onto the spindle. Heating them up in a tin of oil over a flame is the traditional way. An electric heat gun can do the job too, with care. If you do have to gently tap the bearings onto the spindle, support spindle on block of wood and use a tube etc so you are tapping only on the inner beraing race and not transfering shock through the balls.

I would then do a final test with the spindle and bearings installed in the quill and lay the quill down or clamp it down to something or hold it in the four jaw and fixed steady again etc and clock the taper as the spindle turns in its bearings in situ. If runout mysteriously shows up at this stage, it must be due to bearing installation not being quite right somehow. Very unlikely though. But best to check anyway.

Definitely leave the press alone at this early stage!

This video on spindle rebuilding might be of interest:

One thing I forgot to mention in my above post was that there are two diameters to check with an R8 taper not just the cone, or tapered part.

Should the cone be running true and the inner register be running out then this is down to one of two things.

a) The R8 register was never machined on the true centreline of the spindle.

b) There is a bend in the shaft.

This latter condition would show up using a good clock, (0.002 mm or better) and the part set-up in precision Vee blocks as a run across the bearing register. This run would be along the spindle centreline and the check would only be performed on a static shaft.

Rotating the shaft through 180 degrees should show this defect up but in the opposite sense, (Think Banana shape).

regards

Gray,

Some good news. I think I have now reduced runout to a workable level.

I cleaned everything, and ordered new bearings. When testing with various methods and setups I was clocking run out at .005mm on the bench. Fitting the lower bearing to the shaft was difficult. Even having heated the bearing up to 80 degrees C it was still difficult getting it onto the shaft. I drifted it on and it took more effort than i would have liked (from the point of view of risking damage somewhere). I used SKF bearings other than for the thrust bearing where I used NSK as this was readily available. The thrust bearing was subtly different in that the one of the 'washers' was an tight fit on the shaft whereas the other just slipped on. I put the tighter one in the lower position (i.e against the face of the lower bearing) as I assumed that this one should move with the shaft whereas the other contacts the step in the quill.

I tightened the nut the at the top of the quill so is to created just the slightest amount of resistance when turning the shaft by hand in the quill. I then backed it off and re tightened it once again slightly less to again just the start of perceptible resistance. Is that correct? How would i know if it was too loose or tight?

All done I am now getting .01mm runout measured at various points on the lower taper. I don't have a long enough knob piece for my indicator to measure the upper surface on the R8 at this stage. Next step is to use a collet with a ground bar and measure 200mm down from the nose and see how things are looking - but I am really pleased at this stage.

Doing my time again I would have heated the bearing for longer (in the oven) to more like 90 degrees C before fitting to see if that reduced the problems fitting the bearing to the shaft and therefore reduced the chances of introducing some runout at that stage by rough handling of the bearing while drifting it on. I'm not sure its worth stripping again and buying new bearings to find out.