spindle run out

Advert

New Replies

Report Bug

Help/FAQs

spindle run out

This topic has 28 replies, 8 voices, and was last updated 5 October 2010 at 10:23 by Nicholas Farr.

Viewing 25 posts - 1 through 25 (of 29 total)

1 2 →

Author

Posts
28 September 2010 at 22:21 #5201
Ian MILLARD
Participant
@ianmillard76204
Advert

28 September 2010 at 22:21 #56227
Ian MILLARD
Participant
@ianmillard76204
HI I have just measured the run out on my raglan five lathe. With a dti on the machined shoulder where the chuck locates there is a difference of 2 thou as the spindle rotates, the end of the spindle is about 3 thou. Is this normal or do i have a problem. Thanks IAN
29 September 2010 at 02:34 #56232
John Olsen
Participant
@johnolsen79199
Sounds a bit much to me. Is the spindle itself out of true, or do the bearings need adjusting? If it is the bearings you will be able to change the reading by pulling the spindle back and forth or up and down. I just found recently that I had to adjust my ML7, and that was at least part of why I had started having trouble with parting off. Mind you, I think that might have been the first time since it left the factory some time around 1953 or so.

regards

John
29 September 2010 at 09:32 #56238
KWIL
Participant
@kwil
Quoting Tubal Cain from ME 4 Oct 1996 Centre Lathe Inspection Limits

Chuck Seating Register Schlesinger Limit true TIR for 3.5″ Model Engineer’s lathe Allowance 0.0005″ ACTUAL 0.00005″

I think you have a problem to solve.
29 September 2010 at 16:51 #56254
Dinosaur Engineer
Participant
@dinosaurengineer
It would be difficult to do accurate work with spindle run-out this large .
29 September 2010 at 22:06 #56275
Jeff Dayman
Participant
@jeffdayman43397
As John Olsen said above – the spindle bearing clearance should be checked here first. .002 to .003″ spindle runout is a lot, but if the bearing clearance is OK, even a running – out chuck will be able to turn good accurate parts as new work if the job is done in one setup and not removed and re-chucked. Work between centres will also be possible but the cut diameters will not be concentric to the centre drilling. If the spindle is bent, mounting drills or other tools in the spindle will result in a flailing drill / tool and ovesize holes (if the drill stays in the spindle at all.)

Check the spindle bearing clearance, then address the runout later. If everything else on the lathe including bearings is OK, it might be worth making or buying a new straight spindle.

good luck JD
29 September 2010 at 22:40 #56276
Anonymous
My gut feel is that 2 thou runout is quite a lot. If I understand it correctly from http://www.lathes.co.uk the Raglan ‘Five’ has taper roller bearings. Just out of interest I’ve been out in the workshop to measure the runout on my lathe nose spindle. It’s a D1-4 fitting, and the spindle runs in opposed taper roller bearings.

Measurements were made using a Mitutoyo 0.01mm DTI. Measurements are:

Inner MT5 taper: 0.005mm

Short Outer Camlock Taper: 0.007mm

Outer Diameter (plays no part in locating the chucks): 0.01mm

Front Face of Spindle: 0.01mm

These measurements are from a 30 year old lathe.

However, before we panic and start stripping the spindle it might be worthwhile discussing the measurement technique, so as to be sure that the runout you are seeing is genuine. For instance did the needle move smoothly as the spindle was rotated, or did it move in jumps? If the needle jumped is there any corrosion or pitting on the spindle. Were the readings repeatable from one rotation to the next?

Regards,

Andrew
29 September 2010 at 23:15 #56278
John Olsen
Participant
@johnolsen79199
Since I had just adjusted my ML7 recently, I just put the indicator on it to see what I can see. (couldn’t remember what my checks at the time said, except it had about 4 thou of play in the front bearing beforehand.) With the indicator reading off the register, I see a total indicator runout when I rotate the lathe by pulling on the motor belt of less than 0.005 mm. That is interpolating, since the devisions on the dial are 0.01 each and would be about 1and a half mm wide on the dial.

When I put a shaft into the taper and pull, I can get about 0.02mm of total movement either up and down or front and back. But these are plain white metal bearings and there has to be some clearance for oil. When I adjusted it I took one too many shims off and it was too tight to turn. So I think I have it about right now, it is certainly parting better than it was for a bit there.

The dead centre could be trued in place. It would then have to be marked so it can be replaced in the same position. If the chuck backing plates were turned with the spindle off centre they will work OK too. (If they are the type with backing plates.) But I would not care to work like that…if the spindle is actually bent I would want to get it sorted. The more accurately you can work, the easier the later operations down the track are.

regards

John
30 September 2010 at 10:14 #56282
Nicholas Farr
Participant
@nicholasfarr14254
Hi Andrew, when you say your tapered roller bearings are opposed, I assume that there are two single row bearings at each end of the spindle in a back to back arrangement which is normal for a headstock. Back to back arrangement is when the cups of the bearings are towards each other, whereas a face to face arrangement is when the cones are towards each other. The differance will be the adjustment needed. With the back to back arrangement there will be zero or negative clearance (axial preload) as opposed to face to face arrangement there will be a positive clearance. Having a positive clearance in the lathe spindle which results into end float is not desirable in our lathe spindle. I’m sure you understand this, I’ve posted this for those who do not realise the differance.

Regards Nick.
30 September 2010 at 17:43 #56290
Ian MILLARD
Participant
@ianmillard76204
Thanks for the info. I hadnt tightened the spindle bearings up enough, with these tightened further and a centre in the spindle taper the runout is below a thou, however when i fit the faceplate and run the lathe the whole lathe rocks back and forth is this just because the faceplate needs a balance weight attached.
30 September 2010 at 19:51 #56293
Anonymous
Ian,

Definitely doesn’t sound right. The faceplate, by itself, should be reasonably well balanced, certainly up to hundreds of rpm. What speed were you running at?

Nick,

Yes, didn’t explain myself very well. My lathe spindle has two ‘Super Precision’ Gamat bearings, one at each end of the spindle. It’s not clear from the manual, but I presume that the bearings are arranged such that, as the spindle warms up and expands, the axial load on the bearings increases. The manual has a section on how to test the torque necessary to turn the spindle once it is warm, and how to adjust the preload in small steps. It also contains dire warnings about doing so, with strict instructions to contact the factory beforehand. Unless I have an unforeseen problem I have no intention of touching the bearings.

Regards,

Andrew
30 September 2010 at 20:08 #56295
Ian MILLARD
Participant
@ianmillard76204
Andrew

I am running the lathe at its slowest speed in normal range ie not back gear. ian
30 September 2010 at 21:48 #56299
Nicholas Farr
Participant
@nicholasfarr14254
Hi Andrew, it is little long winded to expland but, in a good design the preload will remand the same at whatever temperature the headstock is, within the makers specs for duty cycles ect. as the radial and thermal expansions cancel each other out when in back to back arrangement. It’s all to do with choosing the right bearings for given loads and distance between bearings ect. ect. If one is not familar with adjusting taper bearings it is best to heed makers warnings, because too much or too little preload will shorten bearing life as well as too much lubricant.

Regards Nick.
30 September 2010 at 23:10 #56302
Anonymous
Ian,

That definitely isn’t right. I assume that low speed without backgear is about 150-200rpm. At those sorts of speeds the faceplate must be wildly out of balance to be rocking the lathe. Or there is another, as yet undetected, fault with the spindle. Does the lathe do the same with a normal 3 or 4 jaw chuck, or just the faceplate?

Nick,

Thanks for the information. I hadn’t thought about the axial and radial expansions cancelling out. The headstock itself gets mildy warm (about 30°C) after a few hours of running at 1200rpm plus. So I assume that tends to counter the expansion of the spindle too. I’ve made the assumption that the Harrison design engineers knew what they were doing and I’m going to leave well alone. I have never had any issues with accuracy or vibration that could be attributed to the spindle, so no need to fiddle. Not the real engineers’ way I know, but I believe that if it ain’t bust don’t bugger about with it!

Regards,

Andrew
30 September 2010 at 23:36 #56303
Nicholas Farr
Participant
@nicholasfarr14254
Hi Andrew, believe me, but in industry even some “real engineers” have the same policy especially if routine maintenace cost a lot ££’s or needs a lot of down time.

Regards Nick.
1 October 2010 at 13:42 #56329
Ian MILLARD
Participant
@ianmillard76204
Andrew

I think i have a dud faceplate that came with the lathe. the centre fixing hole is a couple of mm off centre, its no problem the whole thing was bucking about. Thanks for the help IAN
1 October 2010 at 15:08 #56337
KWIL
Participant
@kwil
Super Gamet bearings on the Harrison.

Sorry Nick, you have it the wrong way around. Back to Back is where the outer raceway Cups are facing outwards with the inner raceway Cones facing inwards, so tha axial preload works inwards. Check SKF website if you disagree. That is the arrangement on the Harrison which of course are not SKF bearings!!
1 October 2010 at 16:00 #56340
Nicholas Farr
Participant
@nicholasfarr14254
Hi Kwil, I havn’t got it wrong, We are saying the same thing, it’s our termanolegy that is different, mine was probaly not clear enough. When I said towards each other, I mean the other bearings elements are not inbetween. Here is a simple sketch to clarify what I mean for back to back arrangements, more commanly found in lathe headstocks than face to face.

Regards Nick.

Edited By Nicholas Farr on 01/10/2010 16:30:17
3 October 2010 at 09:55 #56424
Anonymous
Nick,

Thanks for the sketch. That’s just how I imagined the bearings in my lathe would be orientated. It always amazes me that the lathe spindle and bearings are subjected to all sorts of cutting forces, yet retain an accuracy measured in tenths.

Regards,

Andrew
3 October 2010 at 10:11 #56425
KWIL
Participant
@kwil
Nick,

We are in agreement! Good quick sketch for the uninitiated. I used SKF terminology.
3 October 2010 at 10:17 #56426
ady
Participant
@ady
If you can run the faceplate at low speed and it fits the spindle nose well you can skim it until it runs true.

Some faceplates have a “middle bit”(spindle adaptor) and are not a single one piece casting, if this is the case the middle bit probbly needs adjusting.

Suss it out before you start any cutting.

Edited By ady on 03/10/2010 10:18:32
3 October 2010 at 10:17 #56427
KWIL
Participant
@kwil
Andrew,

You only have to look at the precision with which the bearings are made to understand how they serve us so well. The fit of the cup of the Super Gamet bearings on the shaft is not just a slide-on fit, you would need to heat the cup or apply hydraulic pressure to make it move, although the adjusting ring will move it, do not expect it to be easy the other way around.

KWIL
3 October 2010 at 11:05 #56430
Anonymous
Hi KWIL,

I have a basic appreciation of the accuracies needed for precision bearings. Far beyond anything I can achieve at home, and probably explains why precision bearings cost well into the hundreds of pounds. There’s an interesting note on the Gamat website about fitting bearings. They do not recommend mechanical methods, but gentle heating of the inner race.

At one place I worked, there was a need to build a test rig for electric motor systems. This was basically back to back electric motors, one the custom test motor and the other a commercial item. Once running, the system was very noisy. I refused to work in the same room unless I got ear defenders. Then the bearings failed. Being nosy I insisted on being involved in the post-mortem. Turns out the mechanical designers had fitted the ball bearings onto the shaft by hitting the outer ring, doooh! I took one of the failed bearings home and cut it open. It was easy to see where the balls had left small indentations in the rings and the subsequent break up of the bearing surface.

Regards,

Andrew

Edited By Andrew Johnston on 03/10/2010 11:17:02

Edited By Andrew Johnston on 03/10/2010 11:18:13
3 October 2010 at 14:58 #56443
Nicholas Farr
Participant
@nicholasfarr14254
Hi,
Posted by Andrew Johnson. “It was easy to see where the balls had left small indentations in the rings and the subsequent break up of the bearing surface.”

Yes, people don’t realise how easy it is to damage bearings of all types by incorrect mounting proceedures. I suppose you don’t realise they can achieve the same indentations through bad long term storeage either. They sould aways be stored in reasonabley dry condition and without vibration, which can cause indentations even when wrapped from leaving the factory.
Regards Nick.

P.S. glad you liked my sketch.
4 October 2010 at 19:50 #56479
Anonymous
Wow, sounds like they need to be feather bedded!

Regards,

Andrew
Author

Posts