Quill bearing temperature

[David George 1Participant @davidgeorge1]

In my last job as a service engineer I had a bearing heating problem and the end result was cased by the Rubber seal that had worn and with the larger contact area on the spindle there was a huge increase in temperature beyond what I believed possible, so please check the seals as well if you do dismantle. PS if you don't live to far from me I have some Kluber grease you may have. Sutton in Ashfield Notts.

[Martin CargillParticipant @martincargill50290]

As NDIY points out the IR temp guns can produce misleading results.

I recently had a customer call me out to a hot glue spreader, complaining that the doctor roll was running 30 degrees lower than the demanded temperature. He was using an IR gun to measure the temperature of a chrome steel roller. As the roller is heated by hot oil and the other rollers (rubber covered) were showing the correct temperature there was obviously something wrong. It turns out that the IR gun does not like shiny surfaces – two minutes with a temp probe attached to my fluke meter proved that the temp was correct and that the gun was wrong. It cost the customer around £150 to find out that his £25 gun was incapable of producing an accurate result on a shiny surface. IR guns are good for measuring changes in temp or rate of rise of temp but sometimes they are no use for absolute measurement.

Martin

[Neil WyattModerator @neilwyatt]

> It turns out that the IR gun does not like shiny surfaces

The instructions for my Aldi IR gun made that very clear, al;though it seem to give usable results on almost anything else… walking around the garden at night as the temperature falls below freezing seeing how different things cool at different rates (it can be boring telescope-minding)

Neil

[AjohnwParticipant @ajohnw51620]

I adjust my Boxford bearing using heat. Some others do since I mentioned it. The boxford way just doesn't work for many.

Not sure how this could be applied to a miller.On the lathe I can get a finger directly under where the bearing is from inside the spindle – which isn't running of course. I go for what I would call nice and warm. Sort of hold your hands in front of a fire warm when they are cold. It's pointless feeling the temperature of the castings on a boxford. The bearings take well in excess of 10mins to settle.

During training – much bigger lathes. One of the instructors mentioned how hot the area around the spindle on the front of the headstock gets and that it takes them about 20min to reach "operating" temperature. He reckoned that the actual bearings would be running at well over 100C. The warmth this creates is as I described. Also for best work possible some benefited from being allowed to warm up before critical work was done. They did too.

What I usually suggest is warm and see how it goes. What's not wanted is hot.

Edit

I'm pretty sure that my dore westbury bearings need adjusting based on the finish. What I intend to do is use the Boxford method and tighten until I see a slight increase in the torque needed to turn the spindle. They suggest rope round something and measuring the torque with a spring balance, One problem for me is that the only thing I can currently put on the spindle is a Myford chuck which is a bit heavy and may make taking readings more difficult.

John

–

Edited By Ajohnw on 24/12/2016 11:12:39

[John HaineParticipant @johnhaine32865]

The bearing housing end of the quill is shiny and I did get rather variable readings initially. The Maplin instructions suggest sticking masking tape on a shiny surface to make it matt, which I then did, and the readings were then fine. How accurate is another matter, I'll try making a contact measurement too.

[Ketan SwaliParticipant @ketanswali79440]

Broadly speaking, Nigel B's first post is correct.

Some notes:

– I am not so sure that the bearing at the working end of your mill is a sealed bearing. It is likely to be some kind of angular contact bearing (ball), or taper roller bearing. Have you got an exploded diagram to guide you – giving you bearing number for the bearing involved?… before you go any further, this is the first thing to check to see what kind of bearing/s are you dealing with.

– Grease/over torque possible problems. How regularly have you been using the mill before changing over to VFD?.. Is the machine old/second hand… and do you have any history about its use before the change to VFD?

– 2000 rpm is not exactly high speed. If it was a sealed bearing, and machine not really used over time, then you may have a grease problem/issue. What was the speed before, and what is it after?…So, what is the percentage increase? have you checked this?

– When you changed to VFD, did you make any mechanical changes?… I don't know your machine. Is there any belt drive to the spindle?… If so, was there any extra tension introduced introduced to the belt after the change over?

– What is the method of applying pre-load to the bearing? Are there some kind of pre-load nut/s at the top of the spindle which may have been tightened during the change over?

– As Nigel said, less grease is better than more. I would say between 20 to 25% of internal volume… provided it is not a sealed bearing. In your particular case, normal LM grease – Castol NLGI2 or similar is fine. You will be wasting money on special grease which is not really going to do anything special in your particular case. If you were doing 5000 rpm +, maybe more specialised greases would start to come into play.

All said any done, there is no point in disassembling the spindle, until and unless you know what you are dealing with to start off with. The answer can be simple or complex, and as Martin Cargill said, dismantling a spindle assembly should not be taken lightly. Changing the bearings (only as a last resort), re-assembly, torque and running in process there after are another thing altogether.

Ketan at ARC.

[John HaineParticipant @johnhaine32865]

Thanks Ketan. It's a tapered roller bearing at top and bottom of the quill, with preload nuts at the top. Yes there's a belt drive but that is to a separate pulley with its own bearing on the top of the machine. The VMB has a long splined shaft driving the milling head running through this pulley, the head slides up and down the column on dovetails. So the belt tension is not taken on the spindle at all. Having fitted the VFD I'm using higher speeds than before, but also doing more cutting of aluminium and generally being more confident of what you can do with cutters, but still within the design spec of the machine. The bearing type is not specified in the instructions but there is a cross section detailed drawing of the quill assembly. There's a good description of the VMB on lathes.co.uk that describes the construction.

(Later note.  Lower bearing is a type 30205, upper a 30206.)

Edited By John Haine on 24/12/2016 15:23:59

[Ex contributorParticipant @mgnbuk]

Lower bearing is a type 30205, upper a 30206.

"Myford Ltd" are listing P6 versions of these on Ebay – ex-Myford Nottingham stock @ £26.50 each. As use don ML10 spindles.

**LINK**

**LINK**

Nigel B

[David George 1Participant @davidgeorge1]

Hi John

From a drawing I have seen on line there are three bearings in the spindle a 6206zz a 6207zz and a thrust bearing 2907 the first two are double metal shielded and greased for life better the thrust bearing is open and needs grease. The adjustment is with a locking ring and tab washer it is a fairly simple jubilee to drop out the spindle after removing the stop and the quill shaft, leaving the pully in place and check and or replace the bearings. Just be very clean and ask if you need any help.

David

[Mike PooleParticipant @mikepoole82104]

It's amazing how cheap and available IR measurement equipment has become, in the late 70s our department was trying to measure the temperature of the plastic in a vacuum forming machine. Enquiries met with sorry that equipment is classified and not available to industry. How things change.

Mike

[David George 1Participant @davidgeorge1]

Edited By David George 1 on 24/12/2016 20:31:03

[HOWARDTParticipant @howardt]

If the questionable quill bearing design is as the above, ensure that the thrust bearing is fitted correctly. They have a shaft washer and a housing washer. The shaft washer has a sized bore which fits to the spindle, lower washer, and housing washer which has a sized outside diameter which fits the housing, upper washer. Only tighten the locknut to remove end play, don't over tighten. Grease volumes can be found in SKF literature. The failing with this design is the empty space between the bearings into which the grease from the top bearing can quickly migrate.

Howard

[John HaineParticipant @johnhaine32865]

David, I'm afraid that isn't the VMB spindle. I have the VMB drawing and it just uses two taper roller bearings.

[Ketan SwaliParticipant @ketanswali79440]

Posted by John Haine on 24/12/2016 15:19:19:

(Later note. Lower bearing is a type 30205, upper a 30206.)

Edited By John Haine on 24/12/2016 15:23:59

So, if 30205, then it is an open taper roller bearing, rather than a sealed bearing.

Provided this is the case, then read this timken page 47 on operating temperature…. you can carry on reading and get some useful knowledge.. without getting too involved with the technical.

Hope this helps you to get some clarification.

Ketan at ARC.

Edited By Ketan Swali on 26/12/2016 14:44:16

[AjohnwParticipant @ajohnw51620]

Interesting reading Ketan. My understanding for running in excess of 100C is moisture – getting shut of /preventing any that happens to emulsify into the lubricant.

I'm surprised stainless looks so good too.

John

–

[Michael GilliganParticipant @michaelgilligan61133]

The Timken document suggests that the rise in temperature that John is seeing is fairly trivial for the bearings … but it is worth noting this, brief but important, statement [from p47]:

[quote]

The equipment designer must evaluate the effects of temperature on the performance of the equipment being designed. Precision machine tool spindles, for example, can be very sensitive to thermal expansions. For some spindles, it is important that the temperature rise over ambient be held to 20° C to 35° C (36° F to 45° F).

[/quote]

I think perhaps our discussion should be focussed upon the design of the spindle.

… Could you please supply a diagram, if you have one, John ?

… We currently have only a picture of 'what it is not'.

MichaelG.

[AjohnwParticipant @ajohnw51620]

Posted by Michael Gilligan on 26/12/2016 18:26:55:

The Timken document suggests that the rise in temperature that John is seeing is fairly trivial for the bearings … but it is worth noting this, brief but important, statement [from p47]:

[quote]

The equipment designer must evaluate the effects of temperature on the performance of the equipment being designed. Precision machine tool spindles, for example, can be very sensitive to thermal expansions. For some spindles, it is important that the temperature rise over ambient be held to 20° C to 35° C (36° F to 45° F).

[/quote]

I think perhaps our discussion should be focussed upon the design of the spindle.

… Could you please supply a diagram, if you have one, John ?

… We currently have only a picture of 'what it is not'.

MichaelG.

I suspect that it's more important to take note of the lubrication. It's pretty safe to assume that as the mill uses sealed bearings it uses grease for lubrication. It probably does if they are not sealed anyway. Probably a higher temperature one if sealed for life and the fill quantities will be a lot less than mentioned later on in the pdf – p55 onwards. There is also mention of water contamination. Actually I don't think people would generally fill a machine tool spindle from new to the levels mentioned in the pdf or by Ketan. It has the effect of generating more friction and hence more heat.

It would probably be a good idea to look at a parts manual for the machine or it's manual and seeing what it has to say about bearing adjustment.

John

–

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Ketan Swali on 26/12/2016 14:42:52:

Posted by John Haine on 24/12/2016 15:19:19:

(Later note. Lower bearing is a type 30205, upper a 30206.)

Edited By John Haine on 24/12/2016 15:23:59

So, if 30205, then it is an open taper roller bearing, rather than a sealed bearing.

.

Ajohnw … for info. ^^^ in case you missed it.

MichaelG.

[HopperParticipant @hopper]

Posted by John Haine on 23/12/2016 21:00:56:…

… looked as if it would stabilise at about 50C. That sounds a bit warm to me, though a quick web search suggests SKF say that up to 80C is OK. I think I'll investigate the pre-load first.

So it's well within range for normal bearing use. (And home hobby milling spindle use at 3,000rpm is nothing special.)

+1 on check preload if you still have concerns. But, a bit of extra preload might give better rigidity and therefore better surface finish than running a bit low on preload. Be careful what you wish for.

Edited By Hopper on 27/12/2016 02:05:08

[AjohnwParticipant @ajohnw51620]

Posted by Michael Gilligan on 26/12/2016 23:55:30:

Posted by Ketan Swali on 26/12/2016 14:42:52:

Posted by John Haine on 24/12/2016 15:19:19:

(Later note. Lower bearing is a type 30205, upper a 30206.)

Edited By John Haine on 24/12/2016 15:23:59

So, if 30205, then it is an open taper roller bearing, rather than a sealed bearing.

.

Ajohnw … for info. ^^^ in case you missed it.

MichaelG.

I didn't, hence mentioning even if not sealed. I do assume though that some one posted a shot of a parts diagram that shows what should probably be in it.

Personally I will stick with what I was told on normal run of the mill spindle bearing temperature. If you look at the data given they show a drop in hardness of 2 to 3 for a huge increase in temperature from 70C but not on standard bearings even though the top 2 show the same steel. They are also talking about dimensional stability. Many bearing set ups on spindles tighten as they warm up. Measuring temperature can also be a bit tricky as the inner may heat up more than the outer due to heat conduction rates. There is likely to be a large difference between the heat conduction rate of a spindle as against a typical machine tool bearing housing.

John

–

[John HaineParticipant @johnhaine32865]

Here is a C/S of the spindle from the manual. Items 35 and 36 are the bearings. They are unsealed. There is a separate rubber seal (presumably silicone) below the lower bearing to prevent suds etc getting in (not that I use suds). Nuts 37 apply preload. You can see the splined extension at the top that engages with the drive which is at the top of the column. The extension is enclosed in a rubber bellows arrangement as in the pic below.

Many thanks for all the input. On the whole I'm inclined not to worry as (a) the bearings are OK to run much hotter; (b) the temperature seems to stabilise not continue rising; (c) the design of the spindle and head is such that the axial loading will decrease as the spindle gets hotter and expands more than the quill, which is cooled by contact with the head. And maybe the main source of friction could be the seal? So I'll carry on using and keep an eye on the temperature rise (which is easy now I have the IR gadget).

Happt New Year all!

[AjohnwParticipant @ajohnw51620]

Tell you something John. That arrangement is not that dissimilar to a boxford head stock bearing arrangement and lots of other machine's spindles. The bearings tighten as the bearings warm up and the expansion of the spindle doesn't make up for it. Good job too as the bearings would be more likely to loosen as they warm up if that sort of thing actually went on. The spindle cools pretty rapidly away from the bearings so what must actually be happening is that the inner is expanding with the spindle going with it to a certain extent.

Anyway I mentioned Boxfords way of setting preload.. This is it.

Big catch though. It generally doesn't work. Some of that will be down to wear but the most likely candidate is the one they mention – too much grease. They suggest a smear when new bearings are fitted. Something I have also come across else where. That leaves a problem. Grease needs to be added periodically and no way to get the old stuff out other than pumping way too much out. If a spindle has seals on the bearings even if on just one side this may burst them – happened on my Raglan. So what tends to happen is that eventually bearings finish up packed with a mix of fresh and well oxidised old grease full of contamination and the friction goes way up. Well past the figures Boxford mention.

I usually use moly grease but when I replace the bearings I may switch to a fully synthetic long life bearing grease. Can't make my mind up.

John

–

[Mark CParticipant @markc]

John H,

I would agree with all you have mentioned.

I might add to that by suggesting a possible modification you might be able to try; As long as the drive pulley (splined spindle driver) is supported on it's own bearings, you might be able to apply the spindle pre-load through cone springs. This would provide a more linear loading with respect to temperature change (as long as the top inner race is a sliding fit on the shaft). I would look at the side loading before committing but it does look unlikely given it is a light machine.

The spindle design (although similar to a Boxford of which I have two) is really a poor design for a machine tool. As with "all" modern spindles, the base bearing should be a double arrangement to alleviate the problems with thermal expansion with the tail of the spindle being supported by a sliding radial bearing allowing free expansion of the spindle without detriment to pre-load or cutter position.

This is an interesting subject if you are interested in machine tool design and well worth reading about if you have some spare time! It has also been "discussed" on here at least once before with the same unhelpful "assumptions" made by John W regarding thermal behaviour. In the case of this type of design, the pre-load will be the result of finding a "one size fits all" compromise for the type of service the machine designer expected it to see. This would most easily be "standardised" by setting a torque loading for the spindle with the variables removed (drive pulley or belts etc.) It would be interesting to know if the machine service instructions give a torque value for the spindle?

Mark

[Neil WyattModerator @neilwyatt]

A word of caution: "For operating temperatures above 427° C (800° F), consult your Timken engineer."

Neil

[John HaineParticipant @johnhaine32865]

Posted by Neil Wyatt on 27/12/2016 12:32:52:

A word of caution: "For operating temperatures above 427° C (800° F), consult your Timken engineer."

Neil

I'll be sure to do that Neil. (Though given that the VMB is made in Taiwan to Myford specs (and is quite unlike all the other Chinese milling machines around), I doubt that it has Timken bearings fitted!)

[Author]

Posts