How to seal a lathe headstock

[Keith LongParticipant @keithlong89920]

One thing I'm wondering is whether 30mm dia. is really adequate for a 150mm centre height lathe. At that size I'd be expecting 30mm minimum as the through bore. My IXL at that centre height has a 2" dia. nose thread for the chuck so I'm guessing at 2" min dia. for the spindle at the nose end (the lathe takes a 4.5MT adaptor)? If you hope to use the lathe at it's capacity remember chucks that size are getting very heavy as just a dead weight.

Also the seal arrangement as you've redrawn it is beginning to give a significant overhang to the nose of the spindle, mounting a chuck on a backplate will push the load further outboard even before you start adding the work piece.

Just my thoughts.

[RainbowsParticipant @rainbows]

Yeah I went a bit overboard with that in hindsight. Threaded portion is also a bit over long. Can probably cut 40mm from spindle nose to nearest bearing. The collars could be half their thickness were it machined from solid instead of being screwed on.

I would sooner have a thicker spindle but once you get past 30mm single point screw cutting seems to be the only way to go. Anything approaching its maximum capacity will be faceplate work on aluminium castings so hopefully it won't be as strained compared to turning solid steel slugs of 250mm diameters. Most of its life will probably be with a 4" chuck.

You don't need a lathe to make a lathe but it sure helps to. Once I have the lathe made it could start making its own better replacement parts or use it and the expierence to make a new better lathe and give away the old one to starving orphans.

[Raymond AndersonParticipant @raymondanderson34407]

Rainbows, Have you thought about increasing the spindle Ø ? As it is, a spindle of 30mm od is not going to give you a generous hole through the spindle. Usually a lathe of 150mm ctr height would have at least 38mm bore and also a MT5. Do you not have the capacity to single point more that 30mm Ø? is that why you want to keep the spindle Ø down ? I certainly think a Spindle of around 60mm od would be best. There is certainly a lot to get right, and that must start first at the design stage. It could take an age to come up with a suitable design before you even start cutting metal. but time spent at the design stage will pay off handsomely. I wish you all the success.

Edited By Raymond Anderson on 07/05/2016 07:26:57

[John McNamaraParticipant @johnmcnamara74883]

Hi Raymond Anderson

My VDF was made by Heidenreich and Harbeck

In the 60's a group of German machine builders teamed up, they shared common parts and technology while maintaining their identity as separate companies. ​VDF Bohringer was also part of the group.

As you might guess the machine did not look like the picture when I got it. Many happy hours months! have been spent refurbishing it. It is powerful and accurate. The finish is 2 pack epoxy machinery paint from Dulux closely matched to the original colour.

Regards
John

[Neil WyattModerator @neilwyatt]

If you are happy with that diameter and using stud-mounted chucks, a 4" mini-lathe spindle might be suitable and less than £50. 30mm O/D and 20mm bore with MT3 taper.

Personally I think they are a questionable idea for mini-lathes as they mean you can't use an off the shelf faceplate or the standard smaller chucks (which are nice for many jobs) and you can fit a 4" chuck on the standard spindle with a backplate anyway, but it could be just what you need for a slightly larger machine. You could still fit bearings as per the latest iteration of your design, but it would mean a shorter tailstock.

Personally I would use.rubber seals, not labyrinth ones as the job is keeping dirt out.

Neil

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Neil Wyatt on 07/05/2016 09:51:34:

If you are happy with that diameter and using stud-mounted chucks, a 4" mini-lathe spindle might be suitable and less than £50. 30mm O/D and 20mm bore with MT3 taper.

.

It must be me … But this thread seems totally surreal

Witness; Neil's post, immediately following John's photo.

MichaelG.

[Clive FosterParticipant @clivefoster55965]

How strongly are you wedded to the idea of anti-friction (i.e. roller / ball) bearings? Exploiting modern lubricants and tooling it should be quite feasible to come up with a practical plain bearing design pretty much equal in free running to a rolling element set-up and at least as accurate.

Precision roller bearings probably need greater accuracy when machining the mounting bores than equally accurate plain bearings to meet the specified performance. I imagine a decent investigation into modern synthetic oils should turn up something low drag and able to operate with miminscule clearances. Look at the evolution of automotive oils along with the corresponding reduction in running clearances and drag losses to see how far things have come in the last quarter of a century or so. In days of yore instrument makers and similar bench top lathes ran at high speeds for many, many years with minimal wear using simple hardened steel on hardended steel plain bearings. More recently Holbrooks diamond turning capable high speed Model C 10 and C 13 lathes ran happily at over 2,000 rpm on simple, albeit precisely fitted, plain bearings. Modern lubricants should make similar performance routine.

I have along term, probably never get beyond paper unless I can scare up some serious uncommitted money, exercise concerning the design of an improved but much cheaper to make manual lathe substitute using modern technology of around 10" swing. Not CNC as we know it but still a good deal of electronics. Current concept is pump fed hybrid static-dynamic plain bearings running direct in ball sized bores in a mostly solid light alloy headstock. The maths says it will work fine and match my Smart & Brown 1024 for accuracy. Pump has to start before the spindle turns of course. Lateral thinking needed because I want at least 2 1/2", preferably 3", spindle bore.

Clive

[RainbowsParticipant @rainbows]

@ Nigel_G I can't believe I forgot to reply but thanks for the offer. I have an 11/16" bore taper roller bearing I have been using though. Having a bearing in the flesh to understand the shapes and bearing surfaces is pretty useful.

 

@Clive Foster No I don't have any inherent love for bearings. The deal breaker is though that to use plain bearings I would need to harden and grind the spindle. I'm not a stranger to hardening little buts of steel but doing it on a large piece then getting it accurate again is probably beyond me in terms of skills and equipment. There is also more information on installing and using roller bearings compared to plain bearings.

 

Also some people said shaft collars might be inaccurate, others said they would be fine. To err on the side of caution and to allow me a better spindle design I decided that before making my permanent headstock and spindle I will make a temporary one using bronze bushings and a 16mm spindle. Advantage here is I can turn the spindle on my lathe and use taps and dies I already own for the nose. Once made I can turn the permanent spindle between centres and line bore the permanent headstock.

 

 

 

In terms of final spindle and headstock. I bought 2 32307A FAG bearings (oh boy what a bargain they were). If my spindle has a spindle nose with an M30 thread they will be the front bearings. If I make my spindle with a M48 thread they will be the rear bearings (as on the first example in the Timkens bearing link).

Thread cutting is still a problem for the larger spindle. If I get two timing belt pulleys of equal size (this lathe will start off with a 16×2 leadscrew) could I single point turn a 2mm pitch thread? A fully functioning saddle and cross slide will be made before the headstock is so I'm thinking of letting the lathe make itself.

 

 

Anyone know the bearing dynamic load rating of their lathe?  The bearing I bought are 100000 N (radial). Curious how they compare. 

Edited By Rainbows on 08/05/2016 23:10:09

[John Stevenson 1Participant @johnstevenson1]

Why the attraction to labyrinth seals when an off the shelf rubber seal cost 50p ?

You are getting the bearings too close together at the expense of a seal that's not needed.

[JasonBModerator @jasonb]

Rainbows, what are you thinking of using as your "off the shelf" bar as the fit of the taper bearing inner needs to be a good firm fit, any slack and there is no point in using them, you would be better off with split bronze bearings that can be adjusted for diameter.

[Nick_GParticipant @nick_g]

Posted by Rainbows on 08/05/2016 23:07:36:

@ Nigel_G I can't believe I forgot to reply but thanks for the offer. I have an 11/16" bore taper roller bearing I have been using though. Having a bearing in the flesh to understand the shapes and bearing surfaces is pretty useful.

 

.

These have an internal bore of (about) 2.500" for the front and 2.000" for the rear. The cups dia's are ish 3.800" and 3.500"

The cups also have a flange on them to help stop them (presumably) creeping into the headstock.

For your design ref the spindle that these combine with has an internal bore of 38mm.

Let me know if you want them for observation and a 'play'.

Nick

 

Edited By Nick_G on 09/05/2016 11:25:32

[John McNamaraParticipant @johnmcnamara74883]

Hi Rainbows

100000 Newtons to pounds force = 22480 pounds radial

I guess that will do!

**LINK**

Regards
John

[HopperParticipant @hopper]

Rainbows, Have you taken a look at Alan Jackson's homebuilt Stepperhead lathe/multi-machine? It was featured extensively in MEW a couple of years back with all build details etc. Good rundown on it on lathes.co.uk too, here

**LINK**

Looks to me like the headstock is plenty solid yet simply executed.

Page one on the lathes.co.uk link above includes a good description of how and why he designed the headstock. Looks like it uses simple rubber seals. I am sure full drawings are available somewhere, probably back issues of MEW.

Unless you are an experienced engineer with a background in machine tool design/implementation, you really are better off to follow what has been done before and perhaps build upon it, rather than try to reinvent the wheel. Alan Jackson says he based his Stepperhead design on the 1950s Metalmaster machine, then updated the technology to 21st century standard and put his own twist on it once he had the basics down pat. You could do worse than follow his lead, I am sure.

Edited By Hopper on 09/05/2016 13:26:29

Edited By Hopper on 09/05/2016 13:34:02

Edited By Hopper on 09/05/2016 13:35:53

[Neil WyattModerator @neilwyatt]

Metal Master information here:

http://www.model-engineer.co.uk/news/article/the-metal-master-impetus-machine-tool/20110

A modest donation to REMAP is suggested if you find the drawings of practical use.

Neil

[Alan JacksonParticipant @alanjackson47790]

Hopper , Thanks for the plug, I have not fitted any seals at all. The mandrel fits closely in a recess about 0.250" deep in the headstock block bores with about 0,005" radial clearance. The bearings are grease lubricated and as yet it seems to work very well. There has been no grease leakage from the front or rear bearings and running the headstock for extended periods at speeds above 2000 rpm only gets the headstock hand warm. I have dismantled it occasionally and all seems well. I suppose if it was oil lubricated then it would need oil seals but grease does not create these problems.

Alan

[HopperParticipant @hopper]

Posted by Alan Jackson on 09/05/2016 18:03:50:

Hopper , Thanks for the plug, I have not fitted any seals at all. The mandrel fits closely in a recess about 0.250" deep in the headstock block bores with about 0,005" radial clearance. The bearings are grease lubricated and as yet it seems to work very well. There has been no grease leakage from the front or rear bearings and running the headstock for extended periods at speeds above 2000 rpm only gets the headstock hand warm. I have dismantled it occasionally and all seems well. I suppose if it was oil lubricated then it would need oil seals but grease does not create these problems.

Alan

You're welcome, Alan.

Goes to show: simple is often best. LIke the old plain-bearing lathes such as early myfords, Drummonds and most others, no seals were used and swarf entry seems not to have been a problem.

Grease lube, although not so common on lathes, seems to work fine on tapered rollers in wheel bearings on cars and trucks, which I am sure do some pretty good rpm under some pretty stressful loads for some very extended periods — 100s of thousands of miles usually. I found in industry that more roller bearing damage is done by overlubrication than under-lubing. Hooking the grease gun up to pillow block bearings and pumping until grease comes out the shaft seals being the number-one killer.

[Author]

Posts