Spindle design

I've decided to have a go at a micro spindle to take proxxon collets up to 1/8".

I've attached a scale drawing of what I have in mind. Please be aware that the finished housing will be 100mm long. It's just shorter on the drawing for convenience. The grid lines are 5mm and the colours have no bearing on the actual material.

Can anybody see a reason this will not work? Please feel free to comment or ask further questions.

Many thanks

Steve

I would suggest you would need some end float on the rear bearing, make the business end captive, with a cover plate, and the rear bearing needs a bit of clearance so it can float and not bind up on assembly or when running. you may have allowed for this, but it is not obvious from the section.

Chris Gunn

Thanks Chris, I intended to loctite in the front bearing. By clearance and float on the rear bearing do you mean axial or radial?

Cheers

Steve

Hi Steve

The weak link looks like the small size of the belt drive, perhaps a flat or multi vee belt would give better performance.

Emgee

Steve,

I don't believe 'actual' end float is needed - Rather design the shaft/pulley/end nut so that you can apply a small pre-load, and lock the end nut in place with maybe another nut - lock-nut style. And try to make the spindle shaft and the spindle housing of the same material so that thermal expansion coefficients are close to the same - that way the bearing will be treated more kindly.

Also, nothing wrong with your pulley/belt style - I used that on tool post grinders - internal/external, spinning up to 14,000RPM, with a brushless drive motor of up to 800watts..The belt is a sort of neoprene - 6mm diameter, from RS-Components, and you cut to length and heat-join the cut ends. Works very well, and is cheap to replace - mine has lasted around 120 grinding hours...

The spindle takes ER11 collets, but has a hole through for a draw bar that pulls tight extensions to which are fitted grinding wheels, as in these photos. The extension I make up as need for any special wheel - inner grinding or outer with bigger ( 50mm diameter) wheels, etc. The extension where the draw bar pulls is a male ER11 collet fit. You may be able to fit a 3mm or 4mm draw bar to your design as well?

The brushless motor on this unit is 500watts max - 60mm diameter, 70mm long, running of 36volts DC

Joseph, check the belt size as shown, 5mm grid makes the belt diameter about 2mm, not 6mm as you used.

Emgee

I meant axial.

Chris

Man that's a nice job you made of that.

Hi Emgee,

Yes, I realised that, but Steve's spindle would not be absorbing a hundred watts or so as mine does on occasion - a 2mm belt will do just fine for his application - I am sure he can even go up to 3mm. RS does the belting down to 3/32 of an inch and that will do just fine.

Joe

Thank you Pete..was fun making it - I did a second, higher powered one as well, but driven by a toothed belt - that is a mistake which I will retrofit - the vibration from the teeth at high speeds shows through on the grind..

Joe

Steve. Your proposed spindle's diameter is pretty small for its length. I'd guess that you want to be able to run the spindle up to, say, 20,000 RPM. I'd be inclined to make the spindle as large a diameter as you think you can. That would give you the possibility of a central bore, although you probably don't need one for your intended tooling in small collets. However, you never know, and building in versatility is often time well spent.

Angular contact bearings would be better than deep-groove bearings. In either case, setting the right pre-load with nuts on the screw-cut spindle might prove to be frustrating. Make it the finest pitch you can. Probably better to apply axial load with a Bellville washer (or a few). Plenty of spindle designs on the 'net and good stuff in Harprit Sandhu's book, No 17 in the Workshop Practice series.

I think Chris Gunn's point about an axially non-constrained bearing at the pulley end refers to the more complex, but better, design in which there are loaded, opposed bearings at the spindle nose end. It all depends on how 'good' the thing needs to be...

Just to clarify.

The spindle is for use in a tool post and vertical slide on my Sherline lathe. I intend to use it for drilling and very light milling duties for horological type work. I have no intention to use it for grinding.

The belt illustrated is 2mm diameter although, as Joseph mentioned I can easily go up to 3mm.

I was going to make the thread pitch 0.5mm. Do you think this is fine enough? Also, will I have to use locknuts or is there an alternative?

As i am making this on a Sherline, I'm restricted to size. I'm probably going to have to make the housing 80mm long as my centre to centre size will not allow me to drill out any longer.

Many thanks

Steve

Getting the fit of the rear bearing inner race just right is a small issue with this set up. Too tight and you can't adjust the end float, too sloppy and you could get slip between shaft and race. As the axial load will only be one way, what you can do is put a preload ('wave' ) washer between the rear outer race and the housing, then the inners can be press fitted up to shoulders. The wave washer exerts an axial load at all times and takes up any small radial clearance. This is how Quorn does it, but with a more elaborate preload system.

I'd use shielded ball races, not sealed, too much drag,

Edited By duncan webster on 25/10/2018 18:28:35

Edited By duncan webster on 25/10/2018 18:29:18

Thank you Duncan, I like the idea of a wave washer.

I've had a look and the only one available that will fit my assembly just touches both inner and outer races. The drawing below is a section of my bearing and the pink bit is the washer, or rather the space occupied by the uncompressed washer.

I could add a flat washer that only makes contact with the outer race. The pink is the spring and the flat washer is dark blue.

Do you think this could work?

Thanks

Looks very complicated. Surely it doesn't matter where the wavey washer sits, so you could get rid of the pink and blue between the outer bearing race and housing and instead just fit the wavey washer next to your tightening nut?

But unless you define the amount of compression with positive stops, the wavey washer will simply bottom out and you won't have any control of the preload force and might as well not bother with it.

Murray

What;'s with the double row bearings? I'd use a 686ZZ (6*13*5) or similar. The blue bit wants to abut the bearing inner, and the outer wants to be a slip fit in its housing

I made an ER11 spindle a few years ago for engraving and very small milling jobs, it runs at 30000rpm.
Loosely based on something from Harprit Sandhu's book.
I used two ball races for the nose and one for the pulley end. The nose bearings are bonded to the housing and shaft, the pulley end bearing is a moderate sliding fit in the housing and bonded only to the shaft.

It was only an afternoon lash-up to get a job done but it's been running fine ever since, it's easy to over complicate things.

- Nick

Thanks to everybody who's helped with this.You'll have to bear with me - i'm struggling to get my head round it.

Nick, I have the Spindle book but most of the designs use a spacer, something I'm trying to avoid. I really want to make it as simple as possible. Did you make your spindle from an ER11 collet holder with an integral shank?

Duncan, it probably isn't a double row bearing. It's low profile(106ZZ 6*10*3) and I just drew them in for illustrative purposes. I'm quite new to this bearing game! I'm still not sure I understand the positioning of the wave washer. Any chance of a rough sketch?

Many thanks

Steve

sorry it's a bit rubbish, not a lot of time this week. The 6*10*3 is 6mm ID, 10mm OD, 3mm thick, not as you've drawn it. I've also drawn the front bearing wrong but that doesn't matter. The important bit is that you make the housing so that the front bearing takes the axial load as per your original sketch, the rear bearing is trapped hard up axially between the step on the spindle and the pulley/spacer arrangemnent. You arrange the lengths so that when the rear bearing is pulled up tight the wavy washer is preloaded by the right amount. Look in bearing manufacturer's guff about what the right amount is, it is usually more than you'd think. Or give them a ring, they are usually very helpful even if they realise you're only buying one. If the wavy washer overlaps the inner, add another spacer as bottom sketch

As DW above says.

Belville (cone) washers may be easier to find than wavy washers and avoid possible interference with the inner race. The washer contacts the outer race by the outer edge of its face only, and 'points' away from the bearing (towards the other bearing). Therefore install two, 'pointing' towards each other (><; the second one's OD contacts the step in the housing bore, as above. As used by Emco in the Unimat headstock - two, with angular-contact bearings in the early Unimat, 4 pairs, back-to-back*, with deep-groove bearings in the Unimat 3. You need a reasonably firm spring to resist the possibility of the spindle moving axially, should it ever be loaded towards the workpiece by, for example, a drill grabbing in brass.

* thus: >><<>><<