Small Milling/Drilling Spindle (again)

[Steve CrowParticipant @stevecrow46066]

Thank you KB, your reply did indeed enlighten me.

Your description of preload is exactly as I understood it. Since getting the spindle book and studying the designs I've been under the impression that I have been missing something regarding preload but I thought the book had to be correct.

How about the gearcutting frame in chapter 12? With the fully adjustable bearings at both ends, surely this is a legitimate design? I ask as I am just starting to build a slightly smaller version.

Cheers, Steve

[Kiwi BlokeParticipant @kiwibloke62605]

Re the cutting frame design: the bearing arrangement is OK, for the reason you suggest, but it's a pain to have to re-set preload every time the cutter is changed, isn't it? The design doesn't make it easy.

The design is, however, seriously flawed, for another reason. The weakest part of the shaft – the joint – is in the middle of the shaft, and under the cutter. It's exactly where it shouldn't be. It's easy enough to improve this aspect of the design, but why bother at all? I don't see any reason to bother with cutter frames these days. Perhaps I'm missing something…

My understanding is that they are a legacy of the days when you couldn't pop out and buy pre-made bearings. Early cutter frames had (I believe) conical-ended shafts, the cone running in a conical recess in the end of a screwed 'bolt' (for want of a better term) that screwed into the frame. (Hope that makes sense.) This is about the simplest bearing to make, and suitable for very light loads. The 'bolt' could be quickly unscrewed to free the shaft, and also facilitated bearing adjustment to eliminate end-float. One thing you really don't want, if you're cutting tiny gears, is any cutter end-float. With this design, the bearings were necessarily at the shaft ends, so the pulley and cutter had to be between the frame's arms. This can make things a bit cramped.

I can't really see why one shouldn't use a properly designed spindle, with the cutter at one end, and the pulley at t'other. Forget cutter frames – they're history!

Since you are laudably doing your own thing, rather than slavishly following published designs (which we now know to be untrustworthy…), bear in mind that you should only use sealed ball bearings if you are confident that you have sufficient torque to overcome the seals' drag, which can be surprisingly high. You can buy bearings with 'non-contact' or 'low friction' seals which are OK and shielded bearings are fine. Make sure your supplier understands your requirements – to some, a seal is a seal is a seal. Non-contact sealed bearings are no more expensive than 'conventional' sealed, but are perhaps not so readily available.

Edited By Kiwi Bloke 1 on 13/01/2019 21:25:07

[Michael GilliganParticipant @michaelgilligan61133]

A good discussion, Kiwi Bloke

A ball-bearinged cutter frame does seem to introduce pointless complexity.

[ please forgive the inevitable pun ]

I would, however, make a case in favour of the cone-bearinged cutter frame:

It is a simple device [comparatively easy to make]; and the geometry is equivalent to turning between centres

… it therefore has inherent precision.

MichaelG.

Edited By Michael Gilligan on 14/01/2019 07:43:31

[Kiwi BlokeParticipant @kiwibloke62605]

Bearing in mind the 'inherent precision' aspect, I must concede that you've got a point there, MG.

…OK, I'll be off now.

[Michael GilliganParticipant @michaelgilligan61133]

[egaParticipant @ega]

In the introduction to his 1997 book Harprit Sandhu invites corrections. I wonder whether it would be possible for him to be invited to comment?

I bought the book partly because, having faced the challenge of building the Quorn spindle I was interested in knowing if there were easier ways to achieve the desired result.

Edited By ega on 14/01/2019 10:59:05

[Steve CrowParticipant @stevecrow46066]

I agree on the cone bearing aspect but the appeal of cutter frame in the book is the outboard pulley.

I intend to use the frame horizontally on the vertical slide and the only way to drive it is outboard. I think ball-bearings might be the only way to go with that.

As regards the spindle book, I'm rather taken aback that a publication (and probably the only work on the subject available) which has been in print for over 20 years can be so flawed. Also, it made me question my own understanding of the subject for the last 3 months!

There must be other cases where published designs are flawed? Maybe there should be a thread warning of such things? People put a lot of time, effort and money into building such designs.

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Steve Crow on 14/01/2019 18:19:45:

I intend to use the frame horizontally on the vertical slide and the only way to drive it is outboard. I think ball-bearings might be the only way to go with that.

.

In which case, Bill Ooms is the man: **LINK**

https://www.billooms.com/resourceOT.html

MichaelG.

[Steve CrowParticipant @stevecrow46066]

Thank you Michael, some very nice, simple designs there. Such a contrast to the complexity of some others.

It's got me rethinking things again.

Steve

[John PParticipant @johnp77052]

Posted by ega 14/01/2019 10:58:45

In the introduction to his 1997 book Harprit Sandhu invites corrections.
I wonder whether it would be possible for him to be invited to comment?

I bought the book partly because, having faced the challenge of building
the Quorn spindle I was interested in knowing if there were easier ways
to achieve the desired result.

Edited By ega on 14/01/2019 10:59:05

I too have built a Quorn spindle and with care they run well ,it must have done
as i have had it 30 years and it has never been apart.

The photo shows 3 grinding spindles under construction,based on the layout
of the Quorn type spindles with modifications.

The outer sleeve is bored through for a close fit for the bearings,in this way
the bearings will be in line axially.

The shaft is made with bearing seats this gets rid of the bearing spacer tube
and makes the shaft much stronger as it is larger in diameter in the central
portion.

The two sleeves seen in the photo, the shorter one is eventually loctited into
the outer sleeve in the correct position ,the longer sleeve forms the spring box
to push on the rear bearing much the same as the Quorn design.The spacing
of the bearing seats on the shaft is a few thousanth's more than the total
length of the two sleeves limiting the amount of axial movement.

The two screwed end caps similar to the Quorn ,the front one traps the
outer raceway ,the rear cap is clear of the bearing.
In this way a nut fixing is required at both ends of the shaft to trap
the inner raceways to the shaft, as these form part of the labyrinth
seals these are much easier to do.

Perhaps the biggest advantage of constructing in this way is that all
of the turnings are single operations and require no turnarounds for all
of the important features.
The bearings are SKF C3 Eco 17x 35 mm bearings rated for 25,000 rpm
and are more than good enough for the expected loads,i expect that
angular contact bearings a better bet if the spindles were to be used
for milling.

John

[egaParticipant @ega]

John Pace:

Thanks for your very interesting post. If my spindle should fail then I shall know what to do!

[Author]

Posts