Minature Ball Races

Advert

New Replies

Report Bug

Help/FAQs

Minature Ball Races

This topic has 9 replies, 4 voices, and was last updated 13 February 2010 at 09:05 by Peter Bell.

Viewing 10 posts - 1 through 10 (of 10 total)

Author

Posts
7 December 2009 at 11:59 #45959
Peter Bell
Participant
@peterbell11509
Just read some interesting articles about using minature ball races in clocks to reduce friction at the various pivots. See that some high quality clocks (Erwin Sattler?) are quoted as using them everywhere as well as jewels.

Seems a good idea to me but have never used them before—has anyone tried this and what were the results?

Peter
Advert

7 December 2009 at 11:59 #3577
Peter Bell
Participant
@peterbell11509
8 December 2009 at 08:06 #45999
ColH
Participant
@colh
Hi John

I have not tried them myself but I have a copy of an artlicle by a member of the Sydney(Australia) Clockmakers group.

He ran tests to determine the coeficient of friction (CoF) for a couple of types of ball bearings and a conventional polished pivot. He did the tests on the bearings as received ie had some sort of lubricating oil in them and then washed them out with kero (parafin to those in the UK) or acetone and ran the tests again. Results were as follows:

Plain 1mm pivot – CoF 0.16 dry and 0.22 oiled

Economy 2mm bearing – CoF 0.0044 as received

                                           – CoF 0.0023 washed in kero

Stainless 2mm bearing – CoF 0.019 as received

                                            – CoF 0.001 washed in kero

Economy 5mm bearing – CoF 0.0078 washed in kero

                                            – CoF 0.0014 washed in acetone

A US supplier of jewled pivots claimed a CoF of 0.15. Not much better than a plain pivot but would wear much better.

The author also stated that:

Results of the Theoretical Study

With plain pivots, and assuming a pivot C of F of .15, and a gear meshing efficiency of 98%, the losses prior to the escapement are 22%, of which the pivots accounted for 14.8%.

If ball bearings are used on all arbors and using a pivot C of F of .005, with a gear meshing efficiency still at 98%, the losses prior to the escapement are 11.6%, of which the bearings accounted for only 3.4%.

If, indeed, the ball bearing C of F is really as low as .001 as some of the tests indicate, then the losses in the pivots would be virtually negligible. So the only losses through to the escape arbor would be due to gear friction, which, at this time, is an unknown quantity even though an efficiency of 98% was assumed.

As I was not the author I cannot comment on the results but they seem reasonable to me.

regards Col
8 December 2009 at 20:31 #46041
Peter Bell
Participant
@peterbell11509
Hi Col,

Thanks for the reply—I have a copy of that article by Rex, also an article by Dick Stephen here in the UK. Both agree and make interesting reading so I am going to use them in my next clock.

I wonder what the downside is—apart from the cost of buying them!

Regards Peter
8 December 2009 at 20:31 #46042
Peter Bell
Participant
@peterbell11509
Hi Col,

Thanks for the reply—I have a copy of that article by Rex, also an article by Dick Stephen here in the UK. Both agree and make interesting reading so I am going to use them in my next clock.

I wonder what the downside is—apart from the cost of buying them!

Regards Peter
8 February 2010 at 16:53 #48381
Richard Stephen
Participant
@richardstephen45588
Hi All, I only use ball races in all my clocks now . I get mine from SMB Bearings and if possible get the cheaper Chinese ones. These are slightly less orcision and hence have slightly less friction.
9 February 2010 at 02:33 #48399
ColH
Participant
@colh
Hi Dick

How do you mount the bearings?

Col
9 February 2010 at 11:45 #48408
Circlip
Participant
@circlip
Usually from the left.

Sorry, coat on.

Regards Ian.
11 February 2010 at 16:23 #48537
Richard Stephen
Participant
@richardstephen45588
Col, I drill a hole through the plate 0.5 mm larger than the pivot diameter. I thencounter bore the plate to a depth equal to the thickness of the gearing. The diameter of the counter bore should be about 1.5 mm larger than the o.d. of the bearing. I cut off a piece of brass rod bigger than the counter bore Face off the enf and drill and ream a hoe equal to the o.d. of the bearing. Turn down the outside of the brass so that it is a easy fit into the counter bore. Part off a length slightly longer than the thickness of the gearing. Glue the sleeve into the recess using Loctite (326 is best). Clean out all the excess loctite. Use Hamerite solven (Xylene)t to remove the Loctite. File flush with the plate.

This may seem a long winded method Try boring a blind hoe in a piece of scrap brass for say a 6 mm bearibg. See if you get an accetable fit. I never have which is why I do the above

Regards
13 February 2010 at 09:05 #48603
Peter Bell
Participant
@peterbell11509
Hi All,

I have had some success boring blind holes for these bearings. After experimenting I found that I can successfully bore a blind hole for a 5mm bearing using my Abene knee mill and a Kurodo boring/facing (same as Wohlhaupter) head I bought in a sale some years ago.

The secret really is being able to line up to the arbour hole using the excellent centering microscope built to to the Dick Stevens design and the tiny carbide tin coated boring bar from e-bay USA. The Dro’s help and the whole setup looks ridicules really with the size of the machine but it is repeatable and I imagine could be achieved with any boring head with micrometer adjustment and a good sharp tool—or perhaps I was just lucky!

Regards Peter
Author

Posts