This site will look much better in a browser that supports web standards, but it is accessible to any browser or Internet device.

Clubs, glossaries, museums, shops and much more......

...all of your model making needs from one site.

Wooden Clock Bearings

Nigel Taylor


After looking at numerous articles and papers I decided to use ball bearings.

David Lucy has a page around building his own clock, in this he notes that ball bearings are not suited to the stop-start motion of clocks and that the drag is high. - miniature ball bearings in clocks. This article quotes a friction value for the miniature bearings, µ=.002, compared to a µ=.15 - .20 for plain pivots.

I did some simple measurements of the bearings in the pendulum. The basic bearing is a 10mm OD, 6mm ID and 3mm thick ball bearing with metal shields and filled with grease.

I held the pendulum out at 15° from upright and marked the back plain of the clock at this point, I marked again at 10° and 5° using a Wixey angle position gauge attached to the pendulum.

I then started the pendulum just beond a 15° amplitude swing and then timed it from the point it decayed from 15° to 10° and to 5°.

Sealed Bearings0177s554s
Ball Bearings - no seals or grease0201s688s
Ball Bearings - no seals, graphite0167s458s
Ball Bearings - no seals, WD400197s704s

After fitting an exponential decay curve through the 3 points I then differetiated this and calculated the power being dissipated by the pendulum at an angle of swing of 5° which is around the angle the pendulum swings when the clock is running.

This gave:

  • Std sealed and greased bearings = 0.054mW
  • No seals and WD40 = 0.03mW

And so around half the power is dissipated with the seals removed and a very light oil used as a lubricant.

The main clock index page.

If you have made a similar clock and would like to share your experience we would love to hear from you - ed.