Mike's Rodless Pendulum Clock Build

I thought I'd show you guys some of the work I'm doing in making a unique pendulum clock. There's several videos on my youtube channel. Here's some links:

[ame="http://www.youtube.com/watch?v=h0CB8ZoF7RY"]YouTube- Rodless Pendulum breadboard 2, HD, with bob[/ame]

And one of it under electromagnet power:

[ame="http://www.youtube.com/watch?v=tvUU5iRaScU"]YouTube- Rodless pendulum, indicator light added[/ame]

It will eventually be all mechanical, but I wanted to get it oscillating and it ran quite well. The idea is to approach isochronism, where the period does not change with amplitude, making the need for perfect impulsing less strict. The roller approach is a high friction one, but the bob mass is following a near perfect cycloid path, which is part of the battle. Next up is to have the bob on a bearing so it does not rob energy from the system by rotating back and forth.

Here's a wire escapement I'm considering for the all mechanical version of the rodless pendulum clock. I'm pretty happy with Solidwork's ability to run this mechanism. Took some doing, but now runs well.

[ame="http://www.youtube.com/watch?v=jB228ePKrd0"]YouTube- escapement1 animation[/ame]

Hi Mike

Interesting and hence worth turning into conversation pieces. However the resultant time keeping is going to be very variable due. The work on escapements in the past is concerned with minimising or countering the effects of the working atmosphere in the way of dirt - corrosion - oil migration - oil oxidation. The wire escapement has similar issues with constant impulse to the pendulum which might imply that the base rolling line in your novel rodless pendulum should curve at the extremity of the travel so that it will disipate energy and hence stand being over driven.

Like the solidworks stuff - gives a very clear idea of how your escapements would perform.

This animation of the Arnfield escapement may give you some inspiration on how to provide constant impulse.

http://www.clockmaker.it/ingle42animaz.htm

I suspect your nice work on the rodless pendulum is a conversation time piece not a regulator.

Regards

Pat

Hi Pat, yes, anything supported by rollers like my pendulum will emphatically not be in the regulator class. I've been calling it "interesting clock that keeps decent time". The dirty breadboard test was plus/minus 75 seconds a day, so not great. With attention to cleanliness, material and smoothness, I think it's possible to be a few seconds a day.

This animation was an eye opener just because it was so easy to make it run virtually. I just picked one of my sketches and had at it. You've hit the nail on the head and more recent sketches have been gravity impulse like the Arnfield. If I were to actually make this wire escapement, then I'd put jewel rollers where wires drag on other wires.

One recent idea is a nice take on purely sinusoidal impulse, but a constant velocity input is required, which is the subject of the eddy current governor on my youtube channel:

[ame="http://www.youtube.com/watch?v=68nmt3dfwik"]YouTube- eddy current governor[/ame]

Quote:

---

Originally Posted by wildwestpat

Hi Mike

the base rolling line in your novel rodless pendulum should curve at the extremity of the travel so that it will disipate energy and hence stand being over driven.

Pat

---

I forgot to address that snippet. This pendulum is exactly like any other with respect to being impulsed and ignoring the friction, except that the bob follows a cycloid path (which is the interesting part), so no need for curvature at the base. The radius of the arc segment that supports the bob is one quarter of the pendulum length it imitates.

Hi Mike

Yes a fascinating piece for any mantle piece.

My point about curving the base line was just a suggestion on compensating for any tendency to over impulse. I am impressed with your results. makes me wonder if using other materials than steel might get even better results.

Hope you go on to make a clock or at least an exhibition piece out of your discovery.

Regards

Pat

Hi Mike

Sorry did not view you video clip. Fascinating stuff. Again would make a very interesting conversation piece must look further at youtube and go out less!

Thanks

Pat

I think that is a beautiful pendulum movement.

As for accuracy, you could just drive it externally with a xtal locked power pulse so it would be as accurate as any xtal driven clock.

Quote:

---

Originally Posted by RomanLini

As for accuracy, you could just drive it externally with a xtal locked power pulse so it would be as accurate as any xtal driven clock.

---

Now, that would be cheating! True, there's no practical need for precision all mechanical clocks, but there's bragging rights.
Thanks for the kind words.

Yep I understand that, a good mechanical clock is a work of art and very respectable. :)

But from your video above I got the impression you were using an electric pulse to power the pendulum? If that is the case it's not such a big stretch to making that electric pulse locked to an external timebase, seeing that you would have electronics in there anyway.

Yes, I was using an electromagnet to keep it going, but just as a stepping stone to a full mechanical escapement.

I've put up a few more vids on youtube lately, and started a blog on my technical projects, both hobby and professional (see links in my signature).
[ame="http://www.youtube.com/watch?v=9sbbbnflM2U"]YouTube- point mass pendulum[/ame]

I've recently posted pics and vids of an upside down pendulum that you might like, at mikeeverman.com