Clockwork Devices

[Gig...@gmail.com]

I was thinking it would be really nice to be able to build mechanical, clockwork devices.

I've done some mild searching, but have been relativity unsuccessful.

Does anyone know where I can find reading material on how clockwork devices work, and how to design them?

Thanks
Kevin

[ldic...@mindspring.com]

Oohhh 00h 0hh! I need to know a good resource for this too. I have a bunch of antique cuckoo clocks that need to be repaired but haven't found a good resource for building clocks.

Make had this kit for automata. They had a book too, but I can't seem to find it in their store.

http://www.makershed.com/ProductDetails.asp?ProductCode=MKCMT1

[Matt Joyce]

Well, there's plenty of info on gearing, and how that works.  Since your pursuit is a standard rate of speed of 1rpm for a minute hand, 60 rpm for a second hand, and 1/60 rpm for an hour hand.  It's actually just a matter of deciding on a motor, and feeding the motor into a gear box that gives you those speeds. 

This is gearing 101 stuff.  So just look up how to work with gears.  http://www.blueink.com/CLASS/physcom1/gear.htm

That's a great tutorial.

-Matt

[Rob Graham]

Matt, I think when they're talking about clockworks, they're talking about escapements, regulating mechanisms and the like.  Most of the really nice cuckoo clocks are weight driven, not electric.  It's not a question of gear reduction, but of the classic clockworks mechanisms.  I've seen a few wood and paper pendulum clock kits around the internet, so I know they're available.  There was also an article in the NYT a couple of years ago about one of the last watchmaking schools over in NJ.  The students spend their first year just making their own tools (files, screwdrivers, etc.) 

---

From: Matt Joyce <mdj...@gmail.com>
To: nycresistormi...@googlegroups.com
Sent: Friday, July 24, 2009 3:34:04 PM
Subject: [NYCR:Microcontrollers] Re: Clockwork Devices

[Matt Joyce]

They still rely on the basic principals of gearing.  How they drive their gears really doesn't matter.

Be it a spring, or a counter weight, or what have you it's still a mechanical device reliant on gearing for display, and a constant rate of speed for accuracy.

-Matt

[dr...@stonedcoder.org]

If there is enough interest, I can queue up a htink class with a
horologist (I believe that's the term for someone who works with
clockwork).


This is one of my favorite topics, but the bar to entry is kinda high
since you need an enclosure to place the clockwork within, and it takes a
methodical approach to placing the gears and such, let me know if there is
enough interest to find someone to teach a class.

-Eric

[Matt Joyce]

You don't have to build tiny clocks...  you can build very large geared clocks.  They look great on walls.  In fact this discussion is really tempting me to go abuse my laser privileges at NYCR.  I've got spare motors too.  I could make this happen this weekend.

-Matt

[P Leiby]

Matt's answer completes the question IF the desired "mechanical clockwork device" is allowed to use an electromechanical device like a constant-speed motor.

If not, then we need to control gear speeds.  Of course the historic approach to clockwork is gears regulated by an escapement (http://en.wikipedia.org/wiki/Escapement), with a pendulum controlling the movement of the escapement.  The (reliable) laws of gravitation determine the period of the pendulum (e.g. http://dev.physicslab.org/Document.aspx?doctype=3&filename=OscillatoryMotion_BasicPendulum.xml or http://hyperphysics.phy-astr.gsu.edu/HBASE/pend.html) which is constant for a given pendulum and proportional to the pendulum length, provided the angle of swing (amplitude) is small.

Agreed?
Of course I may be just stating the obvious, since a simple search on "how clocks work" shows many useful resources:
http://electronics.howstuffworks.com/gadgets/clocks-watches/clock1.htm

Best wishes,
Paul

[Matt Joyce]

There's also a lot more gears than the round kind, gearing itself can probably mimic an escapements function, but it would be very much a different set of calculations than the basic ones I referenced.  So I will say this.  For any mechanical clock to work, you must first establish a constant to act as your time source.  How you do that, can be a matter of intense creativity and complexity, depending on how exact you must have your clock source, or how insane you want to be.

Sound good?

-Matt

[Rose White]

Yay, Matt -- build a big gorgeous clock and document the hell out of it!

Yay, Eric -- put together a class for htink! Would it be an all-day
class, or a multi-session one? Or a demo with lots of stuff to then do
at home?

-- Rose

[Rob Graham]

My point is that the gearing doesn't matter, how the gears are driven is what's usually most interesting to people researching clockworks.  Electric wall type clocks are generally not considered "clockworks", because they only have gearing from the motors to the hands, and the only thing regulating the accuracy of the clock is the voltage regulator between the power source and the motor.  There's no escapement, no automatic winding mechanism, no jewels, etc.

here's a good site illustrating escapements (which regulate the accuracy of clockworks)

http://www.geocities.com/mvhw/escapement.html

a breakdown of a Rolex,
http://mvheadrick.free.fr/photos/watchesr.html

Wooden clockworks:
http://wooden-clockworks.com/

British Horology Institute
http://www.bhi.co.uk/

The master list, but pretty outdated set of links:
http://www.horology.com

---

From: Matt Joyce <mdj...@gmail.com>
To: nycresistormi...@googlegroups.com

Sent: Friday, July 24, 2009 3:50:20 PM

Subject: [NYCR:Microcontrollers] Re: Clockwork Devices

[Rob Graham]

A clear or tinted acrylic one would be tres' cool.

---

From: Matt Joyce <mdj...@gmail.com>
To: nycresistormi...@googlegroups.com

Sent: Friday, July 24, 2009 3:55:55 PM

Subject: [NYCR:Microcontrollers] Re: Clockwork Devices

[Matt Joyce]

In which case, I'd say I love atomic clocks, using half lives and rates of decay is pretty damned awesome as a constant is pretty awesome.  And in terms of finding constants... I am pretty sure people could come up with some amazing ideas ( a la bre's laser etched finger nails... ).

So I guess maybe we should discuss a specific time constant, and the source mechanics?

-Matt

[Rob Graham]

Yeah, the study of the regulating source is really the interesting part of horology.  I suggest a pendulum, it's more visually interesting.

---

From: Matt Joyce <mdj...@gmail.com>
To: nycresistormi...@googlegroups.com

Sent: Friday, July 24, 2009 4:07:20 PM

[Matt Joyce]

Okay,

   Well I promise to try and build a clock this weekend.  I'll dump anything I come up with on thingiverse and blog detailed info on it.  If anyone else builds one... and feels like documenting it... I'd love to post that as well.

-Matt

[Christopher J. Pilkington]

<thinking type="wishful">I'm SO down for a Cesium Fountain workshop.</
thinking>

>> ------------------------------
>> *From:* Matt Joyce <mdj...@gmail.com>
>> *To:* nycresistormi...@googlegroups.com
>> *Sent:* Friday, July 24, 2009 3:50:20 PM
>> *Subject:* [NYCR:Microcontrollers] Re: Clockwork Devices

>>
>> They still rely on the basic principals of gearing. How they drive
>> their
>> gears really doesn't matter.
>>
>> Be it a spring, or a counter weight, or what have you it's still a
>> mechanical device reliant on gearing for display, and a constant
>> rate of
>> speed for accuracy.
>>
>> -Matt
>>
>> On Fri, Jul 24, 2009 at 3:46 PM, Rob Graham <rgrah...@yahoo.com>
>> wrote:
>>
>>> Matt, I think when they're talking about clockworks, they're
>>> talking about
>>> escapements, regulating mechanisms and the like. Most of the
>>> really nice
>>> cuckoo clocks are weight driven, not electric. It's not a
>>> question of gear
>>> reduction, but of the classic clockworks mechanisms. I've seen a
>>> few wood
>>> and paper pendulum clock kits around the internet, so I know they're
>>> available. There was also an article in the NYT a couple of years
>>> ago about
>>> one of the last watchmaking schools over in NJ. The students
>>> spend their
>>> first year just making their own tools (files, screwdrivers, etc.)
>>>

>>> ------------------------------
>>> *From:* Matt Joyce <mdj...@gmail.com>
>>> *To:* nycresistormi...@googlegroups.com
>>> *Sent:* Friday, July 24, 2009 3:34:04 PM
>>> *Subject:* [NYCR:Microcontrollers] Re: Clockwork Devices

[dr...@stonedcoder.org]

The class is already in the pipeline!

It will occur mid-september when the Horologist I spoke with gets back
from vacation

Awesome!

-Eric

[Matt Joyce]

Anyone have a source on cheap caesium beam tubes?

[Rob Graham]

This may hold you over until then - a guy built a grandfather clock out of Legos - with a Lego Technics mechanisms.

http://www.ericharshbarger.org/lego/clock.html

---

From: "dr...@stonedcoder.org" <dr...@stonedcoder.org>
To: nycresistormi...@googlegroups.com
Sent: Friday, July 24, 2009 4:45:35 PM
Subject: [NYCR:Microcontrollers] Re: Clockwork Devices

[GigaRoc]

I would defiantly be interested in a Clockwork Class.

And although gearing is integral to the workings of clocks, it's how the gears work together that I'm truly interested in.

Kevin

[Nick Pytel]

has anyone experience with getting an apple imac g3 motherboard
working out of the imac case?
i.e. making an atx power supply work with the board...

There are the "iracks" ... but i dont necessarily want a case for it
... i'm just looking for the pinouts of the atx power supply, and the
pinout for the connector on the imac motherboard... which i could
prolly find with a multimeter, but if someone else has already done
it ; )

I have a project I've been waiting a long time to do, and since
people are telling me to clean up my office @ work, it's time. I
have at least 20 imac g3 working motherboards that I want to make
work, all with firewire (400mHz&above). If i can figure out the power
supply issue, I could get them working in a week. cases, enh, i might
just mount them on the ceiling with some kind of airflow ... ; )

the imacs are old, yes, but they still work ...

thanks...
nick