I'm pretty new to electronics and hope that this is the right place to post this. I want to take a battery powered analogue clock and make the hour and minutes hands spin faster. I want to make it seem as though the clock is spinning out of control. What is the best way to do this?

Can I make some change to the circuitry inside the clock or should I use an external gear to manually accelerate the speed of the clock hands?

  • \$\begingroup\$ I would think it is the gearing that controls the speed, you could try changing some out. Or run the 'set' dial with a small motor as when setting the time the hands move much faster. \$\endgroup\$ – rob j loranger May 22 '13 at 23:31
  • \$\begingroup\$ Maybe just hook a motor up directley to a hand \$\endgroup\$ – skyler May 23 '13 at 3:25
  • \$\begingroup\$ You could find a way to integrate a small dc motor into the existing gearbox. Probably somewhere later in the game so that it isn't geared down too much. I guess it just depends on how fast you actually want the clock to spin. \$\endgroup\$ – Jay Greco May 23 '13 at 4:48
  • 2
    \$\begingroup\$ Back in the 70s, when I was an industrial electrician, I modified the clock in the maintenance shop; it had a shaded-pole motor, and I pulled the coil assembly off, flipped it over, and put it back. So the magnetic field rotated in the opposite direction, and the clock ran backwards. Just set the hands to the mirror image of the normal time and everything makes perfect sense. Everyone in maintenance knew how to read it, but we'd get really puzzled looks from visitors who'd say that the clock was wrong. Apparently they didn't notice that the second hand was going the wrong way... \$\endgroup\$ – Pete Becker May 23 '13 at 19:53

I made a video that answers exactly this question called: How fast can a Quartz Clock spin?

What is in the video?

In the video I show:

  • How to deconstruct the clock.
  • What circuitry you can potentially use to make the clock spin faster.
  • What is the maximum speedup you can (roughly) obtain on a regular Quartz wall clock.

Specifically the video first breaks apart a Quartz Clock mechanism, hooks up the solenoid that powers the clock to a H-Bridge driver and then hooks up the H-Bridge driver to an Arduino. Then we proceed to program the Arduino so that the clock spins faster and faster until it can spin no faster without experiencing errors. This is how I worked out how fast a Quartz Clock mechanism can spin and what I recommend you do too. See the video for more details and get a gist for how you might do it yourself.

What were the final results?

enter image description here

The fastest time I managed to get the clock mechanism to spin without errors (slipping or direction reverse) was 156ms per second tick. This means that we can get (1000 / 156) = 6.41 (2dp) ticks per second. If you could get 156ms down to 142ms then you could get in around seven ticks per second but I don't think that you are going to do much better than that on a common quartz clock mechanism.

So the answer to the question "how fast can I make a quartz clock spin?" is:

6 ticks per second on average, 7 ticks per second maximum without changing the mechanics of the internal mechanism. Faster speeds can probably be accomplished with a custom motor / mechanics.

So, finally, we can now ask the question, "what is the minimum time that we can get the hour hand to spin around the clock without rotating the tuning gear?" the answer for me is:

(156 * 60 * 60 * 12 / 1000) = 6739.2 seconds = 112.2 minutes = 1 hour 52 minutes 19 seconds

Which was a pretty interesting result to me; I hope that you think so too.

Hope this helps; good luck!

I think that this really does answer your question perfectly. I hope you enjoy the video. It took me a bit to make. Let me know what you think!

P.S. I made the video first and then saw this question. So this is a fortunate coincidence.

  • \$\begingroup\$ Can you summarize the answer here? Link-only answers are discourated. \$\endgroup\$ – Scott Seidman Feb 7 '15 at 23:08
  • 1
    \$\begingroup\$ @ScottSeidman added in a summary of the video. Does that meet the guidelines? \$\endgroup\$ – Robert Massaioli Feb 7 '15 at 23:56

I've seen something similar for a stage production before - they replaced the clock mechanism with a standard "remote control" servo motor, driven from a microprocessor. The input was a DMX signal from the lighting control desk, which allowed the clock times to be sequenced to the show progression/time travel effect.

The servo wasn't modified for continuous operation - it has high torque one so could drive the "hours" gearing directly, which indirectly caused the minute hand to move.


Analog clocks use a solenoid to advance the second hand, while gears move the rest. If you replicate the pulses in the clock fed to the solenoid, you may be able to go 10x faster depending on the response time of the second hand. You won't be able to go say 100x faster with this type without some serious electro-mechanical modification and thermal cooling.

example http://resonanceswavesandfields.blogspot.ca/2011/03/understanding-quarz-analog-mechanical.html


The "motor" in a quartz clock uses a single winding, driven with a short DC pulse (1.5V, 1/32 or 1/64 second) every second. Alternate pulses have opposite polarity so the rotor (which is just a magnet) rotates a full turn in 2 seconds.

The opposite polarity is arranged by a full bridge driver (or you could use 2 pins from an MPU at these power levels, setting them to '1' alternately. I have just measured the coil resistance of a broken one at 180 ohms, so the MPU should be able to supply 10ma at 1.8V or 20ma at 3.6V). The original drive voltage is 1.5V but you could increase it to 3.3 or probably 5V without burning anything out - allowing shorter pulses to turn the rotor faster.

Replicate this waveform with shorter intervals between pulses and the clock will run faster - possibly up to 32x faster if you drive it continuously or more if you shorten the drive pulses.


Not the answer you're looking for? Browse other questions tagged or ask your own question.