@sil Ah, the lemon, nature's battery.

But really, just take a sundial and a compass and mash them together. Now you have a sundial that keeps to North so you can tell the time, since you're already floating on a giant sundial which does not.

Or better yet, why would anyone want to keep time. You're on a boat with air you can breathe. Enjoy it, and maybe grill up a sailfish or something.

@sil While a clever start, I gather that the slowest resonance that you can get is still in the kilohertz range, so the problem becomes what can count thousands of pulses per second.

You COULD build a simple generator (spin a magnet perpendicular to a wire coil), but if you can do that smoothly enough over time to count the smaller number of pulses, then it's simpler to ditch the magnet and wire to count revolutions directly...

@sil best guess: same as originally happened. Possible alternative: powerful long wave transmissions from shore(s) counting pips on each GMT hour, radio receivers on ships.

@sil I was listening to the In Our Time about this last week and the prize wasn’t just for making the clock, it was for making something that could be mass produced in contemporary factories for all the ships. Get to it!

@sil Nope you'd still need the REST of the electronics and to invent a reliabe electrochemical battery etc..etc...Really the best in 1700 would be a 'Harrison-light' clock like he buit; His was overly cautious - he was after the prize, it was TOO precise (and we know how to make it simpler now) . Contemporary clockmakers COULD make an accurate enough clock (2 seconds a day) but the innovation was the temperature & motion driven innacuracies through his sprung system.
We have better ways to do THAT now (and we know how to use materials better)

@sil Hmm wouldn't you need a switch of some type - transistor/valve? to drive that quarts? I don't know if you could make yourself a point contact transistor.