Hacker News new | past | comments | ask | show | jobs | submit login
First Atomic Clock Receives an IEEE Milestone (ieee.org)
77 points by sohkamyung on Aug 14, 2017 | hide | past | favorite | 20 comments



For anyone curious about atomic clocks, the time nuts mailing list is really interesting imo :)

http://www.leapsecond.com/time-nuts.htm

A while ago I tried to measure the phase difference between two of my Rubidium ones

https://www.anfractuosity.com/projects/rubidium-clocks/

I would like to try that again at some point, if I can fix one of my clocks again, as I don't think I got the timebase setting right.

As an aside the light from the rubidium bulb, glows a really nice purpleish colour :)

I'm sure I read that there are now atomic clocks, which are so accurate, they 'tick' at different rates, depending on their height in a room.

This is what I was thinking of: http://www.riken.jp/en/research/rikenresearch/highlights/828...

It seems it's an Optical Lattice Clock.


And now you can buy a genuine atomic wristwatch. (Not the usual fake "atomic" watches that receive the time from a radio signal.)

https://www.hoptroff.com/collections/atomic-timepieces

They're kinda ugly, and hilariously expensive, but probably the last word in wristwatch accuracy.


They are certainly more practical than the first generation [0] of atomic wrist watches.

[0] http://leapsecond.com/pages/atomic-bill/


>The atomic resonator is set to atomic time, but periodically the atomic physics unit will check to ensure that atomic time is properly locked: when this happens an oven will fix the ideal temperature, the internal laser will strike, the microwave resonator will reach oscillation and atomic time will be locked in.

There's an oven and a laser in that little watch??


They're really quite large, compared to most watches.


But compared to most radioactive-powered devices, they would be quite small, no?


Funny that the design of dial makes it impossible to read to better than the nearest 5 mins.


There's also a pulsar clock, which is interesting for a whole host of reasons: https://en.wikipedia.org/wiki/Pulsar_clock#cite_note-Pulsar_...


Not only are they ugly, but setting the time with NTP seems strange if you want atomic clock level of accuracy.


If your watch is that accurate, you probably want the initial time set to be as accurate as possible, I suppose?


Could you explain? I won't lie, most of the watches are a hassle visually. But, I can read the time from them so long as I ignore all of the extra hands and dials that are there, probably, to lend additional accuracy.


Do you know what "EC" and "Non-EC" refers to? It seems the EC variant of the same watch is 1.2x the cost.


I believe it refers to Europe. It's cheaper if you live outside.


That explains why the difference is 20%. That's the UK VAT rate.


I've always been fascinated with the measurement of time since I was young. Well deserved milestone.


Are there genuine atomic clocks on the market for private computing?


Seems that you can at least buy a Quantum SA.45s Chip Scale Atomic Clock for a mere $928:

- https://www.microsemi.com/products/timing-synchronization-sy...

- https://www.digikey.com/product-detail/en/microsemi-corporat...


What a time to be alive.


You can get refurbished rubidium oscillators from ebay for around 200$. Hooking these up to a Raspberry Pi should be fairly easy (see the pps_gpio kernel driver for that).


You should probably set up a true RT system if you're going to use an atomic clock.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: