Add your atomic clock here / Electromagnetic Radiation / Forums

Forums

4hv.org :: Forums :: Electromagnetic Radiation

« Previous topic | Next topic »

Add your atomic clock here

1 2 next

Move Thread

LAN_403

Conundrum

Fri Nov 28 2014, 06:22AM

Registered Member #96 Joined: Thu Feb 09 2006, 05:37PM
Location: CI, Earth
Posts: 4059

Not mine, but if I ever got around to building one this is probably how it would look.

Certainly very accurate clocks can indeed be made using OCXOs scavenged from old radio gear, I recall sending one to Dave_B a while back.
Pretty sure there is another one around here somewhere as I recognized what it was and saved it before junking the rest of the board.

(still facepalming for trashing the FPGA board, even though they were useless it might have made good wall art!)

Original article link:-

Sulaiman

Fri Nov 28 2014, 10:04AM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

I bought one of those eBay Rubidium oscillators, mine was FE5680A at 10 MHz
at work I compared it with a GPS-locked 10 MHz clock beloging to one of my workmates
using X-Y Lissajous curve on a 'scope ...
frequency so close that a small phase drift in my workmate's GPS pll could be seen !
defiitely sub-0.1 Hz error = 1/10^8 error ...0.3 seconds /year, probably better.

The power consumption of the FE5680A was about 15V @ 800 mA
which is not a problem UNLESS you want to protect against power failures
(no point having a super accurate clock that looses accuracy)

I sold the FE5680A to my workmate (cheaper than I paid) for his clock experients
so I no longer own it
.. in retrospect a mistake as my MkII HF transceiver is getting closer to completion
and it would make a nice reference clock.

OTOH I have a couple of radio synchronised clocks (cheap from LIDL) that display accurate time
at GBP6 for a radio controlled clock (derived from the Frankfurt atomic time) you need a good reason to home-brew an atomic clock

If you are interrested, have a look at 'Time Nuts'

kimbomba

Fri Nov 28 2014, 06:45PM

Registered Member #3854 Joined: Fri Apr 29 2011, 03:45AM
Location: Mexico
Posts: 95

Video has been deleted :(

Conundrum

Thu Jan 29 2015, 07:25PM

Registered Member #96 Joined: Thu Feb 09 2006, 05:37PM
Location: CI, Earth
Posts: 4059

There are now single chip rubidium clocks apparently.

Also

Intriguingly

seems to suggest that detecting changes in the flow of time with reference
to an external clock *might* be able to detect gravitational waves.

-A

Bored Chemist

Thu Jan 29 2015, 07:52PM

Registered Member #193 Joined: Fri Feb 17 2006, 07:04AM
Location: sheffield
Posts: 1022

I guess I'm the only person on the planet who would like to get a rubidium clock and divide the output down to 1Hz then amplify that signal and connect it to an electromagnet in the bottom of a grandfather clock with a small magnet on the pendulum.
I just like the idea of having one of these that's good to a second in a tdecade or whatever.

Sulaiman

Thu Jan 29 2015, 08:53PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

no, you're the second ! (pun intended)

I was trying for ages to buy a pendulum clock cheaply for exactly this purpose

I even put requests on my local 'recycling' lists .. nothing.
too many people are willing to pay reasonable amounts for me to get one cheaply so I gave up.
and I have too many projects already to start building a pendulum clock from scratch.

as above, no point having a clock accurate to a second per decade if there's a power outage
so some form of ups/battery storage is required.

Conundrum; I have a double-oven XTO, the short and medium term stability is excellent but long term the frequency will drift, something to do with metallization diffusion on the crystal.

Incidentally, I read that the better old pendulum clocks were so good
that the limit of accuracy was determined by earth/moon/tide gravity changes !
also
GPS systems are corrected for relativistic effects due to motion of the satellites,
maybe comparing the GPS signals from different satellites could detect gravity waves?

...

Thu Jan 29 2015, 10:12PM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

For building a clock you are better off with a GPS module, since with an oscillator you need to deal with the synchronization to colorado (or your respective timekeeper) to keep accurate time. If you are going to just look at your cellphone and set it by eye you might as well use temperature stabilized quartz, which will be good to within a few seconds over a few years. The one tricky bit is making sure that you can get a GPS signal, but with modern gps chipsets I have had good luck even inside, as long as you are not far from a window in a multistory building. Also, keep in mind that it is much easier to get a timing lock from the GPS constellation than it is to get a position lock, since you only need a single satellite in view. Also, any GPS module will have a fairly decent oscillator onboard (necessary to be able to receive the gps signals) so even if you loose GPS lock for a few days due to poor weather or whatnot you will still be able to keep accurate time (although be careful, some modules are dumb about this and will stop outputting time data altogether upon loss of satellite link).

This cesium clock is not mine anymore, it was found in the trash at MIT; thrown away because the 'continuous operation' light (which lights while the clock is locked to the Cs reference) had burned out due to old age.

Bored Chemist

Sat Jan 31 2015, 01:48PM

Registered Member #193 Joined: Fri Feb 17 2006, 07:04AM
Location: sheffield
Posts: 1022

I'd also quite like on of these if there's a millionaire out there who wants to get me a late Christmas present.

The advantage to the pendulum clock with a link to a rubidium oscillator is that, if the Rb clock fails- the system still keeps pretty good time anyway.

The easy way to do it is to get a GPS module- but then you are just reading someone else's clock. You might as well just buy one of these

great clocks for time keeping- but not nearly as cool.

klugesmith

Sun Apr 10 2016, 03:30PM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1714

I am shopping for a consumer-grade Radio Controlled Clock, the kind almost universally marketed as "Atomic Clocks". Battery powered, LCD display, fixed location in WWVB territory.

Any hints on how to pick one with an exceptionally good antenna and receiver, for freedom of placement inside a concrete building with metal Venetian blinds on the windows? On a wood frame house, does anyone think a sheet metal roof (and exterior walls of stucco over wire mesh) will create a reception challenge?

Here's a nice NIST page about RCC's:

Are there commercial frequency references disciplined to WWVB's 60 kHz carrier and/or time codes? Or perhaps to cellular phone resources? In an urban environment, I bet those are available wirelessly in many rooms where GPS et al don't reach.

Conundrum

Mon Apr 11 2016, 07:26PM

Registered Member #96 Joined: Thu Feb 09 2006, 05:37PM
Location: CI, Earth
Posts: 4059

I am working on a way to boost the sensitivity on one of these, so far my cunning plan is to use a hand wound Litz wire pickup coil made out of clock wire.
This approach while time consuming does yield superior performance and a simple mini drill chuck can wind a couple dozen meters of wire for various projects.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob