hobbyist frequency standard

The usual method is to use WWVB at 60 kHz, with a ferrite loopstick antenna, a crystal filter, and a PLL. You can get nice 2400 mu ferrite from Amidon. The ground wave delay is more stable than the sky wave from the 10 MHz stations.

Cheers

Phil Hobbs

I see, thanks.

Isn't that a once per minute thing? Perhaps combine the OCXO to provide interim precision, tethered to and recalibrated for accuracy periodically by WWVB? Could also make knowing when to wake up for receiving WWVB precise, as well.

Odd thing is, I'm wearing a Casio WR50M that does close to this and I paid a grand total of $14 for it, including the watch band, case, display, etc. Too bad that isn't made into a nice, convenient embedded module.

Jon

I'm in the uk, but bought one of the telco surplus HP Z3816A ones from Ebay US around 5 years ago and been powered up ever since. The Hp active antenna came from an Ebay vendor in Korea. They were very expensive in the uk at the time. There's a similar unit (58114A ?) in a fancy box and expensive, but these are the half rack width x ~4" high industrial finish types with a single huge toggle power switch + leds on the front panel and rf + serial interfaces on the rear panel. Mine is line input, but they come in 48v dc input as well.

There's a utility that talks to the box, which you drive through a serial line from a dos shell. This allows you to get status reports and set up / compensate things like the nS delay from the antenna cable length. It has a couple of 10Mhz sma outputs into 50r, (which drive all the counters and synthesisers in the lab) and another odd ball frequency which I don't remember.

Stability wise, they can be a bit noisy in the short term, depending on the actual unit that you get. Long term of course, they are locked to Caesium standards. Certainly good enough for any work done here.

I have the utility on the server if you can't find the download, though it's been some time since I hooked a terminal to mine. It just works...

Regards,

Chris

Phil Hobbs wrote in news:6qadnTZWGf110H7XnZ2dnUVZ snipped-for-privacy@supernews.com:

I'm surprised there isn't some receiver IC that would do this job,and maybe a companion IC to do the PLL/dividing.

in the late 1980s there was a plan published in radio electronics that took a radio shack "time cube" receiver and mixed some signals back with it to lock a reference.

These days its easier to find a used Jupiter GPS or Trimble Thunderbolt and make a standard. I just bought a ex cell site rubidium system for 70$ a few months ago. Just add 24V at 4 amps per warmup and

The 10 mhz vcxo was 6$ on ebay.

Steve

Oh and if you cant find a jupiter, (there are none on ebay right now) let me know,by posting here, I have a few spares.

Steve

There was, then Bowmar went bust...

Depending on the accuracy you require - and you *did* say hobbyist - the Jupiter GPSr module is my preferred solution. I'm allergic to the loooong conditioning times required by many approaches to GPS-disciplined arrangements. A good example of a Jupiter-based setup (very similar to my own) can be found at

The benefits of this type of approach are quick settling from cold start, and low cost oscillator. Depending on your jitter requirements, a non-ovened VCXO and a simple loop filter can be more than satisfactory. I have checked my own 10MHz Jupiter-based reference against an ovened TC/VCXO counter timebase and cannot discern any phase jitter on a CRO or with DC recovery on a NBFM receiver in a comms test set.

"Usual" is an interesting word. I think it depends on what sort of quality you want.

WWVB at 60 kHz is common for watches and battery powered wall clocks. It lets them set themselves to better than a second each night. It's inexpensive. There are similar signals from other countries.

Somebody asked for a chip. (I think.) Feed CME6005 to Digikey. Others may carry similar toys. $11 includes an antenna. (Not watch size.)

For higher quality, GPS disciplining a good quartz crystal is the current sweet spot. It's widely used for cell phones and 911 type time-stamping. A lot of it is available surplus. You can also use GPS with a Rubidium oscillator.

"Jim Yanik" kirjoitti viestissä:Xns9CAEBF8A4257Fjyaniklocalnetcom@216.168.3.44...

There are for example Micro Analog Systems MAS6180 Stand-alone AM Receiver IC with Differential Input for WWVB, JJY, MSF and DCF Time Code Formats

Available as 150 mm tested wafer or TSSOP-16 from nowhere

-ek

It was used to discipline Rubidium atomic clocks for low frequency continental VLBI in the early 80's. The transmitter - Rubgy in this case has a very good stable carrier wave that covers most of Europe. ISTR the remote VLBI stations could detect dew on the ground at Rugby causing a few us shift in rise time on the seconds pulses.

GPS is probably better these days, but 60kHz may still be cheaper. Unlcear to me why a hobbyist would want such an accurate frequency standard apart from for the challenge of making one. For VLBI it makes life a lot easier having precise timestamps when searching for fringes when the observations are combined at correlator stations.

Regards, Martin Brown