Frequency Standard - Rubidium or GPS?

Think of the GPS disciplined ones as being like an old fashioned AC wall clock--the phase can go all over the place, but over a long period the frequency will be extremely accurate. Rubidium standards have really good short term stability compared even with caesium ones. For calibrating a counter, you're better off with the GPS one, but for making really low phase noise signals, the rubidium is better.

Cheers,

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net

Hell, you can do WWV on the air at 10 MHz within 1 Hz. if you are careful about it.

Jim

Phil Hobbs wrote in news:HM2dnUFHgv3MFwPXnZ2dnUVZ snipped-for-privacy@supernews.com:

I'm surprised no one has cooked up an inexpensive WWVB receiver/freq reference,for hobbyist uses.I used a Spectracom WWVB unit while at TEK. Of course,that had a chart recorder to track the ref.osc. drift.

--
Jim Yanik
jyanik
at
kua.net

Which was acceptable. In 1940.

--
You can\'t have a sense of humor, if you have no sense!

There have to be some construction projects in old ham radio magazines.

After NIST upgraded the transmitter site for WWVB, its easy to get a good signal in Central Florida. Recover the 60 KHz carrier. Divide it by three to get 20 KHz. Divide that by two for a 10 KHz signal, then phase lock it to a 10 MHz OCXO. I have a spare from a damaged HP 5245L that I will use.

--
You can\'t have a sense of humor, if you have no sense!

Build yer own.

A older rockwell/conexant "Jupiter" gps has 10 khz outputs when locked that align with the hydrogen clock in the satellites.

Most surplus Jupiters do just fine, the T model is highly desirable , but not readily available

This worked for me, and there are other articles using similar techniques, some using the 1 PPS signal instead of the 10 khz from a Jupiter.

Trimble Thunderbolts are highly popular as well.

See

for one of many sites.

Since this is mainly used for phase locking 10 Ghz amateur radio, a good search string is

"GPS 10 Ghz Ham GPSDO "

Mine cost 45$ in ebay parts. 10 Mhz VCXOs are all over ebay.

Steve

Oh, and I forgot, a design by Brooks Shera using a pic and the 1 PPS signal

Steve

I still have a TRF "High Fidelity Tuner" made from J.W.Miller parts. Many years ago I used it to make FCC required frequency measurements for local broadcast transmitters.

The TRF design removed the problems of having to compensate for a local oscillator in a super-het receiver. Using WWV, it was easy to achieve the required measurement accuracy.

There was no WWVB, of course.

PS: As a ham operator, I used harmonics of local broadcast carriers to calibrate my receiver. Very easy to do and accurate enough.

Virg Wall, K6EVE

On a sunny day (Wed, 02 Sep 2009 11:13:15 -0700) it happened VWWall wrote in :

If you still have an analog color set, in Europe the PAL color subcarrier, when locked to the transmitter, will be:

4.433 618 75 MHz 4.43361875 MHz = 283.75 * 15625 Hz + 25 Hz. so 4433618.75 Hz

For NTSC 3.579545 MHz or 3579545 Hz

See:

So, almost every home in Europe has [had if digital] a quite precise frequency standard readily availabe. In the US the same but a bit less accurate perhaps? Good enough to check a simple counter against. And plenty of amplitude too :-) No messy recepion issues.

I used to look at the time changing from the Colorado station compairing it to the stations Cesium standard at Goldstone Apollo. That was for getting the right second. Ground Loran D was the prime day to day reference. That was tricky and not very good anyway. The station had Rubidium and Crystal backups, with the last being wristwatch. When i first started working there there was a morse code chart on the receiver console ?? I love backups. I really hated hearing the Sonalert go off from the Collins time machine and figure out whats wrong. That was when sonalerts were not common, but when I went to McDonalds and that thing went off, I would jump. I still have that frequency counter I calibrated from the Collins equipment, and I added a crystal heater before that. Cheap counter at best. I'm sure the crystal aged. The hams used to use the color subcarrier back then for calibrations broadcasted live.

greg

There's a *mostly* complete WWVB antenna/receiver project at

It's a minimal receiver design that would appear to work, although I've never known anyone other than the author who actually built one. The antenna is the main focus of the site, although the receiver is there in schematic form. The posted design uses a 6 MHz VXCO as the disciplined element, but a 10 MHz unit could easily be substituted with appropriate changes in the dividers. There's another loop antenna design at if you like to experiment.

--
Dave M
masondg44 at comcast dot net

Bart,

My company built flight avioinics for GPS satellites now on orbit. We are also building the next generation blockF avionics.

We use a rubidium standard for testing the flight units. There's nothing special about it. The one for engineering checkout is a lab-built unit that just sits on the bench. I've used it a couple of times for other projects -- so they are "GPS calibrated".

Tom Pounds

Well, for what it's worth, my $10.99 plus tax wall clock is sync'ed to WWVB; I'm sure there's a 1 Hz pulse in there somewhere. :-)

I'd imagine it would make a pretty good reference if you wanted to wait, say, 60 seconds or so to get a good reading.

Cheers! Rich

crystal aged.

broadcasted live.

I know what they used. Some had no clue, but insisted that the station was dead nuts on frequency. Read CFR 47 to see what the allowable range was.

The TV color burst was a poor choice after stations started using a local 'Frame Store' to hide switching glitches, over 20 years ago. They used a 4X color burst crystal that was genlocked to the local sync generator. That stripped the network Rubidium standard generated synch and replaced it with a 20 cent crystal used in their sync generator.

--
You can't have a sense of humor, if you have no sense!

If it wasn't approved for that use by the FCC, the test wasn't valid.

AM broadcast stations were allowed +/- 20 Hz variation in the US.

--
You can\'t have a sense of humor, if you have no sense!

I'd like to see the math on how much precision you're gonna get out of sixty samples of a 1hz clock from a $11 wall clock.

How much would it cost to have your counter calibrated at a shop? Is it expensive? Maybe a sympathetic local university can help?

The rubidium oscillator will likely give you the best short term stability, which is really what you are after for your purpose. If you get a rubidium oscillator (or any other reference source), you'd simply leave it permanately connected to the external 10MHz reference input on your counter. No need "recalibrate" and use your internal crystal any more. Rubidium aging would fairly negligable for your purposes (in the order of

10^-10/year IIRC)

Dave.

--
================================================
Check out my Electronics Engineering Video Blog & Podcast:
http://www.eevblog.com

I'd go with the GPS disciplined for your application. The ones on ebay are out of E911 systems, though I didn't know they were that cheap. I bought a few surplus at the Livermore swap meet for $60 each. They were Starloc brand, but OEMd by Trimble., identical to the Thunderbolt. You can even use the Thunderbolt software.