frequency standard from color TV signal?

Seems pretty good, I assume NTSC is similar, but if your video comes via digital and gets converted to composite video in the STB, all bets may be off. And the way analogue Txs are being shut down in favour of digital.....

Have a browse at SMPTE, they may even have the specs for free

martin

in the old days, when transcontinental microwave was the feed from NY or LA to the station, things were excellent, timing being generated by a network's master cesium clock. Now each station gets a digitized fiber feed and its ran through a buffer, so unless for some reason that station decides to have a rubidium master clock for its timing chain, your probably screwed.

some smaller countries still use microwave and things are good, but in the us and mainland europe, its on a station by station basis. Big stations (NY,LA,chicago) probably have a decent clock.

Steve Roberts

Most multiplexing today is SONET. Network sync traceable to Stratum 1 is available from the multiplexer. Don

Any time standard that depends on transmission through the atmosphere must deal with the variabilities of the medium. Unless your receiver is in direct line of sight from the transmitter, the effective length of the transmission channel is not constant.

The color sub-carrier and the colorburst refer to the same frequency in NTSC. 3.579545 MHz.

The nominal colour subcarrier frequency for NTSC is 315/88 MHz; you can do some multiplication and division and get to one of the "standard" reference frequencies like 10MHz = colour subcarrier *

176/63 = c.s.c.*2^4*11/(3*3*7). The FCC tolerance is +/-10Hz, last I knew; heaven only knows under deregulation. Though stations MAY broadcast it with very high accuracy, they may not, too. And if you get your signal via cable, be aware that cable companies have a habit of moving channels around. Even if the signal is in the same channel it was broadcast in, there is no guarantee it will be synchronous with the broadcast signal.

If you want an accurate frequency standard, why not use an ovenized

10MHz crystal oscillator? If that's not good enough, why not discipline it with the 1 pulse per second from a GPS receiver?

Cheers, Tom

This is true, and technically even if you are receiving your signal line-of-sight, there's variation due to changes in atmospheric pressure and humidity, for example. I believe you can find some good information on the accuracy to expect for off-the-air reception of WWV/ WWVB/WWVH on various frequencies, on the NIST website. A few years ago I was playing with how accurately I could calibrate a good oven standard from WWV's sky wave from Colorado to the Seattle area, and found that using FFT techniques I was able to consistently do better than they suggested, but not reliably better than about one part in

10^8, as I recall. That took some understanding of how and when the propagation path experiences the most change.

Cheers, Tom

I think orginally, the live TV shows had direst feeds and the carriers were more accurate. I used to run NASA timming at the Goldstone Apollo site, back when they actually had ground stations before the TDRS satillite. The sky waves were pretty radical. We used loran to monitor drifting. Were pretty close to Colorado, but still a lot of drift. I usd to cal my frequency counter with the stations Cesium clock using the basic 5Mhz sent around the station.

greg

one of my favorites is

martin

I think the WWV signals were only good enough to tell what second you were on. The Collins digital equipment had battery backup, and there was Rubidium and crystal backups. Of, course the Cesium was a crystal oscillator. There seemed an unwritten backup, anyone with a good wristwatch.

greg

My first experience with this was about 1980 while testing some oven- controlled oscillators by comparing to a bench frequency standard that received a reference signal from WWVB. One night I was testing late in the day, and as the sun wnet down my numbers wandered all over the place. The next day I showed the results to my boss and he had a good laugh at my ignorance.

Shera is the granddaddy of all the reflock boards,

but if you want something simple and fast, start with a jupiter gps board (~20$ US) from ebay

and go here:

Jupiter boards output a 1 pulse per second signal and a 10Khz signal

Steve Roberts

Not Found The requested URL /n1jez/ was not found on this server.

Any about theShera Grandad, I've never quite followed P2 on the pdf

"Interestingly, it is desirable to have the frequency of U7 drift slightly rather than being synchronized with the VCXO. A slight random drift averages out the ±1 count ambiguity that is inherent in any pulse-counting device.

My measurements indicate that the simple phase-measuring circuit I use is consistently accurate to 2 or

3 ns (for a 30-second measurement), while without drift, the resolution would be lim- ited to 42 ns. "

Do you have a simple explanation?

martin

Yes, a very good one.

That is one of many places where the Binary Sampling technique can dramatically improve the accuracy and speed of sampling noisy data.

For a description, see "Noise-Rejecting Wideband Sampler" at:

The same technique works in software as well as hardware. A delta value (dv) is selected appropriate for the signal. The system is initialized by averaging, then storing the result in the current value (cv).

At each sample time, the raw input signal is measured and compared to the current value (cv).

If the raw signal is greater than the current value, the current value is incremented by the delta value, otherwise it is decremented:

if (raw > cv) then cv := cv + dv else cv := cv - dv

This allows the software to track the signal much better than conventional averaging techniques allow. The reason is the magnitude of the large amplitude noise spikes are not included in the result, only the direction.

Since random noise has zero mean, the system will track the mean of the signal much more accurately than averaging techniques allow.

The same software can also perform a moving average on the acquired output signal. A simple moving average filter is a very fast method since each sample requires one addition, one subtraction and one division. This can easily be done in integer math, so inexpensive microprocessors and PIC's can be used.

A single stage acts as a low-pass filter with a cutoff floor of about -13 dB. Cascading 4 filters provides a gaussian response with a floor of about -60dB. This is in addition to the filtering provided by the Binary Sampling technique.

Regards,

Mike Monett

Antiviral, Antibacterial Silver Solution:

snipped-for-privacy@spsdialup.com/index.htm SPICE Analysis of Crystal Oscillators: snipped-for-privacy@spsdialup.com/spice/xtal/clapp.htm Noise-Rejecting Wideband Sampler:

ouch, my head hurts.... thxs

martin

Color burst is just a... burst of subcarrier signal where the phase is synchronized with that subcarrier for the picture frames involved. I don't quite remember exactly but the burst phase is offset by 90 degres with the subcarrier phase in NTCS, the correct phase can be obtained with a simple, ajustable RC network in the TV set, the old famous TINT control.The short term precision of the phase is the important factor in television to get correct colors in the following picture frames. Long time frequency accuracy may not be interesting.

Octavio

Most analog NTSC signals are sent through a frame store at the transmitter site to clean up the sync, and to clean up any switching errors. The color burst is regenerated, and must be within +/- 10 Hz of

3,579,54.54545 Hz. This setup was already common 20 years ago.

--
Service to my country? Been there, Done that, and I\'ve got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida

IMHO the downside of the Shera system is the long conditioning times to get a decent accuracy. For many applications the Jupiter-based references will give an acceptable result within seconds of the GPSr achieving lock, due to the higher reference frequency..

I built a 10MHz reference on that basis, and it spends its days turned off because I know it achieves its ultimate performance within three minutes of turn-on. A very similar unit can be seen at

That's quite nice, had a quick browse through the Navman OEM site (dreadful). I couldn't find any GPS modules with 10KHz out, any pointers?

martin