AM Frequency Measurements

[kcbooboo]

"47 CFR 73.1540 Carrier frequency measurements.
[ (a) and (b) not included here ]
(c) The primary standard of frequency for radio frequency measurements is the standard frequency maintained by the National Bureau of Standards or the standard signals of Stations WWV, WWVB, and WWVH of the National Bureau of Standards."

There are tons of GPS-disciplined 10 MHz ovenized oscillators out there. Every IBOC exciter has a GPS receiver and a 10 MHz oscillator, although it may not be ovenized or disciplined that well. GPSDOs can usually provide accuracy to 1 part per billion. That's way more accurate than any transmitter would ever have to be, especially AM where the limit is +/- 20 Hz, about 10 parts per million to cover all stations.

Do other people use GPSDOs as the reference for measuring AM TX frequencies, or does everybody try to pick up WWV on an HF receiver? I know the GPS system is maintained by the US Naval Observatory, and they probably reference NBS somehow, but as GPS isn't mentioned in the above rule, perhaps it's time that the FCC update this. Didn't the NIST replace the NBS? That's another thing they should update.

I wonder what the procedure is to get them thinking about this particular rule?

Bob M.

[Deep Thought]

This is the FCC we're talking about here... . The underlying rule is whatever you use has to be traceable to the stated standard, WWV et. al., which is the nation's frequency standard. How you get there is up to you.

[BroadcastDoc]

That seems to be the rule with the FCC. We don't really care how you determine that you're compliant, as long as you can prove that you're compliant!

[boiseengineer]

Since it's kinda' difficult to get WWB at the site with the ICOM, I fire up the un-used HD exciter with it's GPS and check the freq counter with that.
It's got pregnant caps so it's a wonder that it boots.

[Deep Thought]

That seems to be the rule with the FCC. We don't really care how you determine that you're compliant, as long as you can prove that you're compliant!

Otherwise known as "here's the rope, go hang yourself".
Since the late 1980's the mantra is "as often as necessary to ensure compliance". For the most part that means you are free to determine how often to check these parameters and with what. However, if you are inspected and are found out of compliance you will be asked to document your most recent compliant day. If you can't do that to the inspector's satisfaction, you will be assumed in violation of the rule(s) back to when you *can* prove compliance. If you can't prove compliance at all you will be assumed in violation back to the statute of limitations, with a "willful and repeated" icing on the NAL cake.

[Dale H. Cook]

Do other people use GPSDOs as the reference for measuring AM TX frequencies, or does everybody try to pick up WWV on an HF receiver?

None of the above. I calibrate my counters to WWVB using an HP 117A VLF comparator and matching preamplified loop with crystal filter. By running the calibration for 12 hours and logging it with the built-in strip chart recorder, beginning and ending in daylight, I get an accuracy of about 1 part in 10^10. The strip chart can be attached to the calibration report for documentation.

[kcbooboo]

I was more concerned with the use of a GPSDO as the reference to calibrate my frequency counter or signal generator, and trying to find a way to say that it's traceable to NBS / WWV and meets 73.1545(c). It's convenient, receivable almost everywhere, and doesn't suffer propagation problems like WWV. Granted, if the station is in compliance with the FCC rules and has supporing data that matches current conditions, then there's really not much to worry about. Even the cheapest counter will get you to within +/- 10 Hz and today's TX equipment will probably be within +/- 1 Hz.

I was reading some metrology articles on the NIST web site. While they don't come right out and say that the GPS system is traceable to NIST, they sure talk enough about its use with their frequency reporting system. Now I may not want to use a GPSDO to get down to 1 part in 10^-15, it should be more than good enough for 1 part per million, and that's good enough for government work at any AM station.

Bob M.

[Dale H. Cook]

that's good enough for government work at any AM station.

It is good enough for "government work" (i.e., satisfying an EB inspector) only if you can demonstrate the counter's timebase traceability to NIST to the satisfaction of that inspector. I can demonstrate that traceability for my counters, but then, I have been an instrument collector for more than 30 years, and I have NIST traceable calibration equipment of a few types in my shop. None of it is up to the accuracy of today's calibration systems, but most of what I calibrate is older tubed instruments which I can calibrate to the manufacturer's specs. The standards that I use (such as standard cells) I in turn have calibrated and certified to NIST by a local cal lab. They charge me low rates because they enjoy seeing the older calibration gear that I own.

[GregHahn]

It is good enough for "government work" (i.e., satisfying an EB inspector) only if you can demonstrate the counter's timebase traceability to NIST to the satisfaction of that inspector. I can demonstrate that traceability for my counters, but then, I have been an instrument collector for more than 30 years, and I have NIST traceable calibration equipment of a few types in my shop. None of it is up to the accuracy of today's calibration systems, but most of what I calibrate is older tubed instruments which I can calibrate to the manufacturer's specs. The standards that I use (such as standard cells) I in turn have calibrated and certified to NIST by a local cal lab. They charge me low rates because they enjoy seeing the older calibration gear that I own.

"Demonstrate traceability?" Seriously? That really seemed important in 1974. (And it really was important then.) But now?

Let's look at the whole rule:

§ 73.1540 Carrier frequency measurements.
(a) The carrier frequency of each AM and FM station and the visual carrier frequency and the difference between the visual carrier and the aural carrier or center frequency of each TV and Class A TV station shall be measured or determined as often as necessary to ensure that they are maintained within the prescribed tolerances.

(b) In measuring the carrier frequency, the licensee may use any method or procedure that has sufficient precision to establish that the carrier frequency is within the prescribed departure limits.

(c) The primary standard of frequency for radio frequency measurements is the standard frequency maintained by the National Bureau of Standards or the standard signals of Stations WWV, WWVB, and WWVH of the National Bureau of Standards.


Where is there anything in there about traceability? There is an appeal to WWV, et al, as a standard. But that's different than traceability, which implies a paper trail all the way back to WWV. That simply isn't there.

[Deep Thought]

Where is there anything in there about traceability? There is an appeal to WWV, et al, as a standard. But that's different than traceability, which implies a paper trail all the way back to WWV. That simply isn't there.

Assuming you are asking a legitimate question and are not just here to argue, everyone knows that the FCC's readings are the ones which count, not yours, and if there is a discrepancy which shows the station is out of tolerance then you'd better be able to back up your readings with something more than your own arrogance. The key phrase here is "any method or procedure that has sufficient precision". If you can't trace your instrument's calibration and precision back to a recognized standard than you are not in compliance.

[Dale H. Cook]

If you can't trace your instrument's calibration and precision back to a recognized standard than you are not in compliance.

Mark is old school, as am I. We are used to demonstrating compliance - that is how we were mentored when we started out in engineering decades ago. In the case of frequency measurements that means a paper trail to NIST (in my case, to WWVB).

Calibrating a counter to WWV is simple enough. The question is, how accurate is that calibration when you measure a transmitter's frequency days or weeks later? The underlying question is, what is the ageing rate of your counter's timebase?

[GregHahn]

Assuming you are asking a legitimate question and are not just here to argue, everyone knows that the FCC's readings are the ones which count, not yours, and if there is a discrepancy which shows the station is out of tolerance then you'd better be able to back up your readings with something more than your own arrogance. The key phrase here is "any method or procedure that has sufficient precision". If you can't trace your instrument's calibration and precision back to a recognized standard than you are not in compliance.

I'm asking a very serious question, and am not here to argue.

A follow up question is: Which FCC rule are you not in compliance with if you cannot trace your instrument's calibration and precision back to a recognized standard? If you say, 73.1540, I will politely and respectfully disagree with your interpretation, and opine that you are reading a lot into that rule that it does not state.

If, however, there is an additional rule that sheds more light on it, I'd like to see that.

Or maybe we're mincing words about "trace". What does it mean, exactly, to be able to "trace" your instrument's calibration and precision back to the known standard? A certificate from a lab?

[GregHahn]

If you can't trace your instrument's calibration and precision back to a recognized standard than you are not in compliance.

Mark is old school, as am I. We are used to demonstrating compliance - that is how we were mentored when we started out in engineering decades ago. In the case of frequency measurements that means a paper trail to NIST (in my case, to WWVB).

Calibrating a counter to WWV is simple enough. The question is, how accurate is that calibration when you measure a transmitter's frequency days or weeks later? The underlying question is, what is the ageing rate of your counter's timebase?

Exactly. That your counter was within 1ppm a month ago isn't all that relevant for today. So what if you have a paper trail on a counter that is no longer accurate? As Mark said, "it's the FCC's readings that count" - not the ones from some test lab last October that may or may not have actually tested your counter.

And if your counter agrees substantially with a known standard and you have every reason to believe it's accuracy, and you're going about trying to do the right thing, they aren't going to cite you for willful and repeated violations.

When you have 12-14 working FM exciters and transmitters, and all of them are within 200 Hz of licensed frequency according to your counter, some high and some low... you know what? I think your counter is "has sufficient precision."

C

[Deep Thought]

"if your counter agrees substantially with a known standard"

Now, in my world that is what "traceable" means. But what the hell do I know...I've only been doing this for 35 years. When I started this we would routinely check frequency monitors (remember those?) against WWV. Handheld frequency counters were a decade away and accurate ones rarely left the lab. Most of today's modern transmitters are more stable than the measurement equipment was in the 1970s.

You're the one who introduced certificates and lab tests into this.

Good to see you here but you need to know this is not a place for arguing word semantics. As the logo says, it's "a forum for the people who get it done".

[Dale H. Cook]

I know the GPS system is maintained by the US Naval Observatory, and they probably reference NBS somehow

NIST monitors the accuracy of GPS transmissions and compares them to the national frequency standard at Boulder. The archive of those comparisons is available here:

http://www.nist.gov/pml/div688/grp40/gpsarchive.cfm

That page also has links to papers concerning the use of GPS for calibration and traceability.