Transceiver Frequency Calibration With WWV: Zero Beat and PPM

Your transceiver shows its frequency to the hertz, but that does not mean every digit is true. The number on the display comes from a reference oscillator, and every oscillator is slightly off, drifts with temperature and changes with age. For SSB the error rarely matters, but for narrow digital modes, for CW at the band edge and for anyone who wants to be sure they are inside a sub-band, it is worth knowing how far off the dial is. Transceiver frequency calibration is simple and cheap, because the United States government broadcasts a reference carrier on shortwave all day and night. This guide explains how to use WWV, how to measure the error by zero beat, how to convert it into parts per million, how to adjust your radio and how to judge when the result is good enough.

Why Frequency Accuracy Matters

A reference oscillator that is wrong by a small fraction produces an error that grows with the operating frequency. The error in hertz equals the error in parts per million multiplied by the frequency in megahertz. A radio that is 5 ppm off reads 50 Hz wrong at 10 MHz and 140 Hz wrong at 28 MHz. That is invisible on a voice contact and a nuisance on a narrow-bandwidth mode.

• Digital modes. FT8 and similar modes work on a fixed dial frequency, and stations find each other by the audio tone. If your dial is off by 100 Hz, your transmitted frequency is off by 100 Hz, and some of your contacts will see you at a different place than the one reported in the software. Spots and reports then disagree. See our guide to WSPR, where accurate frequency matters for the reports.
• CW and band edges. If you operate close to the edge of a band or a sub-band, a dial error of a few hundred hertz may put part of your signal outside it.
• Zero beat and netting. When you tune a station to the pitch you expect, a radio with a large offset makes you think the other station is off frequency.
• Beacons and measurements. Your readings of beacons, frequency standards and weak-signal tests mean more when the receiver is true.

The point is not perfection. A modern transceiver with a good temperature-compensated oscillator is often within a few parts per million out of the box, and some are better. Calibration is a check, and if the radio is good, you can stop at the measurement.

The WWV Reference

WWV is a radio station operated by the US National Institute of Standards and Technology near Fort Collins, Colorado, and it broadcasts time and frequency information continuously. According to the NIST page for WWV, it radiates 10,000 W on 5, 10 and 15 MHz and 2,500 W on 2.5 and 20 MHz, and the signals use double sideband amplitude modulation with a steady carrier. NIST also runs WWVH on Hawaii, and the WWV article adds that the stations announce the time each minute in Coordinated Universal Time, with a male voice on WWV and a female voice on WWVH. An experimental broadcast on 25 MHz also exists and may be interrupted without notice.

What matters for calibration is that the carrier frequency is held to a very high accuracy. The frequencies are exactly 2.5, 5, 10, 15 and 20 MHz, so you know what a perfectly calibrated receiver should read. The signal that arrives at your antenna may have a small variation caused by the ionosphere as the path changes, but if you average over a minute or two, the result is far better than any cheap oscillator.

Choosing the Frequency

You need to hear the carrier well, and that depends on your location and the time of day. As a general rule the lower frequencies are stronger at night and the higher ones in daytime, but the actual result varies, so try several. If you are outside North America, the path may be long and weak. 10 MHz is often a good choice because it works at many hours, and 5 MHz or 15 MHz serve at the extremes of the day. If none of the frequencies are readable where you live, look for a different reference in your own country, because many national time services broadcast a carrier in a similar way. For general advice on matching the band to the time, our guide to NCDXF beacons and propagation shows a quick way to check conditions.

Measuring the Error by Zero Beat

The method uses a receive-only measurement, so nothing is transmitted. You compare the true carrier of WWV with what your dial says.

1. Let the radio warm up for at least 30 minutes so that the oscillator has settled. Calibrating a cold radio gives a result that is wrong an hour later.
2. Switch off any frequency offsets, RIT, and any frequency shift settings. Use the same antenna and the quietest possible conditions.
3. Set the mode to CW and tune to the WWV frequency, for example 10.000000 MHz. The carrier appears as a tone in the audio.
4. Tune slowly until the tone matches your CW pitch setting, for example 600 Hz. In CW mode the dial then shows the carrier frequency, so a perfectly calibrated radio reads exactly 10.000000 MHz at that point. The same method works in USB with a known offset, but CW is simpler.
5. Read the difference. If the tone is higher or lower than the expected pitch, that is your dial error in hertz.
6. Repeat on a second WWV frequency, for example 5 or 15 MHz, and compare the result in ppm.

The most accurate way to read the tone is on the screen, not by ear. A waterfall or a spectrum scope shows the carrier as a bright line, and software can measure its position to a fraction of a hertz. Our guide to the panadapter and waterfall shows how to use the display. An older but valid method uses the zero beat itself, as described in the article on zeroing a CW signal: you tune until the beat note disappears, then read the dial. The receive-audio method gives a more precise value, because you can read the tone on a screen.

Convert the Error to PPM

Divide the error in hertz by the frequency in megahertz. If the tone on the 10 MHz carrier is 40 Hz higher than expected, the error is 40 divided by 10, which is 4 ppm. If your radio is wrong in the same way on 5 MHz, the error there is 20 Hz, and the ppm figure is still 4. That agreement across frequencies is the sign that the reference oscillator is the cause, as opposed to some other effect in the radio or the path.

If the two figures disagree by more than a small amount, take more readings. Fading and multipath can cause the carrier to move during a measurement, and a single reading is not enough. The ARRL runs frequency measuring tests twice a year, and the page for the ARRL Frequency Measuring Test notes that modern transceivers can measure frequency quite accurately, which is a good reason to take the result seriously.

How to Adjust the Radio

Many transceivers include a setting for the reference frequency. The menu name varies, and it may be called a reference adjustment, a frequency calibration or a TCXO trim. Look in the manual, and write down the original value before you change anything, so you can restore it. Some radios allow you to enter the correction directly, and others have a trimmer capacitor inside the case. If the adjustment is inside the case, follow the manual and take care with the antistatic precautions, or ask a more experienced operator.

The ranges in the table are a practical guide, not a standard. Every operator has a different tolerance. What matters is that you decide on a limit that suits your operating, test against it and keep a note of the date and result so you can see any drift over time.

Check Again After Adjusting

Measure again after the adjustment, on at least two WWV frequencies, and confirm that the error is now small. Do the measurement a second time a few days later, with the radio fully warmed up, to check that the result is stable. A radio whose frequency moves from day to day or with the room temperature will not be fixed by a single adjustment.

Other References and Tools

WWV is convenient because it requires only a receiver, but it is not the only way to check the frequency. Other time and frequency stations in other countries serve the same purpose, and some operators use a GPS-disciplined reference oscillator that feeds the radio directly. If you use SDR software, a calibration setting is often available that lets you correct a ppm error. Weak-signal operators sometimes use beacons with known frequencies as a quick check, though their accuracy varies. Time-service signals also help your clock. Our article on FT8 time sync covers the computer clock, which is a separate matter from frequency but is equally important for digital modes.

• GPS-disciplined oscillator. A small box that provides a 10 MHz reference accurate to a tiny fraction of a part per million, for radios with a reference input.
• Frequency counter. A good counter with its own accurate reference can read your transmitter, but it needs to be calibrated too.
• Another calibrated receiver. A friend’s radio can serve as a cross-check if it has been checked recently.
• The ARRL FMT. A structured, competitive way to practise measuring frequencies, held twice a year, as described on the ARRL page.

Common Mistakes

• Measuring a cold radio. The oscillator moves while the radio warms, so wait at least half an hour.
• Forgetting the CW pitch. In CW mode the radio displays the frequency of the carrier, and the tone you hear equals the pitch setting. Subtract the pitch from the reading.
• Using the wrong sideband. In USB the tone rises as you tune below the carrier, and in LSB the reverse. Read the direction correctly.
• Trusting one reading. Fading and the ionosphere move the carrier slightly, so average several readings.
• Adjusting without a record. Always note the original setting and the date.
• Calibrating on an offset. RIT, XIT or a frequency shift setting turned on will give the wrong answer.

The Bottom Line

Transceiver frequency calibration is a quick job that pays off for anyone using digital modes or operating near a band edge. Use a stable WWV carrier on 5, 10 or 15 MHz, warm up the radio, zero beat the carrier and read the error. Convert it to parts per million so that you can compare readings on different frequencies, and adjust the reference setting if the radio is more than a few ppm off. Write the result down, repeat the check after a few days and treat it as part of regular station maintenance. A radio that is true to a few hertz gives you accurate reports, clean netting and the confidence that the number on the display means what it says.

Frequently Asked Questions

Which WWV frequencies can I use for calibration?

WWV broadcasts on 2.5, 5, 10, 15 and 20 MHz, with an experimental signal on 25 MHz. Pick the one that you hear best at the time of day, and 10 MHz is a common first choice.

How do I convert a frequency error to ppm?

Divide the error in hertz by the frequency in megahertz. A 40 Hz error at 10 MHz is 4 ppm. The same oscillator error gives 20 Hz at 5 MHz and 56 Hz at 14 MHz.

How accurate does my transceiver need to be?

For voice a few hundred hertz is fine. For digital modes and band edges, an error of a few ppm or less is a good target, and many radios already meet it.

Do I need to transmit to calibrate?

No. The method only receives the WWV carrier and compares it with the dial, so nothing is transmitted.

Why does the measured error change from day to day?

Temperature changes, warm-up time and ageing all move the oscillator, and fading adds noise to the measurement. Warm up the radio, average several readings and check again on another day.

Can I calibrate if I cannot hear WWV?

Yes, with another reference such as a national time-signal station, a GPS-disciplined oscillator or a recently calibrated receiver. The principle is the same as for WWV.