Accurate Zero Beating, another perspective.
When trimming an oscillator so it or one of its harmonics zero beats with
WWV or other standard frequency transmission, much comment has been made
over the ability to approach true zero beat. When the harmonic is directly
zero beat, the stated accuracy is generally in the 1-5Hz range. There is a
technique that allows one to repeatedly zero beat to a much higher accuracy.
The method is called the "Three-Oscillator Method" and dates back to the
1930's, or earlier. The earliest discussion I have found was on page 47 of
Bulletin 10, "Frequency Measurements at Radio Frequencies," published by the
General Radio Company in February 1933. The bulletin states that the
"method has been in use for a number of years." The technique is also
presented in sections II and XII of the 1956 Technical Manual (TM11-2665)
for the AN/URM-18 Frequency Calibrator Set, the military version of the
General Radio Type 1100-A Frequency Standard. More recently, Alan Melia,
G3NYK, reports an accuracy of 0.1 Hertz using the same technique,
The three oscillators are the standard, the unknown, and either another
lesser accuracy oscillator or a receiver BFO. The AN/URM-18 and the General
Radio 1100-A frequency standards utilize regenerative receivers. Using
reception of WWV as an example; in normal practice the unknown or a harmonic
of the unknown is adjusted to zero beat with WWV by injecting a sample of
the unknown source into the antenna of an AM receiver tuned to one of the
WWV transmissions. As the unknown is trimmed or adjusted to match WWV, a
beat frequency will be heard that approaches 0 Hz or zero beat with the WWV
transmission. Unfortunately, the audio bandpass of the receiver and the
observer's ear limit hearing a beat frequency much below ten Hz. It is
possible to reach closer beat frequencies by listening to the background
noise wax and wane, but the results are not readily repeatable. Now, a
third source is introduced when the receiver BFO is turned on or the
regenerative receiver is adjusted to oscillate. With the unknown source
temporarily disconnected, the receiver is tuned to give a nominal 1 kHz beat
frequency while receiving the WWV transmission. When the unknown source is
once again added, the 1 kHz beat will wax and wane at a rate equal to the
beat between the unknown source and the WWV transmission. Changing the BFO
or receiver tuning only changes the frequency of the tone that waxes and
wanes. The waxing and waning rate is determined solely by the beat between
the WWV transmission and the unknown source. It is now easy to reliably
adjust the unknown, or its harmonic, to within a fraction of a Hertz of the
John M. Franke WA4WDL
From: "Hal Murray" <hmurray-8cQiHa/C+6Go9Gemail@example.com>
Sent: Monday, November 29, 2010 3:31 AM
To: "Discussion of precise time and frequency measurement"
Subject: Re: [time-nuts] If there a FAQ
>> I know I can zero-beat it by ear and get within a couple Hz out of
>> That is better then 1E-6 simply by hand, ear and screwdriver. No
> How do you get down to "a couple Hz"? I thought most (young?) ears
> out at about 20 Hz and I expect lots of radios chop off more than that
> to get rid of noise.
> Can you center the screw between two I-hear-it spots?
>> I figure my first upgrade is to replace the crystal with a temperature
>> compensated oscilator chip. Now to go find one.
> Don't get too focused on TCXOs. You are just as likely to find a low cost
> While you are looking, keep an eye open for a TBolt or Z3801A. (or one of
> the Z38xxA variations) If you find one at a reasonable price, they have
> incredible price/performance ratio. If you hang around here for long, you
> will probably get one anyway. Unless you like the sort of fiddling you
> described, getting one now will save you a lot of time/effort.
> These are my opinions, not necessarily my employer's. I hate spam.
> time-nuts mailing list -- firstname.lastname@example.org
> To unsubscribe, go to
> and follow the instructions there.