Initial Accuracy
The output frequency of this TCXO (Temperature Compensated Crystal Oscillator) was factory set and should be very close. But these
are "New Old Stock," DC 99xx, and may have drifted slightly so you may want to tweak them. An internal trimmer is accessible for
making minor frequency adjustments to compensate for aging.
Test Equipment Options
A really good lab quality frequency counter would be convenient, but if you are like me and don't have a high resolution frequency
counter, you can go to plan B and fine tune the oscillator by zero-beating it against another 10 MHz reference signal. Fortunately, this
alternate calibration method of the TCXO does not even require any sophisticated equipment. All you need is an AM shortwave radio
(Note 2) that can receive a good signal from the NIST 10 MHz WWV transmitter located in Boulder Colorado. NIST is the federal
agency that everyone else uses to calibrate their calibrators. Can't get any better than that.
Calibration
Carefully remove the tape over the trimmer access hole, figure 4, and save for reuse. I used an all-metal jeweler's screwdriver to
tweak the frequency and surprisingly I didn't experience any frequency pull when I touched the metal blade to the metal trimmer.
Tune the radio in AM mode to the 10 MHz WWV station. The stronger the signal, the better. Power up the TCXO and use a short wire,
adjusting length as necessary, to produce a comparable signal strength.
It may take a little experimentation to balance the signal levels of both WWV and TCXO but it's really not very critical.
Once you get their levels set, you can easily adjust the TCXO to a fraction of a Hertz by ear. When you are close to synchronization,
you will hear a very slow pulsating or swooshing sound. This is the beat frequency difference between WWV and your TCXO and it is very recognizable
when the frequency difference is several Hertz. Once you
are familiar with the sound, your objective is to slow the pulsing 'swoosh' rate down. With a little patience you can slow the pulsation period to a few
seconds or even make it
appear to freeze. A two second period would represent a 1/2 Hz error at 10 MHz, but remember, short term results are still subject to
the frequency stability specifications shown above so it is not reasonable to expect that calibration to within a few tenths of a Hz will
'hold' over the long term.
A reasonably constant ambient temperature and a well regulated 5 Volt power source will provide best results. Ambient temperature,
supply voltage and aging will cause drift but it should still be less than 1ppm. (See the specifications above.)
Mounting and Wiring
This device has 3 leads. Lead spacing is compatible with the standard DIP-14 IC pattern (i.e. pins 7, 8 and 14) although extra room is
required for the larger footprint of the metal can. Lead length is 3/8" and can be trimmed to fit a socket or left intact for mounting on a
thermal insulator.
Before applying power, verify your connections with our diagram / datasheet provided above. Incorrect wiring will destroy the TCXO.
Notes:
1) This listing is for TXCO(s) only. Crystals and resistors are for illustration only and are NOT included.
2) The actual receiver calibration accuracy does NOT matter since you are mixing the two 10 MHz signals together and only listening
for their relative difference in frequencies. FYI, 1ppm is roughly equivalent to a clock error of about 30 seconds per year.
3) The datasheet is a high resolution scan and may not display properly on some browsers. If you encounter viewing problems,
download the file and open in a JPG viewer.
|
IMPORTANT - USE OUR DIAGRAMS to wire your TCXO
WARNING: There's an INCORRECT DATASHEET on the INTERNET with a WRONG PINOUT!
OUR WIRING DIAGRAMS including our datasheet link are VERIFIED and CORRECT and we test each device before
shipping. Use our diagrams and you will have no problem. If you use the datasheet on the web that is wrong (pins 8 & 14 are
swapped) you will destroy your TCXO!
|
