By Michael Gordon
First day, out of the box, start using it. The user manual is 330 pages but despite that it's pretty easy to use. I had the customary sine-wave of ambient 60 Hz energy displayed in less than 5 minutes from out-of-the-box and touching a probe.
It competes with Siglent and Rigol; is quite a bit more expensive and pretty much they do the same things.
Keysight is the descendant of the engineering branch of Hewlett-Packard "Founded as Hewlett-Packard in 1939 by industry pioneers Bill Hewlett and Dave Packard, Keysight Technologies offers a portfolio that includes countless industry firsts that help our customers innovate to connect and secure the world. " https://www.keysight.com/us/en/about/keysight-technologies-history.html
The UZ7HO Soundmodem is used with FM VHF Winlink and APRS. For calibration purposes it has a menu item labeled "Calibration". Clicking it offers three choices: Low tone, High tone and Both tones.
Low tone is 1200 Hz. High tone is 2200 hz. Both alternate as if sending alternating 1's and 0's, it isn't like a two-tone SSB test. It produces some unusual sidebands so we use the oscilloscope to inspect it.
The Bell 202 protocol used for AFSK (audio frequency shift keying) does not use harmonically related tones. It would seem a lot better/easier if it was; switching from low to high would then always take place at the zero crossing and it would be very easy to figure out if you had one cycle or two in each symbol-time. BUT there's a problem; the simple envelope detector cannot then tell the difference between a harmonic of the low tone, and the high tone.
So the protocol calls for non-harmonically related tones. They still need to be spaced far enough apart to permit determination and at 1200 baud, the low tone IS 1200 Hz, and the high tone a bit less than 2400 hz. The sidebands that arise from switchin from one frequency to another are called 3rd order intermodulation distortion or simply products and are probably necessary for proper decoding. Each tone has about 500 Hz to either side for these sidebands, so the low tone goes from 700 Hz up to 1700 Hz with its sidebands, and the high tone goes from 1700 Hz to 2700 hz and as you can see, 2200 hz was chosen so that its sidebands would fit in the voice bandwidth.
Display screen 1: 2 milliseconds per division, looking at the two-tone (alternating high/low) test.
Display screen 2: Freeze and zoom into a portion of the captured signal.
Display screens 3 and 4: Simple display of low tone 1200 Hz and high tone 2200 Hz. The oscilloscope has calculated and displays the frequency down in the bottom left, the sweep rate is in the upper right, 500 microseconds per division. The vertical is 2 volts per division. The peak-to-peak voltage has been automatically determined as indicated by the dashed lines of the horizontal cursor bars.
Display screen 5: Letting the scope free-run with two-tone test. It's pretty fuzzy since sometimes it's a 1 and sometimes a 0 BUT it still has to be 1200 baud so every so often it aligns and produces a nearly stable image at the 3.5 millisecond intervals.
Display screen 6: A closer view of the two-tone test. You can see the shape of the lower tone is not always the same; when it must transition to the high tone (or FROM the high tone) it is where it is and this produces a somewhat irregular appearance.
Display screens 7 and 8: Exploring the signal using FFT (Fast Fourier Transform). This is a function that many new digital scopes offer, but do so with varying degrees of effectiveness. I chose the Keysight scope as it seemed to be superior in this regard and so far I have been very happy with it. Using FFT is not exactly obvious. It is similar to a Spectrum Analyzer but works very differently. FFT can go all the way down to zero, DC and is exceptionally useful at audio frequencies that are usually too low for a spectrum analyzer.
Also, the FFT can work on a single shot trace where the spectrum analyzer requires a continuous signal as the analyzer sweeps through its range.
In these two screens, the FFT is configured for Dc-to-5 KHz, with 2.5 KHz in the center and 500 Hz per division.
Display screen 9: The FFT showing the two-tone alternating 1's and 0's. You can see definite sidebands, and sidebands OF sidebands due to the intermodulation that results from irregular shaped wave as it transitions from 1 to 0 or 0 to 1. The sweep is 2 KHz wide centered on 1700 Hz, channel 1 is deselected (the signal stream) leaving only the FFT visible. This provides maximum detail in the spectrum. The intermodulation products are clearly visible and distinct. It's the best FFT I have so far seen in this class of scope.
Some views of the W1AW code practice bulletins from the American Radio Relay League (ARRL). I tuned the bulletin using the Northern Utah Web SDR on 20 meters.
If the audio signals driving it are of equal amplitude, and the audio passband in the radio is "flat", you will get the "pinch off" effect seen here. In case the audio passband isn't flat, it won't pinch off; and if the audio is over-modulated it will pinch off and produce a little flat line and that's bad. Say hello to harmonics and distortion if that happens.
Method: Dial the power down a bit, use a 5 watt 50 ohm inline terminating resistor. It will absorb the power and the scope simply reads the voltage at the resistor. In this photo, the scope reports 23.5 volts peak-to-peak, divide that by twice the square root of 2 (2.82) to get RMS (8.3 volts RMS), then square that over 50 ohms for the watts so 69.4/50 = 1.39 watts RMS which is also Peak Envelope Power when measured at the greatest modulation of a modulated signal.
Some experiments with office noise. The building HVAC blowers create rather a lot of noise and much of it is low frequency; in fact, some of it is sub-sonic below 20 Hz. This has been a bit of a problem for Zoom meetings so using a microphone mixer with a low frequency cutoff helps. Here I experiment with unfiltered audio and then with a low-cut filter.