Tuesday June 2nd, 2020
I wish I could report that there has been some miracle happen that made the Phase Shift SSB Transmitter work properly.
I even spent a small fortune for some 1% SMD capacitors to mate with the 0.1% resistors that would be employed in an 8 Pole filter. I didn't even open the box from Mouser.
While one could easily be swayed to thinking this will finally do it. Realistically I just do not think I would see any different result other than poor opposite sideband suppression that I have been experiencing.
I have become very skeptical of those who have reported amazing results with their phase shift networks and put those comments in the same league as exaggerated gas mileage claims.
We have a great word in Italian --Basta. Enough!
But I am happy to report that I have successfully built a 2nd ZL2CTM Teensy 3.5 SDR transceiver. the Hilbert Transforms Rule and indeed opposite sideband suppression!
The why of the 2nd build is to have something presentable and not look like a bunch of wires hanging out all over the place and to eventually "harden" the build so there are no wires waiting to be shorted and no boards with cold solder joints.
Yesterday I had a QSO with a station running a FLEX 6700 --try as he might he just could not denigrate the signal. His parting shot was that it was indeed a solid build and sounded really well --for a homebrew rig!
73's
Pete N6QW
Thursday May 28th, 2020
Here is a series of tests that only leave me totally baffled.
I built a 6 Pole Audio Phase Shift Network using the Tonne Design Software but KK7B's Schematic. For a signal source I have a waveform generator and use this in the sine mode to cover frequencies in the range of 200 to 2800 Hertz at 200 Hertz intervals. The A and B channels were connected to my Rigol 100 MHz DSO set to read Lissajous figures.
There was a two fold purpose to the test the first of which to see if we got a circular pattern which is indicative of a 90 Degree Phase Shift. The second parameter was to see how that pattern held over the range 200 to 2800 Hz using a 200 Hz step evaluation.
Assuming satisfactory results then this piece can be eliminated as an issue in poor opposite sideband suppression. By poor maybe only 5-10 dB. Below is the test setup.
In the initialization stage, the signal level from the source is critical so you don't overdrive the mic amp and the follow on circuitry. It takes very little signal level! Secondly on the APSN is an input balance pot and that was found to greatly affect the "roundness" of the plot. It is not at the center of the range as you would perhaps think.
The following are the test results...
General observations:
- At 200 Hz --a bit ragged but improving thereafter. No distortion, like at 200 Hz, but higher in frequency a definite pattern shift
- Beyond 1500 Hz some shifting from circular to slightly elliptical
- Very likely "not textbook" for 45 to 50 dB of opposite sideband suppression but I would venture certainly better than 5 to 10 Decibels.
- The front end with the DBM's has been tested with a 2nd ZL2CTM Teensy 3.5 and the opposite sideband suppression is very evident.
So now for the head scratching -- why such poor opposite sideband suppression?
Think about the emperor (with a nearly invisible e) citing that there are only 100K deaths, 1.7M infected souls and 41 Million out of work and saying "I have done an amazing job managing this pandemic."
73's
Pete N6QW
Memorial Day ~ 2020
For those who served, Thank You for your service. For those who made the ultimate sacrifice, we are deeply indebted. For those now serving may you be safe and know we stand behind you!
For those who golf during this pandemic crisis while the country suffers an unfathomable loss of life and dire economic impact ... (readers can finish this sentence)
TL084 Eight Pole Filter, two DCR's, Summer, 74AC74, Band Pass Filter 4X7" Board
