The Radio Gods were smiling upon me this morning. I started out on 17 meters and had three nice contacts with European stations: OH5CZ, a young fellow near Helsinki; HB8DQL; then RM2D in Moscow. FB.
Then Pete showed up on the Skype. As he has said on his blog, he is still struggling with a family medical emergency, but I am happy to report that he is coping well, making good use of his can-do project manager background and his good sense of humor. It was great to see him.
Inspired by my talk with Pete, with 40 meter AM playing in the background, I turned to my R2 FRANKENSTEIN phasing receiver. Last night I completed the 90 degree phase shift network. This is built around two quad op-amp chips and is designed to take the audio output from the two DC receivers and create a 90 degree phase difference between them. I tested this stage by sending the same audio into each set of op amps. I then put one scope probe in the output of one chain of op amps, and the other probe on the output on the other chain. Wow. Bingo. 90 degrees of phase shift across the 300 -- 3000 Hz audio spectrum.
Emboldened by this positive result, I put the completed stages together this morning. They passed the smoke test. Then I tuned to 40 meters. Wow again! As promised, opposite sideband rejection without resort to crystal filters. But as luck would have it, I ended up with a configuration that suppressed the Lower Sideband. For 40 meters, obviously I needed to suppress the other side of zero beat. But all I had to do to remedy this was to reach into the DDS box and switch the I and Q jumpers on the M0XPD/Kanga UK Arduino AD9850 shield. This switch put me on LSB. Very cool.
Here is a view from above:
The AD9850/Arduino DDS box is in the bottom center. Above that, near the center of the picture, is the board (from N6QW) with the two SBL-1 mixers and the initial AF amp stages. The small green board above that is the IC phase shift network. At the top of the picture you see the 3000 Hz low pass filter. Below that, the board with the little blue pot has an IC AF amplifier and a 300 HZ high pass filter.
I still have to build the audio amplifiers prescribed by the designer, Rick Campbell KK7B. But obviously I am already having a lot of fun with phasing. Here is the QST article on Rick Campbell's R2 receiver:
https://www.arrl.org/files/file/Technology/tis/info/pdf/9301032.pdf
Posts
.jpg)
Sounds awesome, Bill. I've been reading up on phasing receivers as well. Just one more project to add to the wish list....
ReplyDeletenot bad :) :) :)
ReplyDeletecomplimenti caro amico Bill ;)
This comment has been removed by the author.
ReplyDeleteFB Bill. Can you point to the phase shifter design? Seems it beats the Tucker Tin's passive network...
ReplyDelete73
Dex
Silly me. It's in the link above. Thanks Bill.
DeleteAre you going to post any videos of the radio in operation?
ReplyDeleteURL gives construction and component details - has anyone produced a Spice-analysis of AF phase-amplitude performance? This shifter seems to be three-point where the Tucker Tin was 2-point. I sort-of grasped an idea of that at the time (well pre-Spice), but am interested in this - and/or a comparison if that can be done or found.
ReplyDeleteDex
DeleteI modeled this in LTSpice, but there's also a very powerful free design and analysis software tool called QuadNet from www.tonnesoftware.com. Give it a try...it's probably all you'll need.
Joe
W3JDR
This comment has been removed by the author.
DeleteHi Bill!
ReplyDeleteI too have been pursuing the phasing receiver approach. Mine is somewhat similar to the NC2030 but with some of my own twists. I'm using a Tayloe style QSD based on an FST3257 bus-switch clocked by an Si5351. I'm using a 6 pole all-pass audio phase shift network based on 2 opamp chips and getting minimum 45 dB opposite sideband rejection across the 2.4 khz audio band. I have a switched-in hybrid cascode RF preamp which gets input sensitivity down to near 0.1 uv in an SSB bandwidth and also provides a point to inject an AGC voltage if desired. I'm working on another AGC approach that uses an AVR microcontroller to sample early-stage audio and control a pair of digipots in a feed-forward gain control fashion to achieve over 80 db of software-controlled AGC w/S-meter output. I also developed a very nice color TFT display with 2 digital encoders for frequency and audio-gain, controlled by a cheap Arduino Nano board. Once I have the AGC working, I'll publish the details on my web site (w3jdr.com). Later improvements will include a companion phasing SSB exciter.
The phasing approach has many advantages including easy spur-free multi-band operation and is easy to reproduce on any/all bands since everything after the QSD downconverter is at audio frequencies.
Joe
W3JDR
This comment has been removed by the author.
ReplyDelete