http://nnamd.duckdns.org:8073/
David,
Depending on where in Georgia you might be able to see your signal here.
The ear can deceive. I have several MB8719 boards ranging from the Grant to the 140/142 GTL. The 148 is second gen, not far removed from the BIG crystal.
Schottky diode mod
Hand built crystal filter. Select, sweep, adjust, repeat.
Doubled up 455 filter
Bandpass filter 6db down @ 26.6 and 28mhz
Close attention to receive stage gain.
2SC1306/1307 with active cooling . (old CPU coolers and fans)
Multi turn potentiometers on the driver and final bias.
5 pole Chebyshev filter
AM power adjust only 560 ohms above ground
Limiter intact but a 5k ohm inserted in the sense line.
8 volt supply for AM lifted at the jumper from the MB3756 and B+ from the filter cap applied instead.
Power supply was in a word overkill. 10A draw won't sag.
Driver and final supply chokes rewound for better current carnying capability.
Turner Super Sidekick with a somewhat higher impedance cartridge that is (for my voice) flat from 40-4K
Graphic equalizer between mike and radio. I kill 120Hz and down, 3.2Khz and up.
Receive audio is delivered to either an old Fischer stereo or a somewhat upscale sound card in a workstation.
I really have to stop here and get technical. The computer isn't normal at all . You see there are laptops, desktops, gaming rigs and workstations. Mine is 8 cores@3GHz with 32G of 1.6GHz ram, 4T Raid 6, Nvidia Quadro SLI, 24-bit/192kHz development card. A beast. I have real time audio processing available that most do not. Still no matter what the processing it's still transistor and not tube.
Back to the radio. I had the advantage of some good gear to look at the radio RX and TX. Although it could be turned down to 500MW and swing to the rail it looked ugly. On the other side of things it would go into compression around 7.5W. A bit on it's 1db compression limit, It looked like the audio chip distortion begins to take over around 7W so it really isn't dependent on RF linearity exclusively.
On SSB carrier suppression was excellent for the design and parts used. Sweeping the filter endlessly and careful tweaks help. Wash rinse repeat...
Okay, so that is a heavily modified radio. I have a Grant that is nearly stock save for a tune up and AM power mods.
I also have an unmodified 148 GTL small crystal .
Now back to
http://nnamd.duckdns.org:8073/
It's the closest full, general coverage SDR I could find and this may change by day by day.
These radios are surprising and somewhat predictable on transmit. Even without extensive filter modifications they are "brick wall" on the opposite side (up freq on LSB and down freq on USB) and only slightly less on the near side due to audio frequency peaks . They all predictably suffer from the usual "blowing out the sides" if the mod limiter is removed, if heavy clipping, processing is used, or if there is enough AM output that the distortion limits of the 7222 are exceeded.
On receive there is the inescapable difference between the single and dual conversion radios. The single conversion radios may hear what the dual conversion may miss bu the dual conversion radio will hear through man made and atmospheric noise. Curiously the noise blanker in the single conversion radios works better , to my ear anyway. Only my saturated, over modified, wash rinse repeat filtered, used to be a TR-296 board, Frankenstein radio begins to compare with my , wait for it ...
uPD858 (TRC-458)
Lastly neither the single conversion 8719, the dual conversion 8719, (modified or otherwise) the or uPD858 can be identified or separated from one another or my TS-830S by listening to or looking at the waterfall on a web SDR site.
Back to business. Tube receivers are overrated. Tube audio amplifiers are not . Any of the above receivers will out perform all but my Hammarlund HQ-170 or Hallicrafters SX-42 . I do own one audio DSP "shack in the box" a FT-857D Yaesu that will absolutely wear me out in short order. I have one experiment sitting here that will have to wait for winter. I have a class "A" tube audio amp that I want to graft to low level audio out of any of the above transistor receivers. Look up "tube headphone amp" for reference. I seriously think that not nearly enough attention was paid to the audio output of any sand state CB. Just a look at the white papers for the Ta7222, Ta7205P or the 1156H shows that the designers felt upwards of 10% THD was acceptable at less than 5w output. If anyone (Chris, Andy?) wants to chime in with the differences between voltage amplifier and current amplifier distortion please do . I just know that if I have a choice I'll gladly listen to voltage amp (tube) distortion.
Conclusion
I'll gladly accept advances in transistor receiver design if I can listen to the warm glow of vacuum tubes.
I'll willingly ditch present day audio dsp for a tube audio output. Give me a true IF DSP and all bets are off other than I still crave filaments in my audio chain. My ideal then would be either a dual conversion 8719 or the uPD858 with tube audio output, grid driven (swamped grid input for stability) single 6146 output, direct digital synthesizer tuning with RIT, wash rinse repeat filter (8719) or careful impedance termination (uPD858) in a DAK X case (modern) or Hallicrafters SX-28 case (retro) . No I wouldn't butcher a classic it would be reproduction.
Sorry for the rambling stream of consciousness brain dump (no I'm not) but I wanted to walk the thin line between dry data sheet specs and personal experience. Obviously I've given this some thought but as always life intervenes
