Still no Bitx40 here, it's been over a month. Perhaps somebody out there would be kind enough to try my pet scheme for transmitting CW. Involves 3 resistors, a cap, and a straight key. First off we have to move the BFO down in frequency by a khz or so, such that it is in the crystal filter passband. I assume the boards out now don't have C103 stuffed. So install a small variable cap at C103, likely needs well under 100pf, Set it for minimum capacitance, then tune in a CW station. Find a setting for C103 such that you can tune that station through zero-beat and still hear it as it starts going up the other side. (I suspect the next board rev of the Bitx40 will use the 3'rd Si5351 channel for the BFO, allowing these BFO adjustments to be done with just Raduino code. Reduces part count, and also gives us USB/LSB selection. Or could borrow that MV209 from the analog VFO so you can adjust the analog BFO with a DC voltage from a pot or switched resistor network.) Now take a 10k pot, a 1k resistor, and a 50 ohm resistor, string them up in series. Wire the top of the 10k pot to TX 12v, the bottom of the 50 ohm resistor to ground. Run a wire from the junction between the 1k and 50 ohm resistors to one terminal of a straight key. Run a wire from the other terminal of the straight key to the top of C107 near the balanced modulator. When the straight key is pressed, a DC voltage of about 0.57v max will be given to the top of C107. If you care about key clicks, maybe put a 4.7uf cap between the top of C107 and ground. (You could add a stereo phone jack, with tip to TX 12v, ring wired to the top of C107, and sleeve to ground. Then the 3 resistors and 4.7uf cap can all be out there on the straight key, and when you pull the phone plug this is set back up for SSB operation.) Set the 10k pot to maximum resistance, remove the microphone, and have some way of measuring the transmit power. Turn on the rig, put a jumper across push-to-talk so it is always transmitting. There should be no output to your power meter, because the balanced modulator is still balanced and thus suppressing the carrier from the BFO. Now press the straight key and slowly advance that 10k pot until you see 5 watts out (make sure that IRF510 isn't getting hot). OK. Did you see 5 Watts? What is the voltage at C107 when you do? Once we know that, we can get rid of the pot and just have two resistors. If you zero beat a received signal, pressing the key should transmit on exactly that frequency, which is handy. But normally when receiving CW we'll prefer the BFO at the LSB position for single signal reception. I have a one transistor circuit simulated in LT Spice that allows C107 to be keyed with a clean trapezoid from a Raduino digital pin. But I'll wait to see if the above works before dragging that out. On Thu, Jan 19, 2017 at 07:02 pm, Jerry Gaffke wrote: Easier and possibly cleaner would be to inject DC via a diode into the top of C107 to unbalance the modulator during a dot or dash ? |