I've been working on a new version of an IF amplifier using 2N3904 transistors, trying to build on and modify lessons learned from earlier versions that I've built over the last year or so. This is the result so far:
Lower left: detector and large filter capasitor.
Upper left: AF amplifier board
Lower middle: 2nd IF amplifier module, Lower right: 1st IF module
Upper middle: SSB/CW product detector with BFO in the small shielded compartment
Mixer module, two FET cascode can be seen on the far right.
The product detector is a new addition, I don't know if it works as intended. If it doesn't it can be easily removed as I've mounted each circuits on small boards that can easily be removed. That's another lesson learned, as I don't have to scrap the whole project just because one of the modules fails. I can just replace the module affected. That's what I think anyway.
Schematic of AF amplifier:
Sensitivity is about 20-40mV. That appears to be ideal for the signals coming out of the detector. I've added a quiescent curret adjustment, R3 (pot-meter) and Q3, to the output transistors. 1,5 to 2mA seems to be good, then the visible cross over distortion disappears and it sounds clean and pleasent. (R14 is a loudspeaker, 8 ohm)
The IF amplifier modules are quite standard 2N3904 circuits. But I've added double tuned IF-transformers to the setup to - hopefully - improve utlimate selectivity. Seems to work OK.
Schematic:
The 56 ohm resistor in the emitter is to adjust stage gain.
I've decided to 'take the digital plunge' and orderd a DDS VFO circuit. I guess it takes quite some time to arrive though. I will try a traditional VFO in the mean time.
It's done!
For once I've made a finished project, cabinet and all, 4mm thin plywood.
In the end I made coils for AM band reception, covering 503kHz to 1650kHz. Winter is AM season! And I got a bit bored by short wave only. It turnes out, I could hear a few stations on AM.
The coil calculations, as well as padding capsitors, were done using the nice software: ''Tracker Version 1.0 Copyright 2008, Robert Weaver''
The red button is to activate the frequency counter. It has a separate 9V battery to eliminate counter noise and to prevent it from draining the radio battery. The counter uses a lot of current.
The radio circuit has its own 9V battery.
Rear view:
It needs a back panel - so - almost finished project.
The MW (AM band) receiver ''in action''
The speaker is a bit small though. I would have liked a bigger one for a more full sound.
I don't want to sit idle and wait for the DDS VFO. So, in the meantime, I threw out the BFO and product detector and installed a variable capasitor. Then I made an old fashioned dial drive mechanism - without a dial. I will be using a frequency counter display.
A 6 mm knitting needle works well as a shaft. I filed it down where the tuning cord goes. It gives a 5:1 tuning ratio.
I found a 5420 kHz crystal in my junk box and then I made a pierce oscillator and a tuned antenna circuit for 5 MHz. It didn't quite add up, so I had to parallell each IF transformer with a 30pF capasitor to make it work at the new 420kHz IF frequency. The result is quite satisfying. Very low noise, good sensitivity and it uses only 7 to 8 mA when audio is low. Here's a short video:
I removed the screens between the IF modules. It turned out they were not needed. The circuit is stable. No oscillations, even when I touch components. I've been listening to 5MHz for a while now, there seems to be a lot going on on that frequency! I think I can hear talking, morse code, and maybe some announcements weakly in the background. There's some data signalling too. Quite interesting.
The output from the detector diodes is quite high, approximately 1 volt. There is enough voltage to drive an S-meter directly. I have a 500uA meter, and have a 18k resistor in series.