WEAK-AM

OK, I could not resist… I ordered a Ramsey AM25C and put it together over three lunch hour sessions. There were no major problems assembling the kit. The coils were easier to wind than I expected.

When I turned it on, there was no smoke, but there was no output either… until I found I had made an error setting the DIP switches, and corrected that. Then I got a strong carrier on 1500, accompanied by a massive amount of hum! Keeping the unit powered on for very short intervals, I made all of the adjustments according to the manual. When I was finished, the hum was still very strong, and the output power measured 23.9 dBm into 50 Ohms (over 200 mW)! I also noticed that the regulator transistor (Q9) and the driver transistor (Q5) were both hot.

Thinking to reduce the power a bit, I padded down the drive into Q5 by putting a 1000 pF capacitor across R16. The power went down a few dB, and the hum dropped significantly. I bumped that up to 1500 pF and the power dropped to 19 dBm. The hum is essentially gone, and Q5 is a lot cooler now. The 2nd harmonic is down just shy of 40 dB. I do notice some close-in spurs at around +/- 250 kHz, which I will have to investigate. They are pretty far down, but I would like to eliminate them. I am guessing that they may be caused by instability in the audio circuitry.

The overpowered ac adapter supplied with the kit appears to be the reason for the hot regulator transistor. Even after padding down the RF drive, I am still measuring almost 18Vdc into the board. The adapter is rated at 1 Amp at 15 Vdc, and since it is unregulated, the voltage goes even higher when the current drain is less. So I think the problem here is not that Ramsey supplied a cheap adapter; rather, it is over-rated for this application.

I like the 50 Ohm output. It simplifies designing the matching network for a base-loaded antenna, and making measurements on the transmitter’s output. The audio section of this unit doesn’t look like much to write home about. I have not made any measurements of it with modulation yet, but have read that it’s difficult to get it to hit 100%. I’ll write with more details after further tests are completed.