Repair of an Eddystone Model EC10
The EC10 is one of Eddystone's very early solid state communications receivers. Using only PNP Germanium transistors, it covers 550 kHz to 30 MHz in five switched ranges. The radio is a single conversion superhet, with a BFO for SSB and CW reception and a switchable audio filter for CW. I have a mark-2 version, essentially the same as the original, with the addition of a fine tuning control.
Frequency coverage - 550 kHz to 30 MHz (in five bands)
Intermediate Frequency - 720 kHz
Sensitivity - <5uV for 15dB s/n ratio
Image rejection - 50dB at 2.0 MHz, 20dB at 20 MHz
The EC10 I now own had been 'previously enjoyed'. At some point in its history the audio PA had died, and a Velleman kit amp had been installed instead, this too was dead when I took possession.
The first obvious fault was a large char-mark on the PCB in the audio PA stage. Inspection of the circuit diagram and comparison with another EC10 showed that the immolated part was R48 - a 5 Ohm 3 Watt resistor in the emitters of the PA transistors. Not having a suitable part to hand, this resistor was replaced by four series-parallel 4.7 Ohm 1/2 Watt parts.
With the audio PA working, but no station heard, it was time to work back from the PA. With the audio filter and BFO both off, next back from the PA is the IF strip - specifically TR5 and TR4, both OC171 devices. A preemptive snip of the case/shield wire on both OC171s (in these parts, the Collector or Emitter often shorts to the case and down to ground) proved ineffective. A quick check of the voltage around TR5 showed that its emitter and collector were shorted. The OC171 was stolen from the BFO to replace TR5 - success! The radio began to receive.
I now had a working radio with a dead BFO - somewhat limiting if you want to listen to more than AM stations. An attempt to use an OC71 (more of an audio transistor than anything) in place of the purloined OC171 failed. Zero oscillation. Having no other germanium transistors to hand, I dug into the spares bin for a PNP Silicon transistor instead. The 2N3906 is a pretty general purpose part, sibling to the 2n3904 (NPN) beloved of QRP constructors. A straight forward swap of the OC71 for the 2N3906 did not work. Silicon transistors need about 0.6 Volts to turn on compared to germanium's 0.3-0.4 Volts. The easiest fudge for this is to replace R35, a 47K Ohm part, by a 22K Ohm resistor.
Next up will be to re-align the set and perhaps switch out all the germanium transistors in the RF and IF sections for their silicon counterparts.