I picked this old radio up at a local sale, the guy that was selling it told me that it worked perfectly.

This set runs from AC or DC mains. It can manage this because there is no mains transformer (DC won't go through a transformer). The heaters in the valves and the scale lamps are wired in series, much in the same way that Christmas tree lights work and there is a diode valve that rectifies the incoming mains for the HT supply.

These radios were first sold in 1956 for £25 3s 6d.

AC/DC radios are especially dangerous things to work at. In a traditional valve radio, the circuit is isolated from the mains supply by the transformer, but in these sets one of the mains wires is directly connected to the chassis. It's important to make sure that the wire that connects to the chassis is the neutral and not the live (which would be a more dangerous arrangement). If you're going to do any work on an AC/DC set while it is powered up you should use an external isolating transformer.

I fitted a new mains lead (the old lead was a bit sick looking), making sure the neutral connected to the chassis. I brought the radio slowly to life with a variac and was surprised to hear some stations coming through. All the stations were very distorted and the radio wouldn't work on FM at all.

Click to see troublesome part of circuit in more detail.

Some voltage checks with my meter revealed that there were a couple of volts on the grid of the output valve and that the anode of the audio valve was at a much lower than expected voltage. I changed the coupling capacitor (C52) between the AF and output valves and fitted a new resistor (R25) at the anode of the audio valve. This brought all the voltages into line and the audio quality was a lot better. I left the radio running and was enjoying the nice sound, until after half an hour the audio gradually became more and more distorted. The voltages around the output stage when still fine, so I started to think about the UL84 output valve - these valves have a reputation for this sort of thing.

I had a quick look at the FM section and found the anode voltages of double triode UCC85, which works as an FM frequency changer, were very high. The rest of the FM circuit looked OK, so I suspected the valve was low emission a quick test on a friends valve tester proved I was correct.

I fitted a new UCC85 and FM suddenly sprang to life, FM sounded very good, but suffered from the same problem with distortion after 30 mins. I tried another output valve and it didn't make any difference at all. Eventually I tried another speaker and was amazed when that cured the distortion problem completely. The old speaker must have been affected by the heat inside the case.