[EdWilliams]

I bought this radio because I fancied a go at something pre-war. I originally tried to buy a Philco 444 on ebay but got totally annihilated in the bidding, so on the rebound I had a look on Gumtree and found this Philips 274A TRF set not too far away from me.

I arranged to meet the seller at his house, who turned out to be the grandson of the original owner. The seller plugged in and switched on before I could stop him, which resulted in a mixture of motor boating and very loud mains hum. He wanted £50 for it, but when I explained that I wanted to restore it to working condition his price went down to zero. I've never had the experience of a seller trying to negotiate the price downwards before! So I started the unusual process of the buyer trying to negotiate the price upwards - in the end we settled on £30.

On getting it home I took the back off and had a look inside. First obvious problem - the anode connection to the second RF valve (S4VB) was broken off flush with the valve top leaving no visible wire to connect to. Drat! I then took the chassis out from the case and spotted the second obvious problem - the winding for the volume control was broken in at least two places. Double drat!

As the success of this project was clearly going to depend on the health of the valves, I thought I would do a minimum set of fixes first to see if it could live, and then come back and have second and third goes to be a bit more complete. So first I attempted to fix the S4VB by carefully filing down the glass pip on the valve top until some of the anode connection wire became visible (trying not to let the vacuum out), then blobbing some solder onto the anode wire so a new connection could be made to the top cap connector, and finally I araldited the top cap back onto the valve. Next I replaced the wire wound volume control with a new log pot (shunted by a lower value resistance so that the correct combined resistance was obtained) and finally I replaced the output valve grid coupling capacitor and the two smoothing electrolytics. I finished this work at 11 pm one evening, switched on in great anticipation, and - nothing, apart from a copious amount of mains hum. Went to bed in disgust.

The next evening I had another look and quickly spotted the deliberate mistake. On this radio there is a plug used to select either radio or gram operation, which I had inserted into a completely random socket on the rear of the chassis. I moved it back to the correct place and success, the radio worked, normal mains hum only.

I discovered two things about the set over the next few days: 1) the HT was very low at around 90 V, 2) the volume control behaved strangely - on a strong station it was impossible to turn it down quiet enough.

I then bought some more capacitors and set about replacing those above 1 nF. In the spirit of getting on with things I disconnected those in the metal container and replaced them with 6 new ones connected as I saw fit (see photo). The volume control worked properly after this; looking at the Trader sheet for the equivalent Philips 834A I think that C3 (0.1 uF) must have been horribly leaky and preventing enough control voltage reaching the grid of the MM4V first RF valve.

With hindsight I think that the use of a log pot for the volume control is wrong. Due to the unique way in which the volume control works on this set (adjustment of front end gain) I suspect I actually need a linear pot or even an antilog one if such a thing exists (presumably the fixed tapping on the original wire-wound pot is an attempt to get a decent control law). As it is, as the volume control is increased from zero you get, nothing, still nothing, heck that's loud!

The rectifier valve had been replaced in the past by a DW3 which I thought might be poor and explain the low HT. I got lucky and bought an original 1821 rectifier which is clearly in better condition - I now get 190 V, still lower than spec but a lot healthier and not too bad I hope. The performance is better for it.

The set now has the correct complement of valves except for the 994V triode detector which has been replaced by a Z8154 (does anybody know what this is - I can't find out anything about it?)

I replaced the non-working scale lamp with a new one (sold as a tail-lamp for a 6 V moped), and started using the radio on a daily basis.

The sensitivity and selectivity are pretty good for a TRF set, but the anode bend detector doesn't like the strongest local station - Smooth radio - and distorts noticeably. I don't like Smooth radio either so the feeling is mutual. It's OK on other stations though.

Over the Christmas break I dismantled the metal can containing the six capacitors and fitted the new ones inside. The can was date-stamped February 1934. Not too hard with the aid of a blowlamp to melt the solder holding the base on; the worst bit was fitting the completed can back in the chassis and getting the screws re-engaged.

The wooden case for the radio is in reasonable condition - a few minor dints and scratches but really quite good for its great age. I'm looking forward to it being 100 in 16 years' time!

Some more things to do: 1) I'd like to re-instate the original volume control if I can think of a way to fix it; 2) it's now occasionally showing instability on high volume settings - I think this is intermittency in the metallised screening of the MM4V first RF valve. It behaves itself if I wriggle the valve envelope a bit or temporarily make a connection from the screening to chassis. All good fun though.

Photos below.

Ed