This is a four-valve (including rectifier) receiver. Typically for Pye the cabinet is somewhat uninspired, but the electronics appears to be fairly well designed. Despite being a fairly cheap second-set radio (it sold for £12 in 1951), the circuit includes delayed AGC and negative feedback in the audio stage. The speaker is a respectable size, so I expected this set to sound good.
The vale lineup is UCH42 mixer-oscillator, UBF80 IF amplifier, detector and AGC, UL41 output and UY41 rectifier. Note the lack of audio preamplifier, making the set a short-superhet. This type of set is intended for use on more local stations since it probably would not have sufficient gain to receive weaker signals at adequate volume. The delayed AGC helps in this regard though, since it allows the set to run at maximum gain unless a fairly strong signal is received.
My example cost about £5 at the BVWS swapmeet in Wootton Bassett in summer 1999, and has a small crack in one back corner of the cabinet. As can be seen from the photo, the front was in need of repainting too.
The set uses the infamous UL41 valve in the output stage, and since nobody in their right mind sells a set with a good UL41 for £5, I was fully expecting it to be useless. The cremated state of some high-power resistors also suggested that all was not well around the output and power supply sections, so the first job was to test all the valves.
Not surprisingly the UL41 was useless (excessive grid leakage which is typical for these valves). The UY41 was also condemned because it was flashing over internally. The UBF80 showed very low emission on both diodes, but since it was otherwise serviceable it was put to one side for testing in the set after repair. The UCH42 was fine. So one out of four – but then the set only cost £5 and was sold as-is.
I decided to start by investigating the wirewound resistors. On the top of the chassis is mounted a 150R resistor, tapped at the midpoint. The two sections of this are connected in parallel with the two dial lamps. Meter checks showed that one side was open-circuit, but since I did not have a suitable replacement (and two separate resistors would look untidy) I decided to investigate further. Much of the green coating was missing, which made it easier to find the break. It turned out that the resistance wire itself was OK, but it was not making proper connection with one of the leadout wires. By wrapping a thin piece of tinned copper wire tightly around the body of the resistor and the leadout wire I was able to complete the connection. To ensure reliability I soldered this wire to the leadout wire and the resistance wire.
The two wirewound resistors below the chassis were beyond redemption, so they were replaced. The dropper resistor is mounted on the top of the chassis in a metal surround. This had clearly been replaced with a RadioSpares component in the past, but this replacement was the correct type and size. The only problem was that the previous repairer had threaded the wires through the holes in the back of the metal surround. Because the wires were quite short and the resistor mounting method is not very secure, this caused the body of the resistor to turn so that its tags contacted the metal surround. I decided that the easiest solution was to remove the wires from their route through the holes in the surround, and connect them directly to the resistor. They then naturally prevent the resistor from turning and shorting to the surround. I connected the voltage selector tapping permanently to the 236V-250V setting.
Under the chassis were several of those horrible wax-paper capacitors. The one across the mains switch had blown itself to pieces, and several others had signs of dripping wax and other distress. This may have been due to the heat from the overheating wirewound resistors, but in any event I had no choice but to replace the lot.
A previous repairer had replaced the 32+32uF 350V smoothing capacitor with a 50+50uF 450V component. Although there was nothing wrong with this replacement, I decided to replace it with the correct value – mainly because I thought the 50+50uF component would be more useful for another repair! I have a number of the triple 32+32+32uF 350V capacitors that were sold by On the Air. These are designed for PCB mounting but are the correct diameter to fit into normal chassis clamps. I originally planned to leave the third 32uF section disconnected. However one end of the 20uF output valve cathode resistor bypass capacitor was badly swollen. Although the remaining section in the triple 32uF can clearly had an inappropriately high voltage rating, I decided it was worth trying it in this role. Generally electrolytic capacitors should not be used at less than 5% of their rated voltage – 17.5V in this case. In practice though it worked quite happily at 7V.
A couple of 3.5V 0.15A dial lamp bulbs completed the electrical repairs, and it was time for a test. I fitted known good valves in place of the faulty and suspect ones, and switched on. Nothing happened. No dial lights, no valve heaters, nothing. A few checks with the meter showed that the heater chain was open-circuit. I narrowed this down to the UCH42, but I knew the valve itself was OK.
The contact in the valve holder was broken – this became apparent when I tried to tighten it with a small jeweller’s screwdriver and found one side was completely loose. Fortunately the valve holder was one of the black McMurdo ones which are commonly used in sets of this age. Rather than replace the complete holder, I decided to just replace the faulty contact with one from a spare valve holder. I unsoldered the wire from the tag, then removed as much solder as possible from the tag and straightened it. It could then be pushed upwards with a pair of long-nosed pliers, and the whole contact could be removed. I removed a contact from a spare valve holder (actually a 9-pin type, but the contacts are the same) and fitted this into the holder in the set.
Time for a second test – and now things were much more promising. The waveband switch and volume control crackled, and the alignment was clearly out, but the set worked. The HT voltage and the voltage at the anode of the output valve were about right too. There was some wisps of smoke from the dropper resistor, but since it had been handled a lot during its refitting, this was to be expected while the grease and grime from my fingers burnt off.
The top adjustments in the two IF cans had clearly been tampered with in the past, so these were readjusted by ear on a weak station to bring the set back into alignment. The underneath adjustments on the IF cans and all other alignment adjustments looked undisturbed, so I did not touch them. I left the set running for a few minutes, then decided to try the original UBF80 valve which had been condemned as low emission by the valve tester. It worked just as well as the good one, so I left it in the set. I did not try the UL41 and UY41 though, because the results from the valve tester showed them to have more serious problems than just low emission.
The chassis was then left running for several hours, while I started working on the cabinet. For a short superhet, the performance and sensitivity is surprisingly good. It is not quite as good on very weak stations, but these are generally not worth listening to anyway. On the more local stations the set sounded really good with my test speaker.
As can be seen from the "before" photo at the top of the page, the paint on the front panel was flaking off so there was no option but to clean and repaint it. Before cleaning the cabinet I took a paint colour sample by attaching a piece of clear adhesive tape to the paint and then pealing it off again, complete with some paint.
I took this sample to Halfords, to compare with the coloured caps on the tins of car spray paint. The closest I could find was "Rover Opaline Green Metallic". This is slightly paler than the original green, but I felt it was close enough.
The speaker and tuning scale were removed, before the cabinet was washed in hot water and washing-up liquid. The hot water helps to soften the remaining paint, making it easier to remove. Although much of the paint came away very easily, some sections, particularly those in the speaker grill area, were more stubborn. Soaking in the hot water for half an hour, followed by some vigorous scrubbing with an old toothbrush removed most of this, and the remaining bits in small corners were removed by careful scraping with a small screwdriver. The cabinet was then rinsed in clean running water, and left to dry thoroughly in a warm airing cupboard for two days. The knobs were washed at the same time.
Before spraying, the crack in the top back corner was repaired. By wedging a toothbrush handle between the top of the cabinet and one of the internal supports I was able to completely close the crack. I applied a small amount of super glue to the crack from both the inside and the outside. This was drawn into the crack by capillary action, so after a few seconds I removed the remaining glue from the outside of the cabinet by wiping briskly with a clean rag. After half an hour I removed the toothbrush and the crack remained closed. To add further strength to the repair I carefully roughened the inside of the cabinet with fine wet-and-dry paper, then covered the area with Araldite (epoxy resin) adhesive.
The areas of the cabinet that did not require painting were masked off with masking tape and old newspaper, then the front was carefully spray painted. This was done outside on a dry and relatively still day. The spray had to be applied from various directions to get good coverage into the speaker grille. Fortunately I was able get good coverage with just one coat of paint. The cabinet was then bought back indoors to dry, to reduce the risk of dust etc. being blown onto the wet paint.
The masking tape and newspaper were removed after a couple of hours, and the paint was then left to dry thoroughly for 24 hours. The unpainted sections were then polished, with Bake-o-bryte polish and finished with a spray household polish.
The missing lettering on the knobs was carefully filled in with gold enamel paint. After 15 minutes or so, when it was dry but not fully hardened, the areas were gently cleaned with Brasso to remove the bits of paint from the surface, leaving just the recessed characters gold.
The tuning scale was originally fixed into the cabinet with spring clips pressing against protective rubber strips against the glass. Over the years these pieces of rubber had become hardened with a sticky surface, and were stuck to the inside of the tuning scale. They had affected the printing, so for long-term stability I decided to remove them. This also took away some of the printing as expected, but these areas were just black so it was easy to repaint them with black enamel paint. The spring clips were then fitted with synthetic rubber sleeves which will (hopefully) not degrade.
As usual, the reassembly was straightforward and uneventful, and the set worked well on completion. The second photo here shows the completed set – hopefully you will agree that it is an improvement.
I recently received the following comments from Paul Earland:
I have read your notes on repairing the above receiver. I am doing up the same model. Just a small point are you sure the front colour is a green shade. I note the Trader sheet refers to a silvered finish and had assumed the greenish tinge was more to do with the silvering going off.
It certainly looked like a pale green to me (see the top photo), but Paul could be right. As I mentioned before, the green I used to respray the front was actually lighter than the original, so if the green colour is due to the silver going off it must have gone off quite badly!