U12 Rectifier - purple flash
 

One of my favourite sets - a GEC BC3754 which I've been using every couple of weeks for about five years now - blew its 1A fuse today when I switched it on.

I powered it up with a lamp limiter and the rectifier (U12) was glowing bright purple, and the lamp limiter was bright.

With the rectifier out, the lamp limiter was dim and the rest of the valves lit up as normal. I checked the voltage at the rectifier base and they looked good - about 600 volts and 4 volts.

I've never seen anything like this before, but I've read elsewhere on the forum that purple can mean air has got into the valve, so would it be safe to assume that it's dead?

If this is the case then does anyone know any useful and more available equivalents - a quick trawl on the usual sites suggests U12s are rare and a bit pricey. The Valve Museum has the following suggestions:

43IU
DW4/350
R2
R3
R4
RV120/500
U14
UU5

In the worst case I could always use a couple of diodes, but I'd like to keep the radio original.

Thanks all,

Joe

Re: U12 Rectifier - purple flash
 

I have seen a UU5 go like that.
It was a bright green flash and I gut the power before anything else failed so I was able to fit a replacement.
I would salvage the base from the old one and fit diodes and resistors into it so that when you find another it will be a quick swap.

Re: U12 Rectifier - purple flash
 

Check the transformer windings and the valve holder, a bad connection to one of the anodes would overload the other anode and possibly damage the valve. While unlikely it’s easy to check and may save another valve if there are problems.

Re: U12 Rectifier - purple flash
 

That bright purple glow could well have been a low-impedance discharge between the two anodes, i.e. across the whole HT winding- that would have likely ruined the transformer in short order if the 1A fuse hadn't blown quickly. An object lesson and good example as to why a low-value mains fuse is a wise move! That U12 isn't worth risking a vintage transformer for any longer.

Note that some of the equivalents to the directly-heated U12 listed are indirectly-heated types- these will result in a somewhat delayed HT appearance relative to rapid-heating directly-heated types which is no bad thing, but they also tend to have a lower forward voltage drop, so HT may well be a little higher. Not something too bad under most circumstances by itself, but nowadays mains supplies can be towards the upper tolerance limits, the cumulative effect could well be unexpectedly high running HT compared to anything noted on contemporary circuits,

Colin.

Re: U12 Rectifier - purple flash
 

I would have a look on Ebay for a UU5. You should be able to get a good one for around £10 plus postage.

Re: U12 Rectifier - purple flash
 

You could re-stuff the U12 with something like a EZ81, even with only 4v on the heater it may work quite well and look original. Could you find a 6v heater feed somewhere on the chassis? Sorry I don't know the set you have.
Or remove the base and cut the wires, install 2 1N4007 diodes in the base cap. You then need a series resistor somewhere to drop a few volts

Re: U12 Rectifier - purple flash
 

I've seen the dreaded purple glow a few times. If this happens a fuse in the HT centre tap, or even each HT sec will save the transformer.

Main problem with valve substitution can be the heater current requirement being higher and overloading what may be a "tight" transformer.

Ed

Re: U12 Rectifier - purple flash
 

I think it would wise check that C25 and C26 (smoothers) are OK before replacing the valve just in case any AC got to them?
Perhaps use a couple of UF4007 diodes to 'test' it is all working before going any further, or (sacrilege) wire one in between each transformer secondary and valve base to save future problems...

I will look in my 4 volt valve stash later, some of the numbers ring a couple of bells.

Alan

Re: U12 Rectifier - purple flash
 

Hello Joe,
I have just opened Pandora's Box and discovered a couple of U14s. They replaced the U12 and were listed in Osram/GEC data books as replacements for the U12. It became the U12/14.

I rigged up this horrendous bit of 'test kit' and discovered that both were 100% and remain so on test.

If you are not fixed up and would like one FOC please let me have your details via PM and I will pack and post. John.

Re: U12 Rectifier - purple flash
 

That's interesting, according to the Valve Museum website, 'Initially the U12 and the U14 were the same valve. However, some valves performed better than others on final factory test so the best ones were graded and marked U14 whilst the also-rans were marked U12.'

So a U14 would be a perfect substitute, thanks for the offer John (HeatherCathodeShort). I'll PM you.

I've also gone back to the radio (rectifier removed) and checked the HT between chassis and anodes, in both cases around 300v so it looks like both 2y windings are OK.

I've also connected a resistance meter between the one of the heater connections and chassis, and I can see the resistance build as the capacitors charge, falling to about 100K, which is about right given the values of some of the resistors bridging HT to chassis, which gives me a rough indication that the smoothing electrolytics look OK.

I'll try to get a replacement rectifier and power the whole thing up with the lamp limiter. If it looks like it's drawing too much current then I'll replace all the smoothing electrolytics (I think there are three)

Joe

Re: U12 Rectifier - purple flash
 

Just to conclude this thread: John sent me his U14 rectifier and I powered the radio up on the lamp limiter and it worked well for a minute, then there was some crackling accompanied by the lamp dimming. I suspected one of the windings on the TX might be about to give out, so I shut down.

I then tried removing and replacing the rectifier in its socket, powered up again and it was fine. I ran the radio for about 4 hours this morning direct to mains with no funny noises or smells, so fingers crossed it's all OK.

If the TX does fail I had an instructive experience last summer manually rewinding an HT 2y for a Murphy of similar age - I'd expect a similar job might be on the cards.

Thanks for all the advice, and to John for the valve,

Joe

Re: U12 Rectifier - purple flash
 

I had a similar problem with a U12 in a GEC AC37 a while ago. It had the purple glow but in my case it didn't blow the fuse, it just didn't work (there was no HT voltage).

As a temporary fix I removed the base from the failed valve and mounted a couple of diodes and a suitable dropping resistor to the valve base so it could be plugged in in place of the valve. Of course several months later the "temporary" fix is still in place!

Re: U12 Rectifier - purple flash
 

This sort of suggests to me that there might be a problem with the valve holder...possibly one of the pins not making proper contact. have a careful look and see if any of them are broken or maybe just need clenching together (carefully).

I would have thought that if the transformer was breaking down, more likely the lamp would have brightened rather than dimmed. Of course windings can go open circuit which would probably cause dimming whereas shorted turns would cause the lamp to brighten. Shorted turns are unlikley to 'right' themselves but perhaps an O/C winding might correct itself particularly if the windings are terminated with solder tags and there was a dry joint at the terminal.

I would carefully check the valve holder first and then check connections at the transformer itself.

Re: U12 Rectifier - purple flash
 

I agree with Sideband- of all the valves in a set, the rectifier is usually the one with highest heater current and a prime case for proneness to contact problems in ageing holders and something with a 4V, 2.5A filament/heater needs as much good contact as it can get! A directly-heated valve heats up and cools down very quickly, so producing dips, bumps and splats on the HT if there is iffy contact and the valve is moved around, and consequent changes in transformer input power and limiter brightness. It sounds as if a good re-seating of the valve has solved the problem for now, but I've sometimes ended up either changing the socket for a good quality, tightly-gripping type, or (particularly hollow-pin octal types) having to clean and thoroughly resolder the bulb-to-pin connections.

Re: U12 Rectifier - purple flash
 

Thanks - that all makes a lot of sense. If it plays up again I'll check over the holder.

Re: U12 Rectifier - purple flash
 

When I picked up a valve radio, the previous owner said that they used to tap the bakelite case to get it to work when they used it many years ago.

A gentle squeeze with a radio pliers on one of the contacts of the rectifier valve holder solved that problem.

Re: U12 Rectifier - purple flash
 

I've seen the purple glow in a 4 volt rectifier and I've seen a white glow in a 5U4 rectifier and in both cases I concluded that the valves had gone soft. I fitted replacements that cured the problems in both cases.
Alf