Trio W-38 servicing process

[qualityten]

Just resoldered the rectifier joints and repeated the cathode voltage measurement. Still 120V. But I noticed that the lamp, after being brightish at the start and then dimming as the heaters warmed, then grew bright again. Switched off. I still intend to replace the remaining caps, but why is the DC voltage so low?

[Station X]

I don't think you ever gave us the AC voltage readings for each half of the Mains Transformer HT winding with the rectifier removed. It would be useful to know this.

It looks like the HT voltage is being pulled down by a low resistance to chassis on the HT line. The most likely cause is a leaky smoothing capacitor. Unplug the output valves, as the following test will remove their screen voltage. Disconnect R301 and then measure the DC voltage across C301 (This assumes that C301 is good or has been replaced). You should see a healthy voltage here.

If not check that you have 5V AC on the rectifier heater.

If still no joy disconnect the wire from C301 to the output transformer centre taps.

If still no joy try a new rectifier valve.

[Station X]

The heaters alone in this set consume 55W and the HT is about another 50W, so I would expect a 100W test lamp to glow a bit and the HT voltage to be low, but not as low as 120V.

[qualityten]

If I disconnect the R301 at the rectifier and then measure DC to the chassis, I get 271V DC.

Heaters voltage is 4.9V DC. AC at each of the anodes is 207 with lamp in circuit and 223 without lamp.

[Station X]

That's good. Go ahead with replacing all the smoothing capacitors. That's C301, C302 (triple cap), C303 and C304. Whilst doing that check R301, R302, R303, R304 and R305 are the correct value.

Once that work is complete measure the various HT voltages again and report back.

[qualityten]

Caps already replaced and I've now checked the resistances, which are okay. I have found a dry joint from the 6V transformer tap to the MPX board, but I don't suppose that affects these tests.

Sorry to be dim, but please can you repeat which HT voltages I should check and which valves should be in or out. Thanks.

[Station X]

The HT voltages to be checked are at the postive ends of the smoothing capacitors mentioned in my previous post. You'll find the expected figures on the circuit diagram.

Given that you had low HT readings with the valves inserted I'd be inclined to remove the output valves to start with and see what sort of HT voltage figures you get.

[Station X]

Let's assume the tests are OK. If so reinsert the output valves and recheck the HT voltages. If they're OK check the anode, screen and grid voltages of each individual output valve. Assuming they're OK and there's about 20V on the cathodes ,all the valves should now be lit, including those whose heaters are fed with DC from the output stage.

That's where we came in.................................

[qualityten]

With the output valves in (sorry forgot to check with them removed, but can go back to do so), I get lower readings at all points:
C301 121V instead of 225
C302 111V instead of 200
C303 106V instead of 181
C304 106V instead of 180
Also checked C305 which has +80, not -121V

Something happens when R301 is reconnected that seems to halve the HT.

[Station X]

This is all pointing to something pulling down the HT and I seem to remember that the original fault was a burnt out R301? That would be indicative of excess current having flowed through it

No voltage is quoted for C305. You'd need to check the voltages on the anodes of V13a and V13b. +80V looks about right for a reading on C305.

Try this. Unplug the set and check the smoothing capacitors for any residual voltage. If there is any it should drop to zero as the cap discharges through the meter. Now measure the resistance between the positive of each capacitor and the chassis. The meter will initially show a low reading, but as the caps charge up from the meter it should rise to infinity. If it doesn't there is a problem on the HT line somewhere and you'll need to disconnect wires to localise it.

Even if the reading does rise to infinity there could still be a fault, but the voltage from the meter may not be enough to "break it down". try the test and see what happens though.

EDIT. When trying this you'll need to set S2 and S3 to a position where R306, R307 and R308 are out of circuit or you'll be measuring through them to chassis. These resistors will also provide an easy way of discharging the smoothers.

[Station X]

Here's something else to try. Check that R306, R307 and R308 measure correctly. Then switch S2 through its various positions to see if the HT voltages change. Leave the switch set such that R306 is in circuit. This will remove HT from much of the set.

If you now disconnect R302 only the output valves will have HT. Check their anode, screen and grid voltages. It's possible that C117, C118, C217 and C218 (THOSE capcitors) could be faulty putting a positive voltage on the grids of the output valves and making them draw excess anode current. This won't show up until you restore HT to the rest of the amp though.

[qualityten]

Who's a lucky boy then! Following Station X's hunch that the last immediately leaky caps were C117, 118 and 217 and 218, and having replaced one pair, I went back to my local Maplins for the next best size .068uF.

I put them in and powered up. First with the lamp. All seemed well. Then without the lamp. The speakers were absolutely silent. What? Then I noticed the speaker on/off switch in front--this is a full-featured receiver after all. With the switch on, and full power, the speakers showed signs of life. Switch 2 is faulty, but when it was coaxed into position, I got some SW and FM sound. Plugged in a CD player into Aux and hey presto: full, clear sound! What a pleasure to revive a beautiful radio amp from the early 60s.

There is lots of tidying up to do. The other ELNA caps are assuredly also leaky to some degree and I have replacements on order and the selector switch will have to be replaced or repaired. It's missing the small ball bearings that make for a clear switch from one position to the next.

I would not have got to this point without the help of the knowledgeable and considerate members of this forum. It was fantastic to receive a copy of the circuit diagram. Thank you. My particular thanks to moderators Station X, Michael Morris and Kat Manton for giving me so much help.

[Station X]

Well done. We like happy endings here!

Personally I would have replaced C117, C118, C217 and C218 with yellow polypropylene types of the correct value, but it is Sunday afternoon with only Maplin open.

I have a selection of ball bearings if you know what diameter you need. Look for cracking round the holes the balls sit in though as this can make the balls slip out. Give the switch contacts a clean with Servisol too.

Don't be surprised if more faults come to light as you soak test the amp.

[Michael Maurice]

C117, C118, C217 & C218 are often referred to as that cap their position in the circuit makes it absolutely crucial that they are not electrically leaky and in elderly equipment should be replaced on sight.

The reason for this is that they are there to block DC from the preceding stage whilst allowing the AC to pass.

A valve must operate with its grid negative to the cathode, if the capacitor is leaky, it will allow the DC to flow, the grid's voltage will rise resulting in a huge increase in anode current. This apart from destroying the output valve itself can burn out the output and mains transfomers.

I'm really glad you've got it working and that we've been able to help you.

Michael

[qualityten]

Originally Posted by Station X

Personally I would have replaced C117, C118, C217 and C218 with yellow polypropylene types of the correct value, but it is Sunday afternoon with only Maplin open.

Maplin doesn't appear to have any yellow polyprop caps of the right voltage and value in their whole catalogue. I did buy two 0.1uF ones to put in series if need be, but they cost about £2.50 each, and I need to replace 14 0.05uF ones. The WIMA ones I've used are polypropylene, not polyester, but do you have reservations about their suitability? Now that I know I have a working amp, I might do a proper job later.

Originally Posted by Station X

I have a selection of ball bearings if you know what diameter you need. Look for cracking round the holes the balls sit in though as this can make the balls slip out. Give the switch contacts a clean with Servisol too.

Thanks for the offer. I'll investigate the diameter. I did give all the switches a clean a week ago, but the balance one certainly needs more cleaning.

Originally Posted by Station X

Don't be surprised if more faults come to light as you soak test the amp.

I'm sure you are right. What is the correct replacement cap for C307, 0.02uF across the transformer from 110V to 0? What was there was a 0.1uF 1000V. But should this be a Class Y2 or Class X2?

Thanks again.
David

[Michael Maurice]

RS or Farnell certainly stocks the Polypropylene capacitors as does Cricklewood Electronics.

[Station X]

Another supplier is Just Radios in Canada. I've never bought from them, but they're highly recommended by many forum members.

C307 should be an X2 type, but in the UK it wil be across 230V, not 110V. At least I think that's the case. You need to check, as the diagram is for a 110V version of the amp.

[qualityten]

Originally Posted by Station X

C307 should be an X2 type, but in the UK it wil be across 230V, not 110V.

Thanks for clarifying this. I have found Cricklewood to be very helpful. Maplin's only advantage is that they are local and you can go back when you remember something else!

[argibbo]

So the transformer is OK now then? If so do you want me to look out some parts, like knobs & switches, assuming I still have them. Haven't looked at posts this weekend. Been fighting a Denon DN961 cd player all weekend.

Andy

[qualityten]

Thanks Andy. I'll send you a PM.