Fireworks in my valves!

[Dave the bass]

Originally Posted by andrew ivimey

.

If INfBaff can tell us more, someone will have and answer and I'll stop being silly - promise.

Owwwhhhhh, don't...please. I've just spent 5 mins reading your replies in funny voices and giggling

DTB

[Richard]

Indeed brother Andrew, your sermons against the demon Chinese amps and Russian rects would be sorely missed pray continue lest we go astray Strange though that those winged Svet Russkies are my favourite 5u4 (or should it be 5u3?) sweeter than the GZ34 though not as good as the 37. Might depend on the amp. The stubby US types are flat and boring Never had a problem with any of them though. Mind you I think the data sheets get overlooked these days, often the first cap is too big or the mains supply is too low impedance. Ali, 47R or 100Rs on the plates may help. To Infinitely Baffled I would say find the first cap and change it (assuming it's not a dummy of course ).

Rich

[Ali Tait]

It's a crappy Chinese jobbie thats the culprit.I'll get to try the Svets tomorrow night,so we'll see.I like the look of them though,construction seems excellent.Should cure the problem I hope.

[andrew ivimey]

Yeeerrrrrrrrrrrrrsssssssssssssssssssss oh brother, 5U3 indeed. I like the look of them; all crinkly and delicate; very thin glass. I have a quad from Sverdlovsk or somesuch cold and dreary place.

No, the wide open plains of the USA is the place to get a decent 5U4, (chaqu'un a son gout!) now that the holes in Blackburn Lancashire have all joined together and .
.
.
.
.

The first cap is the most important; best not to use one at all!

We need to know what is going on with this InfBaffamp.

[Infinitely Baffled]

Wow! I'm astonished, amazed and gratified in equal measure, guys. I had no idea this would provoke even a single reply. Small problem is ... you don.t realize what an imbecile you are talking to. I built the amp myself from a kit about 5 years ago with no understanding whatever what I was doing (really!), just following the excellent instructions carefully and being as neat as I could. It is an Audionote Kit One set, bought back in the day when Quortrup's outfit was still selling them in Hove. It has given flawless service for 5 years until a couple of months back when I lost a channel. I faffed around with it and took measurements to compare with the spec sheet - everything OK. Never did find anything wrong with it so I soldered it all back together and, hey presto, both channels back. It was shortly after this that I first noticed some arcing in the rectifier - the original Svetlana Winged C that came with the set and had worked perfectly for 5 years. I swapped it for an aged RCA 5U4. That one was OK at first (but did have a little flicker on one occasion a week or two later). I then acquired a couple more nos Svetlanas (I'm such a cheapskate, I know) for a WAD6550 kit I am about to start building, so I tried both of those - firework night! At the moment, thefefore, I am using the RCA, which - touch wood - is doing just fine, though I would say it has a slightly denser, creamier sound that the Svets, which just sparkle (in more ways than one at the moment!).

So, Chums, there is some history. Thanks for all your suggestions so far. One plea, though - if any of this prompts any further thoughts, can you bear in mind you are talking to an idiot, and keep it simple? Ta! Infinitely Baffled.

[Richard]

Just the thought that it might be Chinese was enough to get us going laughed my socks off when it wasn't .

Given that you've had it a while and it's just happening at switch-on my bet's still with the psu caps.

Do a full voltage check first to see if anything shows amiss.

Electrolytics are the most likely component to fail and at 5 years is quite possible. See if you recognise this which may be quite similar to yours. Compare the caps to those on your circuit diagram. As they are numbered on this one, you'd be looking to change C1 and 2, C8, then C11, 12, 13, 14 in order of best guess.

If the pcb has to come out you might want to change the lot at the same time or if you can easily get at C1,2 then give them a go first as they get the most work (they are the first smoothing cap and suffer a lot of ripple). Do use suitable replacements designed for high ripple such as Rifa or come back to us with the values needed and we can have a look.

[andrew ivimey]

Yes, I'm told that Audionote is quite good. - eats humble pie -.

So a few measurements then. And what about those rectifiers, then eh!

[John T]

Originally Posted by Infinitely Baffled

can you bear in mind you are talking to an idiot, and keep it simple? Ta! Infinitely Baffled.

Congratulations on your honesty!

Quite a few of us who built the WAD6550 amps experienced similar "flashover" problems at switch-on - particularly with Svet 5U4's. The problem was easily resolved by using 5U4GB's. These can be visually identified by their straight sided envelopes, in contrast to the "coke bottle" shape of the standard type.

These have a "ribbon' type heater and appear to be more robust that the thinner type used in the standard valve. As a bonus, most of us preferred the sound from the GB variant.

Richard, no doubt has made a good point regarding the capacitors but personally I'd first try the 5U4GB - it could save you a bit of soldering!

[Infinitely Baffled]

That's fantastic. Thanks for all the help, everybody. The suggestion to use 5U4GBs is very tempting, since I get a queasy feeling about rooting around inside the amp when it is switched on (I know, you might reasonably ask what am I doing this as a hobby for, then? Bit of a morbid fascination, I suppose). But if the capacitors in the power supply circuit are on their way out, shouldn't I really be trying to find out? I have the impression from your post, John, that using 5U4GBs avoids the problem simply by substituting a more robust component. Or are you saying that there maybe isn't a problem anyway with the PSU capacitors?

Richard's circuit diagram (for which, many thanks!) is pretty much a 100% match with Audionote's so would enable me to work my way through each component to find the culprit. At least it would if I knew how to test them (yes, it's that bad). The testing that is prescribed as part of the build instructions is carried out mainly without valves in, and consists largely of taking readings between various points and chassis to determine if the values match the spec. I assume that I test the integrity of a capacitor by taking a reading across its "legs" whilst the amp is switched on - yes? Also, it appears to me that I have got both AC and DC operating at different points inside this amp. In which camp do the PSU capacitors live? (I don't know why I am bothering about testing them - it's bound to turn out to be the big b****r that cost me about 80 quid!).

Worried, with Multimeter.

[systemx]

I am full of admiration for someone who would build an amp by just following the instructions, obviously you take great care with what you are doing and must have read the instructions very carefully. Very well done.
I would go for the valve change first. the capacitors live in the DC camp but it is not reaily possible to test them as such, you would replace them.