Home built amp shutting down with loud hum after 10 minutes

[Clinkous]

Hopefully this is a better pic

[Clinkous]

Ok,
I swapped the first and third filter cap and the same thing happened. There was 2.6V AC ripple on the first filter cap before the noise and 3.9 after it started. The AC voltage taken where the two diodes come together was 267 before and dropped to 207 after. I'm going to get two diodes and try that.

[Phil_S]

Please post a photo of the whole inside of the chassis, not just a section of it. It helps a great deal to have what I call an orientation photo. You are getting some good advice here, but I am thinking this is something simple and it might help to see the whole build.

[Clinkous]

The new diodes didn't help. Maybe I just don't know how to solder! The cathode voltage stayed at 6.8V with or without the hum. Going to reflow connections again.

[Clinkous]

Here is a picture of the entire amp. The bright cap on the voltage divider is intentionally disconnected on one end.

[dorrisant]

The upper lead on C13 looks like it is floating... Maybe you disconnected it to eliminate a possibility.

Is the resistor across the bias points big enough to carry the cathode current?

Looks like the cathode pin (3) of the EL84 socket might be touching the filament pin (4), or it may be too close... Can you confirm or post a better shot of that socket?

[Clinkous]

Yes, c13 is intentionally disconnected, was taking out some highs. Pins 3 and 4 are per the schematic to let the filament voltage ride on top of a dc voltage (don't quite understand that but it's in the schematic).

The cathode resistor is a 5W. The amp did this before I put the variable bias and octal socket in. Thought I had it licked then because it didn't do it for a while.

I tried swapping the uf4007 diodes for 1n4005's and the hum was exactly the same except immediate. I swapped the old diodes back in and I got about 5 minutes of perfect amp behavior and then the hum. I'm going to order two more uf4007's and put them in. Maybe I overheated one of them when I built the amp and it's breaking down when warmed up.

That's where I'm at. Thanks for looking. I made sure no stray leads or close tube socket tabs could touch even when wiggling.

I appreciate everyone's help and input on this!

[martin manning]

1N4005 does not have a high enough inverse voltage rating. Replacing the UF4007 is the right thing to do. You could find another 1000V rectifier like iN4007, or if you have more 1N4005's, you could try using two in series in place of each UF4007.

[labb]

Pull all the tubes, turn the amp on and check for voltage drop across R2, R3 and R4. If you do have a drop then you have a short to ground that is loading up the PT.

[dorrisant]

1N4005 does not have a high enough inverse voltage rating. Replacing the UF4007 is the right thing to do. You could find another 1000V rectifier like iN4007, or if you have more 1N4005's, you could try using two in series in place of each UF4007.

Good call, Martin.

Clinkous - Order more than just enough for replacements... You can always use them somewhere and you wouldn't have to order more now if you had ordered extra before. They are cheap... just sayin'.

[romberg]

Is the resistor across the bias points big enough to carry the cathode current?

I think this may be one of the problems

I am using a variable external bias pot and...

It looks to me that you still have that thing cathode biased with the pot stuck in the bottom of the chassis in series with the bias resistor on the test points. I think 24mm Alpha pots most commonly used are 1/2W (linear taper) and 1/4W (audio).

I'd try putting the 130R 1 watt resistor from the schematic back in place of this arrangement and see if that makes any difference.

Mike

[Structo]

Can you point out your grounds?

[Clinkous]

My grounds are all along the top (pot side) of the board ending at the input jack ground lug. This also happened with the 130R standard setup before modification. Thought I had it fixed with a solder touch up so I went ahead with the mods. It came back though exactly the same.

I haven't deviated from the ax84 HO schematic with the grounds and they all seem fine with the ohmmeter.

I have ordered some correct diodes so we'll see if that was it.

The bias voltage stays the same when the hum starts. I'm getting 6.9-7.1V with no shift when the hum starts. Neither tube has red plated or anything.

[Phil_S]

Maybe it's just the way I'm looking at it...is the board adequately raised off the chassis? I'm thinking one of those eyelets is making intermittent contact.

Also, the high potential grounds are on the same buss as everything else. I think that CT ground needs to go directly to the chassis along with some other stuff. Somehow, is it injecting noise when the amp warms up?

[fishy]

The HO is a cool little amp. It was my first build. I did run separate grounds for preamp side and power. I also used the elevated heaters.

I have had an amp do something like this and it turned out to be filament wiring that had a poor solder joint and it would come in and out once it had warmed up. Since the I have always wired filaments from the first preamp tube back and checking as I go.