Need help: amp blowing fuses

[Saurav]

I have a DIY 2A3 SET that's worked fine for months (if not years, I don't remember the last time I made changes to it). This evening it started blowing fuses. I've gone through 3 3A slo-blo's so far trying to debug it. Here's what I know so far:

* With all the tubes pulled out, the fuse doesn't blow. The main output taps from the power transformer show ~650V, the 5V heater tap for the 5AR4 shows 6.something volts, the 6.3V heater tap for the driver tubes shows 7.something V

* If I plug in only the 5AR4, the fuse blows almost immediately. The DC output got up to about 6V before it died.

* With all the tubes plugged in (which was the first thing I tried, i.e. just replaced the fuse), it takes about 30 seconds - 1 minute for the fuse to blow. I watched the amp until I saw the filaments light up, then I turned on the source and started the music, and by then it was dead.

* I can't measure any obvious shorts from the DC side of the 5AR4 to ground.

Is this sounding like a bad 5AR4? Do those tubes die all of a sudden like that? The first time it died was also within a minute of powering it up, and I thought I saw a bright flash inside the stereo cabinet. Are there any resistance tests I can do on the 5AR4 pins to check if it's dead? I can't see anything obviously wrong with it, but there are 'fins' inside so it's not easy to see if anything's wrong. If the 5AR4 seems like a likely suspect, I'll go ahead and order a new one.

Thanks,
Saurav

[Enzo]

It is not often a shorted tube looks different from the outside.

You can check to se if the plates - pins 4 and 6 are shorted to the filament. But many times a short doesn't eveidence itself until voltage is present.

Certainly a shorted rectifier is the first thing to mind, if the thing holds a fuse tubeless, and blows with the rectifier only. Until you get the amp to hold a fuse with the rectifier, don;t even bother to put the rest of the tubes in.

Make sure your 5v winding on the power transformer has not shorted to ground.

Your main filter caps could be shorted and not show on your meter. Your ohm meter puts out a volt maybe for ohms testing. A 400v cap might be leaky at everything over 20 volts, and your meter wouldn;t know.

I would get a light bulb limiter hooked up in your mains lead. Do a search for that. Personally I prefer a variac and ammeter, but the bulb works.

You could get a diode - 1N4007 for example - and clip it from one of the rectifier plate pins on the rectifier socket, over to the first filter. That would take the place of the rectifier tube for tests. If your fuse still blows, you have a bad cap or something shorting further along. if it holds a fuse, then perhaps it was just a bad rectifier tube.


But you can also put a good rectifier tube in, and disconnect the wire from the filter caps from the socket. That leaves ONLY the high voltage winding and the 5V winding wired to the recto socket. Now fire it up. If that blows, then something fundamental is wrong right there. otherwise you should have unfiltered DC at pins 2 and 8.

If you get that far, you can connect the first filter to this, but disconnect everything further. No output transformer, no lower voltage B+ nodes. Does that filter hold or blow fuses? Keep adding back in the parts until you find what blows.

Cana tube burn out just like that? Sure. Can a light bulb burn out just like that? Everything works until it doesn;t.

[Saurav]

Thanks. Those are all good ideas.