orderly debugging amp that loses power

[Luthierwnc]

Hi All,

I pulled an amp out of the closet I haven't played in about five years. It is a Second Generation D-clone with a few bells and whistles. I made it when those were all the rage here twelve-fourteen years ago. It always sounded really good but I have no excuse for owning a 50-watt head.

At any rate, I plugged it in and let it warm for 5 minutes before switching the standby. Then I played at a moderate volume for about five minutes. Everything sounded fine until the amp lost almost all its power, quite quickly after making a scratching noise. I had to leave and figured I'd mess with it later. Today I did the exact same thing and got the same result. It plays great until almost all the volume dies.

FWIW, the amp is like most D-clones with 434VDC on the plates. TAD 6L6GCs are biased at 33-35 ma. Nothing smokes. The power tubes don't seem too hot.

I'll haul it downstairs for a closer look but thought I'd ask if anyone knows the first place to look for an amp that loses power when it gets warm. It did it twice and sounded fine until it did so it isn't something that died once and stayed dead.

Thanks for looking and cheers, sh

[Colossal]

Hello Skip,

Two things that come to mind to check for are a cold joint on your heater filament string and then a power supply cap(s) going bad (leaking, high ESR, etc).

[R.G.]

Not enough info to make a substantial guess. I'd go along with caps in the power supply being a problem, or some loss of signal or bias, too, and a cold solder joint will do that fine.

The only thing that may be a pointer to the cause is the time delay. Works OK for about five minutes, then almost mutes. Any substantial delay before something happens is usually thermal in nature. Something heats up, then lets go. Power supply caps can do this, decoupling caps can do this, solder joints can do this, bias supply caps or diodes can do this.

The comment ALMOST complete loss of power and non-melted power tubes almost rule out loss of bias. You'd lose all the sound probably. Screen supply cap would lower the power tube gain a lot; could also be the choke heating up and going high resistance. If it has a B+ fuse, that could get thermally intermittent. Normalled effects loop jacks could be corroded or dirty; master volume pot(s) could get funny. Opening ONE power tube might leave the other putting a little audio through.

Best is to haul it to the work bench and start monitoring things. Make it "fail", then measure B+, screen supply, plate voltages on preamp tubes, etc. All else fails, get out the oscilloscope. The 'scope will show you if you have a REAL dark horse - ultrasonic oscillation making it scream inaudibly, modulated by the audio signal, which the speakers will happily lowpass down to very little audio.

[Stevem]

You don't need a scope to tell if the amp is oscillating when it acts up, just hook a voltmeter set for ac volts across the speaker .

If all is well all you will be reading is far well less than .150 volts ( 150 millvolts ) which will be normal power supply noise .

[Luthierwnc]

Thanks gents. I'll haul it downstairs when I get a little time and do a visual for loose parts first. I'll let you know when I catch the culprit. Cheers, sh

[Luthierwnc]

Doesn't it always work this way?

I put the chassis in the cradle and chop-sticked every joint. Checked the grounds. Then I plugged it in and checked voltages. They were all fine so I plugged in and played for half an hour. Sounded great. Left it on for two hours and played again. Sounded great. I can't put it back in the cab knowing something is lurking so I'll re-flow all of the power solder connections and see if I feel better about things. More to come if there is any, thanks again, Skip

PS I've had cars that made noise until I took them to the shop too. sh