Crate GX-1200 - Problems persist after new parts

[JP-Stereo]

One thing to keep in mind is that "sometimes/often" components that test fine on a meter are effectively defective in circuit while working...
An out of circuit test is only detecting obvious defects...

But probing a circuit directly on the pads is risky, i will not try that in a switcher.
When there are not too much noise around and the capacitance is not that critical, one is better to solder a wire to a leg (DUT Off) and "remotely" take readings.

[skidmark]

One thing to keep in mind is that "sometimes/often" components that test fine on a meter are effectively defective in circuit while working...
An out of circuit test is only detecting obvious defects...

But probing a circuit directly on the pads is risky, i will not try that in a switcher.
When there are not too much noise around and the capacitance is not that critical, one is better to solder a wire to a leg (DUT Off) and "remotely" take readings.

I assume your referring to switchers and amps with balanced output, floating grounds, etc, ???
That's what differential probes are for.

[skidmark]

It looks like the amp just may be fixed! Actually the amp may have been fixed for some time and I didn't realize it. I know, sounds silly so here's the explanation.
The amp was probably fixed after replacing C25 and IC5. The reason I didn't realize it was fixed was because I was using either the variac or the limiter to prevent
blowing more finals or whatever.. Well apparently this amp's rail settling time is such that slowly raising the variac voltage trips up the NFB loop to where it was getting
into a self defeating mode. That's the best way I can describe it. As for the limiter rational, I think the 40 and even 60 watt bulbs filament warm up time was shorter than
the amps rails settling time.
Here's what gave it away. While giving the Crate work a sabbatical, and turning my attention to other amps, I noticed that on 2 other solid state amps there was a faint
thump in the speaker when I powered them on. Then my light bulb came on and I realize that for years I've noticed this on most all solid state amps. As soon as I got my
bench clear off, I re-engaged with the Crate, replaced the 3 amp fuse with a 1 amp fuse, ran the variac up to full voltage then flipped the switch. I heard the familiar
faint thump in the speaker and everything went quiet. I touched the center lead of the input jack and hear the ole familiar AC hum. Next I hooked up my Wavetek
waveform generator, set it on 1kc. Amp sounded good. Cranked the volume up a good bit then disconnected speaker and let my 4 ohm dummy load catch the brunt
of what the amp has to offer. Got the heat sink up to second degree burns for few minutes, then throttled it back a bit so as to not push my luck. It's baking in
right now.

Part count:
2 Zeners
1 cap = C25
4 finals
2 TL72

Al

[g1]

I guess you were trying to start it up on the limiter bulb with a load connected. Many amps will not start properly like that, which is why I had stressed not having a load connected earlier.
The 'soft' rails due to the bulb will cause the output to stick to one rail or the other on startup with the bulb. If there is no load, it will not happen. Then while still powered up and the bulb is dim (and no DC on output), you can connect the load.
Not every amp will do this, but quite a few. As a general rule for solid-state amps, don't connect the load until everything checks ok. And don't try to soft start on bulb while load connected.

[skidmark]

Nope, no load connected at all. I haven't had a load on it for weeks.
Al

[g1]

By load I meant speaker or any other kind of load. In post #65 you said you heard the thump of the speaker. I assumed that was not the first time you started it up with the speaker, but maybe I misunderstood.

[skidmark]

Yep. That was when I made my calculates guess to go for it! After staring at the storage scope timing and waveforms of all 4 rails on power up, I decided it was time for a speaker.
Powered it up, got the subtle thump on the speaker and hooked up the function generator and CRANKED IT UP.

[g1]

Very strange. Can only guess the outputs were turning on too hard during the supply ramp up. With no load, there would be nowhere for current to go even if the output went DC during power up.
Whatever it was, glad to hear you got it sorted out.