Soak testing.

[Diabolical Artificer]

Not sure if this is in the right section.... Looked this one up, but couldn't find any specific practical info. When testing audio amplifiers specifically and other circuits/builds what is normal procedure? I thought to test with a sinewave IP at ,one quarter, then half, then three quarter power each for 24 hours whilst monitoring HT, transformer temperature etc, this into a dummy load. Finally I thought about testing at full power with a music IP into a dummy load for 24 hours.

Any thoughts welcome. Andy.

[MotorBikeLes]

Use a dummy resistive load, but don't go near full RMS output. Probably 50% max. In real life NOTHING ever gets continuous 100% audio output.
Les.

[Craig Sawyers]

The amplifier transistor power dissipation (assuming it is class B or AB) reaches its maximum at about 70-75% of rated output, and then decreases again as the output increases towards maximum.

Depending on the precise design it can be quite stark. A Trace Eliot bass guitar amp I fixed for my brother in law, and I figured out the power behavior dissipated almost the same amount of power in the transistors at 10W (40W transistor dissipation) as it did at 160W (46W).

Reason is that at surprisingly low power you end up with a high voltage across the transistors, and lowish current. And near maximum power you have a low voltage across the transistors and a high current. All time varying of course, so you have to do the sums to show what is happening precisely.

Douglas Self goes into that in his book on power amplifiers. Then he looks at what happens to transistor dissipation when the amp is fed into real loudspeaker loads rather than a dummy 8 ohm resistor. It is not pretty.

Craig

[joebog1]

I think Andy is referring to his monster valve beastie.

I test my valve stuff at 50% of power, measured at the load and using E^2/R with an RMS meter. 24 hours?? nope never done that. I test for about 4 hours. After that everything should be at real normal operating temps. This of course doesn't account for cathode emission, especially if you are flogging the output bottles. Modern "pretend" valves from USSR or Asia are nowhere near capable of "full power" as we read in original Mullard or Brimar catalogues.

Just my take Andy.

Joe.

[Diabolical Artificer]

I guess as we are trying mainly to test workmanship, connections etc. any fault should show up at 50%. As it's an over engineered homebrew design I thought it should handle 75% power OK. I've run it at full sinewave power for about 10 minutes and all was well, whichever I'll try a low power for 24 hours first.

Being as it's not practical to sit by the amp for 24 hours I'm going to have to trust to the protection in the amp and house wiring, best make sure the smoke alarms have new batteries though JIC.

This is a valve amp Craig and from memory HT drops around 10v at full power, current goes up from approx 35mA per OP valve ( 6 in total) to a max of 90mA with a total current draw of the HT of 670mA, so if I've understood your post correctly and the electronics a valve amp differs from a tranny amp, though I take your point about testing tranny amps.

Andy.