Leaking power supply channel

[pattox]

I've just completed a stereo power amplier -an Avondale NCC200. Both channels are fine in terms of basic bias current and dc offset measures. It also sounds very impressive with no obvious faults when used with some very old - 30 years perhaps- KEF canatata speakers. The perhaps the speaker crossovers and or drive units may have deteriorated but they sound ok.

The amp has a simple unregulated power (centre tapped transformer (about +/-30 volts) with with 3 4,700uf caps on each side. Also on each side is a 3.3k "bleed resistor" and an LED to visually indicate the voltage levels. With no inputs or speakers connected, turning the power on and then, a little later, turning it off, shows both power channels discharging evenly and quite slowly (using the led brightness as an indicator).

However once the speakers are connected and the power turned off the -ive channel discharges very quickly and then when nearly flat, generated a pop on the speaker. It doesn't matter whether both, or either speaker is connected, its always the -ve channel that discharges quite quickly. The inputs to the amp have no effect on this process, whether connected.

Perhaps somebody with more knowledge about this than I can explain what is going on, and tell me if it is a problem worth worrying about.

[peranders]

Under the discharge process the amp consumes little power but still with offset. The speaker will help discharging and the pop is a typical design flaw. I'll guess the designer didn't had this in focus when he made the design. Some amps have special circuits which really fast mutes the amp.

... so nothing to worry about.

[pattox]

Thanks, but its not the "pop" at the end that worries me.

What I don't understand is why it is only the -ive power supply rail (not the +ive) that discharges relatively quickly whenever either or both speakers are connected and the mains power is switched off. When only one speaker is connected-and it doesn't matter which one-it is the negative (about -42v to ground) rail that discharges relatively quickly.

[pattox]

I now understand what is going on. Apparently what I am experiencing is known as a "power down" -ive dc offset which is a common feature of many amplifier designs.

In this case-with speakers connected-until about 4 seconds after"power off" this offset is negligible but then rises abruptly to around 4.5volts. It then decays slowly over a few minutes to near zero. This dc offset represents roughly 2 watts into 8ohm speakers which I have been told is not a problem-unless, in the case of an 3 way active system this current is going through the mid or treble units alone.

The answer is a "power down" speaker muting or dc detection circuit.