A40 rail voltage on output

[Eric]

Hi Everyone,

A search on this topic revealed over 1300 unrelated posts, so I'm hoping someone can tell me what is going on here...

Last night I was fiddling around with slightly increasing the bias on my a40 amp when I noticed something strage. When the amp was plugged in and turned on with nothing connected to the inputs (they were not shorted either) I got full rail voltage on the output of one channel, but not the other.

The output returned to its normal 30mV DC offset level once the input was shorted, but when the input was open, there was 32v on the output- but only for one channel.

Any ideas as to what causes this? Then channels are identical and when connected to a my CD player, both play music just fine and don't exhibit any other strange effects...

A description of my amp is here

Thanks!
Eric

[Nelson Pass]

You need a resistor to ground on your input. If you already put
one there as seen in the schematic , you'll find it bad or not really
connected.

[Eric]

Thanks, Nelson! I'll give it a look when I get home. Its entirely likely that I knocked it loose while I was adjusting the bias...

Eric

[Eric]

OK, I've done some more poking around and still haven't yet solved this problem...

Nelson, as you suggested, I checked the resistor from input to ground. I assume that you mean R22 (10 ohms) on the schematic that connects the input ground to the chassis. My design does not have this resistor on either channel as my chassis is made of wood.

The problem seems to be related to bias also. When there is no input connected to the bad channel, there is full rail voltage on the output and the output transistors do not heat up at all. The output transistors only heat up when an input is connected, or when the input is shorted. When this happens, the output drops down to around 40mV or so and dV across R16 through R19 (emitter resistors) measures approximately 0.57V (just about normal).

I replaced the feedback and bias circuit (Q6, D3, R11, R12, and R13) but this made no difference. The output transistors themselves seem to be fine (not blown or shorted) as measuring the resistance from base to collector, collector to emitter, etc reveals them to be nearly idential to the output transistors on the good channel.

Does anyone have any other suggestions?

I guess the next step will be a fishing expedition: comparing each component across the good and bad channels, looking for an oddball resistance measurement or something...

Eric

[jam]

Eric,

Look at R2 and R3.

Regards,
Jam

[Eric]

Nelson and Jam,

Many thanks for your help! Jam, thanks for beating me on head with what Nelson suggested! I went back and re-checked R1, R2, R3, and Q11 (which, I suspect, is exactly what Nelson meant - I was just too slow to figure it out...).

As it turns out, there was a bad solder joint somewhere in the mix. I am not sure exactly where it was because I just re-melted each solder joint looking for a cold solder. The bad channels is now behaving better and the DC offset is also much reduced- from 60mV down to 2mV. Both channels now get toasty warm with or without an input connected!

Thank you for your help and have a Happy Thanksgiving!
Eric