DC output from amp with no input

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I woke up one morning last week to the smell of burning voice coils in the woofers of my left channel speaker. The amplifier was on (I typically leave it on), but there was no program input into the amp. In fact, the system was quiet.

I immediately turned off the amp, but the woofers (2) in the left speaker were toast. I deduced that it was a DC signal applied to the speaker because both woofers were stuck in the same position, at maximum deflection. Also, the midrange and tweeter are just fine. I am assuming a voltage potential somehow developed on the bases of the output transistors (I believe they are bipolar).

The amp is an Adcom GFA-585. Has anyone seen this happen before? Does anyone know what the cause is?
 
AX tech editor
Joined 2002
Paid Member
Hi,

There may be many reasons why the output suddenly (or slowly, we don't know that) goes to one supply line. It can vary from a shorted output device, via a shorted cap, an out-of-spec small-signal transistor or anything. What I want to say is that you really would need a schematic and at least a multimeter to chase this up. Undoubtedly there will be advise from people to change all electrolytics or to change all transistors. Although it is possible that it restores the amp, it may also lead to exactly the same result a week from now. There are no easy answers I'm afraid. Can you read a schematic?

Jan Didden
 
Unfortunately almost any component failre that results in an open or short tends to dive many/most test points to the rail voltage. You have one big advantage over someone who encounters this when building an amp -- you know all component were correct and funtioning once. Rsistors almost never change value and are hard to damage- thus narrowing possabilities.

You have a second advantage - the other channels is fine. This gives you a ready reference for correct values. I suggest to start with the unit unplugges and check the resistance across the pins on the transistors on the good channel. Then on bad if they are not approximately the same, you have found a problem. This can be ambigous, however, because something else may be in parrallel to what you are measuring which mean the problem could be either. Also rember not to stop when you find the first bad component - you can't always (usually?) tell the victim from the perptrator!
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.