Hey guys,
So I've been the repair technician at a store for 3 months now and have managed to get to a stage where I can fix maybe 70% for the things that come through but when it comes to debugging Power amps that are either triggering circuit protection or have some distortion the only method I know is brute force testing every component and crossing my fingers. I'm also the only technician at this shop and have no prior practical experience/people I know to ask for advise.
An example of this is the denon PMA 350 I've been working on all morning. All inputs go through a selector switch, then an op amp, then some small transistors followed by the output transistors and relays. There's was distortion on the left side that made a 1k sine look like a square wave. I traced the sine wave through and out of the op amp into the transistor amplifier stage. On the emitter of the first transistor in this second stage The sin wave becomes a square wave. I tried swapping some transistors from the working side with the faulty side but the distortion is still there, making me think there might be some negative feedback in that section, meaning that all transistors in that area will show the distorted signal (I'm guessing?). Now the schematic online seams to have different board markings and a different components in general so I'm stuck there.
To my questions;
1) does anyone have a technique for testing transistors in circuit?
2) if distortion is generated somewhere in a section with negative feedback, how do you trace the signal to a faulty component?
3) if no schematic is available is the a more economic route to help make sense of the circuit than reverse engineering the whole thing?
Thanks for any and all help!
So I've been the repair technician at a store for 3 months now and have managed to get to a stage where I can fix maybe 70% for the things that come through but when it comes to debugging Power amps that are either triggering circuit protection or have some distortion the only method I know is brute force testing every component and crossing my fingers. I'm also the only technician at this shop and have no prior practical experience/people I know to ask for advise.
An example of this is the denon PMA 350 I've been working on all morning. All inputs go through a selector switch, then an op amp, then some small transistors followed by the output transistors and relays. There's was distortion on the left side that made a 1k sine look like a square wave. I traced the sine wave through and out of the op amp into the transistor amplifier stage. On the emitter of the first transistor in this second stage The sin wave becomes a square wave. I tried swapping some transistors from the working side with the faulty side but the distortion is still there, making me think there might be some negative feedback in that section, meaning that all transistors in that area will show the distorted signal (I'm guessing?). Now the schematic online seams to have different board markings and a different components in general so I'm stuck there.
To my questions;
1) does anyone have a technique for testing transistors in circuit?
2) if distortion is generated somewhere in a section with negative feedback, how do you trace the signal to a faulty component?
3) if no schematic is available is the a more economic route to help make sense of the circuit than reverse engineering the whole thing?
Thanks for any and all help!
A square wave at a low level is a limitation due to low power supply or cutoff.
High level limitation - NFB break, signal overshoot or generation.
There may be digital square wave circuits.
Using a multimeter, you can check diodes and transistors (turn off) or measure voltages at test points.
High level limitation - NFB break, signal overshoot or generation.
There may be digital square wave circuits.
Using a multimeter, you can check diodes and transistors (turn off) or measure voltages at test points.
Last edited:
Use a low level 1kHz triangle waveform for signal tracing and general troubleshooting.
It's much easier to see and evaluate clipping and distortion. Within a nfb circuit block,
first test the semiconductors in circuit, and compare the good channel with the bad.
Suspect mainly semis and electrolytics if the solder joints look good, though the emitter
resistors can be damaged by output transistor failure.
It's much easier to see and evaluate clipping and distortion. Within a nfb circuit block,
first test the semiconductors in circuit, and compare the good channel with the bad.
Suspect mainly semis and electrolytics if the solder joints look good, though the emitter
resistors can be damaged by output transistor failure.
Probably the feedback is no longer feeding back... I'd check all the passives in the feedback network, and that its only grounded where it should be.
Its always worth checking all the power rails have the right voltage in that channel.
Thanks for all the suggestions guys! While pocking around with the scope the working side started distorting as well and I was worried I might have shorted something but coming back to your tips, i thought I'd have a look for cold joints. I found a couple hairline cracks but while reflowing them, they completely pulled away from the pins and after this the good side was back to normal.
I've compared all passives on one side with the other and done a multimeter beep test on all transistors and diodes but haven't found anythign there yet
I've also just had a look at the power rails and low and behold, on the faulty side the mid point bias inbetween the supply caps instantly starts climbing so I'll have about +8, -84V where the working side is +-46V.
still scratching my head on this one but hopefully this a good clue?
Thanks again guys!
I've compared all passives on one side with the other and done a multimeter beep test on all transistors and diodes but haven't found anythign there yet
I've also just had a look at the power rails and low and behold, on the faulty side the mid point bias inbetween the supply caps instantly starts climbing so I'll have about +8, -84V where the working side is +-46V.
still scratching my head on this one but hopefully this a good clue?
Thanks again guys!