snubbers
Not really for the oscillations. The transformer secondaries have an associated inductance that forms a nice LC circuit with the winding capacitance. Abrupt changes in current, as you experience when the bridge diodes shut off, causes this circuit to ring and radiate. Adding the snubber circuit will suppress this. I can't guarantee that this will get rid of your problem. It is one source. Knock them down one at a time.
Not really for the oscillations. The transformer secondaries have an associated inductance that forms a nice LC circuit with the winding capacitance. Abrupt changes in current, as you experience when the bridge diodes shut off, causes this circuit to ring and radiate. Adding the snubber circuit will suppress this. I can't guarantee that this will get rid of your problem. It is one source. Knock them down one at a time.
power supply redesign
I learned some things from the supply used in the WHAMMY headphone amp. Wayne used a C-R-C-R-C filter network that does a pretty good job of reducing the ripple. Each C is 3300uF and the series R between caps is 5.1 ohms. Nothing really magic about the values. That amp idles at about 60 mA per channel in the output stage so it draws roughly 125mA through 10 ohms for a loss of around 1.5 volts. Consider it.
I learned some things from the supply used in the WHAMMY headphone amp. Wayne used a C-R-C-R-C filter network that does a pretty good job of reducing the ripple. Each C is 3300uF and the series R between caps is 5.1 ohms. Nothing really magic about the values. That amp idles at about 60 mA per channel in the output stage so it draws roughly 125mA through 10 ohms for a loss of around 1.5 volts. Consider it.
resistances
Well, I would look at the expected quiescent current draw. Decide what a reasonable voltage drop across the resistor might be, say 2 volts. Calculate the necessary power rating and double that number. If the resistors are to be mounted on a PC board just make sure they are away from the board surface a bit because they may get hot.
Well, I would look at the expected quiescent current draw. Decide what a reasonable voltage drop across the resistor might be, say 2 volts. Calculate the necessary power rating and double that number. If the resistors are to be mounted on a PC board just make sure they are away from the board surface a bit because they may get hot.
Just a suggestion... have you eliminated possible simple causes already e.g. tried different signal sources, internal and external cables etc? I guess if the problem is the same on both channels it suggests a design problem but I’ve wasted hours in the past swapping over components only to track the problem down to a dodgy cable or dry joint. Could just be that I’m particularly bad at soldering and making up interconnects though.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.