Preamp Oscillation problems - Suggestions?

[hirscwi]

I've just completed a couple preamp boards by Greg Ball - called the SKA Pre. Each channel is on a separate board. One board works fine. The other seems to be oscillating at about 14MHz but only with the input shorted. In other words, with the volume potentiometer turned for minimum (i.e. input connected to ground) the oscillation is at its maximum. As the volume potentiometer is turned up, the oscillation disappears. Attached is the schematic. Any suggestions will be appreciated.

 

[rayma]

Try an input base resistor, like R18 on the other side. Are the two boards identical?

 

[hirscwi]

Yes, both boards are identical and I have double checked to see if I made any errors on the one that's oscillating. I can't find any errors but that doesn't mean they don't exist.

Thanks for the suggestion.

 

[hirscwi]

I assume the input base resistor would go on the schematic where i've highlighted in yellow. Same value as R18 OK (3k)? I'll have to find a clever way to insert with wrecking the board, because it's fairly tight.

 

[rayma]

Yes, if you stand that new resistor on end, then the end with the short lead should go
to the bases of the input transistors.

 

[hirscwi]

Forgot the highlighted schematic...

 

[hirscwi]

Yes, if you stand that new resistor on end, then the end with the short lead should go
to the bases of the input transistors.

I'll give it a try in a few minutes

 

[hirscwi]

That solved it! Thank you! If there's a simple explanation of why it worked, I'd appreciate it to add to my learning. But I understand it might be too complex to answer.

I'll implement on both boards to be safe and consistent.

 

[rayma]

Oscillation can be caused by feedback that is not carefully controlled. A lower frequency pole
(from the added resistor) reduced the net HF phase shift enough to restore stability.

An introduction: http://ocw.uc3m.es/tecnologia-elect...-1/Lesson2_Stability_in_Feedback_Circuits.pdf

 

[hirscwi]

Thank you very much! I'll read that with interest.

 

[Mark Tillotson]

We all presumed you meant megahertz rather than millihertz, but watch out for those
prefixes, they are case sensitive, just like any SI or engineering symbol.



1mHz = 0.001Hz
1MHz = 1000000Hz

 

[hirscwi]

You are correct. Just sloppy typing on my part.