re: "If you really run into a brick wall I still wouldn't scrap the amp, I'd either get deep into diagnostics trying to figure out what breaks up the oscillations " Of course, NO ONE is going to treat this as a subject of study for forward and reverse gain, under non-powered and powered (but, be careful) conditions. Recall back in school during the study of two-terminal active devices, and the criteria for stability? Reverse gain (S12) must be at least some 10 dB less THAN the forward (S21) gain, but MORE 'reverse' gain means higher stability especially under CHANGING load conditions. Today's two-port VNAs can make these measurements. AND, on top of that, for more than just the narrow ham bands, a complete 'sweep' or characterization, if this was to be done properly (which is why we see so many 'hack' cures on the web) from 2 MHz up past 50 MHz (SINCE we don't know at what frequency the 'oscillation' takes place). Speaking of which - that would be HELPFUL for analysis and diagnosis - WHAT is the frequency of oscillation anyway? Some investigatory work with a GDO can also reveal where *strange* resonances are occurring, LIKE in the PA Plate choke maybe?