This is an amplifier built on a standard G4DDK board.
The amplifier performed OK in the test system where a 3 dB pad ensures a reasonable input WSVR on all frequencies. The amplifier did however show signs of oscillation problems when connected to an antenna so it was tested at high WSVR by use of an un-terminated variable delay line (trombone.) Strong oscillations were then visible around 900 MHz.
Under normal operation the WSVR at 900 MHz is very high so those oscillations present a real problem. Dick, PA2DW modified the amplifier by adding source inductances. That cured the 900 MHz oscillations, but instead oscillations at about 3 GHz occured at high WSVR on the input. That is less harmful because the amplifier is not likely to see a very high WSVR at 3 GHz, but nevertheless it is a good idea to make an amplifier stable for high VSWR on the input on all frequencies. The 3 GHz oscillations were removed by RW3BP by placing two layers of absorbing material on the drain side.
After having oscillation problems fixed Dick tuned the amplifier for optimum NF on the NF measurement system. The tuning was made without the relay.