The situation I think we might have is that both TX_BPF_SEL and RX_BPF_SEL are HIGH during SSB transmit. This lets a little bit of the transmitted signal leak through the TR switch back into the amplifier input through the BPF selection stage, like this:

?

?

These two lines are controller by GPIO register bank A:

?

?

This register is set by the function call ; the code there looks correct and unchanged. Scanning the code, it all appears to be correct. Some judiciously placed statements like this will confirm:

Debug("Set LPF GPA state: "+String(LPF_GPA_state,DEC));

?

The fix that I implemented (major change to radio layout) to cure the short that caused the burning smell has created a display freeze issue, so I'm not able to test this myself right now.

?

?

?

?

?