Just a few months after starting work on this two-transistor device, I have finally finished it .It needs 'setting up' by adjusting a small, 25-turn resistive trimmer (RV1) to balance the two transistors. In the original design, the amp used a pair of 2N2222's, but in line with various recommendations (including the Wellgood site) I used a pair of 2N3866's. Of course, these transistors take appreciably more current and have been fitted with heat sinks. The amp is designed to operate at 12V.The recommended way of setting the bias is measure the voltage across RV1 (see attached) and adjust to minimise it. As a cautious way of approaching this, I started off with just 6 volts on the board and adjusting RV1 went well, and so too at 8 and then 10 volts, but at 12 volts, the minimum setting is not stable.The two transistors are drawing ~127mA (total) and running at 48 and 54'C. RV1 is also getting quite warm, running around 50'C. The voltage across RV1 climbs from ~0mV to around 10mV fairly quickly; I've not left it running very long, so I don't know how far it may go.I'm not quite sure what's causing this but I guess it could be transistors which were not too closely matched (I need to dig out my notes but I think the pair I used add gains of something like 180 and 195) or whether the 100R trimmer is an issue; it's rated at 0.5W, and it's sandwiched between the transistors and so it's being pushed a bit.