My preamp is pure class A, which means it is effectively constantly running at full power, even if the modulated signal is less. Yes, I know that is an over-simplification, but it serves to illustrated why, after running for a few hours, it gets extremely hot. Because Cyprus can also get extremely hot, last spring I put a row of five small PC 12 volt 0.39 amp fans in a row across the cooling grid at the back of the top. There are lots of holes in the bottom of the chassis to allow air to circulate. The fans are powered by a 12 to 24 volt DC adjustable power supply, giving 5 amps at 18 volts, the minimum voltage without all stalling. I wired them up in series so I could keep the fan speed slow and quiet, rather than howling like a 747 on go-around power. It worked satisfactorily with the exception of fan 4, numbered from positive to negative, which stalled on lowering the voltage earlier than the others, thus limiting my ability to go for max quietness, as when stalled it acted as an inlet vent for the other fans, but I lived with it for the summer. Today I thought I would have another look at it. I tried various series/parallel configurations, but to no avail, as either that fan howled or the other fans stalled, so I just decided to put it back as it was. I took fan 4, the problem child, which was on wander-leads and rewired it back into the all-series loom, but to save time I just added it to the end of the current series as fan 5, +ve to -ve. It all worked perfectly. WTF? I rewired it back in the the fan 4 position - and it started stalling again. WTFingF? I have now put it back as fan 5 - working perfectly again! All the fans, which are good quality are exactly the same, bought at the same time out of the same box in the computer store's storeroom, and supposedly from the same manufacturers batch. I have no explanation for this behaviour - does anybody else?