Thanks to all who gave me the benefit of their experience on this. Sadly, all the tests I've done confirm the simple fact that accessing the ADC12CTL0 really kills my Timer_A Uart interrupt. Whether it's a chip bug or what, I can't say yet, but I've just put in a fixed time delay before reading the ADC (it isn't long anyway), and all seems to workd well. If I get time, I'll investigate further, but as long as the bodge works, it won't be high priority!
Paul Burke
