Forgot to comment on the EC2 problem. 

 

Since the B problem was intermittent, there is a 99.99% probability that the flaw is in the relay that switches the injectors between A & B controller injector driver transistors.  This is one of those cases where the complexity of redundancy causes rather than cures a problem.  (See current stories about the space shuttle now on the launch pad.)    This is the reason why I decided not to add the complexity of auto-switching of controllers in the event of a failure.  I get asked about that a lot.

 

 

Tracy  (Wondering how many times I have cited the example of  shuttle launch scrubs due to faulty sensors in the backup systems)