David Dyer-Bennet wrote:

> Jumping 5 or 6 or 8 to 20 doesn't do anything (other than lighting the
> light associated with 5, 6, or 8 on the breakout box).  But it isn't
> detected either by my program or by the UPS monitoring programs I've
> tried. 
> 
> When I run my monitoring program it reports one value, then instantly
> another value, then never reports a change again.  The initial change
> is *before* I do anything to make a change at the breakout box.  It
> very probably represents some sort of bug in the program, possibly
> even a bug that makes the whole test invalid; I just haven't figured
> out what it *is* yet.
> 
> If I control/c out, and then run it immediately again, I get exactly
> the same behavior (and exactly the same values), which I take as
> additional evidence that the output isn't responding to reality.

Try this David,  tie pins 4 + 5 to the computer together and test again.
Most of the newer computers (post Pentium) need 4+5 jumpered or 5 
connected to something else for the serial port to function.  I would'nt 
think you would need this just to scan the port but maybe...........
John