[NCARC] Need help debugging electronics project

[Ed Gage]

Thank you all for your replies.  I see that additional information is in
order.

First off, as AF6KD observed, I am using a "kit" relay board if you will:
   http://www.futurlec.com/Relay_8.shtml .  Sadly, as WA9FBO pointed out,
the solid-state board (http://www.futurlec.com/SSRAC.shtml) did not exist
when I began this project.

However, it was nice to have 50% of all the parts for the project already
mounted, with contact points, etc. so I went with the former board.

I modified this board in 2 ways:  #1, I tapped the 12V with a 7805 (as
WA9FBO already guessed) to power the on-board IC, and the Stamp, and I also
wired the 1N4001 diodes across the coils.

Also, as WA9FBO guessed, there is an on-board power regulator that turns
whatever slop power you put into the chip into clean 5V.  I'm using that
output to power the buttons.

The Basic Stamp II spec says there's a limit of 40mA source, 50mA Sink per
8 I/O pins.  I overcame the source limitation on the source pins by using
transistors to drive the relays (as AF6KD and Steve both had guessed) so
now they don't have to work so hard .  Also, at power-on, all relays are
closed.  The lights are wired to normally-open, and only flash as relays
are "relaxed".

I'm not able to get a good reading of the load as Steve suggests, because
of my digital multimeter.  Oh, how I miss the cheapie analog multimeter I
had as a kid.  However, even with the slow refresh of the reading, I'm
still seeing over 5V on VCC, with no brownout.

That being said, now I'm leaning towards the input being the issue.  The
eight buttons are each at the end of a 30' run of wire.  Although the wire
is shielded, I'm guessing it's not the best quality wire, and requires more
of a current load to operate, even if not engaged.  I'm guessing
the capacitance of the wire is enough to draw the current off the board and
cause a reset of the BSII.  As proof, if I replace the eight button
interface with my test input board (
http://www.futurlec.com/TrainingBoardAccessories.shtml - 2nd one down),
works flawlessly.

I noted that the test input had 10K resistors, so I originally wired the
long-run push buttons the same way.  Did not seem to make any difference.

Still doesn't explain the reason why when touching the + side of the coil
for each relay causes the system to behave, but I like Steve's explanation
that I have an "electric" personality.

--Ed