CM11A lockup

["Robert Green" <ROBERT_GREEN1963@xxxxxxxxx>]

Once again, an X-10 "broadcast storm" detector would have been a very useful
device to have.  I was sitting in my office, surfing and I began hearing
random appliance modules going on and off.  Every ControlLinc Maxi I checked
showed monitor LEDs flashing but this was something different than the
typical stuck button lockup.  The LEDs were not on constantly as they are
when there's a stuck button.  Also, legit wireless and PLC command
*occasionally* managed to get through.

I got the Monterey analyzer and it showed something incredibly weird.  The
rogue transmitter was issuing about a command each second, and it was
incrementing both the house code and the unit code a time.  I'd see a
general pattern of A's, then B's then C's.  Stuck buttons generally appear
on the Monterey as a string of "BSC" messages and fill the unit's memory of
190 commands within seconds.  This data stream was much slower than that,
and was the first time I had seen such a huge variation of house/unit codes
in the PLC data stream.  In addition, unlike the stuck button case, nearly
all of the commands were legit X-10.

Unfortunately, the XTB's powerful signal means I can no longer guess the
general distance from the outlet-under-test to the transmitter by the
voltage reading.  The Monterey just reports 4+ volts for nearly every
transmission.  The ESM-1 could be modified to cope, but the LED bars aren't
precise enough to estimate transmitter-to-meter distances with any accuracy.

This time, though, my memory was good enough to recall that I had been
working in the PC room and had disconnected the CM11A's serial cable from
the PC and left it hanging.  What I don't understand is why it took nearly
12 hours for the CM11A (which got wicked hot!) to begin its bizarre
broadcast?

--
Bobby G.