Not that it's at all related to your problem, but if I'm watching a movie in dolby digital surround, and then flip the ceiling fan on or change it's speed, the amp drops out of digital mode for a few seconds and then resumes.
I think motors generate quite a bit of noise on startup.
I wonder if they make an X10 filter that can be plugged into the receptacle, and then your shredder can be plugged into it?
This one may be tough. A filter may not help. It depends on which of two scenarios (see next paragraph) is applicable.
I suspect the shredder motor is pulling the line voltage down causing a brown-out reset by the MCU (not so) smarthouse (I assume you mean SmartHome) two-way. Or, the motor noise is somehow triggering a reset directly. I doubt a filter will solve the former but may solve the latter. I don't recall whether you can disable local control or even local control reports in these two-ways. I have three but they exhibit different flaws and I don't use them.
This would explain a few similar reports. I just never made the connection until you noted that you have the smarthouse (sic) two-way. They report local control changes, thus generating valid PLC commands.
Given the Manchester encoding used by the X-10 powerline protocol, I think it's impossible that a noise source can generate valid X-10 commands out of whole cloth. I've been taken to task by several people for insisting that there has to be a transmitter that is sending the signals when there are apparently valid signals being logged.
X-10 has documented the brown-out reset case in one of their FAQs.
It might be enlightening (no pun intended) to put a scope on the powerline to see whether it is noise or low voltage. But, whichever it is, this may be the Achilles Heel of the various *Linc two-way switches and modules.
I haven't looked at the >I do think it was a "brown out" type problem...
I dug one of the LampLinc 2000STW modules out of my junk box and checked the MCU. It uses a PIC16C622A so the idea that noise spikes might trigger a reset is plausible. It might also be that the spikes merely cause it to send a false local control signal. The *Linc two-ways send an On or Off when their load is turned off or on. It doesn't send when itself is powered on but only when there is a change in the load state. Was the light itself cycling or were you just seeing the logged commands? That it was sending both its own address and one other address makes me suspect the noise spikes were messing with its mind.
While I had it out I decided to test how the 2000 STW resp>It might be enlightening (no pun intended) to put a scope on the powerline