Sorry, to clarify - (my router is down stairs and my other equipment is upstairs).
For over a year the configuration was : Apple router downstairs giving wireless etc, and plugged into that was a powerline adapter taking a wired connection (I know its not as good as an actual cat5 cable) upstairs, plugged into a switch with my imac, nas, printer and weatherlink. The imac, switch, nas and console powered via a ups. The powerline is because I don't want to run a cat5 through the house.
I was having some performance issues, so I figured I'd try and eliminate the powerline adapters by moving things downstairs to plug directly into the router (my Nas died so that was ignored). My theory being that plugged directly into the router would eliminate connection problems.
Thats when the weatherlink started to be unreliable. No pattern to it.
So I changed the Apple router for the isp supplied one (that i'd never used) because the Apple one is really old, so I wondered if that was faulty, but the problem still happened.
It was then that I was confused.
The router(s) aren't on the ups, just the console/switch/imac (and dead nas).
So it wasn't originally connected directly with cat5....and it was stable, it was (ironically) only after I connected it directly that it started to be unreliable. This is where I was until I read somewhere that the weatherlinkIP is fussy about power fluctuations (may or may not be true) but as i'd previously had it powered via a ups it seems worth a try.
My concern is that the console or weatherlinkip is faulty, so putting things back will help to confirm that...if it is faulty I would expect it to still be unreliable.
The whole reason for moving it in the first place was slow connection to the imac, so that may return, but its worth it for this test.