WONKY performance issues

[IndyMike]

Hello ergodic,

 

I was performing some tests awhile back and came across something that might explain your situation.

 

I was trying to determine whether dual band devices were better able to reject noise than powerline only devices (they are). I used a function generator to inject varying noise levels into the powerline.

 

What I found was that my PLM would interpret 130 KHz noise as powerline communication activate it's collision avoidance. In other words, it would refuse to transmit either powerline or rf communication because it believed there was already traffic on the powerline. This occurred for relatively low levels (25 mv) of noise in a frequency range of 130 to 131 KHz. The 25 mv is far to low to show up on a ELK ESM1.

 

Placing the PLM behind a filter "shielded" it from the noise and allowed it to communicate via RF. If I remember correctly, that fits your described problems rather well.

[Scottmichaelj]

This is a followup post.

 

Well it appears my old PLM was dying so I purchased two more of the new dual band PLMs. One of which was bad and failed already! At first I thought it was something that I did wrong when I replaced it from the old one but after a couple days it kept loosing its links. Replaced it with the backup I bought (just for this very reason) and so far so good. I also purchased 10 filterlincs to put around the house. I had none on any of my UPS backups for the computers, my SHO laser printer or my fridge. So I put them on a couple days ago. Too soon to say if they are helping but when doing a query the same devices are still having lower hops. I also found I getting lower hops on 2477D switches not just the older 2476Ds - so this means that the 2476Ds are not really the issue, so I wont be replacing them. So far things have improved with the new PLM. I am keeping an eye on it all.