Power supply for ethernet switch?

E-net switches do not use a lot of power, demand stay around the same regardless of what is being passed through it.  If it was a POE switch, then a power supply would make sense. But on a regular couple port switch, don't think it's worth the time and money. 

With a cheap switch, it should not be plugged in with your stereo. I'm sure the cheap switches have no noise rejection protection, generally electrical noise is not a concern with switches. Not only that, but the way the switch works, it will not transfer power noise to ethernet. 

Ethernet is not a audio signal, it's digital packets.

@erik_squires Thanks for the power strip recommendation!!!  I was actually just looking into a good/safe one for my system. I'm using a 'good quality' one (basically just solid, all metal, strip) but there's no real surge protection or filtering of any kind. I would imagine the filtering done is some form of high frequency carry on the power line that has an impact on the Ethernet switch/routers?

Hey @amtprod ,

Furman units with SMP are series mode surge protectors.  Like ZeroSurge or Brickwall.  What they all have in common is they have a low pass filter at the start of the surge protection circuit, which is another word for "noise filter."

This low pass filter is active around 3kHz and higher, which is a lot lower than EMI/RFI filters, a good thing. 

Furman also adds LiFT, an active linear filtering technology to further reduce noise.  While the strip is designed to reduce noise coming in, it also effectively reduces noise above 3 kHz going out, making it a great candidate for wall warts.

The low pass filter slows down a voltage surge long enough for the rest of the circuitry to have time to respond to it.  Some parallel mode surge strips also incorporate some sort of low pass filtering for similar reasons. 

@mgrif104 

“Some will still disagree or remain skeptical about 30 picosends of timing error being audible. So be it. But I would suggest that it’s no longer worth arguing over. It can be measured and some of us hear the improvement.”

I was simply saying that it was not clear whether the jitter (30 picoseconds?) was measured on the ethernet side or the output of the data buffer.

You answer does nothing to clarify this, yet you continue to trot out that something can be measured, so someone can hear the improvement.  You seem to have dropped the con from seconds but not necessarily from your post.

@richardbrand 

curious. - I don’t know why the picosecond didn’t copy over from my post. Another mystery.

As to your question - I won’t venture an answer as I’m not the author of that material. The point the author was making is that a switch indeed introduces measurable jitter.  

Before this study, many believe they heard improvements from upgraded switches. I acknowledge it could still be placebo, but at least we now can begin to consider potential deleterious effects of jitter with some supporting data.

The point I was trying to make is that it’s not worth spending time on as it’s a circular argument.  A point you seem to find unsatisfying.