Learned something new today. Thanks.
Galvanic Isolation is DC blocking only normally. The very best isolation may use tuned circuits but they are way expensive and certainly donāt make it into $20 SOHO switches.
1 Like
The galvanic isolation used in ethernet is made to prevent a power surge to kill every router and switch in the network, and while it does remove some electronic noise it does not remove everything.
No one is championing gear that was scooped out of the trash bin and re-shrink wrapped.
I picked up a Cisco 2390 with 48 copper GBe and 4 10GBe SFP+ for $60 shipped on eBay. For $210 Iām running fiber for my endpoint and file server and averaging 332MB/s. Thatās a single ~65MB FLAC track delivered in 1/5th of a second. After that the cable doesnāt really matter.
I have Ā£10.00 ish basic switches and my system sounds amazing. Meridian DSP SE
I expect they must be well designedā¦
2 Likes
Itās probably because of the rock on top of your switch 
This provides excellent common mode rejection and significantly reduces common-mode noise. That is, noise that appears on both conductors in an Ethernet cableās twisted pairs.
Sounds correct, my point being that some people argue like the galvanic isolation removes all noise but leave the actual signal, but thatās not how it works. That also means that some electronic noise produced by the switch, and some RFI that is picked up, will make it into the streamer.
No, this is not strictly correct. Any external noise will affect both wires equally in a differential pair and the receiver [streamer] rejects common signals, so noise riding the signals is identified and rejected. Contrary to what you may believe, common mode noise on Ethernet cables leads to noise emissionsāsomething you [inadvertently] demonstrated in an earlier post.
BTW, differential circuits and magnetics are used because the distances between networks means that nodes may use different power sources, i.e. a ground offset is more likely. So, the principal reason for magnetics is isolation to protect equipment.
1 Like
We agree on that at least. But the emission generated in the post you linked to came from an radio antenna on the roof, and entered the ethernet cable as RFI. Then the author explained how he got rid of it.
Anyway, I always try to understand when something gives an improvement to the sound, and the EtherRegen I tried did give an improvement. Using fiber and a fiber media converter also gives an improvement. My explanation is that the improvement comes from less electronic noise into the streamer.
No, itās the other way around! 
Emissions from the Ethernet cable were picked up by the radio receiver.
I reread it and to be honest I am not quite sure, you might be right. But if it was emission from the cable itself, using STP cables would solve it. I think in his case the switch picks up RFI from the antenna, and feed it to the network and then affect his measurement/radio equipment connected to the antenna.
So the problem is electronic noise in the ethernet network, and if that noise can get through galvanic isolation and produce problems for his measurement equipment, then noise can also get into the streamer from the ethernet network and affect the streamer directly and also pass along some noise to the DAC.
Not that this is news in any way, for example this article published 4 years ago: https://www.audiostream.com/content/electrically-isolate-your-networked-audio
which is the solution I use, except my switch has fiber connection so I only need one fiber media converter.
Very interesting to read. This time coming from Rafe Amott.
More power to BlueJeans Cable, as many people on this forum already know.
2 Likes
BJC locked me in as a lifetime customer when they posted the correspondence with Monster Cable that one time Monster tried to threaten them with a lawsuit without realizing that the founder of BJC is a former trial attorney. His reply to their letter is a work of art.
Seriously, itās a glorious thing to read: https://www.bluejeanscable.com/legal/mcp/response041408.pdf
7 Likes
Thatās brilliantā¦ remind me never to buy from monster 
Yes, a beautiful thing to read. Makes me even happier that Iām a long term BJC customer.
(and Iām referring to the letter responding to Monster cable threat of legal action)
Stopped reading after the mentioning of the positiv effect of āburn inā. On a DIGITAL cable?
Tubes needs burn in and since this time it is in the brain of audiophools that burn in has a positiv effect. 
6 Likes
Appreciate Your subjective opinion.
Suggest to stop using terminology such as audiophools, audiognostics, etc.
Rafe made some comparison and gradations between cables he tested and I will definitely put BJC on my radar.
Trust that vast number of people here already have had BJC in their good systems, although not resisting challenge by replacing generic cables and probably observing SQ improvement whilst justifying investment.
Assume this is all about related to cables and SQ affection and certainly the title of this topic, isnāt it?
Thereās nothing subjective about rejecting the notion of burn in for cables.
The conductive properties of metals do not change by running electricity through them. That is not an opinion. That is a fact.
Or do the lights in a new house burn brighter after a month of occupation?
3 Likes