Be it internet via noiseless fibre optical or noiseless wireless (or satelite), the modem is a source of noise as well as the environment and in my case noiseless optical does not guarantee no noise within the network.
Sorry, but it seems there is no short cuts to quality in this.
PS … well maybe short cuts could be exactly that wireless > modem receiver > 10cm cable > router > 10cm cable > server
Well sort of but not quite. The exact network set up is this.
And with that set up I am happy that if I play local files stored on the Oladra the sound is the same whether the Oladra is connected to my local network or if it isn’t.
I have not reached the stage where streamed music from Qobuz sounds as good as playing the same music files stored locally on the Oladra.
However @clive101 has visited me and he is adamant that when he connects his Dela S100 (AfterDark version) instead of my PhoenixNET then he considers the streamed Qobuz music to be better than the locally stored files.
I disagree with him on two counts in that I think the PhoenixNET sounds better and I think the locally stored files sound better than streaming Qobuz through his Dela S100.
We have sat side by side on my sofa and tried this several times and still we disagree. I suppose it must at least mean that all the options we listened to sound pretty darned good otherwise it would be obvious to both of us what was better.
Thank you. Great drawing! Could it be that it is the part upstream preceding the PhoenixNET that you ‘hear’ when streaming?
For an audition streaming- vs. local files I would connect a 20m CAT6A UTP from PhoenixNET directly to the 4G router. At least try maybe next time? I am aware you give up on fiber isolation, that’s why it’s not an option probably.
Wi-Fi to ipad is on purpose behind fiber isolation as well, instead of maybe picking up less noise when closer to the phoenixNET?
In my current setup:
router ==> UTP1 ==> switch 1==> UTP2 ==> switch2 (2*, LPS, grounded) ==> FTP3 (AQ cat7 )==> K50
I got the best result by using an “unshielded” cat5e cable for UTP2 instead of a shielded CAT6 or CAT7.
UTP3 is shielded and only grounded on the switch 2 side.
Still glad I removed the fiber solution.
Perhaps caused by the wrong choice of hardware used.
Crazy hobby but fun.
Thanks, I think a diagram helps to get the information across in an easily understood way.
I am no fan of fibre in terms of sound quality and indeed I would be inclined to try your suggestion of replacing the fibre link by using a 20m CAT6A UTP direct from the router to the PhoenixNET except that my router only has one ethernet output and that goes to the first Cisco. What I might try though is first doing the Cat6A between router and PhoenixNET and then compare that with using the Cat6A between the two Cisco switches.
The Wi-Fi to iPad is not intentionally behind the fibre isolation (and which I am not convinced by in any event) but rather the router is the only means I have of Wi-Fi communication between the iPad and the network.
If you do not like fiber did you try Wi-Fi isolation with a Wi-Fi range extender or access point connected to the 2nd Cisco 2960 with cat5 or 6?? The only problem is it needs a good LPS and some models have an smps integrated.
FWIW
Thanks.
So I understand the cascading switch 1 and 2 sound better than only switch 2?
What brand switches are they? Are they both on LPS?
UTP 1 = also cat5e cable? And in this setup that sounds also better than a CAT6 or CAT7?
BTW, as far as I understand al AQ cables are cat 7 STP shielded twisted pair. vs UTP Unshielded Twisted Pair cable. What AQ cat 7 cable are you using?
Hi Paul,
Switch 1 is connected directly from the router and is needed to power several devices on the first floor with Ethernet.
I did (cannot) not try without switch 1.
Switch 2 in the diagram are 2 cascaded switches.
1 TPlink and 1 HP both managed feeded with a seperate LPS power supply an both cases are grounded.
UTP1 and UTP2 cable in this case are ordinary CAT5e cables.
For UTP2 CAT5e did indeed sound better in my case than CAT7 (shielded) and that surprised me too.
I must say, though, that this was not AQ.
UTP3 is Audioquest Carbon CAT700 with Telegartner.
Just in case you are not aware, it can make a difference if the shield is connected to ground (usually headshell) at one or both ends. Usual convention is connect the sheild to ground at the sending end only, but that is not always best according to some manufacturers.
Also I note the Web page for Audioquest Carbon Cat700 states it has …
OVERALL SHIELD AND INDIVIDUAL PAIRS
This sounds like S / FTP which is overall sheild plus foil-wrapped twisted pairs
Zelf AudioQuest Cat 7 kabels maken met Telegärtner connectoren. - YouTube
I mounted the cable according to this video.
Unfortunately, the video is in Dutch.
Here an AQ vodka, you can see a main tinned copper metal braided sleeve and an aluminum shield around four twisted pairs once more with aluminum shield. Each twisted pair has a “drain” ground wire running parallel (likely attached to ground on one side). This AQ vodka seems to have solid conductors. Going up the AQ line of cables the silver content vs copper becomes more important. The copper itself can also be of higher purity, not sure if this is the case with AQ.
The direction indication (on all AQ cables) suggest the main shield can not be connected to ground on one side (likely the receiving side as dbastin2605 mentioned). But if you want to “hear” disconnect from ground or with ground on the other side you could just connect it in the other direction.
This definitely is an STP cable as opposed to UTP.
UTP comes in 22, 23 and 24 AWG with either stranded or solid conductors. The latter with superior electrical characteristics that remain stable over a wider range of frequencies. But UTP is unprotected against electrical interference. UTP still has the twist to help to improve its immunity to electrical noise and EMI.
UTP’s lack of protection is not necessarily a disadvantage for audio as some have better results with UTP, specifically with the longer lengths in the upstream section and only using the best short STP for the last network connection. I even heard of users exaggerating the length up to 100 or more feet to an SQ advantage.
I made a mistake in my previous post’s by using the name UTP(3) instead of S-FTP.
Sorry for that. 
Not really, it just helped explain possible effects on SQ for each type and use each to your advantage.
Just wondering if anyone has any experience with Poe switches and sound quality and whether these should be avoided given the potential for extra electrical noise.
In my opinion, I would avoid POE.
I think isolation is the key word.
I achieved incredible results with the iFI LAN ISILENCER (in my setup!!!)
“Dirty network” ==> ISilencer ==> Switch (with LPS) ==> K50
Intriguing that such an affordable device can impress you so much. I would love to learn more. What did you hear from this? Which switch are you using with an LPS?
The peace and ease with which I listen now is enjoyment.
Note I am experiencing this in my current setup!
I am allergic to turmoil in the soundstage.
For me, this is a bargain since I was looking at an audiophile switch.
I installed this isolator with so much improvement as a result that I no longer have the need to experiment.
The switch I use are 2 in cascade.
One TPlink managed and one HP managed (both have only 1 port active). These switches are powered by a decent DIY LPS.
The housing of both switches is connected to the LPS ground.
Normally I hold back on this kind of writing.
But I am pleasantly surprised by taking this gamble.
Wonderful to hear, @Harry! Thanks for sharing your impressions. I will have to try one.
