Tobias
Rank: Quartet
Posts: 320
Member is Online
|
Post by Tobias on Jul 14, 2024 13:54:31 GMT
Thanks. I´m glad someone tried the amazing throttle cable after a few months and posted your findings. Great stuff! The pairing with the WiFi extender, in client mode (as mentioned), might actually be as good as it gets (with clean power) as source into a streamer(?). Well l only wished l had tried it earlier Tobias. Minor update, that may be of interest. l ditched my FMC on my streamer side, and replaced with Throttle on my Ugreen CAT 8 cable, it just sounded so much natural a clear. So today l removed my 2nd FMC that runs my server, for just LAN. The sound took a number of steps back running LAN, going back to FMC, the magic came back. So oddly FMC will stay on the server side. As Tobias and Eric has mentioned,it's all about increasing the impedance. It is actually about reducing noise (the term used by engineers) only, which degrades your DAC performance. The Throttle cable can only passively target a certain amount of noise, as the last step. Your server is likely very noisy, compared to your streamer, and the Throttle cable is not able to clean that up, on its own. Hence your streamer is still getting polluted with noise from your source (server) if removing the FMC. |
|
|
|
Post by stellabagpuss on Jul 14, 2024 15:52:20 GMT
Well l only wished l had tried it earlier Tobias. Minor update, that may be of interest. l ditched my FMC on my streamer side, and replaced with Throttle on my Ugreen CAT 8 cable, it just sounded so much natural a clear. So today l removed my 2nd FMC that runs my server, for just LAN. The sound took a number of steps back running LAN, going back to FMC, the magic came back. So oddly FMC will stay on the server side. As Tobias and Eric has mentioned,it's all about increasing the impedance. It is actually about reducing noise (the term used by engineers) only, which degrades your DAC performance. The Throttle cable can only passively target a certain amount of noise, as the last step. Your server is likely very noisy, compared to your streamer, and the Throttle cable is not able to clean that up, on its own. Hence your streamer is still getting polluted with noise from your source (server) if removing the FMC. Yes that definitely the case, as now l have tried a number of experiments on the server end, and l like to report them back, for other TAS members. Just looking forward to other peoples impressions of it. |
|
Morph
Rank: Duo
Posts: 62
|
Post by Morph on Jul 15, 2024 18:20:07 GMT
Well this has been a very worthwhile experiment! I ordered a ferrite ring and eight clamps from Mouser as listed in the German paper linked by protegimus and previously by Tobias. Decided to track down a 2m ethernet cable that meets the same specs as shown by trhh in his first photo on this thread. Had the cable up and running in a few minutes. Decided to use the remaining three clamps on the Ugreen ethernet from router to EtherRegen. Quite a revelation it must be said. Frankly my descriptions of audio are pathetic so I'll just have to leave it at, the system has come further to life and I'm delighted with the improvement. So many thanks to all of you that steered me towards this happy event. Cheers Morph |
|
trhh
Rank: Soloist
Posts: 24
|
Post by trhh on Jul 16, 2024 7:53:52 GMT
Well this has been a very worthwhile experiment! I ordered a ferrite ring and eight clamps from Mouser as listed in the German paper linked by protegimus and previously by Tobias . Decided to track down a 2m ethernet cable that meets the same specs as shown by trhh in his first photo on this thread. Had the cable up and running in a few minutes. Decided to use the remaining three clamps on the Ugreen ethernet from router to EtherRegen. Quite a revelation it must be said. Frankly my descriptions of audio are pathetic so I'll just have to leave it at, the system has come further to life and I'm delighted with the improvement. So many thanks to all of you that steered me towards this happy event. Cheers Morph Would be great with a PIC :-) Torben |
|
Morph
Rank: Duo
Posts: 62
|
Post by Morph on Jul 16, 2024 10:38:33 GMT
|
|
Morph
Rank: Duo
Posts: 62
|
Post by Morph on Jul 16, 2024 10:43:56 GMT
trhh. Excuse the dust, even I'm shocked!
|
|
Tobias
Rank: Quartet
Posts: 320
Member is Online
|
Post by Tobias on Jul 16, 2024 17:42:49 GMT
This is probably something that most people already know...(and is sort of obvious) but when you are lowering the ethernet noise floor to this extent, as we are doing here (and using other methods) i have noticed that digital cables must be clearly separated from power cables. It is clearly degrading the sound when they are too close to each other, i feel. Here the shielding on the power cables might make a difference, for the other cables?
Does anyone agree on this?
|
|
|
|
Post by John on Jul 16, 2024 18:13:58 GMT
I would add keep any signal cable away from all mains releated cables, including grounding cables
|
|
Morph
Rank: Duo
Posts: 62
|
Post by Morph on Jul 22, 2024 11:45:04 GMT
Sorry if this a bit off topic. After trying and failing to connect the JPlay app to the Melco N1ZH earlier I thought I'd do an experiment using the R26 LAN port. I've always thought that the N1ZH had the edge over the R26 LAN port. So I connected the throttle choke cable to the R26 and and JPlay made the connection with no issues. Qobuz is in use. Well well well, what can I say? Unless confirmation bias has kicked in it's safe to say that the R26 has finally come out on top. Broader more spacious and deeper soundstage. Bass has more texture and resonance. Mids are definitely more pronounced and focused whilst highs are smooth and gentle without being dulled at all. I can't see this uptick being due to the app so it must be the throttle cable/LAN port combination. Most pleasing indeed. So stellabagpuss you're tip re the R26 LAN port has finally paid off here, thank you. Cheers Morph |
|
|
|
Post by stellabagpuss on Jul 22, 2024 15:37:44 GMT
Well that's great to hear that it's worked for you.
I can't take the credit.. as a number of us have been passing tips back forth, l believe we have to thank Eric.
This mod is really too good to be true, but it's has so many advantages, plus you can stick with your favourite LAN cable.
Also don't forget, you may get a slight increase in Siblance, l sorted this by adding a additional single Ferrite clamp before it hits my Cisco switches, and one in between my Server and FMC adapter.
l guess this is in the realms of fine tuning, perhaps it a R26 thing, but l would have never have dreamed of ditching my FMC and DAC cable into the R26, and yet this mod,totally kills it, at my end.
Best the best bit of low priced Hifi fun l've had in years, and just goes to show, that we all have to adjust our mind set,and learn new things.
|
|
|
|
Post by Clive on Jul 22, 2024 17:38:19 GMT
Well that's great to hear that it's worked for you. I can't take the credit.. as a number of us have been passing tips back forth, l believe we have to thank Eric. This mod is really too good to be true, but it's has so many advantages, plus you can stick with your favourite LAN cable. Also don't forget, you may get a slight increase in Siblance, l sorted this by adding a additional single Ferrite clamp before it hits my Cisco switches, and one in between my Server and FMC adapter. l guess this is in the realms of fine tuning, perhaps it a R26 thing, but l would have never have dreamed of ditching my FMC and DAC cable into the R26, and yet this mod,totally kills it, at my end. Best the best bit of low priced Hifi fun l've had in years, and just goes to show, that we all have to adjust our mind set,and learn new things. Interesting that you mention an increase in sibilance. I got this with CAT 5e but with Ugreen CAT 8 it vanishes - this is when used wrapped around the big ferrites. BTW the Jplay app is good and brings an SQ lift due to reduced network traffic and polling. |
|
|
|
Post by stellabagpuss on Jul 22, 2024 17:49:33 GMT
Well that's great to hear that it's worked for you. I can't take the credit.. as a number of us have been passing tips back forth, l believe we have to thank Eric. This mod is really too good to be true, but it's has so many advantages, plus you can stick with your favourite LAN cable. Also don't forget, you may get a slight increase in Siblance, l sorted this by adding a additional single Ferrite clamp before it hits my Cisco switches, and one in between my Server and FMC adapter. l guess this is in the realms of fine tuning, perhaps it a R26 thing, but l would have never have dreamed of ditching my FMC and DAC cable into the R26, and yet this mod,totally kills it, at my end. Best the best bit of low priced Hifi fun l've had in years, and just goes to show, that we all have to adjust our mind set,and learn new things. Interesting that you mention an increase in sibilance. I got this with CAT 5e but with Ugreen CAT 8 it vanishes - this is when used wrapped around the big ferrites. l honestly think it can vary Clive, depending on the cable your using. I'm currently running a Ugreen Cat 8, but the gaming version, that has more sheilding and Graphene sleeve over the CAT 7 or 8. TBH l haven't started swapping cables out,l just reacted to the Siblance, but it's certainly a trait worth mentioning, but it's shouldn't be something to put anyone trying out, as even with the Siblance,any negatives are minor compared to positives in brings. |
|
|
|
Post by MartinT on Jul 22, 2024 18:23:34 GMT
I really dislike sibilance and have always found it to be the result of poor digital noise performance.
Interesting that you have found the ferrite toroid to possibly increase sibilance? I would have thought that to be a backward step.
The toroid is on its way and I have plenty of snap chokes to experiment with. Using 4G+ as I do for internet, I suspect that I'm already getting lower noise than many with my setup. We shall see - more later.
|
|
|
|
Post by stellabagpuss on Jul 22, 2024 18:52:30 GMT
I really dislike sibilance and have always found it to be the result of poor digital noise performance. Interesting that you have found the ferrite toroid to possibly increase sibilance? I would have thought that to be a backward step. The toroid is on its way and I have plenty of snap chokes to experiment with. Using 4G+ as I do for internet, I suspect that I'm already getting lower noise than many with my setup. We shall see - more later. Well hopefully you'll enjoy experimenting Martin. I only mentioned the Siblance, as a possible issue and a solution. lt's worth mentioning,l also have lots equipment burning in at the moment, 3 x Cisco 2960 switches all with cap modes,and two new Ugreen CAT cables,with probably about 8 hours on the clock, new XO components in my MA Studio 1G speakers,so this,could be a contributing factor,after all.. we've all had our burn in issues. Ultimately its all a bit of fun, for peanuts really. |
|
Morph
Rank: Duo
Posts: 62
|
Post by Morph on Jul 22, 2024 19:58:32 GMT
Well the N1ZH has actually shown up on JPlay, so.... why not knock up another throttle cable using the spare Ugreen ethernet cable that's here? Ordered the ferrite ring and clamps from Mouser again and will report back once it's in place. Love/hate this hobby!
|
|
|
|
Post by Clive on Jul 22, 2024 19:58:59 GMT
I really dislike sibilance and have always found it to be the result of poor digital noise performance. Interesting that you have found the ferrite toroid to possibly increase sibilance? I would have thought that to be a backward step. The toroid is on its way and I have plenty of snap chokes to experiment with. Using 4G+ as I do for internet, I suspect that I'm already getting lower noise than many with my setup. We shall see - more later. Sibilance has me climbing the walls. It’s only a fraction more..but it tips the sound away from natural sibilance. I’m not believing it’s down to the ferrite core. I believe it’s due to the unshielded CAT 5e cable which was recommended for easy wrapping around the ferrite. CAT 8 doesn’t exhibit unnatural sibilance….in my system. |
|
|
|
Post by HD Music & Test on Jul 23, 2024 9:32:07 GMT
I have been working with this solution with a couple of clients, I have been measuring what is happening in their own systems. What I can say it is VERY system dependant, those that have light noise removal within the system, benefit the most imho. The throtle cable in my own system killed it stone dead as in sat on the dynamics an reeled in ther soundstage also I lost bass depth and texture. I use a 4G sim card router feed by a QP-2 power supply, 0.75 RJ45 cable with a custom signal conditoner on the RX end, which is why I feel it was negative in my case. Also experimented with many different choke values (38 at the that point, however many combinations are possible!) based on my modelling. I have dedciated RF current probes which can accurately deliver current values in real time as it is happening along the transmission line. We also have physical layer testing of up to 10GB/s ethernet protocol, and can see the end eye pattern and supply rails with a controlled test sequence.
THis allows me to see ALL the signals happening at once in real time very accurately so genuine real world results can be seen as you make changes.
Then add the ferrites and retest, have tried many other methods of emi/rfi control which produce greater & lesser results.
This is nothing more than basic physic's and how current propigates down a transmission line, whether common mode or differential depending on signal transfer protocol. I like the way you chaps are now thinking more outside the box, good work keep it up! |
|
|
|
Post by stellabagpuss on Jul 23, 2024 13:07:33 GMT
Thanks for the confirmation Tony, l guess having the right test gear makes life easier, where us lot are generally tuning via trial and error.
Just a quick question, could you use a SDR, "Software defined Radio" with probes as an antenna to test for noise. Naturally your limited to what bandwidth the SDR can display and capture, just wondering if it could be a possible cheap solution for us cavemen to have, and visual readout for us to work from?
Damien
|
|
|
|
Post by HD Music & Test on Jul 23, 2024 13:20:52 GMT
Hi Damien Yes the equipment will give you the results, although knowing how to intrept those results and setting up the equipment correctly for that task does take a while to master. Experiance helps a great deal as well D. SDR you could use this, for you chaps having 30Mhz BW would more than sufficent, although here the RBW (resolution band width) is important as the DANL (displayed average noise level) are important for quality repeatable results imho. I would suggest at least 140dBm. You could use a Scope's FFT's this has the advantge of starting usually from around DC rather than 9Khz (Spectrum analyser) You an purchase a quality 12 bit Siglent/ Rigol from around £450 now this also has a lot of tools that would be useful plus FFT's you can made your probes but again around £325 for set. Worth considering a shared purschase.
Obviously we are EE chaps for a living and having physical layer testing abilities (all levels) are not a viabale option unless you work with this technology. Although you can make a test fixture for ethernet basic measurments, using impedence matched stripline, 6 layer board with various test points to enable connection to test equipment, this would help you in this area I feel.
|
|
Tobias
Rank: Quartet
Posts: 320
Member is Online
|
Post by Tobias on Jul 23, 2024 14:26:33 GMT
I really dislike sibilance and have always found it to be the result of poor digital noise performance. Interesting that you have found the ferrite toroid to possibly increase sibilance? I would have thought that to be a backward step. The toroid is on its way and I have plenty of snap chokes to experiment with. Using 4G+ as I do for internet, I suspect that I'm already getting lower noise than many with my setup. We shall see - more later. Sibilance has me climbing the walls. It’s only a fraction more..but it tips the sound away from natural sibilance. I’m not believing it’s down to the ferrite core. I believe it’s due to the unshielded CAT 5e cable which was recommended for easy wrapping around the ferrite. CAT 8 doesn’t exhibit unnatural sibilance….in my system. I believe cable shielding also hinders/blocks the ferrite to work as the shielding then also has a negative impact from that aspect... For me it worked best with unshielded cables but then i also don´t have any bare cable exposes, except for around the ferrites. I had sibilance problem when trying a completely different type of ferrite combination, that i thought would be better. But no, it was much worse. It is important that we use the correct recipe for your exact noise profile, which is "impossible" to know without trial and error. This is why i follow their (Eric´s) exact setup from start to finish, into the streamer. |
|
