The EtherREGEN thread for various network, cable, power experiences and experiments

[GryphonGuy]

3 hours ago, lpost said:

Interesting.  I wonder if he found a difference or tried all copper Ethernet.  It's difficult for me to justify how the optical wavelength used could impact the sound and flatten or crush the sound stage; perhaps unless the 850nm SFP were faulty but again, it's a situation of either signal or no signal.  It can be too low or too hot but this will only cause retries or dropouts of the data.

 

It's cheap to experiment but use caution to not to lose the forest for the trees (or music).

 

fyi and fwiw, I have an ER in my system and very much like what it's brought to the system.  I've got copper from a Cisco 2960G feeding it and 850nm multi-mode out to my server with PCIe board then I2S to my DAC.

 

Hi lpost,

 

Yes I did try all copper Ethernet and quite a few different combinations of same brand before and after the ER and different brands before and after the ER. Cables in Cat 6A, Cat7 and Cat8 were used and they varied with the amount of shielding on the cables from none (UTP) to unshielded cable but foil shielded pairs (U/FTP) and fully shielded (S/FTP). Any form of shielding was better than unshielded before the ER. The most pleasing copper connection was Cable Matters Cat8 before the ER and Tera Grand Cat7 after the ER BUT the OM5 fibre cable's transparency (with the 1310nm transceivers) and airiness trounced the copper connection in my system. Of course YMMV. 

 

Unfortunately I do not have an OM4 fibre optic cable long enough to try it to see if it is something special about the OM5 cable but the reason the OM5 cable was introduced, according to sales literature, was to get greater distance using 850nm transceivers.

 

It was a surprise to me too that the wavelength of the transceivers mattered especially since the cable is supposedly optimised for the 850nm wavelength signals. It was not that subtle that the 1310nm transceivers opened the air around the instruments in the mix. The transceivers are from the Fiberstore custom made for my ubiquity networking gear.

 

Regards

GG

[GryphonGuy]

17 hours ago, HeeBroG said:

Some have experimented with various supplies and said LPS-1.2 is beyond its "comfort zone" powering the EtherRegen.

 

As long as you run the LPS-1.2 at 12 volts into the EtherRegen it is comfortable running at that for days on end. Anything less will draw too much current at the lower voltages of 9, 7 or 5.

 

Regards

GG

[GryphonGuy]

On 5/27/2020 at 5:29 PM, trailblazers_song said:

Those of u who supply LPSU to your etherRegen, what’s the prefer voltage ?

 

Paul Hynes Design SR4 @ 9 volts

[GryphonGuy]

4 hours ago, MasterWarzombie said:

Hi , I'm French .

I have been waiting for my next Etherregen ordered for almost 1 month. The wait is long.
I have owned a JS-2 Linear Power Supply for a year
I have 3 questions :
1- What is the voltage setting of the JS2: 7, 9 or 12 volts? thank you. Is there a "risk" if I climb too high?

2 - For the grounded cord, we plug it into etherregen but at the other end, we plug it into which product? A bluray player with a screw for example?
3 - For SFP modules, should I take a "generic"? is there a recommended model?

 

https://www.fs.com/fr/products/75326.html   ?

 with 

 

https://www.fs.com/fr/products/68296.html

 

with 

https://www.fs.com/fr/products/17237.html

 

 

merci

 

 

merci

 

 Voltages: People here have tried all three voltages, As long as you stick to the range of 7 to 12 volts there is no "risk" that you will damage the EtherREGEN. It is just a matter of sound preference if you can detect a change based on feed voltage. I, for example, settled on 9 volts. I haven't tried a different voltage since setting it months ago because I am happy. Some people report a lot of heat with the 12 volt setting but when complaints emerge of this nature, the guys at UA reassure the community that the components are only operating at approx 50% of their temperature tolerances so no damage will be done from heat.

 

As for fibre-optic cables, make sure you have cables with end-fittings that match the expected SFP transceiver endings. A very common connection is LC-LC meaning locking connection (I think). So for example, if you have LC-LC cable, your chosen transceiver should have LC-LC fittings on it.

 

Some people have said that they prefer single-mode cable, make sure if you select that type of cable that you select two single-mode transceivers. Likewise, I chose Multi-mode cable so I use Multi-mode transceivers.

 

Since the EtherREGEN does not restrict SFP transceiver brands or models, I would just get a module to match your transmission side (fibre-port on your switch may be restricted to a particular model of SFP transceiver according to the brand of the switch).

 

If you are going to need to convert electrical Ethernet to optical Ethernet and back to electrical Ethernet just to go to the EtherREGEN, I would reconsider that move unless you have a 10-30 metre distance to between your audio equipment. 

 

I hope this helps and warm regards

GG

[GryphonGuy]

21 hours ago, chungjh said:

At the end, you say "just go to ER." Just to clarify, you are saying to remove the optical fiber unless you have a very long distance for the cable.Correct?

 

Oops, you've misquoted me. 😆 

 

The point of that paragraph was to imply that if you have only a typical 5 or so metre (yards for USA) distance for the Ethernet signal to travel and your switch or home WiFi router does not already have an optical connection, I would start by using electrical Ethernet cables (CAT 6, 7 etc) first before going to the expense of double converting the electrical signal (from electrical to optical and back to electrical).

 

There are rarely right or wrong answers to how we run our lives. Just like if you are hell-bent on running fibre for 3 or 5 metre distance, you can do that. After all, this hobby is for fun and if your definition of fun is converting signals from one medium to another and listening to the effects, go for it!

 

My response was to a forum member who was wanting some advice on optical transmission with respect to the EtherREGEN. That advice was to give him a starting point which is what he seemed to be asking. I was not professing to provide general advice to a more experienced audience.

 

Regards

GG

 

[GryphonGuy]

9 hours ago, Oggo said:

My excuses, if that has been asked before:

 

What gives better SQ?

 

1) router > copper > low-cost FMC > fibre > ER > copper > Streamer/Mediaplayer/DAC

2) router > copper > sonore OM > fibre > ER > copper > Streamer/Mediaplayer/DAC

3) router > copper > ER > fibre > ER > copper > Streamer/Mediaplayer/DAC

4) router with integrated FMC > fiber > ER > copper > Streamer/Mediaplayer/DAC

 

Option 4 is how I have it set up and I found that to be "more quiet" and cleaner than straight copper. However I have not tried multiple ER's in the chain as the quality of straight fibre over 30metres is stupendous.

 

Regards

GG

[GryphonGuy]

2 hours ago, acousticsguru said:

Might add the EtherREGEN is itself one of those products that I got to test in my system and that I would not have considered buying otherwise. Can't honestly say the white paper made much sense to me…

 

Greetings from Switzerland, David.

 

If you have tried different ethernet cables and can hear a difference in each cable's "presentation", then the EtherREGEN will definitely give you a better outcome because of the way it actively operates on the electrical signals in the ethernet transmission. If you cannot hear any difference between different categories of ethernet cables or different brands of the same category, your investment in an EtherREGEN may not be worth it for you.

I know that the EtherREGEN has a huge benefit in my system, so much so, that I have ordered another one so that my PC's audio network card and the digital audio router will both be operating behind an "A" to "B" cleaning service.

 

Regards

GG

[GryphonGuy]

7 hours ago, Superdad said:

We don't modify them (removing the 75-Ohm BNC from the board is quite difficult and risky), we run 10% of each run with 50-Ohm BNC (and matching termination resistor).

We always end up with extra 50-Ohm units. That's probably because 90% of EtherREGEN buyers are enjoying our switch just fine without external clock. 

 

I was one of the providers of feedback regarding Professional audio clocks being 75ohm vs the video industry seemingly using 50ohm.

So is the abundance of used 50ohm clocks making their way into the audiophile/consumer market because of salvage/recycling operations from professional video gear? I don't know the answer to that but I do know that 75ohms is so ingrained into the professional audio equipment market that some suppliers don't even mention the resistance number, it is assumed that all audio clocks for synchronising digital gear in the studio are 75ohm.

 

Whilst I have a studio clock, it does not in itself have a 10M external port, so I am one of the EtherRegen owners currently (but not forever) happily running without an external clock. When (not if) I upgrade my studio clock, I will undoubtedly jump on the 10M bandwagon and EtherRegen will then easily and seemlessly become part of my synchronised audiophile playback system.

 

Regards

GG

[GryphonGuy]

10 minutes ago, R1200CL said:

Is this 10 MHz clocks ?

 

Studio wide.

[GryphonGuy]

6 hours ago, MasterWarzombie said:

SuperDad ,

 

So you do not advise to use SFP on your ER?
is it better to use an RJ45 cable on one of the A ports?

 

I already have an sfp etc module but I wanted to buy this reference from Afterdark?

 

https://www.adark.co/products/copy-of-uptone-etherregen-femto-switch-pre-order-enjoy-worldwide-free-shipping-tax-free?variant=31047093321846

You do not advise?

 

merci :)

 

Laurent

 

 

 

If you have a short run (a few metres or yards) then copper ethernet. If you have a long run (~30 metres) like me, I use fibre optic on those long runs rather than face issues near the limits of old cat 5 cabling in the walls running pasts electrics, metal studs and the like.

 

Regards

GG

[GryphonGuy]

4 hours ago, Duckworp said:

I am using the attached configuration.  I am getting no network connection through it.  If I put a PC in where the SOtM is in the chain it shows a connection is in place but it does not connect to the network (says "unidentified network") - Windows just shows that a cable is connected.  I have checked that the SFP module should, in theory, work with my Cisco switch, but are there SFP modules which do not work with the Ethergen?  I did the infra-red camera test in a post on this page and there is no light, though I am not sure if there should be when it comes off a switch?  Any advice appreciated.

 

You've set up a 100 Mbit/s link when the EtherREGEN's fibre is ONLY 1 GBit/s capable. So get yourself 2 x 1 GBit/s SFP modules (assuming that Cisco 2960 model has gigabit capable SFP port) and you should get a connection. Buy the SFP modules to suit the Cisco 2960 as the EtherREGEN has been designed to work with any SFP module, not specifically branded ones.

 

Regards

GG

[GryphonGuy]

25 minutes ago, Duckworp said:

Thanks for your reply and advice. Where do you see the link speed in the name of the SFP?   The name of mine is below but I don’t see where you see the speed of 100 Mbit/s.  What would be a code for 1GBit/s?  Thanks again. 

Cisco GLC-FE-100FX Compatible 100BASE-FX SFP 1310nm 2km DOM LC MMF Transceiver Module

 

 

Its the 100FX as well as the 100BASE-FX. you want 1G or 1000BASE.

 

Regards

GG

[GryphonGuy]

11 hours ago, MarkusBarkus said:

...the light is "bouncing around" in the single mode cable too, it just has different "bounce" characteristics.
 

Sure, the SM lasers are more powerful/coherent, the cable core narrower/more constraining, but these cables are snaked around through walls, etc. and not in a straight line. 
 

It's not clear to me why folks seem to prefer SM over MM for audio. But, just because I didn't hear a reliable difference, doesn't mean others don't/won't. I'm on SM fiber, it should be noted.
 

I ordered a couple of SFP+ transceivers to test, which is at the heart of the latest dust-up regarding optical stuff. Will it make an audible difference? Will it even work? I'll find out! 

 

My own experimentation in my network found that 1310nm signals were far better than the 850nm wavelength. I do not know whether 850nm is an LED emitter and the 1310nm a proper laser, I just know that over OM5 cabling (I tried OM4 with subtle differences not worth the expense either way) the 1310nm audio sounded much better with deeper blacks more impactful transients etc etc. Everything an audiophile tries to achieve.

 

I have not tried single-mode fibre.

 

Regards

GG

[GryphonGuy]

4 hours ago, R1200CL said:

My reason to prefer and recommend DFB laser and sigel mode fiber 1310 nm is based on research done by people participating in the White Rabbit Project. The interest was at that time phase noise. You will have to do a lot of google and scan to various documents to find this. Phase noise is easier to find info about. They even developed their own switch. 
 

It also seems that manufacturers almost only use DFB laser and single mode support in 10GB and also SFP28. Yes a SFP28 module will fit the EtherRegen. 

 

I also studied several data sheets (including old ones) from Finisar that until some time last year was available at Finisar website  together with a very nice overview about the various all type optical modules. To bad this was removed. Maybe cause I emailed them with questions about the various modules ? 

 

At that time jitter numbers was available. (Still is on the 1421). 

This is why I recommend FTLF1421P1BCL and of cause the lately discovered use of FTLX1475D3BTL. 
I have obtained 10 of the SFP version. I will also test one or two off the SFP+ version. 
 

I have asked JohnS if he can se any technical benefit with one vs the other, since both is 1310nm with DFB laser.

Based on the strict requirements of the 10GB standard, one could assume better quality of SFP+ modules. 

However this also requires the equipment used with the 10SFP+ module to comply with that standard. Not that is should matter that much for our purpose. I don’t know. 

Both modules cost the same new, but the 1421 is widely available on eBay at a reasonable price.

I would assume it’s enough with one FTLX1475D3BTL to obtain any possible benefit of the 10GB SFP+ and it may be beneficial to use the 1421 in the other end, as you then can use a 10GB switch, which you can’t if both modules is 10GB FTLX1475D3BTL

 

https://ohwr.org/project/white-rabbit/wikis/Non-compliantSFP

https://ohwr.org/project/white-rabbit/wikis/SFP

 

 

Your theory needs practical testing. There's something for you to do! 😉

 

With fibre, the golden rule is not to mix and match different model transceivers on the ends of the fibre, as has been stated ad nauseum lately. So when you see something that fits the bill, so to speak, you buy two or in pairs.

 

Regards

GG

[GryphonGuy]

oh,

 

I'll just add, don't buy the 10km type transceivers for a meter or 5 run of fibre. The laser will be too bright for the other end. You would need to buy an attenuator for each transceiver as well if you insist on buying the tens of kilometre range transceivers.

 

I stick with the 1310nm 2km range. No problems yet but I need to run 30 metres of fibre so the losses in that run of round corners etc might be enough to dull the transmission to acceptable levels.

 

I'd be happy to be proven wrong here, as I don't know that technical side.

 

Regards

GG

[GryphonGuy]

On 3/21/2021 at 9:59 AM, R1200CL said:

Actually 1310 is single mode. 1300 can be MM. Name those SFP you’re using.

Anyway that SQ confirmation of 1310nm is nice. 

 

I will. Still nothing wrong in mixing those two specific modules I think. I’m quite sure equal laser wave length and same manufacturer ensures that. Still I totally agree. Testing is good. And just found the 10GB below $40. 

We could even ask @AfterDark. to test some combination of the modules he has, but maybe he won’t. Won’t benefit his sale I guess. 
 

 

This a totally disagree with. Now you’re the one creating confusion about something well tested.

This the same saying don’t purchase Adrian’s nice 10GB modules. 

I’m using this 1421 with a one meter cable. No problem at all. 
And if it really is an issue for someone, just add $10 and purchase 20 meter and hide it behind your rack.

 

I like to hear the arguments about too bright.
 

 

Please stop mixing up the meaning of my words. I am talking about the plastic/glass strand that carries the light impulses when I say the word "fibre". I am not confusing the wave length of the transceiver with the actual plastic/glass fibre as you seem to have done. There are both Multi-Mode and Single-Mode transceivers out there for 1310nm wavelength signals.

 

Also, please don't attribute words to me that I have not said or ever intend saying. Again you seem to have confused distance of 10km with a transceiver speed of 10 Gigabits. Completely unrelated terms for the intent and meaning of my written words.

 

I'm glad to see that you are indeed experimenting. That's how you understand better in MHO. Keep it up and do more.

 

Regards

GG

[GryphonGuy]

4 hours ago, PYP said:

I'm considering further pampering my eR, which is powered by an LPS 1.2, by replacing the 1.2's stock DC cable.

 

Any suggestions for a DC cable with 5.5mm x 2.1mm plugs at both ends?  

 

Thanks.  

 

I use this one with great results.  https://www.ghentaudio.com/part/dc-gac4.html

 

Regards

GG

[GryphonGuy]

4 minutes ago, PYP said:

All systems are different.  In my system the second eR made a greater difference than adding an LPS to the first eR and adding an OM with its own LPS.  But it is a little hard to tell since the changes are additive.    I would certainly make sure the eR has a good LPS as a first step.   

 

My 2nd eR made a huge difference too. Depth and expanse (in all directions) of the soundstage increased massively. Both eR's run on 1310nm transceivers on optical fibre. Through experiments, this is the killer difference in my system. a really nice tweak was to reduce the SR4 from 9volts to 7volts. This was the last tweak and is what snapped the image into much deeper clarity. Still waiting on the March production run of SR4T to add to the second eR that arrived a little over a week ago now. It is still using the supplied power brick.

 

As far as the Gotham GAC4 JSSG DC cable goes, I am putting the difference it makes down to the amount of shielding at has on each strand as well as the cable itself. Apparently Gotham designed that cable as a microphone cable to carry miniscule microphone signals to a microphone pre-amp in a studio.

 

Regards

GG

[GryphonGuy]

6 hours ago, PYP said:

could you mention the entire chain?  would it be:   Router > copper ethernet > er#1  B to A >fiber > er#2 A to B > copper ethernet > ultrarendu > USB > DAC

 

 

I hope this helps:

Avid MTRX is an audio router that has 8-channel ADC and 8-channel DAC in it. Made by Digital Audio Denmark but branded Avid MTRX.

 

As I have mentioned elsewhere, the fibre connections to and from EtherREGENs are running signals of 1310nm from FibreStore transceivers (in Singapore so quite close).

 

Regards

GG

[GryphonGuy]

5 hours ago, JohnSwenson said:

Hi GG,

I just want to make sure you have the correct fiber cables, you say they are OM4 and OM5, these are multi mode (that's what the "M" stands for), but most modules that use 1310 are single mode. Could you check your modules, if they are in fact single mode you should be using OS fiber rather than OM.

 

John S.

 

Hi John,

 

Thanks for your concern. Items are sold as combined Multi-mode or Single-mode. The account manager recommended this for the multi-mode cable from technical team advice.

 

https://www.fs.com/sg/products/37460.html

 

Regards

GG.

[GryphonGuy]

10 hours ago, Duckworp said:

Do you use OM4/5 for any particular reason?  I use OM1 but have not tried OM4/5. 

 

Hi Duckworp,

 

Future-proofing and minimising signal loss is the reason for the OM4/5. Mind you 30 metres is paranoia level at signal loss but, hey, numbers are numbers. 😆 OM5 is actually worse than OM4 for distance at the slower speeds of the EtherRegen requirements but I could not tell the difference in sound quality of the OM4 vs OM5.

 

The reason I went with Multi-mode cable in the first instance was again on a recommendation from the FibreStore. Given that the price of 2x30 metres single-mode cable is trivial even by Malaysian standards, I will order two to see if it makes any difference as the transceivers I use are both multi-mode and Single-mode transceivers and also use the single-mode wavelength of 1310nm.

 

So yet another round of tests. What else is there to do during lock-down anyway? 😀

 

Regards

GG

[GryphonGuy]

9 hours ago, Duckworp said:

 

So are you saying that OM4 will give me better SQ than OM1?  I have a Cisco switch and took ages to find a module that would work in both it and the EtherREGEN.  I am not sure it will work with Single-Mode OM4.

 

No I'm not. I cannot say those words as I have no experience with OM1. I did not find any SQ difference between OM4 and OM5 in the cable itself, just the transceiver wavelength change made a huge difference to both OM4 and OM5 cable. As others have said, there is no such thing as single-mode OM4. OM4 is multi-mode cable.

 

Regards

GG

[GryphonGuy]

21 hours ago, Duckworp said:

OK thanks. Could you let me know what transceiver wavelength you found was the best?  

 

I and many others have found 1310nm wavelength to be significantly better than 850nm, for example. A cautionary YMMV applies here.

 

Regards

GG

[GryphonGuy]

8 minutes ago, One and a half said:

Just weird unexpected behaviour, send me nuts. Like for the jcat net Femto card essentially uses two nics, each one has its own MAC.

 

Wired : Router > Nic 1, Nic 2 > ER > Lumin

 

I’ve bridged the Nic 1 & Nic 2 connection in Windows and the router assigns the IP address to the ‘bridged’ Nic MAC address which is usually Nic 1.  Nic 2 is not assigned and the Lumin is setup for DHCP, so it works.
 

 Now for some reason, on a reboot, Windows assigns the other nic’s MAC and the Lumin doesn’t connect cause it’s not getting the IP address although it remains connected. Lumin then displays network error. Even on a reboot of the server won’t fix, have to find the MAC address that’s active for the bridge , rehashing the router and off it goes.

 

If it's possible to set or spoof a MAC address onto the logical bridge connection, rather than rely on each NIC's MAC address that will be a stable MAC address, at least for the DHCP server to use.

 

It would also be nice to know why NIC 1 is delayed or disabled from requesting a DHCP address on a reboot but not from a cold start. Perhaps when rebooting NIC 1 is not connected but did have an address and windows thoughtfully sees NIC 2 without an address and requests one for the DHCP server?

 

Lot's of troubleshooting ahead and maybe even technical support of either the JCAT card or even windows to stop NIC 2 taking priority.

 

Regards

GG

[GryphonGuy]

On 4/1/2021 at 7:45 PM, GryphonGuy said:

 

Hi Duckworp,

 

Future-proofing and minimising signal loss is the reason for the OM4/5. Mind you 30 metres is paranoia level at signal loss but, hey, numbers are numbers. 😆 OM5 is actually worse than OM4 for distance at the slower speeds of the EtherRegen requirements but I could not tell the difference in sound quality of the OM4 vs OM5.

 

The reason I went with Multi-mode cable in the first instance was again on a recommendation from the FibreStore. Given that the price of 2x30 metres single-mode cable is trivial even by Malaysian standards, I will order two to see if it makes any difference as the transceivers I use are both multi-mode and Single-mode transceivers and also use the single-mode wavelength of 1310nm.

 

So yet another round of tests. What else is there to do during lock-down anyway? 😀

 

Regards

GG

 

Well I have installed the single-mode fibre into the EtherREGENs maintaining the combined 1310nm multi-single-mode transceivers and whilst the mids and highs seemed to be a tad "cleaner", the bass almost disappeared and worse, the soundstage collapsed depth-wise. I let them play for a couple of albums but the lack of bass and sound stage depth was really annoying. So switching back to the 1310 nm multi-mode fibre (OM4) and the bass and depth returned instantly (not subtle!) and the mids and highs seemed to blend back into the mix.

 

Now I realise that some of you will say that my result is because the dual-mode transceivers are simply better at Multi-mode than single-mode so I have ordered some single-mode-only transceivers and will repeat the test with the same single-mode fibre but exclusively single-mode transceivers and see if there is any difference. I'll report back in about another week or so.

 

Regards

GG