Effectiveness of Short Run of Fiber Optic Cable.

asindc

Well-Known Member
Sep 27, 2012
187
20
923
Hello All,

I’m considering re-configuring my streaming setup with the possible use of fiber optic cable. I just bought an Antipodes CX that I will receive next week. I had originally considered making a long run of FO cable to replace my current long run (about 20 ft.) of generic originally installed 16-yr-old CAT5 cable from switch near router to wall plate in listening room. Unfortunately, making that FO run is just simply not practical, if possible at all. Here is my current setup:

Verizon Fios router (also serves as wifi for iPad and iPhone)—CAT5—>Netgear switch GS-116 (all ethernet connected devices in home are plugged into this, including Mac Mini as Roon Server)—CAT5 to listening room—>Wall Plate—WW Starlight 8 Ethernet—>EtherRegen (stock)—>WW Plat. Starlight Ethernet—>Cary DMS-600 Streamer/DAC.


Since I cannot do the long run of FO to the listening room, I’m thinking of doing a short run from an FMC (placed directly after Fios router) to a dedicated switch placed directly after FMC. The switch would then convert the FO back to Ethernet for the long run to the listening room. It would look something like this (changes in bold):

Verizon Fios router (powered by iFi iPower LPS)—WW Ethernet—>FMC—>1m (or whatever length needed) FO Cable—>Buffalo BS-GS2008 switch (powered by Keces P3)—>WW Ethernet—>Buffalo BS-GS2008 switch (powered by Keces P3)—>CAT5 to listening room—>Wall Plate—WW Starlight 8 Ethernet—>EtherRegen (powered by Paul Hynes SR5T)—>WW Plat. Starlight Ethernet—>Antipodes CX (Roon Server)—WW Plat. Starlight Ethernet—->Cary DMS-600 Streamer/DAC. The non-audio ethernet devices would be served via a separate ethernet cable run from the Fios router to the Netgear switch. The Netgear switch and all other devices in house with SMPS will get Jemeco LPS.


So my question is: Would it be worthwhile to run an FMC—>1m (or however how long) FO cable from the router into a pair of switches (or just one switch) just to convert it back to ethernet for a 20ft+ run to the listening room? The idea would be to isolate the audio stream from the router and everything connected to the other ethernet run from the router, but I don’t know if this would do the trick. Any other constructive suggestions for the proposed setup are welcome as well.
 

asindc

Well-Known Member
Sep 27, 2012
187
20
923
Thanks. Does it matter whether it is a short (less than 3m) or long run of FO?
 

microstrip

VIP/Donor
May 30, 2010
20,806
4,698
2,790
Portugal
What the fiber would do is break any ground loop or ground noise from the connection to the router. Whether that helps depends upon whether you have noise problems now.

It should do a little more than that - different type of fibers and converters sound different in my system. Some people have referred that even using different fiber lengths also change the sound, but I have not evaluated them. My experience was with cheap FMC units, not with the more expensive audio recommended ones.
 

DonH50

Member Sponsor & WBF Technical Expert
Jun 22, 2010
3,947
306
1,670
Monument, CO
It should do a little more than that - different type of fibers and converters sound different in my system. Some people have referred that even using different fiber lengths also change the sound, but I have not evaluated them. My experience was with cheap FMC units, not with the more expensive audio recommended ones.

Color me skeptical but I've almost no experience with fiber in this application. You wanna' talk SONET or 24G SAS I can provide numbers... Have you measured to see why the differences? IME the difference is largely in the converters, and in audio systems how well grounds are isolated and how well signal bleed through the transformer is suppressed and how well the ultimate DAC is isolated from the digital noise. There are of course bad connections -- a clean fiber connection can be tricky, especially when doing it manually.

For the record, specs and measured performance of high-speed data links show much higher jitter for fiber than for coax, but you can't do 100+ m runs in copper either, let alone long-haul km+ runs even with repeaters.

My previous design experience was with VCSEL drivers, not sure what lasers are used in the cables under discussion, and that was over ten years ago (more recent experience is as a user, not a designer). I'll leave audible differences to you and the other experts.
 

nonesup

VIP/Donor
Feb 15, 2017
1,044
623
355
Spain
No. Once you introduce fiber you've broken the noise path.
Sorry, regardless of any technical explanation, my experience is that it does matter a lot.
 

nonesup

VIP/Donor
Feb 15, 2017
1,044
623
355
Spain
I have purchased the following cable sizes: 1, 2, 3, 5 and 10 meters. As the measurement was smaller, the sound became more "nervous". In my system I preferred 5m.
 
  • Like
Reactions: Thuan and asindc

StreamFidelity

Well-Known Member
Jun 30, 2020
110
121
115
Rostock, Germany
griggaudio.de
Would it be worthwhile to run an FMC—>1m (or however how long) FO cable from the router into a pair of switches (or just one switch) just to convert it back to ethernet for a 20ft+ run to the listening room?

The fiber optic cable should usually be at least 2m long. But you can't say that in general.

If the transmitter couples light into the fiber at -20 dBm and the weakest signal that the receiver can still recognize must have a power of -30 dBm, the difference between the two values is 10 dB. This means that the system must have an attenuation of at most 10 dB before the signal becomes too weak and the receiver can no longer process it.

But what if the receiver were coupled to a transmitter that has a power of -5 dBm? The signal would be too strong and the receiver would be blinded. With optical systems, care must be taken not to overload the receivers, as this would be just as harmful as a signal that is too weak.

I bought a 2m and 20m fiber optic cable. For me the 20m fit better. You have to try it.
 
  • Like
Reactions: asindc

djsina2

Well-Known Member
May 30, 2019
1,125
964
213
You are not going to have any attenuation problems with short distances inside a home using LX optics and single mode fiber. As you say, likely you will be running too hot or at least close to the saturation point on the receive. I would be using multimode SX optics which are designed for shorter range in building networks. Or, I would be putting a 5dB or 7dB attenuators on the single mode setup.
 
  • Like
Reactions: asindc

kennyb123

Well-Known Member
Nov 30, 2012
856
796
1,155
Kirkland, WA
No. Once you introduce fiber you've broken the noise path.

As others have pointed out, there can be more to it than that. Attenuation can be beneficial. Coiling a longer run has been shown to increase attenuation. This could be an alternative to using attenuators.

My ears have told me that it’s best to avoid attenuators. Even better to not use an SFP that requires attenuation to avoid saturation.
 
  • Like
Reactions: Thuan

DonH50

Member Sponsor & WBF Technical Expert
Jun 22, 2010
3,947
306
1,670
Monument, CO
You add attenuators to a fiber-optic cable?

In any event, what I described was noise coupling through the power/ground lines, which any length of fiber should break since there is no longer a galvanic connection between source and load, only optical.

I am not sure what sort of fiber we are talking about here. My current experience is with high-speed data links, and the transceivers work with from 1 m to >100 m of fiber with almost the same performance from roughly 3 to 22.5 Gb/s or more per lane. My previous VCSEL experience to ~25 Gb/s may or may not be relevant, I don't know, but I was designing the ICs and measuring performance using a DSO and JBERT since the application was not audio. I don't have any recent experience with long-haul fiber (multi-km runs) and little with Ethernet. Given the way the transceivers and CDR (clock and data recovery) circuits work, followed by buffering and reclocking to drive a DAC, I do not see how fiber length affects the sound, but I've no experience with that as I said above.
 

djsina2

Well-Known Member
May 30, 2019
1,125
964
213
I think most audio people are using standard 1Gbps 10km LX optics. The receive window is likely something like -22dB to -3dB. With it being good for 10km your receive power will be closer to the top threshold. That’s why putting in 5dB or 7dB attenuators will land you right in the sweet spot.

If you are using a Ubiquiti Edgerouter you can check the receive power on it. It will report in mw which you need to convert to dB. If your device doesn’t report on receive power you can buy an inexpensive power meter from FS.com.
 

asindc

Well-Known Member
Sep 27, 2012
187
20
923
You add attenuators to a fiber-optic cable?

In any event, what I described was noise coupling through the power/ground lines, which any length of fiber should break since there is no longer a galvanic connection between source and load, only optical.

I am not sure what sort of fiber we are talking about here. My current experience is with high-speed data links, and the transceivers work with from 1 m to >100 m of fiber with almost the same performance from roughly 3 to 22.5 Gb/s or more per lane. My previous VCSEL experience to ~25 Gb/s may or may not be relevant, I don't know, but I was designing the ICs and measuring performance using a DSO and JBERT since the application was not audio. I don't have any recent experience with long-haul fiber (multi-km runs) and little with Ethernet. Given the way the transceivers and CDR (clock and data recovery) circuits work, followed by buffering and reclocking to drive a DAC, I do not see how fiber length affects the sound, but I've no experience with that as I said above.

I think most audio people are using standard 1Gbps 10km LX optics. The receive window is likely something like -22dB to -3dB. With it being good for 10km your receive power will be closer to the top threshold. That’s why putting in 5dB or 7dB attenuators will land you right in the sweet spot.

If you are using a Ubiquiti Edgerouter you can check the receive power on it. It will report in mw which you need to convert to dB. If your device doesn’t report on receive power you can buy an inexpensive power meter from FS.com.

What about this SFP?: https://planetechusa.com/product/mgb-tlx-mini-gbic-lx-module-20km/

It seems to be popular among members here.
 

djsina2

Well-Known Member
May 30, 2019
1,125
964
213
  • Like
Reactions: asindc

djsina2

Well-Known Member
May 30, 2019
1,125
964
213
As others have pointed out, there can be more to it than that. Attenuation can be beneficial. Coiling a longer run has been shown to increase attenuation. This could be an alternative to using attenuators.

My ears have told me that it’s best to avoid attenuators. Even better to not use an SFP that requires attenuation to avoid saturation.

Attenuation at 1310nm is .5dB or less per km. So unless you’re coiling up kilometers of cable it’s essentially nothing.
 

About us

  • What’s Best Forum is THE forum for high end audio, product reviews, advice and sharing experiences on the best of everything else. This is THE place where audiophiles and audio companies discuss vintage, contemporary and new audio products, music servers, music streamers, computer audio, digital-to-analog converters, turntables, phono stages, cartridges, reel-to-reel tape machines, speakers, headphones and tube and solid-state amplification. Founded in 2010 What’s Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals, we enjoy learning about new things and meeting new people, and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing