Optical cable

[DonDish]

Typical end reflection of a straight (glass fibre) surface to air is around 0.3dB (4-7%) as there are 2 you end up with a 0.6-1dB loss (or thereabouts).
It is even less when broken at an angle.
When the broken fibers are spaced apart the losses will be higher of course and also depends on the Numerical Aperture.

POF = 1dB attenuation per meter (for visible red) for your typical 1mm POF and is by far the biggest contributor in attenuation.

Nice! ,but end reflection will be inevitable anyway. Add to that several differently angled and spaced cleveages. Deformities in plastic. Maybe a big guy put the leg of a chair and full body weight on top, or your dog chewed on it, or booth. Cheap materials and whatnot. These things can surely degrade and break.

Pof = plastic optical fiber?

So if 0.3db attenuation is 4 - 7%, 1 db/m is how much in lost light?

 

[DonDish]

@solderdude and @MaxwellsEq I see im surrounded by paid professionals and born geniuses again. My example with the surround prosessor is 20y ago, but it was like that. I never really had any toslink issues than that except with those small extension and 3.5mm combi laptop minijack adapters, those buggers sure came with on or off state. I might add the cable wasnt stiff like usual tos cable, action was more like loose elecrical wire. Just a discount product from a household store, next to scented candles.

@MaxwellsEq "it will either work flawlessly or you get dropouts / splats." I would compare the effect, in my case with playing music from a locked but not synced source with ADI2pro. I humbly disagree.

 

[solderdude]

Pof = plastic optical fiber?

Yep.

And just like glass fibre they hate abuse but glass fiber (with Kevlar sleeving) can handle a bit more abuse other than being bent in small angles.

1 db/m is how much in lost light?

20% per meter.
One should realize that the receivers usually have a certain range (with internal AGC) that have a dynamic range (power budget) of 10-15dB (depending on the receiver, synchronous or asynchronous transmission, light source intensitiy and BW)

 

[DonDish]

Yep.

And just like glass fibre they hate abuse but glass fiber (with Kevlar sleeving) can handle a bit more abuse other than being bent in small angles.


20% per meter.
One should realize that the receivers usually have a certain range (with internal AGC) that have a dynamic range (power budget) of 10-15dB (depending on the receiver, synchronous or asynchronous transmission, light source intensitiy and BW)

So 20%/m. Thats not a linear falloff i belive. Not much left at the end of 15m then. I must say I got lost between abbrevations and power budget here. Its OK, im not gonna tortue your fingertips. I can look this up.

 

[solderdude]

POF is not intended for longer distances.
You can get much lower damping using a bunch of glass fibers instead of POF and bridge longer distances if needed.
Not all POF is created equal either.
Some receivers still work O.K. at lower levels and some sources (LEDs) give a bit more light so the max. distance (at highest speeds) may vary.

 

[DonDish]

POF is not intended for longer distances.
You can get much lower damping using a bunch of glass fibers instead of POF and bridge longer distances if needed.
Not all POF is created equal either.
Some receivers still work O.K. at lower levels and some sources (LEDs) give a bit more light so the max. distance (at highest speeds) may vary.

That makes sense, but isnt every sender based on a switching led bulb at the right frequenzy. For Toslink

 

[solderdude]

For Toslink yes, 650nm (visible red)

For glass fiber it can be LED (820-850 - 1300) or laser (1260-1360nm window and 1490-1625nm in several channels)

 

[mns3dhm]

If you've set your WiiM to 192kHz fixed sample rate and it's playing audio fine, then the WiiM, Optical cable, and Optical receiver all support native 192kHz.

There are indeed differences in quality between Optical cables, which become more important with increasing cable length.

If you get smooth audio playback without stutter, then there's no reason to upgrade the Optical cable.

Let me ask a question here if you will. I did experience stutter when I connected a Panasonic BDP9000 Blu-Ray player to a Parasound Hint6 integrated with a coaxial digital cable sourced from Blue Jeans Cable. Do you know what might have caused this problem? I was able to solve the immediate problem by using the Panasonic's internal DAC and connecting it to the Parasound via RCA cables (also BJC), but I never could make the connection work via the coax digital out.

 

[staticV3]

Do you know what might have caused this problem?

Poor wire joints, incorrect characteristics impedance, or too high jitter from the Panasonic for the Parasound's PLL to handle.

 

[MaxwellsEq]

"it will either work flawlessly or you get dropouts / splats." I would compare the effect, in my case with playing music from a locked but not synced source with ADI2pro. I humbly disagree.

The clock is part of the signal. If the bits are randomly messed up enough, the receiving DAC won't be able to extract the clock. What happens next depends on a whole bunch of stuff. In theory, a very simple DAC won't know what to do. If the "content data" is OK, but the clock is messed up, sample(s) will get discarded. Most DACs have their own clocks and can "flywheel" through an occasional clock error. Some DACs have Sample Rate Converters, in which case they were already playing out using their own clock. , The receiving DAC may be dumb and not work properly without the incoming clock, but many operate a Phase-Locked Loop which allows them to cope with some clock drift and clock jitter.

The reality is that what happens at the receiver depends on the quality of the parts and the quality of the implementation. Some devices simply won't output anything. Others may make a very good job of covering over the mess. Others may discard samples, have brief silences, generate splats etc.

At what is known as the PHY layer, BERT (Bit Error Rate Testers) let you know what's really happening. An extremely low rate of bits lost or corrupted can be tolerated, but watched to see if degradation is increasing. As the rate of BER increases, the higher layer protocols start having to work harder to correct "packets", but eventually they can't fix things so higher layer protocols have to either start discarding or requesting resends. S/PDIF has some tolerance but not much.

 

[DonDish]

The clock is part of the signal. If the bits are randomly messed up enough, the receiving DAC won't be able to extract the clock. What happens next depends on a whole bunch of stuff. In theory, a very simple DAC won't know what to do. If the "content data" is OK, but the clock is messed up, sample(s) will get discarded. Most DACs have their own clocks and can "flywheel" through an occasional clock error. Some DACs have Sample Rate Converters, in which case they were already playing out using their own clock. , The receiving DAC may be dumb and not work properly without the incoming clock, but many operate a Phase-Locked Loop which allows them to cope with some clock drift and clock jitter.

The reality is that what happens at the receiver depends on the quality of the parts and the quality of the implementation. Some devices simply won't output anything. Others may make a very good job of covering over the mess. Others may discard samples, have brief silences, generate splats etc.

At what is known as the PHY layer, BERT (Bit Error Rate Testers) let you know what's really happening. An extremely low rate of bits lost or corrupted can be tolerated, but watched to see if degradation is increasing. As the rate of BER increases, the higher layer protocols start having to work harder to correct "packets", but eventually they can't fix things so higher layer protocols have to either start discarding or requesting resends. S/PDIF has some tolerance but not much.

So your saying with my old surround system, any scenario could play out, but its unlikely because of narrow tolerance, and no resend capability of spdif. It mostly either work or not. I can buy that. Changing the ****** tos cablle just made difference, I probably changed it back just to check. Evidence is lost any way, electronics graveyard for that. Its not often I find a ghost in the machine. I had a experience with a TV locking at wrong samplerate, dont ask how, but I was hooking up to other gear. That metallic sound still hurt my teeth. Another story. Thanks for enlightenment.

 

[somebodyelse]

Would using coax be a better connection choice ?

Maybe, maybe not. Toslink definitely has electrical isolation, but might not manage 24/192. Coax might be transformer isolated (should be in pro gear at the transmit end, might be in some domestic gear too) but will do 24/192 so long as both ends are capable. I'd say isolation is more important than 24/192.