What cables do you use in your systems?

[Julf]

The context was AES3.

So why would AES3 need 100 MHz?

 

[Timcognito]

I use Kimber from the days when I was drinking the audiofool Koolaide. They sound great still so why change. I just add Monoprice or Amazon if I need more now.

 

[Rja4000]

For Ethernet signals and transceivers, yes. That doesn't mean some other encoding will necessarily work.

Well, that's basically the same thing:
you send bits through kind of a serial connection through a balanced line.

If I'm not mistaken, the AES standard is up to 192kHz.
If you imagine 32 bits, you need around 12Mbps for stereo.
You have way more.
And even if that's 768kHz, you have room left.

Actually, on a Cat5 cable, you may easily transport 8 channels at 768kHz 32 bits over 100m, since you have 4 pairs.

 

[Angsty]

Maybe.
If you want, you may still use S/FTP cable.
But any Cat5 cable is at least 100MHz certified up to 100m.
So..

The AES/EBU standard does call for shielding. It’s always possible that another cable configuration can work, but do recognize that all digital cables are not interchangeable by design.

https://tech.ebu.ch/docs/tech/tech3250.pdf

 

[Angsty]

Why? Balanced is way less susceptible to induced noise.

1800F is designed for use with balanced cables.

 

[mansr]

Well, that's basically the same thing:
you send bits through kind of a serial connection through a balanced line.

If I'm not mistaken, the AES standard is up to 192kHz.
If you imagine 32 bits, you need around 12Mbps for stereo.
You have way more.
And even if that's 768kHz, you have room left.

Actually, on a Cat5 cable, you may easily transport 8 channels at 768kHz 32 bits over 100m, since you have 4 pairs.

What works and what doesn't depends on signal level, receiver sensitivity, and other things. AES3 is robust enough that in practice it works over almost anything. S/PDIF, while using the same encoding and thus bandwidth, probably won't work (reliably) over 100 m of any cable due to the much lower signal level.

 

[Speedskater]

So why would AES3 need 100 MHz?

It seems like a reasonable number to me. Old 2 channel 44.1x16 SPDIF had content up to almost 15 MHz.

 

[Rja4000]

What works and what doesn't depends on signal level, receiver sensitivity, and other things. AES3 is robust enough that in practice it works over almost anything. S/PDIF, while using the same encoding and thus bandwidth, probably won't work (reliably) over 100 m of any cable due to the much lower signal level.

Well, the cable won't change the signal level, receiver sensitivity or other thing, will it ?
What I said is that, as a cable to transport signal according to AES/EBU standard, a Cat 5 cable will work.
And it does.

 

[Rja4000]

The AES/EBU standard does call for shielding. It’s always possible that another cable configuration can work, but do recognize that all digital cables are not interchangeable by design.

https://tech.ebu.ch/docs/tech/tech3250.pdf

Of course not.
You can't swap cables between coax and twisted pair, as an example, without adapters and there are other limitations.
But 2-7V serial signal over balanced twisted pair will, for sure, work on the over-performant Cat 5/6/7 cables.
Which are also very common, cheap to buy, easy to build and easy to check/certify.

They have limitations, but those are mechanical: as an example, I wouldn't use them on tour.

 

[deniall83]

I recently grabbed a Blue Jeans LC-1 and a World's Best Cable with Canare wire and Neutrik Rean connectors. Hoping to use the LC-1 as my phono cable which worked well for me in the past. Probably going to make my own ground wire as my phono stage has a weird ground lug.

 

[Angsty]

I recently grabbed a Blue Jeans LC-1 and a World's Best Cable with Canare wire and Neutrik Rean connectors. Hoping to use the LC-1 as my phono cable which worked well for me in the past. Probably going to make my own ground wire as my phono stage has a weird ground lug.

I also use LC-1 for my phono cable, keeping in mind the desired capacitance for my cartridge and phonostage. My current cable config with low capacitance works much better with my MM cartridge than the high capacitance cable I used previously. Since my phono does not have adjustable capacitance, I used a length of cable to provide "optimum" cartridge matching per the Hagerman Labs calculator. I don't swap MM cartridges often, so I was okay with this tradeoff.

Hagerman Technology LLC: Cartridge Loading

 

[mansr]

Well, the cable won't change the signal level, receiver sensitivity or other thing, will it ?

All cables attenuate the signal to some extent. If the attenuation is too strong, the signal strength at the far end will drop below the sensitivity of the receiver.

What I said is that, as a cable to transport signal according to AES/EBU standard, a Cat 5 cable will work.

Yes (most likely), but not for the reason you said it would.

 

[Julf]

All cables attenuate the signal to some extent. If the attenuation is too strong, the signal strength at the far end will drop below the sensitivity of the receiver.

Sure, to some extent, but at reasonable distances it is not really an issue.

 

[Angsty]

Sure, to some extent, but at reasonable distances it is not really an issue.

It’s the “reasonable” part that can be at question. For long runs, balanced cables would be preferable to unbalanced RCAs if for no other reason than higher nominal signal voltage.

 

[Rja4000]

If the attenuation is too strong, the signal strength at the far end will drop below the sensitivity of the receiver.

Insertion loss is part of the Cat5/6/7 certification criteria.

 

[Julf]

It’s the “reasonable” part that can be at question.

If you are wiring stadiums, fibre is the way to go.

For long runs, balanced cables would be preferable to unbalanced RCAs if for no other reason than higher nominal signal voltage.

Balanced would definitely be preferable, but more for the reasons of common mode noise rejection and avoidance of ground loops than signal voltage. You could easily boost the voltage of an unbalanced connection too.

 

[Angsty]

Balanced would definitely be preferable, but more for the reasons of common mode noise rejection and avoidance of ground loops than signal voltage. You could easily boost the voltage of an unbalanced connection too.

Agree that all of those matter, too. The "ease" of boosting an unbalanced connection output also comes with greater complications in many setups.

 

[mansr]

Insertion loss is part of the Cat5/6/7 certification criteria.

Those cables are also specified with 100 Ω characteristic impedance, whereas AES3 wants 110 Ω. That mismatch will cause higher losses than when the same cable is used for Ethernet.

 

[Julf]

Those cables are also specified with 100 Ω characteristic impedance, whereas AES3 wants 110 Ω. That mismatch will cause higher losses than when the same cable is used for Ethernet.

An impedance mismatch causes reflections rather than losses. Of course some of the power reflected results in less power transmitted, but that 10% difference less than 1 dB of loss...

 

[uwotm8]

PC -> DAC: Generic USB 3.x
DAC -> Sub (line in): Profigold PGA3000
Sub (line out) -> Dynaudio MC15 (active): AudioQuest Chicago (factory terminated; amazed how good it is and considering upgrading to bulk-DIY Mackenzie)
Also considering upgrading speaker power cables from noname PC-grade garbage to THICC 3x2.5 "just normal AC cable" with Chinese fake Oyaide connectors (all that for roughly $20 per one cable). Just because WHY NOT