[johnjen]

AES/EBU uses three conductors irrespective of shielding:

Pin 1: GND
Pin 2: Signal + (hot)
Pin 3: Signal - (cold)

Anyone that builds an AES/EBU cable utilizing the shield as the third conductor is taking a shortcut.

So I went looking for 3 wire 110Ω rated cable and couldn't find any.
Anywhere, which leads me to view this as an exercise in theory rather than practice.
And as such is un-testable and so of no real impact in the real world.
So I'll look into the AES spec and verify this 3 wire w/shield thing, just to find out for myself.

But it does beg the question that if the AES spec does call for 3 wires w/shield instead of 2 w/shield, why isn't it available?
But maybe I missed a manufacturer who does make a 3 wire 110Ω cable, but they are mighty hard to find if that's the case.

Makes me go hmmmmmmm…

JJ

 

[johnjen]

So I checked my copy of the AES/EBU Engineering Guidelines 1995 edition and all I could find were these 2 bullet points.

"- Multicore twisted pair cables with a overall screen are best for installations where runs do not exceed 150 m. (Overall double screens give better EMC protection.)"

"- Multicore cables with individually screened pairs have higher capacity to the screens and hence greater loss but are satisfactory for shorter runs. Cables intended for data have some advantages even for short runs, and they also work quite well with analogue signals. This approach could therefore be considered if new cabling is needed."

(Added Emphasis Mine)
So they only talk about twisted pairs, no 3 wire configurations were mentioned.
Perhaps there is another AES/EBU technical manual that goes into more detail.
If so does anyone know of it, and can I get a copy?

JJ

 

[winders]

There are also companies that make XLR interconnects that only use two conductors. Again, a shortcut.

I will have my AES/EBU cable customer made using OCC copper and 3 conductors.

However, look here:

http://www.mogamicable.com/category/bulk/dig_interface/aes_ebu/

It appears the Mogami AES/EBU cables all have a third conductor specifically for the drain wire (GND).

And here:

http://www.beldencables-emea.com/en.../audio_cables/audio/audio-digital/index.phtml

Belden's AES/EBU cables all have a third conductor specifically for the drain wire.

Do a search on:

AES/EBU and drain wire

See what you find.

 

[earnmyturns]

So they only talk about twisted pairs, no 3 wire configurations were mentioned.
Perhaps there is another AES/EBU technical manual that goes into more detail.
If so does anyone know of it, and can I get a copy?

Look at the circuit diagram in section 6.1 here. Shielded twisted pair, not shielded 3-wire.

 

[winders]

Look at the circuit diagram in section 6.1 here. Shielded twisted pair, not shielded 3-wire.

Look at the diagram again. Do you see the third wire? I see twisted pair with a drain wire....

 

[earnmyturns]

Look at the diagram again. Do you see the third wire? I see twisted pair with a drain wire....

That's not what I see. I see just a a schematic representation of the conducting shield. That's at least how such were represented back in my teens when I used to read circuit diagrams to build gear. If that wasn't the case, it would be very surprising that all commercial AES/EBU cable is twisted pair + shield.

 

[winders]

That's not what I see. I see just a a schematic representation of the conducting shield. That's at least how such were represented back in my teens when I used to read circuit diagrams to build gear. If that wasn't the case, it would be very surprising that all commercial AES/EBU cable is twisted pair + shield.

Did you see the links I posted with the AES/EBU cables from Mogami and Belden?

 

[Defiant00]

Hello!

(I'm new to the forum, but a long-time reader. I posted a message on the introductions thread. Firstly: thank you for your insights! I have been following these forums quite closely.)

So: I am an aspiring musician, I play the piano, and have started to experiment with virtual instruments. I planned to have a sound system with which I can listen and study music, that I could as well connect through midi to my keyboard, using a sampled instrument (Synthogy Ivory II). I also wanted to have a such a system that would not need to be replaced in the near future, that would produce the sound of the piano in a most realistic way possible. Based on my readings here, I sold my Apogee Duet, and acquired a set of an Yggdrasil and a Ragnarok, both in black.

My experimenting is half way there, I have a good feeling about this.

However, I noticed that there might be some latency that the Yggdrasil produces, that seemed to be more that I was getting from a Modi Multibit. I contacted Schiit to ask about this, but they said they don't test latencies with their products.

I wonder if somebody can help me with this:

1) does anybody know how much latency the Yggdrasil produces? What about the Modi Multibit?
2) Are there any differences in latency between USB, AES/EBU and SPDIF connections?
3) What about using converters, i.e. Singxer SU-1, or the possible Eitr?

Thank you in advance! I do not know if latency is the exactly favourite subject on these forums, but many of you seem to know about the functionalities of these systems.

M.Bouche

Lastly: I will gladly share my experiences and lessons I have got from my experiments with virtual instruments.

I cannot tell you specific numbers on Yggdrasil vs Mimby, but from my general experimentation and understanding having used a number of VSTs and dealt with trying to reduce latency, your output method and buffer size is by far the biggest contributor to latency (or the lack there of).

Theoretically the different input types could have different latencies; USB would probably be the worst since it is an extra step, but it is unlikely that it adds enough extra samples of latency to be noticeable.

Any converters will also add some amount of latency, although again, it's likely to not be any worse than using the USB input in general (or at least, not perceptible).

By far, the most important thing you can do to reduce latency with VSTs is to 1) use ASIO (if on Windows), and 2) set your buffer size as small as you can go without getting pops and crackles.

 

[Astral Abyss]

Look at the diagram again. Do you see the third wire? I see twisted pair with a drain wire....

This is what I found:

"Foil shielding used a thin layer of aluminum, typically attached to a carrier such as polyester to add strength and ruggedness. It provides 100% coverage of the conductors it surrounds, which is good. It is thin, which makes it harder to work with, especially when applying a connector. Usually, rather than attempting to ground the entire shield, the drain wire is used to terminate and ground the shield."

The drain wire is attached to the shield and its only purpose it to make grounding the shield at the termination point easier and less likely to fail over time.

 

[earnmyturns]

Did you see the links I posted with the AES/EBU cables from Mogami and Belden?

The confusion in this discussion is that even the drain wire cables have only three electrically distinct conductors: the insulated twisted pair, and the drain wire+shield, which are electrically coupled to each other. The Canare DA202 cable I use has a drain wire too, but I'd never think of it as doing anything except making it easier to terminate the cable.

 

[winders]

Here is an unshielded AES/EBU cable with 3 conductors:

http://www.vovox.ch/en/professional-audio/products/link/digital-cable-aesebu

 

[johnjen]

Look at the circuit diagram in section 6.1 here. Shielded twisted pair, not shielded 3-wire.

Thanks for that link!
I snagged a copy for closer examination.
It goes into more detail than my Guidelines document.

JJ

 

[johnjen]

There are also companies that make XLR interconnects that only use two conductors. Again, a shortcut.

I will have my AES/EBU cable customer made using OCC copper and 3 conductors.

However, look here:
http://www.mogamicable.com/category/bulk/dig_interface/aes_ebu/

It appears the Mogami AES/EBU cables all have a third conductor specifically for the drain wire (GND).

And here:
http://www.beldencables-emea.com/en.../audio_cables/audio/audio-digital/index.phtml

Belden's AES/EBU cables all have a third conductor specifically for the drain wire.

Do a search on:
AES/EBU and drain wire

See what you find.

I usually just include that drain wire which is electrically the same as shield since they run adjacent to each other and is not insulated, along with all of the shield wires.
But since that drain wire is wimpy and the spiral or woven shield is a much better conductor (larger gauge) I don't use it as the sole ground connection.
I figure that the drain wire is there to make it easier to terminate the ends since dealing with that many strands of tiny gauge wire is a PIA so they just cut off all the shield wires and use the much easier to deal with (and much smaller gauge) drain/wire,
IOW it's a practical (cheaper to manufacture) work around.

Unless you are dealing with cable that only has a foil shield, in which case the drain wire is the only choice.
But I generally don't use an AES cable with only a foil shield as they are not optimal (ie less expensive).
The only exception is a multichannel cable which I don't use either.

And the AES schematic shows a pair of twisted conductors inside a shield that is grounded at each end.
If you want to include/use a bare wire that is electrically the same as the shield you can certainly wire it up that way.

I prefer to use all of the shield wires as the ground connection because it's beefier from end to end.

So the AES spec is 3 conductors 2 of which are the insulated (balanced) pair and the shield with or without the drain wire as ground.

That explains why there aren't any 3 conductor 110Ω cables with shield.

JJ

 

[johnjen]

Look at the diagram again. Do you see the third wire? I see twisted pair with a drain wire....

That's not what I see. I see just a a schematic representation of the conducting shield. That's at least how such were represented back in my teens when I used to read circuit diagrams to build gear. If that wasn't the case, it would be very surprising that all commercial AES/EBU cable is twisted pair + shield.

That schematic is using the standard way of showing continuity of the shield and is not a 'separate wire'.

I can see how it could be interpreted as a separate wire, but that isn't what it represents.

JJ

 

[johnjen]

Here is an unshielded AES/EBU cable with 3 conductors:

http://www.vovox.ch/en/professional-audio/products/link/digital-cable-aesebu

If you read their product description carefully it never claims it's an AES cable.
Indeed it does not meet the AES spec as there is NO shield, so it truly isn't an AES cable.
Their cables might work in isolated circumstances but it would be susceptible to ANY stray RFI/EMF/EMI in its vicinity.

AES cables were designed to meet the needs of ProAudio where shielding is critical to avoid ALL forms of noise and signal degradation.
Which also helps explain why the AES spec calls for usable cable performance up to 150M lengths, or greater.
Without the use of a shield there would be no way to meet those specs.

And since AES cables are performance rated for MHz frequencies, I wouldn't use that cable because it would be far to easily influenced by any nearby cables or devices that emit RFI/EMF/EMI, of any sort, especially in the MHz range.

JJ