Jump to content

Recommended Posts

Posted
think about it. if the AES/EBU connection sounds vastly different to the other digital connections, something is badly broken.

I disagree. One could have a properly terminated AES/EBU connection and horrible coaxial/toslink implementation. For argument's sake, the coaxial connector could have an incorrectly populated voltage divider resistor on the output. Would this be noticeable? yes.

Posted

Try to pronounce it.... ;D

In all seriousness the connector used does matter in the same way a RCA plug isn't an ideal choice for the 75ohm of the S/PDIF coax. The BNC is actually 75ohm but for some strange reason many of the "audiophile" connectors are the 50ohm variety. :palm:

Posted

50 ohm BNC are used on scopes, etc. and seem to be more prevalent. Identified by more material around the center on chassis jacks. 75 ohm have next to nothing around the center. The XLRs used for AES/EBU are actually ~110 ohm (which is the standard), but are horrible for RF connections (i.e., S/PDIF or AES/EBU).

Posted
Try to pronounce it.... ;D

In all seriousness the connector used does matter in the same way a RCA plug isn't an ideal choice for the 75ohm of the S/PDIF coax. The BNC is actually 75ohm but for some strange reason many of the "audiophile" connectors are the 50ohm variety. :palm:

XLR is 110ohm, which meets the spec for AES/EBU, it is that the connector has issues with RF signals, which is why they had to crank the voltage so high, which causes other issues...

Bah, beaten by Chris.

Posted

This keeps coming up on the slim devices forum.

Anyway I see two main reasons for the inferiority of AES/EBU.

1. XLRs are horrible RF connectors. In order to send a square wave fairly faithfully the interface must support a bandwidth many times higher than the frequency of the square wave. For the signals in question that is getting well up into the RF spectrum where the XLRs are terrible. The impedance varies radically with frequency which will cause all kinds of bizarre reflections. The choice of XLR was a very poor choice.

2. Output voltage. The S/PDIF electrical spec is 0.5V into 75ohms, but the AES/EBU is 3-5V into 110 ohms. Think about that for a second, what happens when you put 5volts across 110 ohms? You get almost 50mA of current flowing. This means the driver sitting in the source box has to be able to dump between 30-50ma into the cable. That causes huge current spikes in the power and ground pins of the driver chip which is going to cause big noise spikes in the power and ground planes of the board. If you are not extremely careful that is going to cause significant jitter in the output signal.

All modern high speed interfaces use less than 0.5V signal.

As far as I can tell the XLRs were chosen because studios had lots of microphone cables and wanted to use them. Because they are such lousy RF transmission lines they had to go with high voltages to make sure there was some signal left at the end.

Maybe Tyll can tell us whats in this paper?

AES E-Library: Is the AES/EBU/SPDIF Digital Audio Interface Flawed?

Posted
in systems that aren't broken/really badly engineered, are going to give any kind of meaningful difference.

fair enough. my point is that most are badly engineered, hence the prevalent statement that one is better than another*. By badly engineered I mean not designed by a RF engineer.

* I don't believe this 100%, who am I kidding. People hear what they want to hear.

Posted
eggi.... eayja.... iegau...

volcano.

everything is a compromise. i'm highly doubtful that any of these compromises, in systems that aren't broken/really badly engineered, are going to give any kind of meaningful difference. meaningful being audible to human beings on any kind of consistent or predictable basis.

Nope, how about "Island mountains glacier" since that's what it means in a direct translation? :)

There are always compromises and to be honest, how many commercial pieces out there are well engineered from start to finish? There is always something rotten, somewhere... :(

This keeps coming up on the slim devices forum.

Anyway I see two main reasons for the inferiority of AES/EBU.

1. XLRs are horrible RF connectors. In order to send a square wave fairly faithfully the interface must support a bandwidth many times higher than the frequency of the square wave. For the signals in question that is getting well up into the RF spectrum where the XLRs are terrible. The impedance varies radically with frequency which will cause all kinds of bizarre reflections. The choice of XLR was a very poor choice.

2. Output voltage. The S/PDIF electrical spec is 0.5V into 75ohms, but the AES/EBU is 3-5V into 110 ohms. Think about that for a second, what happens when you put 5volts across 110 ohms? You get almost 50mA of current flowing. This means the driver sitting in the source box has to be able to dump between 30-50ma into the cable. That causes huge current spikes in the power and ground pins of the driver chip which is going to cause big noise spikes in the power and ground planes of the board. If you are not extremely careful that is going to cause significant jitter in the output signal.

All modern high speed interfaces use less than 0.5V signal.

As far as I can tell the XLRs were chosen because studios had lots of microphone cables and wanted to use them. Because they are such lousy RF transmission lines they had to go with high voltages to make sure there was some signal left at the end.

Maybe Tyll can tell us whats in this paper?

AES E-Library: Is the AES/EBU/SPDIF Digital Audio Interface Flawed?

That's exactly what my mind was struggling to remember.

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.