The Black Hole......

Moderator
Joined 2012
Staff member
Paid Member
This new thread is intended to supplant the recently closed Blowtorch thread which the Moderation Team had felt had run its natural course.

This new thread will be moderated very differently and will be subject to the normal diyAudio rules:

RULES

with which we would ask all those who intend to post here familiarise themselves.

In addition to the above rules this thread will also be proactively moderated meaning that posts and consequent replies that we feel are going to veer off course or that may lead to personal insults will be deleted.

If the message doesn't get through then those continuing with such postings will find penalties are swiftly handed out. For some of you even a 1 point infraction will automatically place you in read only mode. All those effected by that have already been informed.

What we would like to see here is thought provoking and interesting discussion of anything technical that is related to audio (and other spheres) but we do not want rehashing of material that has been flogged to death in other areas of the forum including the now closed BT thread.

Enjoy, have fun and keep to the rules...
 
AES67

Can I kick off by trying to continue the discussion on AES67 and whether we should be trying to use this for domestic audio.

I hope there be more tech talk in this here thread than the last waste of time...may I also make a small suggestion? I stopped reading the previous thread because it was impossible to quickly tell what a post was addressing unless you read it. I run ham radio forums, and they have headers with the thread title in them allowing you to follow a specific discussion without having to read through others you are uninterested in. Here we can use the title bar to do a similar thing as I just did with AES67

And addressing the AES67 thread: It certainly has become the standard in broadcast studios. Many new studios have just a few short patch cables of analog audio or a few dongles of XLR to multipins, and the rest of the facility is wired with miles of CAT6 variants.

IMHO, the big advantage for a home audio system would be the fact that many AoIP units use transformers on the Ethernet inputs, effectively reducing audio-band common-mode signals. Other than this, the complication of using browsers to configure devices on the AoIP network makes (IMHO) configuring what should be a hardwired audio path into a exercise in software design and fun. In the broadcast industry we are used to that, but I am not sure how Harry Homeowner will view it.

Cheers,
Howie
 
I hope there be more tech talk in this here thread than the last waste of time...may I also make a small suggestion? I stopped reading the previous thread because it was impossible to quickly tell what a post was addressing unless you read it.........
Good idea, OT ;), I'm reminded of Jeanette Winterson quoting her mother from whom she had to hide her books under her mattress : "The trouble with a book is you never know what's in it until it's too late"
 
  • Like
Reactions: 1 user
Can I kick off by trying to continue the discussion on AES67 and whether we should be trying to use this for domestic audio.

This AES67 thing seems a highly complicated matter building on several existing protocols.
What aspect has your attention, the use of it as a consumer, or going into more technical detail ultimately leading to some DIY product ?

Hans
 
Disabled Account
Joined 2012
PCB layout---

finishing comments on that. I may be old but fundamentals dont change. I suggested EAGLE as an intro for a DIYer to do his her own PCB and there are others, of course. What someone laying out a LSIC for microwave freqs would use is vastly more complex and expensive. Never-the-less, 90 degree trace corners are not used in RF nor microwave frequencies for layouts. A 45 degree would be used at a corner.

Here you will see many such patterns of layouts on IC and off on board;

high speed digital circuits and pcb layout - Google Search

As for timing issues due to reflections of poor pcb layout, it is written about extensively. It isnt the freq but the edge which needs to remain unmolested. 90 degree pcb corners would be a point of reflection/mismatch.

Resolving the Signal Part 10: How Clock Signals Affect Precision ADCs
November 06, 2018 by Bryan Lizon, Texas Instruments

Part 10 of the Resolving the Signal series covers how clocks affect precision ADCs, touching on clock jitter, clock intermodulation and best PCB layout practices for clocking.


Many books on the subject and for the ADC/DAC clocks et al, it would be best to use some of these best pcb practices.

Mark4..... do your pcb or DAC boards from Mfr use best practices? Take a look at your clock lines in particular.
Again, I say look to TI and others, not here on this forum, for best pcb practices. Avoid some really bad advice.
You may be able to improve on the pcb layout and get better sound.


THx-RNMarsh
 
Last edited:
Can I kick off by trying to continue the discussion on AES67 and whether we should be trying to use this for domestic audio.

It is a bit of overkill for a home system, but it should work quite well. It was developed by an AES committee I am not participating in.

In the next week or so I will be sharing my experience in building a version of the BT preamplifier. It does not have a phono stage as I already built a simpler JC version.

The case for the audio bits is made from solid bits of aluminum "Hogged" out with a milling machine. The power supply is not in a matching case. I will probably make more than one, but instead of starting with solid aluminum I will use C channel.

I am using a relay based attenuator. I do have a selection of potentiometers to compare.

In the same vein I will try a different buffer than the BT clone. BTY the DC servos are off the main preamp card so I can play with them.
 
Last edited:
Member
Joined 2004
Paid Member
Quad 63 impedance measurements

Here are several superimposed impedance plots.

The taller one is at a 10 db higher drive level. the lower drive level plots are with bias on and bias off. I could only turn up the AC to about 40% before the discharge internally became excessive. As such the clamp circuit was not powered and shutting everything down at any higher drive level.

I think the drive levels are around 10V and 3V. The annoying ticking from the bad panel persuaded me to stop pretty quickly.

There really is no difference bias on and bias off. There is some difference related to drive level in the LF resonance.

This speaker is too sick to do much more testing until the dead panel is replaced.
 

Attachments

  • Quad impedance test.PNG
    Quad impedance test.PNG
    83.4 KB · Views: 1,571
Member
Joined 2004
Paid Member
My understanding about right angles on PC layout was more an issue of etching the copper than RF. Sharp angles with narrow traces and have etch problems either not removing enough copper or etching through a narrow trace. However this was from about 15 generations of PCB technology ago. Laying out for high speed differential signals (LVDS like HDMI) does require some special care with the traces needing identical length and impedance.
 
Member
Joined 2014
Paid Member
AES67

This AES67 thing seems a highly complicated matter building on several existing protocols.
What aspect has your attention, the use of it as a consumer, or going into more technical detail ultimately leading to some DIY product ?

Hans


It may seem complicated, but given my (digital) music starts on a server that has ethernet ports it seems logical to have the next stage connected the same way. Other than all in one streamers there doesn't seem to have been much work done in the DIY space on it. Or I've been looking in the wrong place.



MiniDSP appear to have an 8 channel AVB box on the cards but other than that slim pickings.
 
Disabled Account
Joined 2012
My understanding about right angles on PC layout was more an issue of etching the copper than RF. Sharp angles with narrow traces and have etch problems either not removing enough copper or etching through a narrow trace. However this was from about 15 generations of PCB technology ago. Laying out for high speed differential signals (LVDS like HDMI) does require some special care with the traces needing identical length and impedance.

No, the 90 degrees will cause problems. See examples/sources and books on it. What you are referring to is true also but a separate problem to be dealt with.


THx-RNMarsh
 
Last edited:
It may seem complicated, but given my (digital) music starts on a server that has ethernet ports it seems logical to have the next stage connected the same way. Other than all in one streamers there doesn't seem to have been much work done in the DIY space on it. Or I've been looking in the wrong place.



MiniDSP appear to have an 8 channel AVB box on the cards but other than that slim pickings.

Is streaming from network (if connected by Ethernet) AES67 or is it something else? If it is would networking your server with a streamer/dac do it?