Re: [Jack-Devel] Ethernet-based audio interface using (net)jack idea

PrevNext  Index
DateThu, 18 Aug 2011 14:51:15 +0100
From John Rigg <[hidden] at jrigg dot co dot uk>
Tojack-devel <[hidden] at lists dot jackaudio dot org>
In-Reply-ToDan Swain Re: [Jack-Devel] Ethernet-based audio interface using (net)jack idea
On Wed, Aug 17, 2011 at 08:15:05PM -0600, Dan Swain wrote:
> Whilst doing a bit of Googling, I came across this manufacturer:
> 
> http://www.vectron.com/products/vcxo/vcxo_index.htm
> 
> Would some of their ICs be of use for this project?

It depends on whether the required frequencies are available.
Audio ADC and DAC ICs typically require 128x or 256x fs for MCLK.
It would be worth reading data sheets for some of the widely
used converter chips to get an idea of what is required. The
AK5394A (ADC) and AK4395 (DAC) from Asahi Kasei EMD (www.akemd.com)
would be good examples as they are so widely used in pro audio 
equipment.

The VCXOs from Tent Labs are very good but expensive:
http://www.tentlabs.com/Components/VCXO

They are also available in a complete PLL PCB which is useful
for R&D purposes:
http://www.tentlabs.com/Components/DACupgrades/xo23dac

I've used some of their XO2 modules in converters I've modified, and
am planning to use a couple of their VCXOs in a mastering grade converter
I'm designing for my own use.

> I can't find the price
> of a DICE chip, but it may be possible to find a more specific solution, so
> maybe we wouldn't need to compromise with a more expensive option.
> 
> I'll also take a look at the circuit boards of the video capture cards we
> have at work, and see what they have. 

Be aware that video capture has different requirements which might
not be appropriate for high end audio.

>It may also be worth us studying some
> of the other audio manufacturers' implementations, to see if there's a
> particular trend, and why.

IIRC, DICE is used in the Presonus Digimax FS and Focusrite Octapre II.
Both of those are ADAT interfaces, but until recently equipment at that
price level would not have had such sophisticated PLLs for clock
recovery. 

My SSL Alpha Link uses AK4117VF chips for external clock sync
and my old Lavry AD-10 used an AK4115VF (I'm not sure if the Lavry
had additional jitter attenuation stages after that, as it sounded
very good whatever the clock source). Both of these use a pair of 
fixed frequency XOs for internal clock.

John
PrevNext  Index

1313675279.11495_0.ltw:2,a <20110818135115.GA2804 at localhost0 dot localdomain>