• Disclaimer: This Vendor's Forum is a paid-for commercial area. Unlike the rest of diyAudio, the Vendor has complete control of what may or may not be posted in this forum. If you wish to discuss technical matters outside the bounds of what is permitted by the Vendor, please use the non-commercial areas of diyAudio to do so.

exaU2I - Multi-Channel Asynchronous USB to I2S Interface

I hope USB High-speed is isolated in the new design by embedding such a technology.
Core Modules, Daughter Boards, Core Development Kits for USB Extension - Icron Technologies
Why? This would seem to add unnecessary expense to the product.

Perhaps I am biased, since I made a feature request that the new hardware be based on a single PCB without a daughter board. Single-board reduces the number of connections, which are a point of failure. Single-board also alleviates the need to buy a retail product as a manufacturing component of another retail product - it's much cheaper to just buy the individual electronics and place them directly on the main board compared to buying a sub-assembly that is separately manufactured.

I also have my doubts that the iCron board can be programmed with custom firmware as easily as a proprietary design from exaDevices.
 

IMO, products of exaDevices have marked breakthroughs in their performances; 386kHz/32bit 8 channels, DSD512 play, etc. That's the reason why people are interested in them. Isolation of High speed USB in an audio application domain is not achieved yet. This is a challenging thing to do.

From the point of view of one of audiophiles who always pursue the best sound quality, I hate to have my audio master clock oscillators share a ground ( a source of electron flow ) with a noise contaminated PC.

However, from the point of view of audio device selling business, your thought is reasonable, of course.

Bunpei
 
Isolation of High speed USB in an audio application domain is not achieved yet. This is a challenging thing to do.

From the point of view of one of audiophiles who always pursue the best sound quality, I hate to have my audio master clock oscillators share a ground ( a source of electron flow ) with a noise contaminated PC.

Bunpei

@Bunpei
I am playing (have been doing that for a good period of time already) with isolation of USB (480MB/s) with both exaU2I and other USB implementations.
This is the greatest upgrade you can implement with PC / Mac USB based systems.
I have isolated USB connections with copper wire, silver wire and pure gold wire.
Isolated USB with pure gold wire insulated with pure silk are by far the best sounding.
These observations are based on my setup where the internal cables in the DAC are pure gold with air as isolation, the interconnects to the power amplifier are pure gold with pure silk isolation, the PCB´s are gold plated in the whole system, the speaker cables are a combination of pure silver and pure gold - isolated with pure silk.
Power amplifiers are pure JFET, speakers are pure AMT fullrange.

And my Mac OSX player system have no fans, SSD HDDs, custom PSU with JFET regulation and more capacitance
than most power amplifiers, and runs from a 12 volt battery source.
 
@Bunpei
I am playing (have been doing that for a good period of time already) with isolation of USB (480MB/s) with both exaU2I and other USB implementations.
This is the greatest upgrade you can implement with PC / Mac USB based systems.
I have isolated USB connections with copper wire, silver wire and pure gold wire.
Isolated USB with pure gold wire insulated with pure silk are by far th best sounding.
These observations are based on my setup where the internal cables in the DAC are pure gold with air as isolation, the interconnects to the power amplifier are pure gold with pure silk isolation, the PCB´s are gold plated in the whole system, the speaker cables are a combination of pure silver and pure gold - isolated with pure silk.
Power amplifiers are pure JFET, speakers are pure AMT fullrange.

And my Mac OSX player system have no fans, SSD HDDs, custom PSU with JFET regulation and more capacitance
than most power amplifiers, and runs from a 12 volt battery source.

Hi RayCtech,

So I guess you're kinda "purist" person... :D

As much as I respect your approach and would love a setup like yours, most of us cannot afford the type of cables you're using... Pure gold with silk, how much per foot? :eek:

Ciao!
Do
 
Hi RayCtech,

So I guess you're kinda "purist" person... :D

As much as I respect your approach and would love a setup like yours, most of us cannot afford the type of cables you're using... Pure gold with silk, how much per foot? :eek:

Ciao!
Do

I hope exa065 will tolerate this OT postings :D

The largest contribution to cost is the odd 40 hours of work pr. meter of cable (silk insulations) as they are 100% very perfectly made by hand...
 
I hope exa065 will tolerate this OT postings :D

The largest contribution to cost is the odd 40 hours of work pr. meter of cable (silk insulations) as they are 100% very perfectly made by hand...

Hi guys,

As you know I wear two hats - I am a DIYer by destiny and I produce products for profit. It is fine to post here for the benefit of both causes. Just manage your conflicts of interest as I manage mine.

I enjoy tremendously XXL volume controls with ball bearings and gold and silk wires. I don't believe in supernatural cables, but handmade gold and silk cable - this is a work of art, it is like a 13th century katana with a silk handle...it is a weapon for sound supremacy.

I wonder, how come it is an honourable occupation to search for the illusive sound in all areas of R&D, except proprietary sound drivers. Double standards....
 
I would like to see:

1. Very low phase noise oscillators, like Crystek CCHD series-reason being that I prefer to synchronously clock my DAC, providing masterclock from the USB interface, so phase noise/jitter in the I2S output is critical.

2. Ultra low noise/impedance local shunt regulators, one for each onboard oscillator to allow the oscillators to perfrom their best, and not back modulate the power supplies.

3. Users choice to use the GMRs or not, with separate output headers, and I would prefer micro bnc (for U.FL) for the non-isolated I2S output.

4. Option to send masterclock back to the exa from the DAC over I2S.

5. Oh yeah, prefer two channel only here.

Thank you, barrows. Your suggestions are on the task list.
 
Here is my input of what I would like to see in a revised design:

- Ability to support synchronous operation of the Buffalo DAC. The exa-device has two oscillators. It should therefore be the clock source when operating the Buffalo-DAC in synchronous mode with its ASRC disabled, since it can generate both multiples of 44.1 and 48 Khz. This means that the oscillator on the Buffalo needs to be removed.

- Because of the above, a galvanic connection is needed for this clock. The galvanic isolation should therefore be moved to the USB-side.

Since the synchronous option appears to elevate the sound quality yet another notch (RayCtech, Bunpei), such an implementation would be very desitrable. I also support the suggestion by other posters to incorporate high quality regulators, if this increases the sound quality.

Hi Javin5,

We will use top quality clocks and we will make the clocks and the clock power sources replaceable. Synchronous operation is also on the task list.

-Best,

exa065
 
I am playing (have been doing that for a good period of time already) with isolation of USB (480MB/s) with both exaU2I and other USB implementations.

Oh, that's great!
You must have completed the work quoted below that had required exa065's help.

I tweaked some of the component values and now the galvanic isolation of the USB works reliable, but some more work are needed to perfect the solution.
I have asked exa065 for his help as the component values on the exaU2I and the programming of the ASIO driver together with the component values of the galvanic isolation can for sure be optimized as my implementation are not fine tuned.
...
The differences are easy to verify in all resolutions from 44.1k/16bit flac and up to 384k/32bit upsampled DXD...

However, it's pity that no one other than you can reproduce the way you have established in-house.

Bunpei
 
Oh, that's great!

However, it's pity that no one other than you can reproduce the way you have established in-house.

USB isolation works..
SATA / eSATA and I2S isolation works..
The method I now use (there have been some technological improvements) do not add jitter or noise, but actually reduces lower frequency phase noise etc..
The next I now will work on are isolation of clocks (master clocks) as I may be able to reduce the low frequency noise / phase noise by more than 60dB at 10kHz and more than 100dB at 10Hz if the "isolation" performs as expected.

I will assemble some few USB, I2S and master clock isolation modules and let some hardcore DIYers evaluate the performance.
As isolation can easily be switched on / off I can also implement a remote control to do this switching in realtime...
 
USB isolation is very exciting news. I hope we can hear more about your approach (at DIYaudio.com) in a dedicated thread.
Exa, sorry for the slightly off topic comment.

There will be no more dedicated threads regarding any of my ideas

Exa065 - this will be my last OT posting in your threads...

As this may be my last posting on diyAudio.com except in those threads I have started myself I will use the opportunity to say goodbye to all the great personalities I have encountered here :)
 
Member
Joined 2009
Paid Member
I have isolated USB connections with copper wire, silver wire and pure gold wire.
Isolated USB with pure gold wire insulated with pure silk are by far the best sounding.
These observations are based on my setup where the internal cables in the DAC are pure gold with air as isolation, the interconnects to the power amplifier are pure gold with pure silk isolation, the PCB´s are gold plated in the whole system, the speaker cables are a combination of pure silver and pure gold - isolated with pure silk.
Why do you think you find that gold sounds "far better" better than copper or silver, when it is actually a worse conductor than both?
 
May I ask about a conventional channel assignment rule on DSD playback?

ExaU2I supports a DSD play. The connection diagram for DSD signal output is shown on this ExaDevices web page.
Native dsd playback from WinXP PC using the exaU2I

The notion, "DSD data 1" and "DSD data 2", is used here.
When a stereo play is applied, which channel is to be assigned to "Left" according to a conventional rule?
 
May I ask about a conventional channel assignment rule on DSD playback?

ExaU2I supports a DSD play. The connection diagram for DSD signal output is shown on this ExaDevices web page.
Native dsd playback from WinXP PC using the exaU2I

The notion, "DSD data 1" and "DSD data 2", is used here.
When a stereo play is applied, which channel is to be assigned to "Left" according to a conventional rule?

Sorry, I don't understand the issue here. Can you explain?
 
exaU2I Project Contest - Win a free exaU2I Version 2!

Hello exaU2I Owners!

This is a call to showcase your projects for the opportunity to win an exaU2I version 2!

Post pictures of your project and explain your goals and results on the http://www.diyaudio.com/forums/exadevices/200281-exau2i-projects-gallery.html. The best project, as judged by the exaDevices Team (with the help of the DIY collective), will be declared the winner of a free exaU2I version 2. The contest closes at the end of February. Only DIY, non-commercial contestants are eligible to compete.

For those of you that don't own an exaU2I, but are interested in competing, we are offering a $100 discount on the original exaU2I.

-exa065