[DaveBSC]

Quote:

 
But OR5 and Alpha USB are not in the same price range.  When people consider Audiophilleo 2, it is all about price.
 
There is nothing dumb about powering the input stage by the USB bus.  Anedio D2 (U2) does that too.  The question is how you isolate the ground in the later stages.

The Alpha USB is $1895. The Off-Ramp 5 with Turboclock is $1999. Same price range. Using bus power for anything at nearly $2K IMO is unacceptable. It would've been quite easy for them to include a second linear supply running off of AC for the input stage. They were lazy/cheap. The Off-Ramp 5 is isolated, and uses no bus power. All of the circuitry is powered by the 12V DC power supply, and yet it still sounds better with a common mode choke that cuts the 5V bus line entirely. Bus power is lousy. There is no reason to use it for anything, and the further you can get it away from the converter or DAC, the better.
 
I wouldn't be surprised if the Alpha USB benefited considerably with an AQVOX, Vaunix, or Y-cable with a battery or LPS supply for the 5V line.
 

 

[Riker]

Quote:

 
Hi Riker,
 
It depends on how "attached" you're with the Sonos.  I'd think about getting a SBT and use the EDO plug-in to drive the D2.  It maybe quite a bit better than Sonos driving D2 via coax already.
 
Presently, I'm working on a clock module with custom PCB to replace the clocks on U2.  The crystal oscillator I use has half the RMS jitter compared to Audiophilleo 1/2.  I shall have it installed into my D2 by late next week.  Let's see how close it gets compared to the Legato

 
I've not seen the inside of Synchro Mesh to know what EA is doing.  But such idea has been done by others before: Apogee Big Ben, PS Audio Genesis, etc.  If I were to do such thing, I'd take a buffer to hold the incoming S/PDIF data, and use an ultra low jitter clock to clock out the data from the buffer.

Hi Viper,
I have a significant investment in Sonos throughout my house and no intentions of replacing it anytime soon. The Sonos to D2 via coax spdif sounds miles ahead of just Sonos to my receiver. I got curious when I read that a spdif to spdif reclocking interface such as the EA Synchro Mesh did improve the sound. So I am trying to determine if this is a device that would be good for me to purchase in order to further improve the already really good sound quality.
Riker

 

[Riker]

Quote:

Hi Viper,
I have a significant investment in Sonos throughout my house and no intentions of replacing it anytime soon. The Sonos to D2 via coax spdif sounds miles ahead of just Sonos to my receiver. I got curious when I read that a spdif to spdif reclocking interface such as the EA Synchro Mesh did improve the sound. So I am trying to determine if this is a device that would be good for me to purchase in order to further improve the already really good sound quality.
Riker

I should also add, just in case I was not clear enough, I do not use USB in any way. All of my music is on a Seagate Black Armor Network Attached Storage device (NAS) connected to my network router and the Sonos ZP80 is also connected to my router via RJ45 Ethernet. 

 

[Viper2001]

Quote:

The Alpha USB is $1895. The Off-Ramp 5 with Turboclock is $1999. Same price range. Using bus power for anything at nearly $2K IMO is unacceptable.
 

 
When I auditioned the Alpha USB a couple months ago, the dealer was offering something like 15% off the price.  OR5 after you add the Turbo Clocks and Hynes regulator for S/PDIF output, the price becomes $2249.  Which is why I keep thinking they are not quite in the same ballpark.
 
Anyway, you're right that there is no reason why the Alpha USB cannot do better by isolating the USB port.  Consider the price they are asking.

 

[Viper2001]

Quote:

Hi Viper,
I have a significant investment in Sonos throughout my house and no intentions of replacing it anytime soon. The Sonos to D2 via coax spdif sounds miles ahead of just Sonos to my receiver. I got curious when I read that a spdif to spdif reclocking interface such as the EA Synchro Mesh did improve the sound. So I am trying to determine if this is a device that would be good for me to purchase in order to further improve the already really good sound quality.
Riker

 
I'd say hold your horses, since you're happy with the sound of D2 with coax connection from Sonos.  If EA offers something like 30-day trial, then I'd give Synchro Mesh a spin.

 

[LeoDuran]

Quote:

I should also add, just in case I was not clear enough, I do not use USB in any way. All of my music is on a Seagate Black Armor Network Attached Storage device (NAS) connected to my network router and the Sonos ZP80 is also connected to my router via RJ45 Ethernet. 

FYI: The SBT can also be network attached (so music can be served from a NAS), the USB interface is simply one of the output options of the SBT transport...with other output options being SPDIF (RCA & TOSSLINK), and Analog (from internal DAC).
Like this:
[NAS/Music server]-->Network (wired or wireless)-->[SBT+EDO]-->USB-->[Converter]-->SPDIF-->[DAC]
 
--EDIT--
Here's a rear view of the SBT (Note the RJ-45 for network connection, plus it also has a wireless receiver)

 

[LeoDuran]

Quote:

 
There are two aspects, internal, and external (apperance / conveience features).
 
Internal:
- U3 to use CCHD-957 clocks or better clock x'tals for the clock circuit.
 
- Redo the interface b/w U3 and D3's main PCB (pulse transformer change, better impedance drive to the pulse transformer, direct coax connection to WM8805 or ESS9018 with a SMA connector).
 
- May want to eliminate WM8805 and go directly to the ESS9018, as it already has got 8 S/PDIF inputs.  Maybe WM8805 is still useful for the external coax inputs in terms of reducing jitter, but the output from U3 can go directly to ESS9018 in case a better clock is used on U3.
 
- May want to add I2S output on U3, and D3 to accept I2S connection.  This would allow DSD playback.  Need to think about how to isolate the ground from the USB port to the rest of the DAC.  May want to use high speed optical isolators, but there could be other compromises.  This would be a major redesign.  The question is, how many people would want DSD playback?
 
- If I2S is used for connecting U3 to D3, may want to consider putting a RJ45 I2S connector on the back of D3.
 
- Eliminate the RCA coax input, and change it to AES input.  Include a RCA to BNC adaptor for people with RCA coax cables.
 
- Improve IC decoupling and isolation with better film caps/non-magnetic inductors (I'd not bother you guys with the detail).
 
External:
- LEDs to indicate input sampling rate and DSD signal (if DSD is supported)
 
- Ability to dim/turn off the volume LEDs (timeout jumper set internally?)
 
- Switch to invert phase.
 
- Switch to bypass volume control (ie: fixed at 99).
 
- Remember volume setting even if it is >81.
 
- Studier and better fit and finish casing.  The cover is just too thin and light and the fit is not great.
 
- Optional black casing.  That probably means the voulme LED color would need to change from blue to say orange red for a better match.
 
Now only if James has time to read and reply to his emails...
 
EDIT: Someone actually suggests to me to use synchronous mode of ESS9018, bypassing the ASRC in there and just work with the incoming sampling rate as it is.  This is another major design change.  Would be nice if you can toggle b/w the 2 modes via remote control or front panel button.  It is possible that the transients and attacks are better with synchronous mode.  But I suspect the improvement would be minimal, knowing how good D2 already sounds with the Legato, when compared to MSB DAC IV Signature.

Wow, that's quite a list!

... I'd like to see at least a subset of that:
 
* Internal Async USB to I2S interface
* Input options: SPDIF/RCA, SPDIF/BNC, I2S/RJ45, USB2
* Volume Control Bypass (LineOut Only): Sets volume at 0dB ("99"), and disables Headphone amp
* LED's for sample rate
* External power supply
* De-emphasis switch
* Digital Filter switch
 
Ummh...that looks a lot like the "EA Overdrive", plus LED's and support for Linux USB drivers.
 
--EDIT--
How about an extra Line-Out only version:
- Remove volume control, and headphone amp
- Remove SPDIF/RCA, and add AES/EBU

 

[DaveBSC]

Quote:

Wow, that's quite a list!

... I'd like to see at least a subset of that:
 
* Internal Async USB to I2S interface
* Input options: SPDIF/RCA, SPDIF/BNC, I2S/RJ45, USB2
* Volume Control Bypass (LineOut Only): Sets volume at 0dB ("99"), and disables Headphone amp
* LED's for sample rate
* External power supply
* De-emphasis switch
* Digital Filter switch
 
Ummh...that looks a lot like the "EA Overdrive", plus LED's and support for Linux USB drivers.
 
--EDIT--
How about an extra Line-Out only version:
- Remove volume control, and headphone amp
- Remove SPDIF/RCA, and add AES/EBU

An external power supply isn't necessary for the D2's design, or for most DACs. Steve is a big believer in signal paths that are as short as possible, and he's not a believer in linear power supplies. There's no room for a power supply inside the Overdrive case, which is why it has to be external. It's switch-mode because that's what he prefers.
 
In a traditional full size case, there's more than enough room for two or more toroid or R-core transformers (for analog/digital, or L/R channels, or however you want to do it).
 

 

[Viper2001]

Quote:

An external power supply isn't necessary for the D2's design, or for most DACs. Steve is a big believer in signal paths that are as short as possible, and he's not a believer in linear power supplies. There's no room for a power supply inside the Overdrive case, which is why it has to be external. It's switch-mode because that's what he prefers.
 
In a traditional full size case, there's more than enough room for two or more toroid or R-core transformers (for analog/digital, or L/R channels, or however you want to do it).
 

 
The idea of an external PSU is to move the transformer out of the way, to avoid stray magnetic field interfering with the internal circuitry.  For example, if you put a 50VA R-core transformer a few inches next to the clock circuitry, the clock circuit will degrade the sound (ie: it'll sound soft, losing dynamics).
 
The downside of moving the PSU away is that you now have much higher inductance due to the cabling between the PSU and the DAC.  That is going to hurt voltage regulation.  So the proper way to do thing is to do just a pre-regulation in the external PSU (a choke + cap would do), and put the voltage regulators inside the DAC.
 
Switch mode supplies can provide better dynamic current, so the sound is also more dynamic.  Linear PS is subject to the rectifiers and regulator used.  Many well designed switchers can sound better than the average LPS.  So it takes a bit work to get a LPS to beat a switcher.  Usually, the better LPS will give you better mid-range weight and less glare on the high end.  But again, it all depends on the design of the LPS and the switcher, things can fall either way if done correctly.

 

[Sapientiam]

Quote:

Switch mode supplies can provide better dynamic current, so the sound is also more dynamic.  Linear PS is subject to the rectifiers and regulator used.  Many well designed switchers can sound better than the average LPS.  So it takes a bit work to get a LPS to beat a switcher.  Usually, the better LPS will give you better mid-range weight and less glare on the high end.  But again, it all depends on the design of the LPS and the switcher, things can fall either way if done correctly.

 
You omitted to mention the biggest negative against SMPSUs - their common-mode noise generation. The CM noise is in my experience why they result in more glare at the top.

 

[DaveBSC]

Quote:

 
The idea of an external PSU is to move the transformer out of the way, to avoid stray magnetic field interfering with the internal circuitry.  For example, if you put a 50VA R-core transformer a few inches next to the clock circuitry, the clock circuit will degrade the sound (ie: it'll sound soft, losing dynamics).

 
True, but many DACs (but not the above Bricasti M1) use shielded mini enclosures inside the main case to control interference from the power supply to the most sensitive components in the DAC. The Calyx Femto DAC takes that idea farther than most. The design is similar to their standard DAC, a fairly small, one piece I/O board that leaves most of the case empty, so there's room to isolate the power supply inside almost entirely.
 
http://www.wickeddigital.com.au/index.php/2011-09-04-22-32-08/news/142-calyx-femto-dac-arrives

 

[veritas aurum]

Would there be any improvement in SQ to be had by wrapping the D2's transformers in mu-metal? Inquiring minds want to know. Thanks and regards

 

[Sapientiam]

If there's any audible hum from the D2 then, yes subject to being able to get the mumetal in place without it losing its magic. Which is rather hard - normally mumetal is annealed after being bent to restore its permeability. A DIYer with no prior experience of mumetal would have a really tough time getting decent screening. If no hum is audible then its a wast of time I reckon.
 
If the D2's using toroids there's prrobably some improvement to be had by going to a lower-noise type of trafo - R or EI core. But since these have more radiated flux this could be a two edged sword.

 

[veritas aurum]

Perhaps some other material might be more suitable. Any suggestions? We need a metalurgist here, Any in the audience? Regards

 

[Sapientiam]

Steel's the best bang for the buck in terms of screening magnetic fields.