Schiit Wyrd USB "Decrappifier"

[jcwc]

Yes it does, BUT.

 

FEATURES

 

 

• USB isolation 1000VDC between USB host and USB device
  
• Fully USB 2.0 Compliant
  
• Full speed data rate 12 Mbps
  

 

 

12Mbps will give you max 96fs, no higher, so no chance to do DSD on DoP which needs 176.4fs. A lot of USB isolators behave this way. USB Audio Class 2.0 requires 480Mbps compliant devices and the Olimex misses out.

 

Great for Redbook only though.

 

Some more here and here. Same story, 12MBps

 

After reading a bit more, it seems for USB galvanic isolation it will cap the transfer to 12 Mbps. As you said many isolators behave this way. Any idea what's the technical limitation behind this?

[Superdad]

I frequently wonder when a Fiber SFP connection will be universally used on the Input side for the DAC connection instead of USB, SPDIF and Ethernet. Such a connection would go a long way in eliminating most of the concerns with power induced noise in computer audio. These connections are used all the time in non audio related computer connectivity so the technology is there.

 

I think the only thing lacking from what I have seen is software capable of using a fiber connection on the media server side. One could use an available PCI Slot on the computer side to install such a card but I haven't seen any music library related software capable of sending data out thru a fiber connection yet.

 

Maybe it was over in the other thread (on LAN Input DACs) where I mentioned having some clients who are using the Adnaco PCIe fiber optic extender system (Adnaco-S1B: 5Gb/s Over Fiber Optic Expansion System – 4 PCIe Slots) with a Adnaco/JCAT USB card. With that system you at least get true isolation from the computer. And I also suggested that with a system like that someone could design a really nice DAC on a PCIe card so USB, etc. is entirely out of the picture (sorry, I am not thrilled by the existing computer PCIe soundcard DACs that are out there; I'd want something very special).

 

Anyway, I don't no quite which driver's my customer in Poland is using with his Adnaco fiber set up (he is using my JS-2 LPS to power both the far end Adnaco 4-slot PCIe board and the JCAT USB card plugged into it).

[Superdad]

In relation to the matter of "transmission of ground noise from the USB source to the DAC", is galvanic isolation the only way to stop this?

 

Would you know if the Wyrd or ifi USB stops such ground noise?

 

No they are not able to do so.

 

After reading a bit more, it seems for USB galvanic isolation it will cap the transfer to 12 Mbps. As you said many isolators behave this way. Any idea what's the technical limitation behind this?

 

Yes, the chips used by all these isolators usually are the kind with either tiny transformers between the two sides, or with RF transmitters/receivers on each side (wild but true)--and neither of these methods can support the higher speeds. Not only that, but in a lot of cases (the Analog Devices ADuM chips), the cure is worse than the disease. Some of these devices add a ton of jitter! Some opto-isolators can do high-speed, but these can be nasty for other reasons.

 

Just to be clear "galvanic isolation" is simply the blocking of any DC path from source to receiver. That's all it means, DC blocking, it doesn't mean anything about high frequency noise. That doesn't mean galvanic isolation is useless, just that it is not the end-all of isolation. Galvanic isolation is very good at getting rid of low frequency ground loop noise, but that is it, it doesn't by itself block high frequency noise such as the noise produced by processors doing their thing.

 

It seems to me that now would be a good time for me to repost an excerpt from something my dear friend John Swenson wrote a while back, focused on this very subject:

=======

"….As Alex mentioned I hate opto-isolators, there are much better isolators out there, I use the GMRs exclusively. Unfortunately they are pretty expensive.

 

Even using isolators does not completely block jitter. I'll try and get this across without pictures. A signal goes into the USB side of the isolator, current flows from the driver, through the input side of the isolator (whatever it is) and back through the plane to the driver chip. That signal passes through the isolator some how (light, magnetic field, radio waves, whatever) (yep one of the isolator technologies actually sends radio waves between the sides) and causes the receiver side to do something, which changes the signal on it's output. That output then drives the DAC chip or reclocking flop, which then sends the current back to the isolator output on the groundplane. The current ALWAYS goes in loops, thus the signal going to the DAC chip creates noise on the DAC side groundplane.

 

Thus any jitter on the signal crossing the isolation barrier is added to the inherent jitter of the isolator and that shows up as noise on the DAC side ground plane, even with the isolator! What the isolator does is prevent OTHER GPN such as being produced by the USB receiver itself from getting into the DAC groundplane. It's definitely worth it, but you still have to deal with the jitter on the I2S signals themselves which cross the barrier.

 

Because the I2S signals are fairly jittery after the isolators, you usually should reclock them before sending them to the DAC chip. Why do you need to do this? Isn't the jitter on the clock the only signal that matters? Because GPN also happens INSIDE the DAC chip. Jittery input signals generate noise on the ground traces in the DAC chip, which change how the clock signal is received. You can have an extremely low jitter clock going into the DAC chip, but if the I2S signals are very jittery, the GPN inside the chip will cause that ultra low jitter clock to look MUCH worse.

 

So you still have to look at the jitter on the I2S signals, even with a perfect clock.

 

There are ways to cut down on these issues by careful board layout, but you have to include these as part of the overall design from day 1 to make sure they will be effective. But even with this, some influence still gets through.

 

John S."

=======

[jcwc]

Very interesting. Thanks for the explanation.

[jabbr]

Indeed, ground plane noise (whether heard or not) is one of the most pernicious factors limiting the performance of any DAC. Once it gets in, it can sail past almost every type of isolation and reclocking and make its way to the master clock.

 

Assuming you use a dual headed USB cable in which only the +/- data lines are attached to the USB source and both the +5V and Ground are attached to the linear power source, shouldn't that also provide a clean(er) ground?

[Superdad]

Assuming you use a dual headed USB cable in which only the +/- data lines are attached to the USB source and both the +5V and Ground are attached to the linear power source, shouldn't that also provide a clean(er) ground?

 

That is an interesting question. Are you certain that the "dual-headed" cables (that is 2 USB 'A" plugs to 1 USB 'B' plug) are truly running just pins 2 & 3 (data) on the data side and not connecting the shell at all at the computer end?

 

If that is the case, there is also the possibility that the USB won't work properly if the other 'A' plug (power side) is plugged into a completely different 5V source. Even DACs that don't use ANY USB power (and the computers they are attached to) are not very happy if you try to just run them on a single pair of wires with no shell or ground attachment whatsoever.

Maybe someone here with a dual USB cable to do some tests with a continuity meter.

 

However, if those cables really do have the data leg with just two wires and no grounding, and if that works as you propose, then ideally you would want to connect the 5VBUS leg to a power supply whose output "ground" side is not common with the house mains ground. That is not typical with most LPSs.

 

It just so happens that John Swenson and I went to a lot of trouble to totally isolate the "ground" side of the DC outputs of our JS-2 dual-output, choke filtered LPS from the chassis and AC ground.

For safety the chassis is grounded via the mains AC cord ground pin, but the PCBs and voltage regulators are insulated/isolated from the chassis. (Finding thermal pads that both isolate and perform well enough in mating the regs to the heatsinks was a BIG headache at the beginning; and I still have to hand cut and punch this expensive material for each TO-220 device).

Although the JS-2 does not have a USB jack, either of its outputs can be set by the user to 5V, and someone with a dual-headed cable just needs one of these adapters:

[tranz]

That is an interesting question. Are you certain that the "dual-headed" cables (that is 2 USB 'A" plugs to 1 USB 'B' plug) are truly running just pins 2 & 3 (data) on the data side and not connecting the shell at all at the computer end?

 

If that is the case, there is also the possibility that the USB won't work properly if the other 'A' plug (power side) is plugged into a completely different 5V source. Even DACs that don't use ANY USB power (and the computers they are attached to) are not very happy if you try to just run them on a single pair of wires with no shell or ground attachment whatsoever.

Maybe someone here with a dual USB cable to do some tests with a continuity meter.

 

However, if those cables really do have the data leg with just two wires and no grounding, and if that works as you propose, then ideally you would want to connect the 5VBUS leg to a power supply whose output "ground" side is not common with the house mains ground.

 

Mentioned this before on the Gemini dual headed cable, only the shield drain wire (outside usb shield) is not connected. All 4 pins are connected hence the benefit of blocking pin 4 (signal ground) on the computer side, ground filter of the IFI, and sending a clean 5v ( pin1) if needed downstream, or blocking it if not.

 

The signal and power wires just do not run next to each other until the very end at the connector where they come together again.

[mordante]

I recieved my Schiit Wyrd last weekend. Since I celibate my birthday party Saturday I enjoyed a hangover Sunday. Which means I haven't done any serious listening yet. But as far as I could tell in the few minutes between headaches that I did listen the Wyrd did indeed improve the SQ. Less grainy, less background noise. Maybe I'll do a more thorough review later.

 

My setup:

Synology NAS

Laptop in docking station, windows XP

Foobar using M2tech Hiface kernel streaming

Schiit PYST

Schiit Wyrd

M2tech hiface (directly into the Wyrd)

pro audio DIN 75 Ohm cable

Wadia 12

Pink Faun 5th anniversary interlink

Symphonic Line La Musica

Audio Quest rocket 88

Master One speaker by Rick Paap

[goalline92]

Wow, this little device has just decrappified a really noisy USB output. I can do anything on my PC and the sound coming out of my Bifrost is as quiet as a freaking church mouse, even into headphones. I don't know if it sounds better, and quite frankly, I don't care. Just to be rid of the noise...