USB type Interconnect quality for a Mac or PC to a DAC ?
Posted by: Disposable hero on 28 September 2015
Dear knowledgeable minds,
At the moment I'm using a NaimDAC, which might be replaced with a Chord 2Qute or preferably Hugo TT (all depends on the outcome of the next NaimDAC firmware update). In terms of interconnections between a MacBook Pro (or any other computing source) through USB cable to DAC, does the interconnect quality really matter? Essentially it will handle data transmission, so would it be sufficient to use any computer-grade USB type A-B which can be picked up in Maplin or found on amazon e.g. LINDY USB type A-B cables?
Are there really any sonic benefits to investing in audio-grade USB interconnects, as produced by Audioquest, Chord Company or WireWorld and what are your experiences with these for your computer to DAC setups?
Thank you.
Yes, Supra USB 2, and I can confirm that inside the jacket there are two pairs of cables, both pairs individually screened, no overall screen around the two bundles. The two screens are joined together at both ends of the cable and at both ends are joined to the metal shrouds of the USB connectors. The metal shrouds are joined to chassis earth in the MacMini and the DAC, so very poor from a ground noise isolation point of view.
^ Let us know how the surgery works out and which end is the floating shield. I had constructed a similar cable for the Hugo, but found the Belkin Gold (even with a USB A/micro B adapter) to perform better.
On the DAC-V1 critical listening tests in my main system, the C2G Ultima has held my attention, clean lean perhaps, but with micro dynamics in a familiar recording previously unnoticed! I love listening 'into' the soundscape and it helps with that quite well. I would not have thought that was possible via USB
^ Let us know how the surgery works out and which end is the floating shield. I had constructed a similar cable for the Hugo, but found the Belkin Gold (even with a USB A/micro B adapter) to perform better.
Following the effort to figure out precisely how and where the earth was being delivered in my setup, then taking steps to come up with what should be a much improved arrangement, expectation bias is very high. However, I genuinely believe that this has made quite a significant improvement in my system. The sound is more solid, weighty and forward/present in the room, it's more nuanced and natural sounding.
I've cut the shields at the mac-mini end, leaving them connected at the DAC only. Not sure if either end makes a difference but just felt that connecting the shields to the DAC would result in less noise on the shields that could be induced into the signal wires
Did your customised Hugo cable just have the two signal wires connected and the shield at one end only?
^Don't forget Hugo is a 'portable on batteries' so quite different from 2Qute in this regard! The shields were connected at the 'A' shell (PC side). I'm not sure I gave that cable a fair 'break in' period (was ok with SPDIF)! I will re-visit it later.
Best is to listen blind for a while (without pressure- say 30 mins) with music you know very well. Have someone switch the cables for you - if there are differences then focus on those
^^ power connected at both ends - I believe Hugo needs to sense the 5V
Are you able to check with a multi-meter to see if the metal shroud of the USB socket on the Hugo is connected to the ground connections of the RCA sockets? If it is, then it's likely that signal earth in the pre-amp is still connected to chassis earth of your PC/Mac. Even with the shield cut, the ground wire in the USB cable is connected to chassis ground inside both the source and receiver devices. Also, if that is the case, depending on what other sources you have connected to the pre-amp, if any of them have an earth connection to the pre-amp then there will be an earth loop which will impact performance of all sources.
Next steps, I think I'll order a length of Belden single twisted-pair foil and braided shield cable and a couple of solderable USB connectors and make up a data-USB cable from scratch (I.e with just the data pins connected and shield connected at the DAC end only) and see how that sounds compared to the modified Supra. While I'm at it, I'll add a short length of cable soldered to the gnd and +5v pins of the type-b connector with an inline 5/2.1mm dc socket on the end. Then if I try out a 2Qute or Hugo in the future, I can plug a simple, clean, 'floating' 5v supply into the inline socket to get the DAC up and running.
I have no drain wire in that assembly, just shields over the twisted pairs. The USB connector shell is not connected to RCA signal out ground . The Vbus ground is (connected to Hugo signal ground) so yes gnd loop possible preamp/PC but not detected.
I used Alpha wire (2811/2) which I had lying around, but it's characteristic impedance (64 ohms) is a bit below USB spec (of 90 ohms) and missing the inner foil shield - you can do better
| Component | 2 X 1 COND | |
| a) Conductor | 24 (19/36) AWG SPC | 0.025 |
| b) Insulation | 0.010" Wall, Nom. PTFE | 0.044 |
| Cable Assembly | ||
| a) Twists: | 9.6 Twists/foot (min) | |
| b) Shield | SPC BRAID Shield,90% |
| Ground Capacitance | 41 pf/ft @1 kHz, Nominal |
| 4) Characteristic Impedance | 64 Ω |
| 5) Inductance | 0.18 µH/ft, Nominal |
| 6) Conductor DCR | 22.4 Ω/1000ft @20°C, Nominal |
| 7) OA Shield DCR | 7.6 Ω/1000ft @20°C, |
Then if I try out a 2Qute or Hugo in the future, I can plug a simple, clean, 'floating' 5v supply into the inline socket to get the DAC up and running.
I got mixed up somewhere above - sorry I thought you had the 2Qute, hence my comments vis-à-vis Hugo
Its been interesting playing around with USB(DAC-V1 & Hugo) - at times I think I have it sorted with this cable or that gadget, but then I play one other recording and it all sounds wrong! I then switch to SPDIF and feel relieved! YMMV
Well, several weeks later and a bit of experimentation ...
I made myself a data-only cable using a length of belden foil and braid single twisted pair cable, shield connected at the DAC end only, and using solderable USB connectors. The braid is especially dense and tight and I'm sure very effective at keeping RF out, or indeed preventing leakage from the cable itself.
However, the proof of the pudding is in the eating, and I was never entirely convinced, always had a feeling it sounded a bit hard and clinical. It also brought with it clicks and pops with lights and appliances being switched elsewhere in the house and a mysterious ground noise interference problem that I could never really track down but only occurred when the mac-mini or the monitor was in standby.
So for now I'm back to a fresh Supra cable with only minor surgery (the 5v line cut internally near the source end), and a minor change to the earthing arrangement for the monitor connected to the mac-mini so that it is no longer the source of signal earth from the DAC to the pre-amp. This is the best overall sound so far and I appear rid of strange clicks, pops, chirps and buzzing.
It's still not ideal as the chassis ground of the monitor and the mac-mini are still directly connected to the hi-fi system, so I suspect still some gains to be had here. I could mess around with USB isolators but they seem a bit limiting regarding highest sample rate that can be passed through.
It's all confirming to me that USB is just not that good an idea for Hi-Fi applications. I think what I need to try next is a really good USB to SPDIF converter with galvanically isolated coax and TOSLINK outputs so I can do a bit of comparison of the two connection types.
Interesting.
For reference the USB isolator - SPDIF convertor I use between Mac Mini and Hugo (Gustard U12) goes up to sample rate of 384 kHz.
Thanks, that looks like a well spec'ed device. I was thinking about a Teddy U2S converter due to high praise here on the forum. However, it's just for tinkering and experimenting with so don't want to spend too much on something new. The U12 looks like a better bet from that perspective
Can understand why DAC opinions vary so much when using USB. It's all too easy to create ground loops, even if inaudible, and your opening yourself up to the variable electrically noisy PCs, laptops, Mac's, etc.
Allan Probin posted:
It's all confirming to me that USB is just not that good an idea for Hi-Fi applications. I think what I need to try next is a really good USB to SPDIF converter with galvanically isolated coax and TOSLINK outputs so I can do a bit of comparison of the two connection types.
Shouldn't a good USB DAC have that kind of isolation built-in? Not requiring USB to S/PDIF converters or lifting grounds of your computer and monitor? Or is that too much to ask?
I think at least the Naim DAC V-1 attempts to isolate the USB in, I suppose it succeeds better than that (Chord?) dac that you're now working with?
I use a Wireworld purple USB cable between my computer and Audiengine D1 and that was after an extensive listening of various other USB's. It was the only one that made the music sing.
Felix H posted:Allan Probin posted:
It's all confirming to me that USB is just not that good an idea for Hi-Fi applications. I think what I need to try next is a really good USB to SPDIF converter with galvanically isolated coax and TOSLINK outputs so I can do a bit of comparison of the two connection types.
Shouldn't a good USB DAC have that kind of isolation built-in? Not requiring USB to S/PDIF converters or lifting grounds of your computer and monitor? Or is that too much to ask?
I think at least the Naim DAC V-1 attempts to isolate the USB in, I suppose it succeeds better than that (Chord?) dac that you're now working with?
It's all a compromise ![]()
Felix H posted:Shouldn't a good USB DAC have that kind of isolation built-in? Not requiring USB to S/PDIF converters or lifting grounds of your computer and monitor? Or is that too much to ask?
Maybe the question is the other way round, why do DACs have USB inputs with all the interference implications that has when connected (typically) to computer equipment?
One answer to that might be so as not to close the door on people using computers and for whom adding an external DAC is such a step up in quality compared to the computer's own DAC that the limitations aren't evident.
Innocent bystander, I think nowadays DACs are being bought and used mainly for the purpose of getting good audio from a computer (while being also useful with many other sources).
Check out this quote from the Naim DAC V-1 white paper:
"A decade ago the outboard digital-to-analogue converter (DAC) – once a popular means of upgrading a CD player – had virtually disappeared as an audiophile product category. Today, with burgeoning interest in computer audio, the outboard DAC is enjoying a new lease of life.
But with the computer as a music source comes the need for a new audio interface. The S/PDIF and AES/EBU digital interfaces which used to suffice for an outboard DAC are not generally available on a computer, and where an S/PDIF output is provided it is usually via the Toslink optical interface and may not be capable of working at up to 192kHz sampling rate.
The obvious interface to use with a computer music source is USB – Universal Serial Bus. Widely employed to connect all sorts of peripheral, it is as universal as its name suggests and capable – in 2.0 or ‘Hi-Speed’ form – of a maximum data rate of 480Mbits per second, which is ample to handle the 9.2Mbits/sec required by 24-bit, 192kHz stereo."
Felix H posted:Innocent bystander, I think nowadays DACs are being bought and used mainly for the purpose of getting good audio from a computer (while being also useful with many other sources).
Check out this quote from the Naim DAC V-1 white paper:
"A decade ago the outboard digital-to-analogue converter (DAC) – once a popular means of upgrading a CD player – had virtually disappeared as an audiophile product category. Today, with burgeoning interest in computer audio, the outboard DAC is enjoying a new lease of life.
But with the computer as a music source comes the need for a new audio interface. The S/PDIF and AES/EBU digital interfaces which used to suffice for an outboard DAC are not generally available on a computer, and where an S/PDIF output is provided it is usually via the Toslink optical interface and may not be capable of working at up to 192kHz sampling rate.
The obvious interface to use with a computer music source is USB – Universal Serial Bus. Widely employed to connect all sorts of peripheral, it is as universal as its name suggests and capable – in 2.0 or ‘Hi-Speed’ form – of a maximum data rate of 480Mbits per second, which is ample to handle the 9.2Mbits/sec required by 24-bit, 192kHz stereo."
In that case I wholeheartedly agree with your original sentiment: DACs intended to enable computer rendering to achieve the highest quality of sound of which the DAC is capable should indeed include whatever isolation may be necessary to cope with the notoriously unclean USB digital outputs from computers. Not to do so makes a nonsense of the intent outlined in that white paper.
My £0.02, I use Moon Audio silver dragon cables to connect my system together. I have a USB A-B from my MacBook Pro to my DAC-V1. For MacMini (running iTunes with Audirvana 1.x) to my SuperUniti I have a Bel Canto USB-SPDIF converter, Moon Audio cables either side, way better than using the hideous optical TOSLINK connection built-in to the MacMini. All (Bel Canto, Moon Audio cables, Audirvana) highly recommended.
Shouldn't that be £0.01 (2 ha'porth)? Or technically 0.0021 as I don't think 1/2p coins were ever called ha'pennies.
yes, the optical output of the MM is pointless with Audirvana as it still uses the MMs soundcard, which I believe is avoided using the USB output, which is the reason for using a convertor if you want to use the better spdif input if using a Hugo
ewemon posted:I use a Wireworld purple USB cable between my computer and Audiengine D1 and that was after an extensive listening of various other USB's. It was the only one that made the music sing.
Yes, I use the WireWorld Ultraviolet cable as well. Good VFM as well.