Jump to content
IGNORED

My own USB cable experiment .. not so good..


m5sime
 Share

Recommended Posts

Hi everyone.. I found this forum as I was interested in the Brooklyn DAC and have just bought one..

 

I was considering what USB cable to use with this DAC and have never had the chance to experiment much.. Being an engineer at heart and working in Pro video, then I was interested to see what Cat7 ethernet would be like as a USB cable for audiophile use.. Basically its 4 twisted pairs, each screened individually.. So duly bought and connectors cut off and new USB connectors soldered on, I had a listen...

 

Hmm... now - many of the people that will say cables make no difference... please do not reply to this post.. I wanted to see if there was a difference..

 

The Cat7 was audibly quieter and less impact and much less bass (partially im sure to lower volume).. Now this was a quick experiment against a factory bought USB cable of no HiFi merit.. but the real USB cable sounded much much better..

 

So for something with a pure digital transfer function it was very very different..

 

So that experiment is over and the Simon cable isn't much use!.. I am interested to know what USB cables (3M length or so) anyone would suggest for the DAC.. I really do not want to spend hundreds on a USB cable..

 

I am using Belden power and SPDIF interconnects.. The DAC is burning in with some tracks currently and looking forward to a listen..

 

Best wishes

 

Simon

Link to comment
Share on other sites

There are two or three general alternatives:

 

- Get a good normal USB cable (people say good things about Supra, Curious, ...)

- Get a Corning optical USB

- Continue to refine your DIY build (including trying with a hard connector)

Dedicated Line DSD/DXD | Audirvana+ | iFi iDSD Nano | SET Tube Amp | Totem Mites

Surround: VLC | M-Audio FastTrack Pro | Mac Opt | Panasonic SA-HE100 | Logitech Z623

DIY: SET Tube Amp | Low-Noise Linear Regulated Power Supply | USB, Power, Speaker Cables | Speaker Stands | Acoustic Panels

Link to comment
Share on other sites

Hi Simon,

 

First off Cat7 is 100 ohm differential impedance and USB is 90 ohms, so there will be an impedance mismatch. It is not THAT big of a mismatch so it may not be too much of an issue.

 

How did you connect the shields of your CAT7? This is critical for the cable to get things done right.

 

The connectors have a metal shell, this needs to be connected to the shields of the Cat7 on both ends. Do NOT connect the shield to the ground wire (pin 4).

 

So shields of the twisted pairs to outer shell of connectors on both ends, pins 2,3 on one twisted pair and pins 1,4 on the other pair.

 

You might want to try a 1M set first and see how that sounds.

 

What you do with the two unused pairs might also be important. If it was me I would connect all the shields form ALL the pairs to the shells of the USB connector, even if you are not using a pair. If you don't the unconnected shields in the cable could be adversely affecting the properties of the shields in the pairs you ARE using.

 

John S.

Link to comment
Share on other sites

Hi Simon,

 

First off Cat7 is 100 ohm differential impedance and USB is 90 ohms, so there will be an impedance mismatch. It is not THAT big of a mismatch so it may not be too much of an issue.

 

How did you connect the shields of your CAT7? This is critical for the cable to get things done right.

 

The connectors have a metal shell, this needs to be connected to the shields of the Cat7 on both ends. Do NOT connect the shield to the ground wire (pin 4).

 

So shields of the twisted pairs to outer shell of connectors on both ends, pins 2,3 on one twisted pair and pins 1,4 on the other pair.

 

You might want to try a 1M set first and see how that sounds.

 

What you do with the two unused pairs might also be important. If it was me I would connect all the shields form ALL the pairs to the shells of the USB connector, even if you are not using a pair. If you don't the unconnected shields in the cable could be adversely affecting the properties of the shields in the pairs you ARE using.

 

John S.

 

Hi John, thank you

 

The Cat7 has a sink wire per twisted pair that connects to the shielding per twisted pair (x4).. I connected the drain wires together and soldered them to the source metal shell (the computer end).. I left the DAC end metal shell 'floating' as per the typical setup for shielded power cables.. (not wanting to make another earth loop).. Perhaps this is an issue..

 

I used all twisted pairs connected together (so in effect doubling the cable surface area).. 4 x 2 pairs and each go to each pin of the USB (pins 1-4)..

 

Like I said.. it sounds definitely quieter and nowhere near as much impact.. The impedance mismatch as you say I looked at and assumed for a 3M run might not make a huge difference.. I wonder if having both shields connected to the grounds will make this better..

Link to comment
Share on other sites

Hi John, thank you

 

The Cat7 has a sink wire per twisted pair that connects to the shielding per twisted pair (x4).. I connected the drain wires together and soldered them to the source metal shell (the computer end).. I left the DAC end metal shell 'floating' as per the typical setup for shielded power cables.. (not wanting to make another earth loop).. Perhaps this is an issue..

 

I used all twisted pairs connected together (so in effect doubling the cable surface area).. 4 x 2 pairs and each go to each pin of the USB (pins 1-4)..

 

Like I said.. it sounds definitely quieter and nowhere near as much impact.. The impedance mismatch as you say I looked at and assumed for a 3M run might not make a huge difference.. I wonder if having both shields connected to the grounds will make this better..

 

So each pin of the connector got its own shielded twisted pair? Are both wires in each pair shorted together?

 

At least the data pair (2,3) should be in a single twisted pair, that way the return current (mostly) goes through the data wires and not through the shields.

 

For power and gnd, using a twisted pair for each may work well, I have not tried that.

 

If both ends of the shield are not connected the shield does not behave properly and can do very strange things.

 

John S.

Link to comment
Share on other sites

So each pin of the connector got its own shielded twisted pair? Are both wires in each pair shorted together?

 

At least the data pair (2,3) should be in a single twisted pair, that way the return current (mostly) goes through the data wires and not through the shields.

 

For power and gnd, using a twisted pair for each may work well, I have not tried that.

 

If both ends of the shield are not connected the shield does not behave properly and can do very strange things.

 

John S.

 

Hi John yes all connections (data and power) have their own twisted pair connected together as a single wire. Individually screened

Link to comment
Share on other sites

I have also tried making USB cables. Those ends are a real challenge to assemble without any shorts.

It was not worth it to me.

Get a Supra Cable and be done with it. I think I paid a reasonable $64 for 3 meters.

Been happily using it for a year now. Also stated by them to be at the correct 90 ohms USB.

Link to comment
Share on other sites

Hi John yes all connections (data and power) have their own twisted pair connected together as a single wire. Individually screened

 

I'm amazed this worked at all. In that configuration the return current from the data wires will be going through the shield, and with the shield disconnected on one end there is nowhere for the return current to go. It must be taking some other highly un-optimal route.

 

Try putting both pins 2 and 3 on on the the same pair and connecting both ends of the shield, I'll bet this sounds WAY better.

 

John S.

Link to comment
Share on other sites

I'm amazed this worked at all. In that configuration the return current from the data wires will be going through the shield, and with the shield disconnected on one end there is nowhere for the return current to go. It must be taking some other highly un-optimal route.

 

Try putting both pins 2 and 3 on on the the same pair and connecting both ends of the shield, I'll bet this sounds WAY better.

 

John S.

 

Hi John,

 

I do not think this is the case.. Each twisted pair is being used as a single cable with the ends soldered together.. Each twisted pair carries one of the 4 USB signals (power (vcc)/data ground, Data+, Data -).. The four twisted pairs each have a screen.. each of these screens are connected together at one end (the drain wire) to the shell of the connector (chassis ground).. So unless data ground = chassis ground then nothing is going via the screens or shells.. these are pure chassis ground or should be to my knowledge.

 

All that I have that could be changed is that chassis ground/shell for the DAC end is floating. The data ground is of course connected

 

Br

 

Simon

Link to comment
Share on other sites

Each twisted pair is being used as a single cable with the ends soldered together..

 

You seem likely to have very little chance of getting even close to the specified 90 ohm impedance for the USB data pair.

 

How a Digital Audio file sounds, or a Digital Video file looks, is governed to a large extent by the Power Supply area. All that Identical Checksums gives is the possibility of REGENERATING the file to close to that of the original file.

PROFILE UPDATED 13-11-2020

Link to comment
Share on other sites

You seem likely to have very little chance of getting even close to the specified 90 ohm impedance for the USB data pair.

 

Because of the twisted pair being connected together?

I could try and change the cable setup and have one twisted pair for Data + and Data - and leave the others for ground and V+

Link to comment
Share on other sites

It seems my research was not enough to identify that the data + and data - typically share a twisted pair in a differential transmission similar to ethernet.. So my use of a cable for data + and data - separated wasn't the right way to go..

 

I will change the connections and report back.. Putting Data +/- into the one twisted pair..

 

Br

 

Simon

Link to comment
Share on other sites

Because of the twisted pair being connected together?

I could try and change the cable setup and have one twisted pair for Data + and Data - and leave the others for ground and V+

 

That sounds like a much better idea.

 

How a Digital Audio file sounds, or a Digital Video file looks, is governed to a large extent by the Power Supply area. All that Identical Checksums gives is the possibility of REGENERATING the file to close to that of the original file.

PROFILE UPDATED 13-11-2020

Link to comment
Share on other sites

I will change the connections and report back.. Putting Data +/- into the one twisted pair..

 

My own DIY cable is two separate twisted pairs: one pair for the signal, and one pair for +5V and GND, these two pairs are each separately double-shielded, and also separated physically.

 

Additionally, make them as short as you can.

 

Depending on your system, you can also experiment with putting switches on the +5VDC and GND to test the SQ and operation when those are OFF.

 

Add an input for your own clean power as well.

 

Yet another thing is to use a Common-Mode choke on the power-GND lines, especially with your own power.

Dedicated Line DSD/DXD | Audirvana+ | iFi iDSD Nano | SET Tube Amp | Totem Mites

Surround: VLC | M-Audio FastTrack Pro | Mac Opt | Panasonic SA-HE100 | Logitech Z623

DIY: SET Tube Amp | Low-Noise Linear Regulated Power Supply | USB, Power, Speaker Cables | Speaker Stands | Acoustic Panels

Link to comment
Share on other sites

My own DIY cable is two separate twisted pairs: one pair for the signal, and one pair for +5V and GND, these two pairs are each separately double-shielded, and also separated physically.

 

Additionally, make them as short as you can.

 

Depending on your system, you can also experiment with putting switches on the +5VDC and GND to test the SQ and operation when those are OFF.

 

Add an input for your own clean power as well.

 

Yet another thing is to use a Common-Mode choke on the power-GND lines, especially with your own power.

 

Hi thank you!... I was thinking to drop the power connections as these are an area of issue.. but wasn't sure the DAC would function without them (hence I guess the need for the ghost power from a better source than my MacBook Pro)..

Link to comment
Share on other sites

Hi thank you!... I was thinking to drop the power connections as these are an area of issue.. but wasn't sure the DAC would function without them (hence I guess the need for the ghost power from a better source than my MacBook Pro)..

 

Once you've tested with the switches you'll know what your DAC allows, then you can remove the switches and hard implement the changes or leave them so that you can re-use your cable with other DACs.

Dedicated Line DSD/DXD | Audirvana+ | iFi iDSD Nano | SET Tube Amp | Totem Mites

Surround: VLC | M-Audio FastTrack Pro | Mac Opt | Panasonic SA-HE100 | Logitech Z623

DIY: SET Tube Amp | Low-Noise Linear Regulated Power Supply | USB, Power, Speaker Cables | Speaker Stands | Acoustic Panels

Link to comment
Share on other sites

I do not think this is the case.. Each twisted pair is being used as a single cable with the ends soldered together..

Agree, I don't think this is an issue.

 

Each twisted pair carries one of the 4 USB signals (power (vcc)/data ground, Data+, Data -)..

The USB signal (and Ethernet) are differential - there's a reason they are done in a twisted pair: this minimizes differential interference (ie external signals affecting D+ or D- differently) so you should use the pair as such.

mini+Roon > dCS Rossini DAC + Rossini Master Clock 

SME 20/3 + SME V + Dynavector XV-1s > vdH The Grail

Audio Note Kondo Ongaku > Avantgarde Duo Mezzo G2

system pics

Link to comment
Share on other sites

No USB cables over there. Unless I missed something?

 

I see usb 3 on Amazon.

 

At BJC the USB cables are in the "Data Cables" section.

 

You do not want a USB 3.0 cable unless the DAC manufacturer specifically tells you to do so. If you get a cable with a 3.0 B plug it will not fit into a DAC with a 2.0 B jack (which is by far the most common USB connection on DACs) There are a couple with a 3.0 B jack, but they are very rare.

 

John S.

Link to comment
Share on other sites

  • 2 weeks later...

So the plot thickens a little with my experiment.. I have yet to re make the cable but as I just bought my Mytek Brooklyn I didn't want to experiment with the risk of damaging it.

 

So my first tests as described in this post were with a USB connection to some hifi computer speakers. These speakers have a definite issue with this cable and does not sound right at all.

 

I tried the cable on the Mytek and no issues at all. So the Mytek USB input stage must be a lot better (which it is). Than the amplified desktop speakers and a lot more tolerant.

 

I will still re make the cable and test, but interesting that the Mytek is a lot more robust

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share



×
×
  • Create New...