[NCSUZoSo]

I just took some oscilloscope measurements with gain at 99% and nothing playing with mouse moving. I am waiting to see if these are in line (or close) to Burson Audio's own measurements or not, I may still need to tweak my scope. I am using the Cable+ to connect between the Play and my SDS-1202X-E due to impedance matching properties, should give cleanest signal. Compared to normal RCA, this was much better. I changed the scale going to Ch 1 because I didn't understand why Ch 1 and Ch 2 would be so different right out of the Play's own RCA outputs. Playing with their properties in sound made no impact it didn't seem.

I just realized my horizontal scale may have been too small to capture what we are after, I'm going to play with some settings and see if I can get the 2 channels to come closer together in amplitude.

 

[raoultrifan]

[...]

When I mute all the inputs from my system, even unplugging mic, it doesn't help.
When turning off all the others audio devices, noises are still present.

I think the problem comes from the PC so I don't want to make an offtop. However, the XMOS reciever could filter such noises if it's that good

I'm going to try the AudioQuest Jitterbug soon.

It’s good to know all these details, thanks.

PLAY gets thepower from a 2-poles SMPSS external PSU, so if noise gets from computer’s Ground there’s notmuch to be done here. However, it might worth a special USB cable with GND wire cut-off. Also, is this happening when using a laptop (first powered from mains, 2nd powered from its internal battery)?

Jitterbug is filtering only the USB +5V, if I remember well, so don’t think will help much here, but please give it a try and post back your thoughts.

 

[NCSUZoSo]

There is a 12V Molex on the back to easily power it with any PC PSU. I have been doing it this way for a while, despite it not being in a PC. You'll need to trick the PSU to kick on or buy a small module to act as a switch (or even just a paper clip).

An option that doesn't involve cutting into any cables. When using a PC PSU you are getting a dedicated PSU and some like the Corsair HX750i don't even have their fan turning except when necessary,. In other words, extremely clean and reliable power through a device many probably own already (almost any PC PSU has enough wattage to push the Play).

I talk about this in more detail in my review. I met with my editor tonight and I am restructuring it to flow with the pictures. Taking a while, but so far all who have read the pre-published versions have liked it.
Edit
I have held off publishing as I update my review to include the digital noise issue. Hopefully we can find the cause of this issue and fix it.

I am using one close to this to turn my PSU on and off (don't currently need it, since I have a better lab PSU now). Each of the pairs of + and - terminals have their own slow blow fuse.

https://www.ebay.com/itm/24-Pins-AT...071398?hash=item4d4892f166:g:hqQAAOSwIFtaBWxJ

 

[raoultrifan]

I was reffering to the USB ground only...just for testing if noise gets lost or not.

 

[NCSUZoSo]

Without a USB ground you will not have USB power and thus the Play's XMOS will not show up at all. I tested this already using the Sbooster VBus2 that kills both GND and +5V lines from the USB cable. If we had an XMOS device that didn't require power on the USB line, you could do what you are saying. Perhaps Burson just put a capacitor across the optional DC input, it's common on standalone XMOS modules.

The Burson Play requires voltage on the XMOS and the main power in order to work. To my knowledge you can't send +5V only without a GND to accompany it, unless ...

You could possibly tie into the Play's main ground? Burson would need to chime in on this one.

I have noticed that I get much less digital noise/mouse noise (~75% less) when running the 1/4" headphone output on the Play instead of the RCA outputs/Pre-Amp. Can anyone else confirm this observation?

Note: The ES9018 (for the mic) also gets its power from USB's +5V.

 

[Dżoneł]

Without a USB ground you will not have USB power and thus the Play's XMOS will not show up at all. I tested this already using the Sbooster VBus2 that kills both GND and +5V lines from the USB cable. If we had an XMOS device that didn't require power on the USB line, you could do what you are saying. Perhaps Burson just put a capacitor across the optional DC input, it's common on standalone XMOS modules.

The Burson Play requires voltage on the XMOS and the main power in order to work. To my knowledge you can't send +5V only without a GND to accompany it, unless ...

You could possibly tie into the Play's main ground? Burson would need to chime in on this one.

I have noticed that I get much less digital noise/mouse noise (~75% less) when running the 1/4" headphone output on the Play instead of the RCA outputs/Pre-Amp. Can anyone else confirm this observation?

Note: The ES9018 (for the mic) also gets its power from USB's +5V.

Actually I have made a custom USB cable with ground wire cut off and it works. The +5V remain connected. I don't hear any difference, any downgrade in sound performance. The noise is gone and I can say that might be a solution to do something with xmos grounding.
In my setup there's no audible noise from Bursons front jack when I use headphones. Just speakers.

 

[raoultrifan]

Glad you had the time to test this! Most likely the noise could be now seen with a scope connected to both grounds (measured between the 2 hales from the cut wire).

Keith, this is about the ground that covers the USB cable and not the -5V black wire within the cable itself. If you have this "mouse movement noise" then try the above mod and see if it works for you. If it will, then place the scope on the open-circuit ground wire and let us know what you see, please.

 

[NCSUZoSo]

Glad you had the time to test this! Most likely the noise could be now seen with a scope connected to both grounds (measured between the 2 hales from the cut wire).

Keith, this is about the ground that covers the USB cable and not the -5V black wire within the cable itself. If you have this "mouse movement noise" then try the above mod and see if it works for you. If it will, then place the scope on the open-circuit ground wire and let us know what you see, please.

I didn't think USB had a -5V line. Last time I was modding USB cables there is only the 2 data lines and the +5V and GND. You can also tie a ground to metal shielding (like with CAT6) if the cable has it. So I can think of a few ways it could complete the circuit without the USB GND wire. I will try what you suggested and get back to you all. Below is the standard pinouts for USB:

 

[NCSUZoSo]

Yep, it seems like removing the USB ground may actually be the fix!

Trying to capture this on the oscilloscope, but still playing around with settings. I know roughly what I should see on the scope if there is noise present one way and not the other. However there are a load of settings to go through, so I'll figure it out soon and post them.

I just don't want to post misleading scope captures with different time/div settings by using the auto button. If I use the auto button for both with and without GND, it shows a Pk-to-Pk drop of about 35mV. I can't be sure this is accurate though, talking with my old Elect Engineering roommate about it.

Thanks for telling us about this! BTW, did your USB cables have metal shielding/braid? Mine did and I was just curious what your USB cables had. I was using a semi-good Belden USB A to B cable.

 

[raoultrifan]

I meant the GND from the USB's power, indeed, thanks for pointing this out.

I have 3 USB cables with GND cut-off, all of them are having foil shielding and braid too. I was getting lower noise with other DACs, but never tried them on PLAY yet. I'm glad Dżoneł and you had the time to test this.

 

[Dżoneł]

Hi there,
Few more things to lay out about those noises.

When I plug the Burson to my laptop, which is 5 or more years old I hear no noises, it doesn't matter if I run from battery or the power brick.

Also got the AudioQuest Jitterbug and I have done some short tests. Unfortunately all the noises from my PC goes through this thing and are audible on the same level as they are on normal USB cable...
I have not noticed any sound improvements from the Jitterbug, so i'm gonna to return this as fast as possible
So far the custom USB cable is the best solution but it ain't perfect becouse I hear that after playing music there's some kind of barely audible hiss comming from tweeters, which fades out after few second after pausing the music.
It's probably caused by the groundless communication between PC and DAC (Burson of course)
I made my cable by simply unsoldering the black wire (ground) from USB type B plug-end and it's not connected to shield.
Can't wait for the USB measurements.
Cheers!

 

[raoultrifan]

Oh, so you basically disconnected the GND from inside the USB cable and not the outer shield? Perhaps a macro pic might help best here, thank you for sharing this with us.

 

[selvakumar]

I didn't think USB had a -5V line. Last time I was modding USB cables there is only the 2 data lines and the +5V and GND. You can also tie a ground to metal shielding (like with CAT6) if the cable has it. So I can think of a few ways it could complete the circuit without the USB GND wire. I will try what you suggested and get back to you all. Below is the standard pinouts for USB:

please post an actual pic will help a lot

 

[raoultrifan]

It's probably like he said: "I made my cable by simply unsoldering the black wire (ground) from USB type B plug-end and it's not connected to shield". So, it's probably the black wire from Keith's pic from above.

As stated in my review: "There's a low-power USB hub controller on the USB module, GL850G connected to onboard dedicated 12 MHz crystal. This has an 8-bit RISC processor inside that quickly responds to USB host requests. This USB hub should minimize PC's USB host ripple and noise and also to power the USB chip via the internal low-noise regulators". So I'm pretty sure that the PLAY's USB controller voltage is getting from dedicated regulator and not from computer's USB. Although, for some DACs, handshaking might not be possible, but per Dżoneł tests seems that PLAY can play just fine with USB's ground black wire disconnected.

 

[Dżoneł]

Yeah, exactly, I unsoldered the black wire from the USB type B plug, so the pin number 4 has been disconnected and the interference noises has no way to go through. The pin 4 is still connected on the other side of the cable though.
It stops when using custom cable so my conclusion is that it goes through ground wire
On the XMOS module there's a blue diode that lights up when you connect the USB cable,
It lights up even when the custom USB cable (ground cut off) is being used.
When I disconnected the +5V wire (red colour on the scheme) the device doesn't show up in the control panel at all, Burson doesnt work I made this little test before.

Now I'm thinking about some USB filter/isolator with external power supply or something, I'm a beginner at electronics.