• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Cleaning USB for Bus Powered Audio Devices: Discuss

L5730

Active Member
Joined
Oct 6, 2018
Messages
135
Likes
45
Location
East of England
#81
If your PC uses an earth grounded (3 prong) power brick, try a doubly insulated 2 prong power brick. Breaks earth ground while meeting regulatory safety requirements.

I tried a 5v disconnected usb cable and two different usb optical isolators (both based on the analog devices chip) without success before this solved my usb issues (noise and bit hits)
The PC PSU has a 3-pin IEC connector which must be grounded, no question about this (with exception of star grounding, but I ain't going there).
It's a desktop tower with a 750w ATX PSU.
All laptops and Raspberry Pi each have a 2-pin (plastic earth) to AC/DC transformer for their power, and so have no connection to ground, and therefor no loop. However, connecting the DAC analogue output to the same computer that is connected via USB does produce some noise.

The D10 digital and analogue grounds are common and this helps to cause a ground loop across the device. However, if the device connected via USB and the device(s) connected to analogue out do not share the same ground, there is no problem. If one is mains earthed and the other not (eg. amplifier and laptop) there is no problem.

I am thinking of grabbing an ADuM4160 based isolator form ebay for £8, de-soldering the DC-DC transformer (because they are crap noisy ones and ~200mA is useless). Then I could add a micro USB B port for external (not mains earthed) PSU power.
 

L5730

Active Member
Joined
Oct 6, 2018
Messages
135
Likes
45
Location
East of England
#82
As @AnalogSteph said here I could just use quasi-/pseudo-balanced cables and not have the shield connected at one end or the other.
I bought some bits on the cheap and tried it, disconnecting pin 1 (ground) at the balanced amp/speaker connection end. No buzz.
I am not adding in the resistor as I could not find enough online mention of it, and it is easier to hook up/fit without it.

So instead of screwing around with the USB ground and possibly causing the device to have a significantly different potential to that of the USB Host driving it, I am just doing what a great many studios have done, and that is to cut the ground link between DAC and Amp.

If this DAC had balanced outputs, it would be arguably easier still, or maybe the buzz wouldn't be audible at all, however it is unbalanced.
In the case of an unbalanced output just follow Lavry advice.

If the sending end is the unbalanced two-conductor connector; both the shield and the “-” conductor of the pair are connected to the “shield” conductor of this connector. The “+” conductor of the pair is connected to the signal conductor of this connector. The shield is connected to nothing in the T.R.S. connector (receiving end) and the pair is connected to tip and ring similar to a balanced connection.
Of course, a balanced cable (two pole + shield) should be used as the shield could be a different resistance/impedance.
Luck would have it the long unbalanced cables (RCA-RCA) I have that I was using are actually balanced cables inside, with the shield attached at the source end - probably why they are labelled as directional. So All I have to do is remove the RCA connectors on the destination end and pop on a couple of Neutrik XLR or TRS jacks. Leave the ground connection on the destination end unsoldered and insulated.
 

Gregorius

New Member
Joined
Jan 7, 2019
Messages
3
Likes
8
#83
What brand of power supply is in the PC? So many bad power supplies out there, and a place people skimp on.
 

L5730

Active Member
Joined
Oct 6, 2018
Messages
135
Likes
45
Location
East of England
#84
I didn't skimp on the PSU.
Seasonic Ultra Prime Titanium 750.
As a series, it's one of the most expensive consumer PSUs around, and measures very well.

Simply put, it is a ground loop induced by having the tiniest of voltage change between two points the USB ground (mains earthed PC) and audio ground (active speakers, mains earthed).
However, this DAC will also show similar noise if it is connected via USB to a non-earthed Laptop and the DAC analogue output is fed back into the same laptop audio input. Because there is some, even small, potential difference between the USB ground and audio ground, there is a loop, which puts current where it doesn't want to be and the DAC amplifies it.

I read an interesting paper from Analog Devices all about where to put the ground and how to try to stop ground loops and ground noise causing issues between digital and analogue sides.
 

L5730

Active Member
Joined
Oct 6, 2018
Messages
135
Likes
45
Location
East of England
#85
Hmm...if I was to pursue the USB>USB ground loop breaking idea, I think I have a possible solution to run by the electronics folks.

Hows about a pair of diodes facing opposite directions and in parallel on the ground line from USB Host.
Something like this diagram, but instead of analogue/digital ground, say USB Host and USB Device grounds.

As long as the diodes could be sourced that open at less than say 100mV, the two grounds should not drift too far apart to screw up transmission, yet are still linked should something funky happen.
One issue is that they may start switching if the two ground potentials normalise and drift often.

I am content, for myself, with using an analogue solution. But it is still nice to continue discussion of a purely USB solution.
If it is this easy, how come there aren't solutions like this commercially available? Because isolators exist and do better job and are safer - we wouldn't want to think we have a separated ground, only to find it becomes connected when the voltage jumps up, like when using a probe or something.

However, might be perfectly fine if devices on either side are connected to the same house mains, and only a few rooms apart max.
 

solderdude

Major Contributor
Joined
Jul 21, 2018
Messages
1,186
Likes
1,822
Location
The Neverlands
#86
Schottky diodes like the BAT85 have a low voltage drop but higher than 0.1V, closer to 0.2V.
As soon as the voltage becomes higher the ground loop is present again.
Another disadvantage could be HF noise being rectified and cause even nastier 'nasties' as not only AF noise could be in the loop but HF noise will become audible as well.

But you won't know until you tried, might work in this case, might not.
 

L5730

Active Member
Joined
Oct 6, 2018
Messages
135
Likes
45
Location
East of England
#87
You have a balanced input on your monitors that is completely going to waste right now. Why not make use of it?
[...]
2. Solder up your own RCA to XLR adapter cables. Required per cable: 1x male XLR connector, shielded TP cabling, 1x RCA connector, 1x 100 ohm resistor, heatshrink tubing or similar insulating material. Connect as follows:
RCA signal - hot (usually red cable) - XLR pin 2
RCA ground - 100R - cold (usually black) - XLR pin 3
RCA ground - screen - XLR pin 1 (may need to be left open at XLR end)
Note that the 100 ohm resistor is chosen to match the D10's nominal output impedance. If the output is AC-coupled, further refinement may be possible.
I skipped the resistor on the fact I couldn't find enough examples using it nor the impedance of the XLR inputs on the active monitors.
My friend, who has designed and built audio electronics in the past advised me to try without the resistor.
So we have:

RCA tip (signal > XLR pin 2 (hot/+ve)
RCA sleeve (ground) > XLR pin 3 (cold/-ve)
RCA sleeve (ground) > cable shield
XLR pin 1 (ground) unconnected.

No ground loop, buzzing, funny sounds or anything.
Unplug the cable, turn the speaker up to max volume, gentle soft hiss. Plug in XLR, same hiss, no different.

Ground loop solved, in my case.

However, now that I have the cheap ebay USB isolator ADuM3160* board, I am looking for a cable like @graz_lag posted earlier.
The little isolator boards have a low end B0505S 1 watt DC-DC isolated unregulated transformer, which means a max. of 200mA output and some voltage drop from the component too.
Best thing to do is to just take it off, and maybe the opto isolator and low-speed/full-speed switch, as they only allow the low-speed operation, else pointless.
Then grab a USB-Y cable, get external power, and we have an isolated USB port with plenty of current.

*same as 4160 just half the kV protection

Reading John Swenson's blog at UpTone, it's a shame his isoRegen product didn't make more of a difference in testing with the Mytek.
 

solderdude

Major Contributor
Joined
Jul 21, 2018
Messages
1,186
Likes
1,822
Location
The Neverlands
#88
Something like the isoRegen will only show differences when there is a ground loop.
When someone measures it in a no loop situation there will only be downsides.
Way too expensive for what it does in your case.
When you have a k$ rig the price for lowering loop noises is more balanced.

Ground loop problems do exist and isolators work best and simplest in these cases.
Be them analog (audio line transformers) or USB isolators.

kV protection is not needed here.
Capacitive coupling is the biggest issue. Less pF in series means less HF ground loop noises.
These are inaudible by itself but could under certain circumstances affect the digital side before the DAC chip.
 

warp2600

New Member
Joined
Mar 26, 2019
Messages
2
Likes
1
#89


This is basically the ifi defender.
The ground is 'broken' here, which is what USB should have, but reckon when there is a ground loop the 'common' of the PC is still there but running through the mains.
Problem that might be there is if there is a voltage present between the 2 'grounds' (which makes the current loop flow) that exceeds that of the USB input of the DAC and is not 'protected' in the DAC it might destroy the DAC.

Perhaps measure the DC resistance (while everything is connected but the USB is disconnected) between the USB screen of the PC and the USB screen of the D10 input. If that reads close to 0 (or perhaps 100 Ohm or so) it may be safe to use the circuit above and save money on buying an ifi ?
@L5730 nice, I am going to DIY this implementation on my basic layout, which in fact equals the iFi defender as pointed out by @solderdude ...
I like a lot this thread, you go very depth into a very specific issue, which many folks try to solve with some fancy and expensive-considering-the-application devices ...
Thank you so much ! ;)
Hi Solderdude, Graz_lag and everyone,

This is my first post. I found this great forum site when doing a search on Korg DS DAC100.
I have got this DAC and saw that graz_lag feeds his Korg from a power bank battery through an altered Y USB cable.
I wanted to try that without the hassle of DIY-ing the cable so I ordered the iFi iDefender and the same powered battery graz_lag has.
I don't have any ground loop hum or other noise needing to be addressed, I just thought I might get even better SQ if I disconnect the 5V feed coming from the Mac Mini.
The battery and the iDefender have not arrived yet but I am a bit deterred now from using the iDefender for fear of destroying the DAC if the iDefender cuts the ground from the Mac.
You say that the iDefender is wired the same as the Y USB cable in the pic (i.e. the PC sends only the 2 data lines and there is no ground or shield).
The iDefender product page says that the device cuts ground loops if it finds one (with DACs that require the 5V power feed from the PC, it can be done only if external 5V power is connected through the mini USB input). I wanted to use a USB A to USB mini cable provided with my Olympus digital camera for connection between the powered battery and the iDefender (thought it would be cheaper than an iFi iPower) but now I am afraid to use this new setup and send it it all back when delivered. I already have great sound and there is no need to fix what's not broken...
 
Last edited:

L5730

Active Member
Joined
Oct 6, 2018
Messages
135
Likes
45
Location
East of England
#90
From what I understand of the iDefender, the external power supply powers a bunch of transistors which allow for the ground from USB Host to USB device to be disconnected. However, I am under the impression that there is some kind of cleaver circuit employed which detects the difference in voltage between the grounds on the two sides, making it safer.

There was an odd review on Amazon or somewhere where the person said that the iDefender caused damage, but I doubt it is to blame.

If you don't have a ground loop problem, chances are the iDefender won't do anything except make you 50 quid poorer.
iFi have other products which may have a larger effect on sound quality, such as the iPurifier.
 

solderdude

Major Contributor
Joined
Jul 21, 2018
Messages
1,186
Likes
1,822
Location
The Neverlands
#91
When you aren't plagued by weird sounds I recommend not to use the idefender.
It won't detect a groundloop issue so won't improve nor 'fix' anything as nothing seems to need fixing.

try it and maybe return the item.
When you hear an audible improvement you can keep it in there but be almost assured it's placebo or a really crappy DAC that needs an external power supply.
 

Shadrach

Active Member
Joined
Feb 24, 2019
Messages
205
Likes
256
#92
I built a power supply based on the ADuM4160 many years ago.
The problem I was trying to fix wasn't so much noise, it was the problem USB had in providing the 4.5 Volts needed to power my HRT Pro.
While some USB ports will provide a maximum voltage of 5 Volts others if my memory serves me would only provide 3.5 Volts.
I haven't read the whole thread carefully but I think you've already discovered that the handshake across devices needs to be made with every re connection.
Although the ADuM chip blurb says it can handle 24/96 I never managed to get it to handle anything more than redbook.
 

mansr

Senior Member
Joined
Oct 5, 2018
Messages
355
Likes
749
Location
Hampshire
#93
While some USB ports will provide a maximum voltage of 5 Volts others if my memory serves me would only provide 3.5 Volts.
The USB 2.0 spec says high-power ports must supply 4.75-5.25 V. Low-power (≤ 100 mA) devices must function with a Vbus supply of at least 4.40 V. High-power devices need only function with a minimum supply of 4.75 V.

Although the ADuM chip blurb says it can handle 24/96 I never managed to get it to handle anything more than redbook.
The ADuM chip supports only low/full-speed USB, not high-speed. The USB Audio Class supports up to 24/96 over a full-speed connection, so that should have worked.
 

Shadrach

Active Member
Joined
Feb 24, 2019
Messages
205
Likes
256
#94
The USB 2.0 spec says high-power ports must supply 4.75-5.25 V. Low-power (≤ 100 mA) devices must function with a Vbus supply of at least 4.40 V. High-power devices need only function with a minimum supply of 4.75 V.


The ADuM chip supports only low/full-speed USB, not high-speed. The USB Audio Class supports up to 24/96 over a full-speed connection, so that should have worked.
Do you know when the specification for USB Voltage you stated came into being?
I remember measuring mine, say 15+ years ago and only one port could supply a nominal 5 Volt.
Yup, I know it should have worked, but it didn't. I read the specs to.;)
 

mansr

Senior Member
Joined
Oct 5, 2018
Messages
355
Likes
749
Location
Hampshire
#95
Do you know when the specification for USB Voltage you stated came into being?
I remember measuring mine, say 15+ years ago and only one port could supply a nominal 5 Volt.
Yup, I know it should have worked, but it didn't. I read the specs to.;)
The USB 2.0 spec was initially released in 2000. I haven't checked the old versions, but I doubt there have been any changes to the voltage requirements. USB 1.1 might have been worse.
 
Last edited:

L5730

Active Member
Joined
Oct 6, 2018
Messages
135
Likes
45
Location
East of England
#96
I haven't tested myself, but have read that USB 2.0 ports have been pretty lousy at adhering to specification. It's only really now when there is design on the motherboards to deliver more current for USB 3.0 devices that the power issue seems to have gone for USB 2.0.
Desktops have been better than laptops, no surprise there really.

I put 24/96 through the ADuM 3160 isolator into a USB headset, worked fine. The headset was rated for 100mA so plenty safe to use here.
Current draw of the Topping D10 DAC (320mA) is a bit too high for the poor quality DC-DC isolator (1watt = 5v @ 200mA ish) plus losses.
 

mansr

Senior Member
Joined
Oct 5, 2018
Messages
355
Likes
749
Location
Hampshire
#97
I haven't tested myself, but have read that USB 2.0 ports have been pretty lousy at adhering to specification. It's only really now when there is design on the motherboards to deliver more current for USB 3.0 devices that the power issue seems to have gone for USB 2.0.
Desktops have been better than laptops, no surprise there really.
Oh, there's poor adherence to spec aplenty, from both hosts and devices. The spec says current draw exceeding 100 mA must be requested and approved by the host. A device may never draw more than 500 mA. Hosts must limit current draw to no more than 5 A for protective reasons.

Now many hosts are lazy and don't enforce the initial 100 mA limit. Consequently, many devices are lazy and draw far in excess without asking, sometimes more than 1 A. For instance, every device with a Y cable for extra power is in violation of the spec.

Battery powered hosts (laptops) are often stricter in their implementation, which is why naughty devices often fail there. Many hubs also come with power supplies unable to provide full power to all ports.

It's all a mess, and USB 3 has only made it worse.
 
Joined
Apr 7, 2019
Messages
5
Likes
0
#98
D10 was powered by USB port of his PC when he did that test and noise floor level was at app -140dB.

Of course, it may very wel be the case that your USB port is particularly noisy.

When I crank volume of my D10 to 100% I have to get app 5cm to the tweeter to hear the noise and that is probably coming from my amp, not from D10. Now it's connected Vidmate iTunes Notepad++ to Dell PC but it was the same with RPI.
Take a USB cable and cut the +5v (red) cable. Run an external +5v power supply into the device-side of the UBS cable red wire that has just been cut.
 
Last edited:

daftcombo

Addicted to Fun and Learning
Patreon Donor
Joined
Feb 5, 2019
Messages
560
Likes
194
#99
Hi,

I have a problem at my office with this DAC: https://www.audiophonics.fr/en/dac-...e-ampli-24bit96khz-sa9023-es9018-p-11911.html
and my 1more Triple Driver Over-ear headphones.
There is a nasty treble noise, which is modulated when the headphones cable is moved.
Is it the same kind of bad noise that I have at home when I plug my AKG K701 into my mixing table. Following the cable position, if I touch it, hold it, etc. the noise is present or not.

When I plug the 1more Triple Drive Over-ear directly in the PC soundboard, the noise is absent. But I want to use the Audiophonics DAC because it has a convenient volume knob.

It seems like a ground issue.

What can I do?

Thanks.
 

DDF

Active Member
Joined
Dec 31, 2018
Messages
166
Likes
132
Location
Ottawa
The PC PSU has a 3-pin IEC connector which must be grounded, no question about this (with exception of star grounding, but I ain't going there).
It's a desktop tower with a 750w ATX PSU.
All laptops and Raspberry Pi each have a 2-pin (plastic earth) to AC/DC transformer for their power, and so have no connection to ground, and therefor no loop. However, connecting the DAC analogue output to the same computer that is connected via USB does produce some noise.

The D10 digital and analogue grounds are common and this helps to cause a ground loop across the device. However, if the device connected via USB and the device(s) connected to analogue out do not share the same ground, there is no problem. If one is mains earthed and the other not (eg. amplifier and laptop) there is no problem.

I am thinking of grabbing an ADuM4160 based isolator form ebay for £8, de-soldering the DC-DC transformer (because they are crap noisy ones and ~200mA is useless). Then I could add a micro USB B port for external (not mains earthed) PSU power.
My suggestion only works for external power *bricks* as I said and is 100% safe in that application. Of course don't cut off the earth ground for a desktop internal supply, that would be unwise and I'd be the last one to recommend that
 
Top Bottom