PC subwoofer noise - EMI or something else?

[surroundman]

Hello. I have bought a great subwoofer for my needs, but I experience some noise when it is connected to my PC. However, when I connect it to my phone there is no noise. Here are the tests I made -

Phone to apple dongle to subwoofer- no noise.
Phone to tempotec sonata hd pro dongle to subwoofer - no noise.
Phone to either dongle to Fosi V3 to subwoofer (via the pre-out of the V3) - no noise.
So to me there is no issue with the subwoofer or the V3's pre out.

PC is much different:
PC (rear motherboard 3.5mm out) to Fosi V3 to subwoofer (via the V3's pre-out) - subwoofer noise AND speaker noise (even if the subwoofer is turned off; just having the subwoofer's power cord plugged introduces noise to the SPEAKERS too).
PC via the 2 dongles previously mentioned to Fosi V3 to subwoofer - same result as the previous.
PC straight to subwoofer via either front or rear 3.5 port - noise.
I managed to eliminate the speaker noise by plugging the subwoofer to the front case 3.5 mm port while the Fosi V3 is plugged to the rear 3.5mm motherboard out (PC audio driver has a setting to output simultaneously to both front and rear 3.5mm outputs).

Apparently anything I try on my PC will not solve the subwoofer noise. So I have a few questions-

1. What is causing the subwoofer issue?
2. How do I solve it? If the dongle DAC's did not help, how unlikely it is that other desktop USB-powered DACs won't face the same issue (like SMSL SU-1 or M300SE) if powered from the PC? I have seen people mentioning that powering a DAC separately from PC might solve noise issues, but is that truly tested and verified? I do not have a way to test optical, but I really prefer to use USB dac/integrated amp if I am to upgrade.

Thanks!

 

[staticV3]

1. What is causing the subwoofer issue?

Most likely a ground loop

2. How do I solve it?

Galvanic isolation.
Either using a USB Isolator like a Topping HS01 to connect the dongle to your PC, or using an isolation transformer like the one tested here:

 

[antcollinet]

just having the subwoofer's power cord plugged introduces noise to the SPEAKERS too).

Definitely a ground loop.

Optical is IMO the easiest and most foolproof way of implementing isolation. And it is audibly perfect, just as USB is. If I was buying new kit to solve this I would make sure optical connection is an option.

 

[DVDdoug]

I have seen people mentioning that powering a DAC separately from PC might solve noise issues, but is that truly tested and verified?

USB power tends to be noisy and noisy USB power CAN sometimes get into the analog-side of the DAC. That can be "ground loop noise". But you won't know if a separate power supply will solve YOUR problem until your problem is solved.

Is it AC power line hum or higher-pitch whine? USB power noise is usually high-pitch "switching noise". But most subwoofers are low-pass filtered and I wouldn't expect whine to get-through.

 

[antcollinet]

USB power tends to be noisy and noisy USB power CAN sometimes get into the analog-side of the DAC. That can be "ground loop noise". But you won't know if a separate power supply will solve YOUR problem until your problem is solved.

Is it AC power line hum or higher-pitch whine? USB power noise is usually high-pitch "switching noise". But most subwoofers are low-pass filtered and I wouldn't expect whine to get-through.

That high pitched noise is also normally coupled into the signal via a ground loop. It can go through the DAC PSU, but that would imply a pretty poor PSU design. Filtering PSU noise is easy (and a pretty natural consequence of bulk capacitance in the PSU)

If the noise is ground coupled then using a separate PSU won't help if USB is still used for data - as you are still connecting the PC ground to the DAC.

 

[surroundman]

I thank all of you for your insights and suggestions! I will see which solution will be the best for my case.

 

[surroundman]

I would like to share some more information.
In my first post I probably missed an important detail - the noise from the subwoofer is not a constant, one, and uninterruptable noise. It varies - the closest way I can describe it is what you would hear as wind blowing when watching a movie with a subwoofer - it varies. When connected to the speaker amp, the noise that is introduced to the speakers too varies - small, tiny spark-like noise. And, perhaps the most important and likely known to you - the noise is very audible when moving the mouse of the PC. I then decided to experiment some more - I purposely ran a game where a stutter would occur (short one, a second or perhaps even less) and I would repeatedly make this stutter happen - the noise disappears during the stuttering (or at the very least decrease massively). I have read about GPU-induced noise, so I made sure to run the GPU at 100% for a while (with a benchmark program) - the noise persists. However, I did the same experiment with a CPU benchmark program - the moment the CPU started to run at 100% the noise disappeared (or, again, the most audible part of it). I then checked if running BIOS only would have noise present - there was no noise at all.

With this information, is ground loop still most likely reason behind the noise issue? Anything else I can try to fix it before I decide on trying USB istolator/isolation transformer or a DAC with optical?

 

[antcollinet]

With this information, is ground loop still most likely reason behind the noise issue?

In my opinion, yes.

A ground loop is, in my experience, the most common coupling mechanism for noise (of whatever source) to couple into an analogue signal.

The noise can be mains frequency - often magneticaly coupled into the ground loop, or it can be RFI from the components themselves coupled to ground by internal capacitive filtering.

And with PC's in the system a very common source is noise coupled onto the ground from (as you have pointed out) graphics hardware, but can also be any other (especially power hungry) subsystems, including the CPU. This could** by via the PC PSU filtering again to ground - or same process in the monitor itself, or it can be magnetically coupled into the ground via (for example) monitor cables running close to mains cables/usb cables/audio cables etc.

**could: I've never attempted to identify the coupling mechanism when I've experienced the problem, so I can't definitively say where it gets in. I'v just isolated the PC from the rest of the system by using an optical connection and poof: Problem solved.

 

[Cbdb2]

Does your pc power cord have a ground pin? Are you using the same power outlet for everything? If not try that.

 

[surroundman]

Yes, it has a ground pin, so does the subwoofer, with every single piece of electronics plugged in the same outlet.

If I don't find some magical solution until Black Friday, I'll very likely buy a DAC with optical.

What I am wondering is, do AC powered DACs like SMSL C200 or DL200 have any means in their circuitries to break ground loops? Is this a thing at all? If not, is it theoretically impossible to implement or perhaps it costs too much to bother?

 

[Matias]

Get a cable with no ground pin for the sub. See if it helps. KEF KC62 even has a ground lift switch.

 

[Cbdb2]

Get a cable with no ground pin for the sub. See if it helps. KEF KC62 even has a ground lift switch.

Don't lift ground pins, they are there to save your life. If you mean just to check if thats the problem should be ok but don't leave it there, please!

 

[Cbdb2]

Yes, it has a ground pin, so does the subwoofer, with every single piece of electronics plugged in the same outlet.

If I don't find some magical solution until Black Friday, I'll very likely buy a DAC with optical.

What I am wondering is, do AC powered DACs like SMSL C200 or DL200 have any means in their circuitries to break ground loops? Is this a thing at all? If not, is it theoretically impossible to implement or perhaps it costs too much to bother?

Probably a shot in the dark, but free and easy. If you have any 2 prong plugs, see if there not polarized, can you only plug them in one way? If not try flipping the plugs.

 

[surroundman]

Thanks for the suggestions. Yes, I am not a fan of lifting the ground too. Flipping the one 2 prong plug (the Fosi V3's) did not help. However ....

Changing Windows' Power Plan from "Balanced" to "Power Saver" reduced the noise subjectively by half. Interesting ... I guess I'll keep trying ...

 

[surroundman]

A small update in case someone faces a similar issue.

Getting a standalone DAC (SMSL DL200) > USB input did NOT solve any of the issues.
Using the DAC's optical input also did NOT solve the issue.
The Topping HS02 however did make a difference - the subwoofer (now 2 of them) CPU-induced pounding is almost (or possibly completely) gone.
The only remaining issue is the noise from the speakers, which occurs when the subwoofers' are connected to the Fosi V3's 3.5mm output (again, even if they are powered off).

So my noise issues are partially solved. The questions I have now are
1. Why optical connection did not reduce the noise at all?
2. If I am to replace my Fosi V3 with the upcoming ZA3 (using balanced connection), can I hope for getting rid of the speakers' noise? The ZA3 will make subwoofer volume easier to control too as the SUB-OUT will be affected by the volume knob.
3. As the Topping HS02 solved at least the subwoofer noise, can we certainly conclude that this is indeed a ground loop issue?
4. Do you believe that there is anything more that I can do? For example, would a better quality power strip make any difference?

Cheers!

 

[staticV3]

1. Why optical connection did not reduce the noise at all?

Because the C200 is also mains-powered. You may have just moved the GND loop from PC->Fosi to DL200->Fosi

If instead you do PC->HS02->Apple dongle, then at least this GND loop may be broken completely.

2. If I am to replace my Fosi V3 with the upcoming ZA3 (using balanced connection), can I hope for getting rid of the speakers' noise?

Yeah I think so.

4. Do you believe that there is anything more that I can do?

You can try grounding the Fosi V3's chassis to protective earth.

 

[audio_tony]

USB power tends to be noisy and noisy USB power CAN sometimes get into the analog-side of the DAC. That can be "ground loop noise".

I have tested several motherboards and mini PCs such as the Intel NUC, GigaByte Brix etc. and not one of them had a noisy USB bus.

It's 99% certain the OP as ground loop issues.

Here's a small sample of my tests, and this also demonstrates what happens when there is a ground loop.

USB noise measurements

USB noise measurements, Raspberry Pi, Intel NUC, PC, USB DAC

gtkc.net

 

[antcollinet]

USB power tends to be noisy and noisy USB power CAN sometimes get into the analog-side of the DAC. That can be "ground loop noise"

Typically this is PC ground noise getting onto a ground loop via the USB ground, rather than it being specifically USB power. Every ground connection to the PC will be susceptible to the same noise - it is just that USB tends to be the connection to our audio systems. It could just as easily be the ground on an HDMI connection.

Note also that the noise is not (normally) getting onto the "analogue side of the DAC" - ground loop noise is a system problem rather than a component problem. It results in voltage differences between the grounds of two components on the two ends of an analogue interconnect. It is here that the noise gets injected into the signal, because the ground reference at the receiving end is different from the ground reference at the sending end. This is why balanced connections work to eliminate the noise, because the ground is not used as the analogue reference.