Interference noise from PC

I’m getting a lot of nasty noise that sounds like electromagnetic interference from my desktop PC when I’m trying to record guitars directly into my audio interface. The sound has a high pitched whine and a “frying” sizzle in it. It gets significantly worse while I’m moving the mouse. When I browse folders I also hear short glitchy noises. It’s not really audible with most cleans, but high gain sounds are not usable with this issue.

I included a sound clip of staying idle and then moving the mouse, and another clip with a few chugs for reference.

So far I’ve tried:

  • putting everything into just one power strip on one socket
  • disconnecting every peripheral and leaving only the mainframe case
  • disconnecting ethernet cable and disabling wi-fi from my phone
  • other electric sockets in other rooms, also in the kitchen where proper grounding should be guaranteed
  • taking the GPU out entirely and running Windows on CPU’s integrated graphics chip
  • disconnecting power cables from hard drives
  • different guitar cables
  • active and passive pickups
  • sitting in front of a Mac laptop; the noise is lesser and almost usable
  • running the entire setup on laptop battery, fully isolated from electric sockets

Disclaimer: I know the kind of noise that computers cause into guitar pickups, and this is not that. I confirmed with a friend that noise of this kind is not normal, it does not happen to him and it hasn’t happened to me in the past when recording with different computers. The piezo pickup in my Ernie Ball Musicman doesn’t create this problem, further pointing towards EMI I think. With all this said, I’m not very skilled in electronics or understanding how ground loops work, so hopefully somebody can guide me a bit!

My system is an Nvidia GTX 1080, i7-8700K, Seasonic 650W Prime Ultra Gold power source, and MSI Z370-A Pro motherboard. My audio interface is an SSL2+.

To test this issue more freely, I connect my audio interface into my laptop instead (which is on battery, unplugged from wall sockets) and run the guitar amp sim there, and then sit about a metre away at my desk with the guitar in playing position. The desktop PC is under the table near my leg. The way that I’m pointing the guitar has a noticeable effect on the noise, but not enough, and it’s not really a solution since I need to sit at the desk to control the DAW.

My laptop also produces this same interference problem, but since I’m far away enough from it, the guitar isn’t picking that up, it only picks up the interference caused by the desktop PC. This setup lets me test what happens to the noise while the desktop PC is various states like booting up, in BIOS screen, etc.

I found that the noise disappears when I disconnect the GPU power cables and let the PC hang on the bootup message that prompts you to connect the cables. But the problem isn’t the GPU card, because if I remove it and run on the integrated graphics chip and go to Windows, the noise is back.

Edit to reflect current situation:
I’m currently suspecting it has something to with USB being polled. If I go to BIOS and move the mouse, which is over USB, it causes sizzling noise. If I have the DAW open in Windows (which probably polls the USB connection to the interface continuously), it sizzles constantly.

Long thread with the same issue over at GearSlutz.

What could I try next? Thanks for any help guys!

Is it a Logitech mouse?

I had this, though not as bad as yours sounds, and switching the mouse out for a cheapy generic one did the trick. Something to do with how different mice handle their usb power.

There remained an irritating low level high pitched sort of noise that I eventually determined was…tinnitus.

It’s a Corsair. I’ll try with any mice I might have, though disconnecting the mouse doesn’t help either.

I used to get this back in the day when I would cheap out and just record directly into the sound card or plug right into the mic jack on the laptop for convenience. I have not actually experienced this since switching to external preamps and audio interfaces.

If you have a laptop you can try using the SSL with that and see if it goes away. This will also let you unplug the laptop power from the wall power to test for ground loops. if it goes away when running on battery power, then it was an electrical issue. Your GPU box is externally powered? Maybe it needs isolated / filtered electrical power.

If so, I’ve used these to eliminate those types of loops:

https://www.amazon.com/Ebtech-Hum-Plug-Style-Eliminator-Cleaning/dp/B07W682STV

Of course lots of other ground loop type tweaks others can suggest.

I tried that but it does the same thing. In fact if I unplug the laptop, the noise gets worse and a bit different too, which is weird.

The GPU is a normal gaming Nvidia graphics card, you attach power cables to it from the computer’s power supply unit.

Meaning just the laptop by itself with no GPU, just the SSL device? Got me then, I’m out of ideas!

1 Like

Yeah, just the SSL and no other devices plugged in - but the laptop also has a dedicated GPU, in addition to its processor-integrated graphics chip. That one I haven’t turned off yet. I’ll check right now if the laptop does the mouse movement hissing.

The laptop does the same pretty much - moving the mouse actually doesn’t make the noise worse, but clicking or dragging a selection box does. Whenever something moves on the screen, it screeches like in the clips I posted. Plugged in is marginally better than unplugged.

It doesn’t seem to matter which computer the SSL is connected to, either. As long as the guitar is near a computer, it picks up the noise.

Someone said that it’s a grounding issue most likely and they fixed theirs by disconnecting the ground pair wires from their monitors’ XLR cables that were fed by their interface. I don’t have anything connected to the interface except headphones, just to eliminate variables. So does this mean that there could be a grounding issue somewhere inside the computer?

Can a guitar have a grounding issue? I did test with two guitars, but still.

If I try this with the laptop in another room, unplugged from wall socket, I still get the problem. Where are the possible ground loops in that scenario?

From the electrical engineering perspective, guitar circuitry is really bad, as it’s not differential. This failure is well understood, and some people have tried XLR-outputs, etc., but with the right hardware things can be made to work, so nobody cares. Therefore, I suggest that you adopt a best practice and don’t try to solve your current problem. The simplest is probably a trusted ADC that will directly feed digital signals into your computer, so would something like this do?

https://focusrite.com/en/usb-audio-interface/scarlett/scarlett-solo

1 Like

Is the computer connected to a grounded outlet?

2 Likes

This was my first thought as well. Had this problem several years ago, apparently there is such a thing as “dirty power” that makes recording a nightmare. In my case I had a bunch of clicks and pops as I recorded.

Yeah, exactly! The Faraday cage that is the chassis works a lot better with a solid ground connection, I believe. Clicks and pops sounds more like a buffer underrun issue though, but it could certainly be caused by interference causing packet drops.

@Shredd the interference you’re getting sounds exactly like the sounds I’ve been recording of my desktop computer, but in my case I was actively seeking them, using a coil as microphone. In fact, very similar to a guitar pickup now that I think about it. I think it it simply noise from the PCI and USB buses. Perhaps moving the computer farther away from you desk would help?

You may want to try using ReaFir plugin in Reaper.

Thanks for the replies guys!

Situation in short: Desktop PC tower emits specific kind of noise into pickups, having DAW open or moving the mouse makes it worse, therefore possibly a USB issue.

My audio interface is only a couple of months old, and I really like it, so I’m not going to replace it unless it becomes a very clear suspect. I’ll see if I can borrow other models from friends, but I don’t think it’s the interface that’s causing the issue since the noise is radiating from the motherboard. Focusrite’s Scarlett series are recommended by many and they are good, but just like my SSL2+ which is more than twice as expensive, they’re not above the prosumer-grade as far as I know. Do you have advice to why a Scarlett Solo could possibly fix this?

I read through this long thread where the OP has the exact same situation: PC tower radiating noise into pickups. It wasn’t resolved even by a completely new computer system and audio interface (he went from Scarlett 2i2 to Focusrite Clarett 4), but a lot of ground was covered from USB power isolation to RF shielding clamps on cables. It’s the most meticulous source of discussion I’ve found, and while it unfortunately turns into some bickering, a lot of good questions are asked and answered nevertheless. Lots of seemingly misinformation debunked too. It’s worth a cautious read if anyone is interested in this problem in general.

Just to clear up: Until this problem turns out almost impossible or extremely expensive to fix at the root cause, I’m going to be staying away from partial fixes and workarounds like moving the PC around, using less gain or other amps, noise profiling and treating the DIs after recording. I appreciate these ideas though and they will be useful if I can’t get this fixed otherwise. If a fairly cheap purchase like a hum destroyer or DI box can provide a robust solution, then cool.

I’m looking for help and offering support to find the root causes, be it USB isolation problem, faulty PSU power rails or motherboard built to inferior USB audio specs etc. My understanding of electronics is very limited but I know computer hardware relatively well. I’m willing to test a lot and learn what causes the issue in my situation, and what serves as the most robust fix. I’m ready to replace parts to get this fixed, but preferably after understanding why it’s being replaced.

I hope to help others who might have the same problem while I’m at this. I can see this issue being a nightmare for anyone who’s doing high gain guitars at home. The more awareness about this gets spread, the better for everyone.

The sockets in my room are not grounded, unfortunately. I moved the test setup into my kitchen because the sockets there are required to be grounded by law, even if other sockets in the apartment weren’t. I don’t know how to test sockets and see for myself though.

One of the tower’s sides is a glass window, with some aluminum trims. Does that break Faraday’s cage? I did get the issue replicated on the laptop though, not sure if it counts.

It definitely would help, but I can’t move the tower far away enough for it to make a satisfactory difference, there’s just no space. It would also limit the possibility of recording somebody else playing, since the PC tower would then likely be too close to them already and they’d have to sit at the desk instead and operate the DAW themselves.

Here’s how I tested today:

Guitar > cable > interface > laptop (test 1) or desktop (test 2).

The guitar was always set at about 60 cm from the desktop PC tower’s front panel. Roughly the practical distance when I’m sitting at the desk.

I put the laptop and interface about 3 metres from the guitar to avoid any noise from them. Only headphones to interface, no monitors or other gear.

The desktop PC was connected to a grounded socket, laptop was on battery power. All peripherals (display, mouse, keyboard) disconnected while testing, except when testing for their effects specifically.

I noticed that plain DI has the same noise issue when amplified enough, so I ditched the amp sim. Just the guitar near the PC tower.

TEST 1: I recorded a noise test with the laptop, DI only.

Sounds like this:

Looks like this:

Moving or clicking the wired Corsair mouse on the desktop PC caused spikes at every 1000 Hz from 1k to 10k (the higher pitched whine in the sound clip). A wireless Logitech mouse caused this less prominently when moving, more so when clicking. The laptop being plugged or unplugged didn’t change anything.

Interestingly, the same noise was audible acoustically from the underside of the laptop chassis while record arm was on, even with no guitar (or cable) connected to the interface. I really don’t understand how this can happen as I don’t know what component is underneath that particular spot in the chassis where the sound was coming from.

Additionally, if I sit with the guitar in front of the laptop and anything moves on the screen, the 1k-10k spikes appear for the duration of the movement.

TEST 2: I connected the interface into the desktop PC and recorded there, removing the laptop fully from the equation.

Sounds like this:

Looks like this:

That is with the DAW running. When the DAW was started up, the 1k-10k spikes became a constant screeching and mouse usage introduced some kind of modulation into the noise. Without the DAW, both mice caused the same symptoms as on the laptop, meaning the noise was lesser overall but using the mouse caused the kilohertz spikes. Using keyboard caused a slightly different tone of noise that started on every key press and lasted for a few seconds.

Oh, I’m sorry, I mistakenly thought that you were trying to use your audio card! (Yeah, that was Troy experiencing that, a long time ago, and I got confused.) I checked out your interface device, and indeed, it should work well, and, as you argue, there should be no reason for another ADC…

One question: Have you tried to interpose a USB hub with a big power supply to power your SSL2? (It would be so interesting to put test equipment on your USB power feeding into your SSL2, but I suspect that you don’t have any.)

One more question: When you turn the volume of your guitar to zero, does the noise disappear?

1 Like

Can you return the audio interface, it’s worth getting a replacement to eliminate any possibility of manufacturing defects. Or try to borrow another interface for a few mins to A/B test.

1 Like

Yeah, that’s similiar to my house. I would love to get grounded power in my “music room”. Did using a grounded outlut make any difference?

I would guess so. As for the laptop, perhaps the case isn’t all metal on the laptop either?

Have you tried different USB cables? Do you have a different interface you can test with?

I haven’t tried that. It’s a really interesting idea though because I do think there’s something wrong with the USB power feeding, from an audio perspective. I don’t think I have the equipment to do that right now but it’s something I could look into, if I can make sure it’s safe to do.

I have my old interface (Line6 POD Studio UX2) and I can borrow an SSL2, which is the slightly smaller version. I’ll try those at least.

No, the noise was identical.

Yeah, the top is aluminum and the underside is plastic.

I haven’t tried exhaustively with different cables or USB slots, I’ll do that. Another cheap thing to try out is clamping RF ferrites on cables, probably gonna go for those as well.

I’ll do more tests with my girlfriend’s Mac laptop as the recording machine and as the noise source. I wasn’t able to pick up as much noise when sitting in front of it earlier.

1 Like

Do you use Speakers or headphones? What kind of guitar cables do you use and do they run close to the power outlets? Any LED lights in your PC / room?
I always had the same noise problem. It got significantly less when I put a hum suppressor between my interface and my studio monitors and it completely disappeared when I switched from HDD to SSD in my Desktop PC.
Now they only appear from time to time when I got my cheap ambient light in the studio on and forget to turn them off before recording guitars with passive pickups.

1 Like