• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

High-pitched speaker noise from ground loop?

Loopnoob

Member
Joined
Feb 18, 2024
Messages
6
Likes
7
Hello, long time lurker here. Decided to make an account because I've been having some issues with my setup after a change in gear where I can hear a high-pitched noise from both my passive speakers. I figured out a sub-optimal solution after reading the ground loop thread, but thought I'd check in here to see if someone here has a better solution.

For context, this is my setup: I have a 17 or so years old NAD C352 Integrated Amplifier with all analogue inputs/outputs. It's powering a set of passive stereo speakers. It's connected to my TV with a RCA to 3,5mm cable. The integrated amp is also connected with a RCA to RCA cable to a Fiio K11 amp/dac that I just got today, which in turn is connected to the rear IO of my desktop PC with the provided USB cable. Finally, the graphics card in my PC is also connected to the previously mentioned TV by a 5m long HDMI cable (Besides being connected to my desktop monitor by DisplayPort). The amplifier is connected to the same power strip as the TV, while my PC and K11 is connected to a seperate power strip. I've tried plugging in the K11 to the same power strip as my NAD, but it didn't make any difference, and plugging in the PC to the same strip as the others is unfeasible. My issue is that I hear a very high-pitched ringing reminiscent of tinnitus from the passive speakers increasing in volume as i turn up the volume on the NAD. The issue is present regardless of any audio actually playing, and whether selecting the input from the TV or the external DAC on the NAD, although the volume knob needs to be turned up about twice as high for the tinnitus noise to be audible when selecting the input from the TV. To the best of my understanding, the K11 doesn't have fixed line level mode and thus functions as a preamp when connected to the NAD, but while turning up the volume knob on the K11 increases the volume of the audio from the speakers, it has no effect on the volume of the tinnitus noise.

Now, as I mentioned at the start of the post, I did find an impractical solution. Disconnecting the HDMI cable running from my graphics card to the TV got rid of the noise altogether with either of the inputs selected on my NAD. I bought the Fiio K11 used to replace my old Dacmagic Plus and Atom Amp combo because I wanted to simplify my setup and be able to seamlessly switch between using my headphones with the Atom, playing games on the TV with stereo audio, and listening to audio on my stereo with the TV off, without having to switch cables on the unbalanced output port of the Dacmagic Plus. So, my questions are: Is this likely a ground loop issue, is there a simple solution where I can keep the HDMI plugged in and get rid of the noise without buying new cables, or should I just buy an optical cable to connect my K11 to my PC to (hopefully) bypass the issue?

Cheers
 
Hello and welcome! Generally in audio, ground loops result in mains frequency hums and buzzes, such as 60/120Hz or 50/100Hz. These are obviously low frequency. Your description is of a high frequency. This can happen with computer ground loops. High frequency noise can also be caused by leakage from devices and cables or radio type pickup.

The fact that you can eliminate it by having only one thing connected at a time does suggest a ground loop. Having one connection optical will certainly benefit you if that's the case. If it's not a ground loop and is (e.g.) leakage from something on the HDMI side, and optical connection may also be a benefit.
 
Since PC->USB->FiiO->RCA->NAD doesn't seem to be an issue, I think you can focus your efforts on the PC->HDMI->TV->RCA->NAD connection.

An easy way to break the electrical connection between your PC and the NAD would be to add a DAC like the SMSL PS100 between TV and NAD, connected to the TV via Toslink.

However, using Toslink, you will no longer be able to control volume using the TV remote.

If you want to keep the TV remote's volume control, then I think you'd have to look into optical HDMI cables.
 
Thanks for replies everyone! I just realised today that none of the of the power sockets in the room are grounded, but is my understanding correct that a ground loop can still occur even if none of the components in a loop terminates in a grounded power socket?

Hello and welcome! Generally in audio, ground loops result in mains frequency hums and buzzes, such as 60/120Hz or 50/100Hz. These are obviously low frequency. Your description is of a high frequency. This can happen with computer ground loops. High frequency noise can also be caused by leakage from devices and cables or radio type pickup.

The fact that you can eliminate it by having only one thing connected at a time does suggest a ground loop. Having one connection optical will certainly benefit you if that's the case. If it's not a ground loop and is (e.g.) leakage from something on the HDMI side, and optical connection may also be a benefit.
Thank you! Can confirm it's a high frequency sound. The passive speakers are floor speakers with 4 drivers (1x Tweeter, 1x midrange, 2x woofers) and I can hear that the noise is coming from the tweeters when I put my ear to the speakers. Can't rule out leakage or radio type pickup. I have a lot of cables routed together on the underside of my desk, including the HDMI cable, the USB-C cable going to the K11, and some power cables powering my monitor and the K11. Moving the USB-C cable out of the bundle of routed cables didn't solve it though. I am using a ASRock X570 Phantom Gaming X motherboard with inbuilt WiFi/Bluetooth with a solitary, generic antenna connected to one of the two antenna ports, so leakage from radio type pickup could be a possibility.

Thanks for the link. From what I've gleaned from the link, there could be a problem between the K11 and the NAD, since the volume control on both devices can be used to change the volume, but upping the volume on the K11 doesn't increase the volume of the noise. The noise only changes in volume when using the volume control on the NAD. But I'm assuming the K11 just works as a passive preamp in line out, which I'm guessing could explain why changing the volume on the K11 doesn't increase the volume of the noise. Switching between the video (TV) and AUX (K11) input on the NAD does change the volume of the noise at the same setting on the NAD volume knob, but the frequency of the noise seems to be the same.

Since PC->USB->FiiO->RCA->NAD doesn't seem to be an issue, I think you can focus your efforts on the PC->HDMI->TV->RCA->NAD connection.

An easy way to break the electrical connection between your PC and the NAD would be to add a DAC like the SMSL PS100 between TV and NAD, connected to the TV via Toslink.

However, using Toslink, you will no longer be able to control volume using the TV remote.

If you want to keep the TV remote's volume control, then I think you'd have to look into optical HDMI cables.
Thanks, I might give that a go. I could try buying a Toslink cable and using my old Dacmagic Plus between the TV and the NAD. Volume control is no issue as I'm already using the NAD remote for volume control, and the Dacmagic Plus does have a fixed line level mode so I wouldn't ever have to bother with its volume knob either.
 
There's recently been an unexpected development, so I decided to write a little update on my issue.

I bought a used graphics card on Sunday, upgrading my older RX 6600 XT to a RX 7800 XT. After hooking it up, making my cable routing a little neater (outside the pc case) and generally moving cables around a little bit, I was about 80% sure that the volume of the noise was audibly lower than before at the same turn on the NAD volume knob. I started thinking about what MaxwellsEq said about the issue potentially being related to leakage, and I wondered how and if the power draw of the components in my PC had an impact on the issue. My GPU(s) had been idling through all my previous testing, drawing only around 10 Watts or so with the newer 7800 XT being capable of drawing around 30 times that. I decided to run Unigine's Superposition 8K optimized performance benchmark to put my pc through its paces and increase the power draw significantly. Lo and behold, the noise changed characteristics completely! It went from being a high-pitched tinnitus like sound to sounding more like a low-quality PC fan struggling with the lower end of its operating speeds. Putting my head to the speakers, it was hard to distinguish spatially if the sound was coming from the tweeter, the midrange, or both. I'm guessing it was something like a mix of predominantly low high-end and high mid-range frequencies.

On a side note, I ordered a cheap TOSLINK cable online last week that will hopefully be here within the next few days. When it gets here I'll try testing it between both my TV and a DAC and between my PC and my Fiio K11 and update the thread.
 
Last edited:
I've not experienced a PC based earth loop or interference, but people on here who have often talk about GPU load affecting and sometimes mouse movements!
 
Thanks for replies everyone! I just realised today that none of the of the power sockets in the room are grounded, but is my understanding correct that a ground loop can still occur even if none of the components in a loop terminates in a grounded power socket?
a] Do any of your audio components have 3 pin plugs on their cords? Are all of your wall AC receptacles 2 pin?
b] Yes, ground loops can happen, even if none of your components have a Safety Ground/Protective Earth connection.
c] Ground loops are mostly associated with power line related hums & buzzes.
 
There's a TOSLINK cable waiting for me when I get home from work, so let's see what kind of effect replacing one of the electrical links in the chain with an optical one will have :D At this point I think I'm as invested in understanding what the hell is going on as actually solving the issue lol

I've not experienced a PC based earth loop or interference, but people on here who have often talk about GPU load affecting and sometimes mouse movements!

Haven't heard anything with mouse movement, but GPU load definitely has an effect. Tried firing up a game on the TV afterwards, and outputting audio through the Fiio to the NAD replicated what I saw in the stress test with the lower-pitched noise, while choosing to output audio to the TV through HDMI just produced the original high-pitched sound at higher volumes. At this point I'm thinking the discrete GPU is the primary culprit, and it's generating a lot of noise or interference that's primarily affecting output from PC → Fiio → NAD, and to a lesser extent PC → TV → NAD. What I find odd is that the latter is not affected by GPU load considering that the HDMI connecting the PC and TV is plugged directly into the GPU port, while the Fiio is plugged into a USB port on the motherboard I/O. I would have thought the output plugged into the GPU would be more affected, but here we are haha.

a] Do any of your audio components have 3 pin plugs on their cords? Are all of your wall AC receptacles 2 pin?
b] Yes, ground loops can happen, even if none of your components have a Safety Ground/Protective Earth connection.
c] Ground loops are mostly associated with power line related hums & buzzes.

A] Not any of the plugs that go into outlets or wall sockets. All my wall sockets are rectangular 2-pin, so no ground. Both the NAD and my PC have built-in power supplies connected with standard 230v power cables (3-pin appliance connector to 2-pin wall plug).
B] Is that because the connected devices are essentially floating at different voltage levels, or have I misunderstood?
C] Pitch definitely becomes lower with increased system power draw, but I don't think it's anything as low-pitched as that.
 
HDMI cables are a perennial noise source, grounding and otherwise, as is leakage from USB, monitor, and other cables.

Ground loops lead to HF noise fairly often, either by modulating the power ground so it gets mixed higher into the audio band, or through noise coupling through an unexpected path (via the ground loop). But other noise coupling/leakage paths can do the same... And note that "noise" in this case is a desired signal, being injected where it is undesirable.

I may have missed it, but I would try spreading all the cables apart and moving them around to change their physical orientation, outside and if possible inside, to see if that helps isolate and reduce the noise source (if it is a cable). Often enough a desire for a clean look leads to bundled cables that then couples noise to each other.
 
I only had time for a quick test with the TOSLINK cable last night, but using it to connect my PC to my Fiio, I no longer get any noise at even the highest turn of the volume knob when AUX input (the RCA ports connected to the FIIO line-out) is selected on the NAD. When I select video (RCA ports connected to the TV phono) on the NAD, I'm still able to reproduce the original high-pitched sound, but now it requires a turn of the volume knob that would make my ears bleed if audio was playing. Stress-testing the PC, and by extension increasing its power draw by a factor of 10-20 or so compared to idle, does nothing to increase the volume of the noise or change its pitch in either of the above scenarios. Only difference besides the input (USB-C to USB-A vs TOSLINK) is the OS sample rate, as the optical receiver on the Fiio is limited to 24-bit/96KHz whereas I had selected 24-bit/192KHz through Windows when I was using USB-C (Max supported 32-bit/384KHz was giving me grief with one particular song from a couple of sources, going down to 24/192 solved it).

I don't know why TOSLINK gets such a bad rap on a lot of forums. Being able to solve such an annoying problem with a 4€ cable is brilliant. Listening to some 24/192 tracks on Tidal with my AKG 712 Pro, I must admit I'm unable to hear any sort of difference when switching between the inputs despite the resampling needed with the TOSLINK input (Granted, it's not exactly the most rigorous way to test it). I might also test using the TOSLINK cable to hook my TV up to the analog integrated amplifier via a DacMagic Plus when I get a chance, see if results are similar with regards to the noise.

HDMI cables are a perennial noise source, grounding and otherwise, as is leakage from USB, monitor, and other cables.

Ground loops lead to HF noise fairly often, either by modulating the power ground so it gets mixed higher into the audio band, or through noise coupling through an unexpected path (via the ground loop). But other noise coupling/leakage paths can do the same... And note that "noise" in this case is a desired signal, being injected where it is undesirable.

I may have missed it, but I would try spreading all the cables apart and moving them around to change their physical orientation, outside and if possible inside, to see if that helps isolate and reduce the noise source (if it is a cable). Often enough a desire for a clean look leads to bundled cables that then couples noise to each other.

Thank you for the explanation, I'll try routing the power and the USB-C cable for the Fiio away from the others in so far that it's possible with my setup when I have the time, if nothing else to get a better idea of what was causing the issue.
 
Only difference besides the input (USB-C to USB-A vs TOSLINK) is the OS sample rate, as the optical receiver on the Fiio is limited to 24-bit/96KHz whereas I had selected 24-bit/192KHz through Windows when I was using USB-C (Max supported 32-bit/384KHz was giving me grief with one particular song from a couple of sources, going down to 24/192 solved it).
You won't hear any difference between 24/96 and 24/192, so I would not worry about this.
I don't know why TOSLINK gets such a bad rap on a lot of forums
I think that many years ago some Toslink implementations had jitter issues which may have been audible.
 
Many people thinks that being limited to a transfer rate good for DSD 64 or PCM 24-192, TosLink is somewhat less HiFi...
 
Maybe due the hype of extreme oversampling affecting large part of the market, honestly don't care so much.
I'm fine living within the Toslink capabilities, some other people just don't.
 
s this likely a ground loop issue, is there a simple solution where I can keep the HDMI plugged in and get rid of the noise without buying new cables, or should I just buy an optical cable to connect my K11 to my PC to (hopefully) bypass the issue?
I also had a noise issue when connecting my television to my audio system. In my case it turned out the noise was on the cable from the cable television provider (Comcast/Xfinity). I solved the issue by running the cable to a filter/surge suppressor on a power strip. Actually, I tried two different power strips. One worked and one did not. The one that worked is made by Moster Power. It is the Home Theatre PowerCenter HTS800, but it is fairly old and my not still be available.

Using a fiber optic Toslink cable also may solve your issue.
 
Last edited:
Sorry for the late reply, I didn't get around to doing some more testing before last night, but I did discover a few things.

HDMI cables are a perennial noise source, grounding and otherwise, as is leakage from USB, monitor, and other cables.

Ground loops lead to HF noise fairly often, either by modulating the power ground so it gets mixed higher into the audio band, or through noise coupling through an unexpected path (via the ground loop). But other noise coupling/leakage paths can do the same... And note that "noise" in this case is a desired signal, being injected where it is undesirable.

I may have missed it, but I would try spreading all the cables apart and moving them around to change their physical orientation, outside and if possible inside, to see if that helps isolate and reduce the noise source (if it is a cable). Often enough a desire for a clean look leads to bundled cables that then couples noise to each other.

I wasn't able to completely separate all the cables, but I did redo most of my cable management last night. Power cables are now routed separately from video/audio cables and other types of input (With the exception of the power cable and DisplayPort cable for my monitor). I connected my Fiio with USB-C to the motherboard I/O with the cable dangling a bit in the air to get as much distance as possible within my setup between it and the other cables. I also connected it to the front I/O afterwards because it's easier to avoid it coming into proximity with other cables there. The pitch of the sound changed to a more low-pitched hum with some crackling, and the sounds produced were more inconsistent than the previous high-pitched noise. Motherboard I/O vs front panel I/O seemed to make no difference. Moving my alarms power cable that wraps around the side of my speakers seemed to change the sound a bit, but unplugging it entirely didn't remove it. When I was lying on the ground underneath the table and redoing my cables, I did notice the power supply to my standing desk was giving off a low-pitched hum, but plugging it out didn't make a difference.

Since PC->USB->FiiO->RCA->NAD doesn't seem to be an issue, I think you can focus your efforts on the PC->HDMI->TV->RCA->NAD connection.

An easy way to break the electrical connection between your PC and the NAD would be to add a DAC like the SMSL PS100 between TV and NAD, connected to the TV via Toslink.

However, using Toslink, you will no longer be able to control volume using the TV remote.

If you want to keep the TV remote's volume control, then I think you'd have to look into optical HDMI cables.

I did try this out today, and going TV -> TOSLINK -> Dacmagic Plus -> RCA -> NAD removed the problem entirely, much like using the TOSLINK to connect my Fiio and PC. So it seems swapping out an electrical audio cable for an optical one anywhere in the chain works well enough. Now I just have to decide for myself if I want to sell my Dacmagic Plus and connect to the Fiio with TOSLINK, or keep it as an external DAC for the TV. Thanks to everyone here helping me to see the light (pun intended), and I hope others having similar issues will find this thread helpful.
 
So it seems swapping out an electrical audio cable for an optical one anywhere in the chain works well enough
That pretty much confirms an earth loop.
 
Back
Top Bottom