Ground Hum/Noise in my Scarlett interface

[lightcrystal]

Warning: extreme engineering jargon to follow with added Linux jargon. Fun may be had for audio engineers. Total confusion for anyone else in all probability. Including me because I have the engineering knowledge of a mongoose.

About 2 weeks ago I bought a Scarlett Focusrite 2i2 3rd Gen interface and a Rode NT 1 XLR microphone.

My intent was to improve upon my Yeti Blue. I take my narrations seriously, especially my Solo Yorkshire Battles which I have done to 30%. Right now my Aunt has sadly passed away aged 84. I will return to the narrating. I also run Ubuntu Studio. At first I had to upgrade my Ubuntu version. Then the Scarlett interface worked. I used the Unfa video to get the reversed drivers to get the Scarlett working. Then I began to do some recordings with the Rode NT1 in Reaper.

I liked the sound. It is better than the Yeti. BUT it has one massive issue: a constant hiss like white noise/rain. The noise is like the video on YouTube called Fix for a noisy Focusrite Scarlett 2i2 Gen 3

Um. OK. Houston we have a problem. I looked up the Scarlett help pages and two that were relevant:
https://support.focusrite.com/hc/en-gb/articles/211615185-Why-is-there-unwanted-hum-noise-in-my-monitors-
https://support.focusrite.com/hc/en-gb/articles/207547015-Troubleshooting-USB-interface-connection-and-performance-issues

Thus I took out most of the other USB connects to my desktop and plugged my Scarlett into alternative USB desktops [my PC offhand has 2 in the front and about 6 USB slots in the back.] Some slots noticeably reduced the hiss. But it is still there and unacceptable to me. It would pass PL OK. It's not THAT atrocious and my test files still pass the checker. But I want my recordings to be clean.

I have no engineering training. This ground hum issue is new to me. What I gather is that the Scarlett interface when it gets the phantom power from the PC is also picking up noise from inside the PC? Yes, I know that I would get a fail grade in an exam on physics for that. But I am new to this.

There are things that I have not tried. I am using the USB Cable from the box. I could find another one to try. It also doesn't matter how long that cable is; I can control the Scarlett from my Alsamixer in Linux. Interestingly for such a common issue [there are loads of comments below the YouTube video https://www.youtube.com/watch?v=lhzeCacFJbQ ] for both novices and audio pros, I had not heard of this Ground Hum issue.

Has anyone else here had this issue? If so how did you fix it? I know that I could eq it out, noise remove tool etc but I do NOT regard that as an answer. I am told that older houses are more problematic for ground hum. My house was built in the late 70's. The floor is carpet if that matters.

Thank you.

[InTheDesert]

Could you upload an unprocessed sample?

[GrayHouse]

A sample of the noise would be useful.

These sorts of noise problems on USB audio devices were more common years ago. It's usually related to the computer supplying dirty power to the USB mic or USB audio interface. Over the years, designers have got better at supplying clean power from computers, so it's less of a problem these days. Also, good audio interfaces like the Focusrite have a lot of power filtering in them. If you look inside the Scarlett interface, about half the circuitry relates to power - both cleaning it, and generating the 48V needed by an XLR mic.

So it's probably 'digital noise' from dirty USB power rather than actual 'ground hum' (which usually shows up as a low frequency hum on audio equipment powered directly from the mains).

One of the solutions used in the past is to use a powered USB hub for your audio interface. In that video, it's what that iDefender thing is doing. It's splitting the USB cable, and passing all the data wires back to the computer, but diverting the power wires to an external power supply, so the Focusrite is no longer powered from the computer. A powered USB hub does the same thing (and usually more cheaply) - but there are some badly designed ones that you need to avoid.

Do you have a laptop or another computer you could test the Focusrite on to see if it eliminates the audio problem? That would establish more clearly that the problem is with your computer.

[lightcrystal]

Could you upload an unprocessed sample?

Do I upload a sample the same way that I upload a narration and post the link?

[lightcrystal]

Now I can't get the Scarlett to record anything or pick up anything; if I clap my hands next to it the green circle doesn't happen.It was recording yesterday.

[InTheDesert]

Do I upload a sample the same way that I upload a narration and post the link?

Into the tests folder... if you can get it to record. Or do you have an old recording you made when it did work?

[lightcrystal]

Here is a recording from yesterday. Not the worst example; I had used a usb port that was reducing the noise a bit.

https://librivox.org/uploads/tests/myfirst_recordscarlett.mp3

Note that it is all fine in the checker. It has to be the usb connect OOTB that has gone to bits? I haven't got another one like it offhand. I have triple checked everything; I am connected like yesterday yet if i clap my hands next to the scarlett nothing happens unlike yesterday.

Or the scarlett seems to register nearby noises intermittently. I don't understand this.

[lightcrystal]

Another one:

https://librivox.org/uploads/tests/groundnoises.mp3


Some idiot noob questions:
1. I have trouble seeing where exactly the pins are in the microphone to put my connect cord into. Sometimes I seem to get the connect better than others. When I get it right I can clap my hands from a fair way away and the scarlett amp ring lights will go green easily.

2. Does it matter where I have the focusrite? Should I have it as far from my PC as I can or doesn't that matter?

3. Are some differentcolours of usb interface on my PC better than others to use?

4. Is there a safe way to turn off the focusrite? Can i blow the amps? Should i for instance turn down theamp/s and the 45 v phantom power before unplugging the usb connect?

0.0000000000000000000000000000000000000000000000000001% of an audio engineer.

[GrayHouse]

If you plot the spectrum of that noise you can see spikes on the 1 Khz harmonics. That's characteristic of USB noise. Every millisecond the USB issues an End of Packet signal which is (probably) finding its way in to your USB power. The solutions would be those shown in that video you mentioned, or possibly a powered USB hub.

That doesn't explain the soft thumps that seem to occur exactly every second throughout your recording. I'm not sure what's causing those.

I'm not quite sure the problem you have lining up the pins on the XLR cable. I think it's just a matter of practice. Most XLR connectors have a little locating 'key' (a raised bump) that lines up with a corresponding groove on the socket.

It's worth trying all of the USB ports, some may be better than others from a noise position. They should all work with your interface.

I'd be surprised if your focusrite was picking up interference directly. You should be able to position it anywhere convenient. But it's worth experimenting. Stranger things have been known.

It's usually considered to be bad practice to 'hot-plug' any XLR microphone. You should turn off the +48v power and leave it for a few seconds before (dis)connecting the XLR cable. Most equipment has protection components to prevent anything bad happening, but you don't want to rely on that. In my experience you're more likely to damage the preamps in your audio interface than the microphone itself. Plugging and unplugging the USB cable is less of a problem.

[lightcrystal]

If you plot the spectrum of that noise you can see spikes on the 1 Khz harmonics. That's characteristic of USB noise. Every millisecond the USB issues an End of Packet signal which is (probably) finding its way in to your USB power. The solutions would be those shown in that video you mentioned, or possibly a powered USB hub.

That doesn't explain the soft thumps that seem to occur exactly every second throughout your recording. I'm not sure what's causing those.

I'm not quite sure the problem you have lining up the pins on the XLR cable. I think it's just a matter of practice. Most XLR connectors have a little locating 'key' (a raised bump) that lines up with a corresponding groove on the socket.

It's worth trying all of the USB ports, some may be better than others from a noise position. They should all work with your interface.

I'd be surprised if your focusrite was picking up interference directly. You should be able to position it anywhere convenient. But it's worth experimenting. Stranger things have been known.

It's usually considered to be bad practice to 'hot-plug' any XLR microphone. You should turn off the +48v power and leave it for a few seconds before (dis)connecting the XLR cable. Most equipment has protection components to prevent anything bad happening, but you don't want to rely on that. In my experience you're more likely to damage the preamps in your audio interface than the microphone itself. Plugging and unplugging the USB cable is less of a problem.

Thank you! The soft thumps were from the metronome playing in the recording.

[lightcrystal]

Here is my attempt to EQ out the ground noise. I used ReaEQ and the stock setting "move to the back". I use Reaper as my DAW.

https://librivox.org/uploads/tests/groundnoise1.mp3

I have talked to some people who have been professional audio engineers. They suggested using an EQ. They told me that the USB noise blockers etc are basically quackery. The only real ways to deal with ground noise, USB noise whatever it is are:

1. Try different USB slots. Tried. About the same
2. Find the source of the ground loop. Can be impossible.
3. Buy a power conditioner. Expensive but would probably work. Solution that's used in a Pro studio.
4. Put the power connect and the USB connect at cross angles [90 degrees?] Might not be easy. I have the whole hamburger: fibre to premises internet box in the room doesn't leave any spare mains power outlet. I am already using a surge protector/adapter board.
5. Putting the Scarlett Interface as far away from my PC as possible. I might buy a longer USB connect to try this.

I have even tried to find someone who's a audio engineer to come and look at all this. It would make more sense to pay someone and find out what is going on rather than waste money on quack usb bock pseudio-cures. Even when those USB cleaner things work I am not convinced that it did anything; it was probably some other accidental change. Correlation ain't causation folks.

This will be an on going project. Meanwhile I'd like to get back to reading my chapters. All suggestions welcome.

[GrayHouse]

The eq solution is probably the easiest. For anyone using Audacity, the Notch Filter on the multiples of 1 kHz will work well. Once you've found the combination of filters that work for your noise, you could make them in to an Audacity macro for easy use in future.

There's no mystery about the cause or the hardware solutions of your problem. I suspect many audio engineers don't know enough about the internal workings of USB to understand what's happening.

A mains power conditioner won't work because the noise is infecting the USB power inside your computer - it's not a mains-borne noise problem. And it's not a ground loop, even though people often call it a "USB ground loop", so there's no point looking at ground loop solutions. And the Scarlett interface is well shielded in a metal case so it's unlikely the problem will be cured by moving it farther from the computer.

The hardware solutions that work are those that involve disconnecting the (dirty) USB power between the computer and the Scarlett interface, and replacing it with an alternative (clean) power from another power supply. The iDefender would probably work, although I think it's vastly overpriced for what it is (and I think you have to provide your own power supply). And a powered USB hub would probably work because it addresses the problem ie the dirty power.