Would getting a non-bus-powered audio interface solve my audio interface problems?

I've done a lot more testing today. I have put my multimeter between a USB port on the computer and a ground pin on a power bar and got a reading of around 2 ohms. And I have done the same thing with the outer casing of my bass amp and got roughly the same resistance.

I have tried connecting everything through a different wall socket but it makes no difference.

As I said in my previous post, I have put my (schuko) socket tester both ways up in a schuko to type E adapter, and got a reading of "live neutral reverse". I get "live neutral reverse" when I place the socket tester in every wall socket in the apartment.

I have tried connecting the bass amp and bass to the computer via my DI box while disconnecting everything else on the computer that connects to a device that is powered from a wall socket, including the HDMI cable, speaker cables, ethernet cable, and every USB cable except that for the 2i2, but I still get computer sounds loudly through both the bass amp's speaker and the headphones that are plugged into the 2i2.

If I just plug a guitar and nothing else into one of the 2i2's inputs, select instrument-level input, and turn the input gain up to full, I hear a prominent 50Hz hum.
 
...
If I just plug a guitar and nothing else into one of the 2i2's inputs, select instrument-level input, and turn the input gain up to full, I hear a prominent 50Hz hum.
If it's an electric guitar with single coil pickups this might not be unusual, but I don't know what "prominent" means to you.

Things like lights (CFL bulbs for instance) and other devices (light dimmers, etc.) on the circuit can introduce noise or get picked up by the guitar. If you rotate around while holding the guitar, does the level of hum change? If it does, then the guitar is contributing. Try something with humbucker pickups to see if that changes things.

Hope the electrician can determine if you have a line noise or even wiring issue.
 
We need a clip to establish a 'Signal to Noise Ratio' Please proceed as follows.

Setup as normal with guitar in DI jack. Put guitar controls on max. Set the gain on the AI to give -20dB fs in your DAW whilst 'picking a riff'. Stop playing for 20 secs with hand on strings. Take hand off strings for 20.

Attach clip as 320k MP3 please.

Dave.
 
We need a clip to establish a 'Signal to Noise Ratio' Please proceed as follows.

Setup as normal with guitar in DI jack. Put guitar controls on max. Set the gain on the AI to give -20dB fs in your DAW whilst 'picking a riff'. Stop playing for 20 secs with hand on strings. Take hand off strings for 20.

Attach clip as 320k MP3 please.

Dave.

I haven't got permission yet to attach audio files, so I've uploaded the clip to OneDrive. I recorded it on bass. Unfortunately, my bass has a problem with its electronics (unrelated to the one I am posting about here) stemming from a failed repair and needs to go back to the luthier because it distorts at certain pitches. This limited the notes I could play in my sound clip.

I did try plugging in the guitar first, and something odd happened. The guitar has a three-way switch for "rhythm", "treble" and a middle position. I only got the horrible computer noises when the switch was in "treble" position, not when it was in the other two positions.

This recording is what I hear when the ground is lifted in the DI box. When ground is enabled, the unwanted noise is much louder.
 
Classic data bus noise. The noise travels from the computer down the USB cable to the interface and is usually worse when the USB 5V rail can't supply quite enough power and the voltage drops a little, creating conditions making the noise worse. Sometimes though, it's just the interface and computer, and there's little you can do. You can try chokes - they sometimes help by creating a low pass filter that gets rid of some of the hash travelling on the cable data line. An old loudspeaker magnet can work pretty well. Remove it and wrap a number of turns of the usb cable through it and then insulating tape it to fix it in place.

My colleague had an Lexicon interface that was making this noise, and nothing I tried cured it. On my computer - totally silent. On his horrible - every disk access made the data shifting around go down the USB cable and then the interface let it through into the audio circuitry.

Using a interface with a power supply means the computer doesn't have to supply power, so can often improve things, but in all honesty - if the computer is prone to sticking this rubbish down the USB DC supply, there's not a lot you can do.

It may mean buying multiple interfaces till you find a stable and silent one. Noise on the computer ground is very common - especially when people use the cheap and nasty internal sounds that pop out of the sound card. The minute you compound it by having different potential grounds, and your 2 Ohm difference is quite capable of injecting these noises because of ground differential voltage - it produces current flow, and if the ground of the computer is dirty - it travels.

They're not that common, but you can get transformer coupling boxes that provide complete isolation between input and output and they can be very effective. DI's simply lift the ground, and this won't remove noise on the actual signal, just breaks the multiple grounds.

Things like this can be very useful for quietening down nosy audio kit - but again, not guaranteed.
https://cpc.farnell.com/art/dti/converter-audio-signal/dp/DP28393?mckv=s05xLZWXO_dc%7Cpcrid%7C224680541983%7Ckword%7C%7Cmatch%7C%7Cplid%7C%7Cslid%7C%7Cproduct%7CDP28393%7Cpgrid%7C47726778220%7Cptaid%7Cpla-890742062373%7C&CMP=KNC-GUK-CPC-SHOPPING&gclid=CjwKCAjw5Ij2BRBdEiwA0Frc9SjyYA214KfmAyEOPqQavJZXole7ne-ABkrlRgk773Ey07vSa3fJhBoCRXMQAvD_BwE
 
To reiterate, this data noise only happens on the laptop, right? What model laptop do you have? Are you able to try a different one? Borrow from a friend? Fellow isolationist? i don't know enough about european wiring to be of help with the apartment's electrical issue, but I say try another laptop. I have heard this problem on cheap laptops before.

Also, I thought the bass audio might be pretty low. Where are your meters at when recording? You can post a screenshot.

As for posting to clips into your posts, you should be able to do so and you do not need any kind of authorization or permission.
 
data bus noise = major irritation.

First computer with ISIS interface late nineties = no noise
Other computers with Firepod at turn of the century = no noise
Newer way more powerful computer = noise
Old laptop = no noise
New laptop = noise
New laptop on battery = no noise

However:
Computer with balanced cables from interface to speakers = no noise
Laptop with balanced cables from interface to speakers = no noise
 
Thanks for your explanations.

The problem is with my desktop computer, not my laptop. When I first posted, I thought I had problems with the laptop too, albeit much less severe and not of exactly the same kind, but now I think that that may have been because of logical errors on my part when I was trying to do my diagnostics. There is certainly nothing that the laptop does that is as in your face as the bus noise from the desktop machine (except for the very loud 50Hz buzz when it is connected to the bass amp via the DI box and the bass amp is powered off). When I recorded the sample on the desktop computer that I posted above, I tried to record playing at around -20dB, as instructed. The bus noise really is that loud, and without the ground lift in the DI box, it is even louder.

The motherboard of the desktop computer is an ASRock 980DE3/U3S3. The power supply is a Corsair CX430. I have googled for audio problems related to this motherboard, but I haven't found anything. I have googled reviews of the power supply and the general consensus seems to be that it is cheap and mediocre, but not terrible. One important question for me is: Is the problem more likely to be with the motherboard or with the power supply? If it is the power supply, I can replace that.

I actually looked at the ART DTI the other day, but I wasn't clear whether it was suitable for mic-level or instrument-level signals. Are you saying that it is?

I already have a USB cable with a choke, but that has made no difference.

And after writing the above, I have discovered something new. I only get bus noise with my digital piano and my percussion pads when those instruments are connected to the computer both via an audio cable and a USB cable at the same time. If I disconnect the USB, the bus noise disappears. This means that the problem is more limited in scope than I had thought. For instance, now I can be reasonably confident that if I ever get hold of an old analogue synth, I should be able to record it with my present setup. However if the ART DTI is capable of eliminating the bus noise, there are still present and possible future scenarios, particularly the computer/bass/amp/DI box one, where this would be important.

And about posting audio files here, I don't see a button that enables me to do that, which is why I assumed that I'm not yet allowed to.
 
I have put my multimeter between a USB port on the computer and a ground pin on a power bar and got a reading of around 2 ohms.

Is that not the source of the problem then? USB ground should read < 1 ohm to PC chassis and, by extension, anything connected to mains earth, as far as I know.
 
And about posting audio files here, I don't see a button that enables me to do that, which is why I assumed that I'm not yet allowed to.

Click on the Go Advance button and you'll see more options. You can add it as an attachment or embed from soundcloud or some other place.
 
Is that not the source of the problem then? USB ground should read < 1 ohm to PC chassis and, by extension, anything connected to mains earth, as far as I know.

I think I need to replace the leads on the multimeter first. They are not in good condition and if I just touch them to each other they tend to report a bit of resistance. I didn't realise until now that even 1 or 2 ohms can be significant. Clearly I need greater accuracy than they are capable of.
 
I think I need to replace the leads on the multimeter first. They are not in good condition and if I just touch them to each other they tend to report a bit of resistance. I didn't realise until now that even 1 or 2 ohms can be significant. Clearly I need greater accuracy than they are capable of.

They'll pretty much always report some resistance but it should be less than an ohm.
My cheap-ass meter reports 0.3ohm for lead-to-lead and 0.6ohm for usb ground to PC chassis.
 
Pull the leads out of the meter and give them a buff with a scouring pad, squirt some De-Oixt in the meter sockets if you have it. WD-40 if you don't. Push and pull the plugs a few times.

The lead resistance does not matter so long as it is fairly constant, just subtract it from the reading.

To measure sub one Ohms accurately you need a four wire system with 'at the component' sensing. But you don't need that!

Dave.
 
Pull the leads out of the meter and give them a buff with a scouring pad, squirt some De-Oixt in the meter sockets if you have it. WD-40 if you don't. Push and pull the plugs a few times.

The lead resistance does not matter so long as it is fairly constant, just subtract it from the reading.

To measure sub one Ohms accurately you need a four wire system with 'at the component' sensing. But you don't need that!

Dave.
The leads are a bit further gone than that. They broke many years ago and are held together very poorly with sticky tape, which makes it hard to get consistent readings. I didn't know until I googled yesterday that the plugs on the leads are a standard size and can be easily replaced, which is what I'm going to do very soon.
 
Thank you all for the effort you put in a couple of months ago to help me try to solve my bus noise problem. Here's an update:

I finally found the solution after I had given up on getting my current hardware to work properly and had posted on another forum asking for recommendations for an audio-friendly motherboard to buy as a replacement. Someone suggested checking that the little tabs on the plate that fits over the IO ports at that back of the machine were making contact with the motherboard components behind them. So I had a look and found that the plate was jammed over the serial port instead of seated neatly around it, with a couple of the tabs touching some of the serial port pins, and other tabs just sticking uselessly into the air. So I took the machine apart, positioned the plate correctly, and the noise has now become a tiny fraction of what it was before. I used to be able to see it on the level meter of my DAW when my bass was plugged in, but now it is so quiet that it no longer shows. There are still a couple of situations where it remains loud enough to spoil a recording, but for most of the things I would want to do, it is no longer a problem.
 
Damn, You'd think, or hope, it would disappear completely but I'm glad you saw a substantial reduction, at least.
I don't think I would ever have thought to suggest checking something like that!

Great catch, whoever it was!

Happy recording and thanks for coming back to let us all know. :)
 
Well, the reduction in noise for when my bass is plugged into the audio interface has the bus noise now at barely more than the noise floor of the bass and audio interface, so it is extremely faint and I would have to do some really aggressive compression for it to be a problem. It is still quite noticeable and intrusive (though not nearly as bad as before) when I have my digital piano connected by an audio cable and a USB cable at the same time, but that is not something I am ever likely to have a real practical need to do. If I ever do find that I want to record a MIDI-controlled synth, I will use a DIN-5 cable rather than a USB cable for an electrically isolated connection and therefore no loop. I've also got an isolating transformer, so that gives even more options for bus noise reduction if I ever need them. So I don't feel inhibited about using my desktop computer anymore.

One of the nice things about this is that I can now use Jamorigin MIDIBass without it mistaking all the fizzes and squeaks for notes.

The person who made the suggestion about the I/O shield said he had first read about it only the previous day, so the timing of it all has been pretty cosmic.
 
Back
Top