Audacity "cuts" make pop sounds, not DC correction or zero-crossing

[ohthehugemanatee]

Any time my Audacity 2.4.1 on OSX 10.15.5 cuts or otherwise splits a track, it seems to generate a noise at the spl,it tracks' end/beginning. The noise has no visible amplitude in the waveform, but it is visible in the spectrogram, and audible of course.

Context:
I have a problem with clicks/pops from my analog recording equipment. Suggestions there are welcome but it's not why I'm posting. The real problem is, I can't seem to remove them with Audacity.

In the attached .wav (

Damaged Besame Mucho cut.wav

(352.94 KiB) Downloaded 14 times

) you can see/hear two of them. They are not visible in the waveform (actually the second one is BARELY visible), but in the spectrogram they are clear as day. Since it's not an amplitude click, the "Click Removal" feature doesn't find them.

They are very short (~10ms) so I thought I would try simply silencing them. But select > zero crossings > silence produces a new (also zero-amplitude) pop on either side of the effect. Same when I "cut" them out, or "split" the track.

I've validated that it's not a DC correction issue - both visually and by running Normalize > remove DC offset before trying my edits. I'm selecting at zero crossings... and I am stumped. Suggestions? Thanks in advance for any help.

[DVDdoug]

The noise has no visible amplitude in the waveform, but it is visible in the spectrogram, and audible of course.

I can see something at around 1.052 and another at around 1.172 if I zoom-in AND amplify. (Actually, I'd forgotten that you can zoom vertically by clicking right-clicking on the scale to the left of the waveform. It's a little tricky but you don't have to amplify.)

I don't know what caused that and I didn't try to fix it. If I get some time later I'll play around a bit.

[steve]

The glitches are easily seen in track spectrogram view:

First Track000.png (540.72 KiB) Viewed 335 times

After labelling the position of the clicks, I've switched back to normal waveform view and zoomed in close on each of the clicks. I then apply the Repair effect.

• For editing, I prefer the "join the dots" representation of samples.
• I use "Advanced Vertical Zooming" (enabled in "View menu > Zoom") so that I can zoom in quickly and easily with the mouse.

Before Repair

First Track003.png (20.71 KiB) Viewed 335 times

After Repair

First Track004.png (20.08 KiB) Viewed 335 times

Before Repair

First Track005.png (21.48 KiB) Viewed 335 times

After Repair

First Track006.png (20.37 KiB) Viewed 335 times

And this is the result:

Damaged Besame Mucho cut.wav

(352.94 KiB) Downloaded 13 times

[kozikowski]

I have a problem with clicks/pops from my analog recording equipment.

It's oscillating. It's easy to think of analog defects such as background noise and distortion, but there's a third defect. They can make up their own trash. Basically, they stop following your show in favor of their own show.

One common cause of that is bad power supply. Something wrong with the wall brick or the interconnection cable. If you opted for the convenience and affordability of USB power, not all USB systems are clean and well-behaved. More than one home performer has gotten into trouble by having a USB microphone and the computer not get along.

We publish special purpose effects and filters to mop up some of the more common sound damage.

You can also have simple bad design, but that's far less common.

What's your system?

Koz

[kozikowski]

You can have more exotic problems, too. Do those glitches and noises happen about the same times your phone goes on and off or takes some action? Phones do radiate energy and they can get into analog devices.

It's less common, but there used to be a signature beep-buzz sound that a phone would make while it was negotiating with a tower. Dead givaway.

There are recording crews that won't allow cellphones on the set.

Koz

[ohthehugemanatee]

@Steve - holy cow that is exactly it. I forgot about vertical zooming too, but I can see what you did and I'm sure I can replicate. Thank you!

@Koz - I need a bigger sample size to be sure. Here's the setup in case it says anything to you:

7 X Shure SM58 for vocalists (and one as a room mic)

2 X condensors for the piano - sorry I don't remember the make/model, i'm out of town at the moment and can't check.

All directly into a Behringer Xenyx X2442 USB board. The board is connected via USB to the recording laptop, but as this has now occurred on two different physical machines with different software stacks (down to the OS level), I'm confident the problem is not there.

* I've validated that the mic cables aren't moving during the pops.
* The power supply occurred to me, too. We're recording in an old building in East Berlin, so the power supply is definitely suspect. Not sure how to test this theory, though.
* All cell phones are in airplane mode for exactly the reason you mentioned.
* All the XLR cables are new as of 2 months ago. No visible damage.
* One other possibility: the head of the dual-condensor mic stand we use for the piano is missing a nut. So for the last two sessions, they duct taped it on. It occurred to me that we might actually be hearing slippage from the duct tape very close to the condensors. That might explain why it's not the same in each stereo channel. BUT looking at @Steve's magnification above, that really doesn't look like a by-noise. Even all the way up the spectrum? That's an equipment issue.

> We publish special purpose effects and filters to mop up some of the more common sound damage.

Who is this "we" you speak of? Where can I find out more?

[kozikowski]

Who is this "we" you speak of?

Audacity, but it's not for what you have. A bad combination of microphone and computer can give "frying mosquitoes" or "whine" sound. It turns out if you subtract the most significant overtones and harmonics from a common USB digital signal, you can just about get rid of it. Yes, that's what the digital signal in the USB cable sounds like when it leaks cross-cable into the analog system.

https://www.kozco.com/tech/audacity/cli ... neClip.mp3

I can't find the official one. Here's my copy.

Mosquito-Killer4.ny

(363 Bytes) Downloaded 11 times

that really doesn't look like a by-noise.

No it doesn't. It doesn't look like any other common noise, either.

How close to an airport are you? That could be Ground Control Radar coming around. Radar transmits very brief, very high energy pips and then listens for the return echoes from metal objects.

I got called into a movie production to fix some sound problems. It turns out the show got a super good deal on very nice studio space at reasonable cost on west-side Los Angeles. It was also inside the ground radial system for the 50,000 watt KABC Radio antenna.



You could listen to the station on your teeth.

I couldn't fix it, either. Wrap the actors in lead blankets?

Koz

[kozikowski]

Screen Shot 2020-08-21 at 8.21.08 PM.png (160.29 KiB) Viewed 308 times

I think the personality of your damage and the action of those antennas happens to match. It rotates rapidly and fires pip pip pip pip. If it happens to be aimed to you when it fires, you get one of those holes.

Koz

[ohthehugemanatee]

nah, we're nowhere near an airport (about 11km to Tegel airport, as the crow flies). If there's interference from ground radar, it would be messing up every recording in the city.

We had another recording session at the same space, and tried using a different outlet. No pops this time. Absence of evidence is not evidence of absence, however. We will keep experimenting. Any suggestions on how to rule out power issues as a source?

[kozikowski]

Any suggestions on how to rule out power issues as a source?

No practical ones. I have a voltage stabilizing "Sola" transformer. They "know" what pure wall power looks like and only let through power that conforms to the specification. They make hummy noises, get hot, they're large, and weigh as much as a Land Rover.

And they're expensive.

Screen Shot 2020-08-24 at 3.16.13 AM.png (283.78 KiB) Viewed 288 times

They can be popular in places where wall power isn't stable. I've seen them in Caribbean island towns.

There are extension cables and power strips with filters inside claiming to "clean" the power. They mostly depend on solid state devices that can prevent surges, spikes, and other over-voltage events. Doesn't do a thing if the power itself is "dirty" or inconsistent.

https://www.amazon.com/Surge-Protectors ... ode=761520

Look around and see what other people are doing.

Since you're not running drill presses or other heavy machinery, you might be able to experiment with those devices that make "wall power" from your car.

https://www.walmart.com/browse/auto-tir ... 3947_96391

You'll need to start looking at the watt ratings of your stuff, but I can't imagine this would be all that hard to do. The cheap ones aren't perfect—they make wall power "equivalent" and you may create more noise than you help. But there are some that do a really good job. The phrase that pays is "Sine Wave Power."

Screen Shot 2020-08-24 at 3.38.16 AM.png (63.7 KiB) Viewed 288 times

Even if they're not perfectly stable, if your noise problem vanishes, then you have your answer.

And since there are no problems a good engineer can't make a lot worse, you can get a cheap wall-powered battery trickle-charger for your car in addition to the inverter and do that for as long as you want. You're using your car's battery to filter the wall power. I know it's silly, but that does work.

Koz