I simply have the Zoom sitting on a desk, running on AAs. The room is silent. Can anyone identify what might is generating this sound?

http://www.legacyhdv.com/noise.mov

Thank you!

What sound? It's usable Playback problem on the OPs end? I opened it in Sound Forge and do not hear (or see on the waveform) any major issues. What do you mean by "silent" ?

I hear a tiny bit of noise, under the voice, near the end of the clip. There is not enough to identify.

Please post just a mono audio file that is longer, taken from your original audio recording, that has a significant amount of noise. Don't waste file size by including the video... that is entirely useless.

Here is the wav file with the time codes where you can really hear it. My best guess is that the wiring in the Zoom is going out. Also I noticed that it's a straight Zoom H2. Again, no lav, running on batteries. I have the H2n in front of me on the desk.

http://legacyhdv.com/eric.wav
02:45-02:52
03:20-04:00
33.35-34.02

• The first two examples are a high frequency buzz, and because I don't hear or see any on/off pulse modulation, I would suspect it's harmonics from a lower frequency source, rather than bleed from some digital signalling system. Thus, I would rule out things like cellphone interference.

The period is exactly 960 samples, which translates to a period of exactly 0.020 seconds, and a frequency of exactly 50 Hz. Thus, I'd suspect this is harmonics from some 50 Hz source. I see you're posting from the US, but if this was recorded in a country with 50 Hz power (e.g. the UK or Europe), I'd immediately suspect harmonics from the power distribution system. (Harmonics like this can be generated by solid-state lighting dimmers, or similar equipment.)

Another possibility is a monitor that's running with a vertical sweep rate of exactly 50 Hz, although I'd think that's rather slow for typical modern monitors. (60 Hz was common 15 years ago, but the frequency has generally gotten higher over the years.)


• The third example is different. It is not steady; it's pulsed. I hear more of this from 34:50.8 to 34:51.3, and in the following minutes of the file. This may well be interference from a cellphone, sending out RF to stay in contact with the cellular base. It may be some other form of RF communication, such as a Wi-Fi signal. It may even be crosstalk from a poorly shielded cable carrying some other form of digital signals.


• Then from 35:29.4 to 35:29.9, there is more steady-state low frequency noise. This spectrum is different from the buzz at the first two instances, there's a lot more lower-frequency content, and a lot less of the upper harmonics. It sounds more like a hum, rather than a buzz. Even so, the frequency is again very close to 50 Hz, so I'd suspect the same source that caused the earlier problem: power line noise, or vertical sweep from a video monitor.

• Also, observe the section from 08:10 to 08:45. I know I asked you not to send video... the file would be much too big. But can you tell us what was happening at that point in time? It sounds as if the talent is moving around, or possibly writing something... possibly manipulating something... possibly touching a keyboard, monitor, or some other electrical equipment. Whatever is happening there, the noise pickup changes from none, to very obvious, back to none, a few times. Can you check your video file and see what the talent is doing, what he's touching, where he's positioned, etc.?


•• Does the above give you any clues?


Finally, since there are two voices, I'm guessing that this was recorded with two separate mics. You posted a two-channel file. Yet the two tracks are exactly identical. Can you please explain, in detail, how this file was created? How many mics? Plugged into what? How and where was the mono mix created?

Thanks.

Thank you Greg!

Evidence at the crime scene:
Two 20" 60 Hz monitors five feet away on in sleep mode
Mac Pro ten feet away in a closet on
Cell phone two feet away
My mic (Zoom H2n) two feet away
No nearby power lines
AC fan off
Six of those curly light bulbs on (brain tired)
No monitoring - using camera mounted shotgun and my mic as redundancy

The mono-mix was done by the Zoom itself as a menu option.

>observe the section from 08:10 to 08:45
He was writing notes from his cell phone which was 2-3 inches from the mic.

From what you said, it sounds like either cell phone interference, "crosstalk from a poorly shielded cable carrying some other form of digital signals" or both. The Zoom is old and road weary too. Thoughts?

If he was playing with his cellphone from 8:10 to 8:45, then it appears the 50 Hz buzz is coming from the cellphone. Is it a smartphone? Was it on during the entire recording (perhaps used as a mini teleprompter)? If so, that's apparently where the buzz originates. Cellphones are usually "bad juju" where audio equipment is concerned.

(You say "AC fan off" but there is significant 120 Hz background throughout a lot of the file. Presumably this is a motor running somewhere... even if it was a refrigerator in another room. So there is certainly some 60 Hz equipment running. But that is not the source of the 50 Hz buzz.)

Please clearly and thoroughly explain what mics were used! You say the Zoom was "two feet away." Away from what???

Were you using only the internal mic(s) in the Zoom?

Were you using one or more external mics? If so, what kind? Connected with what sort of cable/connectors?

I really want to understand what you used for the sound source that was recorded on the Zoom... every tiny detail.

Once we know this info, we can figure out how the buzz got into the recorder, and how to avoid it in the future.

I'm glad to blame the cell phone. That's an easy fix.

We did have the fan on before we started the interview officially. I'd forgotten about that.

He used a Zoom H2 with nothing plugged into it, and I used a Zoom H2n with nothing plugged into it. No lavs. The Zooms were two feet away from each other, as Eric and I were two feet away from each other. We only used the internal mics. Both mics were on and recording.

Can the buzz be from the cell phone too?

If I understand you correctly, you initially had two separate recordings from two separate Zooms.

But the file you posted was a two-track file, with two identical tracks.

How did you get from two entirely separate recordings, to a single mixed recording?

The Mono-mix function in the Zoom H2's menu creates just one wav file.

Yeah, but you said you had TWO zooms sitting on the desk. And you posted ONE file.

So I don't understand about the file you posted.

Did it come from just ONE of the Zooms? Or did you take the two files, from the two Zooms, and mix those two files in post?

Sorry if I'm being dense here, I don't like to make any assumptions, I like every detail spelled out -- twice I guess. ;-)

Thanks Greg. I haven't done any mixing. The Zoom H2n wav (the Zoom for my voice) was fine. I can upload it if you'd like. The one that you've listened to is entirely from the Zoom H2 (Eric's mic) with no lavs and nothing plugged in. It sat on the desk. Eric's cell phone was in his pocket once we started the interview.

The is after the fact but I'd advise of NOT using the H2's mono mode with only one input. It mixes in the noise of the unterminated channel, which can pick up RF spill and other extraneous sound. It would be better to record in the normal stereo mode and remove the unused channel in post.
Some recorders (like the Tascam DR40) have a single channel/file mono mode where this would be not an issue.
One could try Noise Reduction on the current issue or replace/overwrite good 'room tone' in pauses to make it less noticeable. An expander processor may help as well, attenuating the BG noise in those pauses.

Rick,

I was wondering about exactly the same thing... mixing an open input into the final file. That's why I repeatedly asked what mic(s) he had used, and how he had arrived at the mixed-down file. (BTW, I think Zoom is pretty stupid to mix down to a stereo file with identical channels. Why not mix down to a single mono file, and use half as much memory?)

But the OP says he's using the Zoom's internal mics. So in that case, wouldn't he be mixing together the two internal mics, into one mono mix? Nothing would be unterminated.

So in that case, wouldn't he be mixing together the two internal mics, into one mono mix? Nothing would be unterminated
Right you are Greg! I was thinking "external input" Duh..
Could it be some kind of artifact from Zoom's mono mode summing architecture?

However my comment does still apply to the external input scenario.

Yeah, I don't use mono anymore, so I'm not sure why I had it turned on.

Rick,

90% of the noises I heard were almost exactly 50 Hz. There was a lot of higher-harmonic buzz, but if you looked at it closly, the spikes were 960 samples apart. 960/48000 = .02 seconds = 50 Hz. That ruled out USA power line as a source. When Dana said the loudest noise happened when the talent was using/touching his cellphone, that pretty much clinched it. Apparently his cellphone has something that's scanning at exactly 50 Hz (probably the refresh rate for the screen).

There was a bit of predominantly low frequency noise, here and there; that, too, was exactly 50 Hz. Same source.

Then there were occasional bursts of noise that do sound very much like a cellphone when it's communicating with the cellular tower. So I'd say those were pretty unambiguous.

As a lot of us know, cellphones are really bad juju in an audio environment. Turn it off, and move it ten feet away, before you start recording!

And yes, Rick, you are absoluely right: If you recorded from the mic jacks, and had only one mic plugged in, the other (unterminated) input could be very susceptible to noise pickup. And if you recorded in "mono mix" mode, that noise would be permanently mixed with your good mic channel. IMHO, that's a very good reason to never use the "mono mix" mode in the field.


Dana,

I'm saying it's dumb of ZOOM to design their recorder that way. (NOT that it's dumb for you to use the mono mode.)

There may sometimes be a valid reason to produce a file that's a mix of the two channels. But once you make a mix in the field, there's no good reason to record exactly the same information on two channels of a stereo file. Zoom should have designed the firmware so the "mono mix" mode records a single-track mono file. That way, the file size will be half as big, and you can get twice as much time on a given card.

Anyway, I'm 99% certain that the cellphone was the source of all your problems. (And the file from your Zoom, which was two feet farther from the talent, had no noise at all, right?) Of course if the recorder can pick that up, it might also pick up other noise from something like an old CRT monitor, or... who knows what?

But for sure, before you start recording, be sure all cellphones are completely off not just in "silent" or "airplane" mode. And then put them on a table that's at least ten feet away from all your audio gear!

--

BTW, I've attached two files to show what I looked at. The first one is a short section of noise. You can see the time scale at the bottom, and find the exact same thing on the original file.

The second one shows the results after I ran it through a HPF to get rid of the LF component, and make it easier to locate and mark the spikes. I used the negative spikes, they're obviously much more apparent.

Note that I selected an area that's exactly 100 cycles long. Note that the total length of this file is 96,012 samples. So 96,012/100 = 960.12 samples per cycle. 960.12/48,000 = 0.0200025 seconds/cycle. That's the same as 49.993751 Hz. As I said, almost exactly 50 Hz. (If I hadn't checked that, I might have mistakenly thought it was harmonics from the AC power line; but since I know our power line is 60 Hz, obviously the power line was NOT the source of the problem, which leads us to look further... then we find the cellphone as the culprit.)

Wow. I learn so much from you guys. Do you think having the two 20" 60 Hz monitors 3 feet away from my H2n would cause problems in the future. They're both fairly new, and I use them while recording.

If they were old CRT monitors, there's a fair chance they might cause problems.

If they are LCD monitors, I think you're probably safe.

Different mics will have different amounts of noise immunity, too. In general, if you're using an external mic, you should use one that has balanced wiring (so it will definitely need an XLR connector). Unbalanced wiring is much more susceptible to electromagnetic noise pickup.

And there's a sad little footnote to that. Almost all modern clip/lav mics have electret capsules, and the wiring from the capsule back to the power pack is almost always unbalanced, which makes them much more susceptible to noise. Usually the wiring from the power pack back to the recorder is balanced, which is much safer. A cellphone in the pocket, a few inches from the wiring of an electret clip mic, can be a real disaster.

My best advice is: test in advance. Pick the mic you're going to use with those new monitors, and record some test tracks while moving the mics around at different distances and locations in relation to the monitors. Do your best to keep the mics away from any power supplies, power wiring, motors, etc... and cell phones! If your test environment is exactly the same as your actual recording situation, and the tests are 100% clean, then your final tracks should be clean, too. And if you catch a problem during the tests, at least you have some time to work it out before your actual recording session.

I agree with your suspect cellphone Greg.
FWIW, Way back when... I recall getting frequency spikes around 15kHz when close to (60Hz) CRT monitors. As well as electromagnetic interference when in close proximity to transformers and A/C lines, depending on certain (questionable) cables. At one point I picked-up the Imus show's IFBs when working across the street from 2 Penn Plaza, but that was the hotel's mic cables.

Rick, that would have been the horizontal sweep, at 15,750 Hz. (30 x 525 lines)

I remember hearing the horizontal sweep on some early computer monitors, years ago, especially in POS terminals in stores. Either the deflection yokes on the CRTs, or the flyback transformers, were pretty badly made. I doubt that I could hear that today, and anyway they're all LCD now.

Yeah Greg, 15.7kHz, it was always clearly visible on spectral graph, it was not very audible though, at least to me, but I always used a LP filter on dialog anyway.

I don't mean that I heard it when I played back an audio file.

I mean I heard it audibly coming from the CRT displays! Apparently some of the flyback transformers or horizontal deflection yokes (both of which operated at 15,750) were not well made, and they actually vibrated mechanically at that frequency... so I'd hear this very HF whistle/buzz actually coming out of the monitors when I got close to them. Some were almost painfully loud. I'm sure glad that's a thing of the past.