An Ominous 21 Hertz Peak (question)

[Ron-Milazzo]

What lurks in the shadows of our perception?

This is about a 21 hertz peak…

See picture below: spectrum analysis of XY mic unedited recording

QUESTION:

“The XY is a condenser mic and of course picks up a lot of sound but how do I know if this 21 hertz peak is part of my performance (vocal/guitar) or if it’s coming from some outside source, or some, internal to the ZOOM, electrical noise?”
__________________________________

I started noticing a reoccurring 21 hertz peak in all my recordings (using a ZOOM H4n portable recorders’ built-in XY mic) recording vocals with guitar, sometimes electric guitar, sometimes acoustic guitar didn’t matter. I’ll start there (at this 21 hertz peak) because it comes through as more ominous than the other peaks which I assume are more normal, but not sure on those either.

My recent recordings are just one track, using the ZOOM’s built in XY mic but on previous I did also use a dynamic mic (input 1) with the XY together, and looking at the raw waveform of just the dynamic mic track, there is no 21 hertz peak … so the 21 hertz peak is coming from the built-in XY mic only.

The XY is a condenser mic and of course picks up more sound…but how do I know if this is part of the performance or coming from some outside source, or some internal to the ZOOM electrical noise?

I can edit it out (and it’s better) but if it is part of the performance I probably should just use EQ or high pass filter to ease the whole sub-bass area down only as needed.

Thanks, … Ron.

[DVDdoug]

The lowest note on a guitar is about 80Hz (about 40Hz on a bass guitar).

It would be a good idea to filter it out.

It’s pretty common to filter-out the low frequencies in voice, and to filter everything except the kick-drums and bass guitar in multi-track recordings.

20Hz is the “traditional” low-frequency hearing limit so unless you’re trying to shake the walls with an explosion effect, anything below around 20Hz should be filtered-out. (Even if you want to shake the walls, most people don’t have woofers/subwoofers that can reproduce it.)

If you are recording in a big room 21Hz could be a room resonance. 21Hz has a wavelength of about 54 feet and I THINK you can get a resonance with a room length/width/height of 1/2 wavelength.

It could also be “rumbling” of heating/air conditioning, or something like that.

[Trebor]

[quote=“Ron-Milazzo, post:1, topic:120074”]
but how do I know if this is part of the performance or coming from some outside source, or some internal to the ZOOM electrical noise?[/quote]

My money is on 21Hz being an artifact, rather than a sound picked up by the mic(s).
If I’m right the 21Hz peak will be on all recordings no matter the location.

[Ron-Milazzo]

All sub bass filtering failed IMO (on this one, my recent edit), the 21 hz peak was best left alone, seemed to just thin out the sound too much: a one track (stereo) recording with live voice and guitar. I also experimented on this one using the ZOOM’s built-in compressor with a 90 rec. level which was a mistake because it really sounded harsh (crushed), and that’s without any editing, and any editing did not improve anything (ie., EQ, filtering, notching). So all I did was balance it and fix clicks etc.

(But) I think I do have this 21 hertz peak on all my recordings with this ZOOM built-in XY mic and tend to agree with Trebor that it’s an artifact, though to be more specific: internal noise, vibration, something, from the recorder itself (though this is unconfirmed until I can record something somewhere else). I will continue to experiment to confirm this and will hopefully be able to update here but probably not because these are being closed to further replies for some reason after a certain amount of time. I guess I could make a new thread with the same title, with a “Part 2.” I record a lot and will probably next try just using dynamic mics (2 of them), one for the voice and one for the guitar…though if it’s acoustic that’s a bit of a chore, getting the level up for an acoustic guitar with a dynamic mic.

(Also) about the sub-bass frequencies: No doubt some of these are from the performance: guitar, vocal, even resonance as DVDdoug suggested; So, if in that range are also ‘other’ noises, I assume there is no way to remove them short of also removing what should be there (?). So, this may be why one would want to up their game and buy better equipment, a mixer etc. (the next level up whatever it is)… and/or build a studio off grid away from whatever may be lurking just outside our perception, in the Sub-Bass Zone … dodododo dodododo.

Fun stuff, Thanks-all! - Ron (ZOOM H4n/Windows 7/Audacity 3.0).

PS, for those who don’t know, a dynamic mic picks up less of the far away noises than a condenser mic, of the two mics usually used in music recording. And in my case, with my portable recorder, the dynamic mic is (like usual) on a cable out away from the recorder as opposed to the built-in to the recorder condenser mic. I think condenser mics are superior though in sound reproduction except when there are other unwanted noises nearby, like in a live setting… where I think they usually use all dynamic mics… but in the old days I think they used condenser mics more (or that’s all they had) and they had to strategically position each player, or the mic, to get the best mix/recording.

PS 2, I can buy a condenser mic and use it with my recorder on a cable… away from the recorder too; … and I think all condenser mics require power (a.k.a., phantom power) unlike a dynamic mic, and my recorder has a phantom power option for this, as I think all recording equipment should.

[DVDdoug]

@Ron-Milazzo

All sub bass filtering failed IMO (on this one, my recent edit), the 21 hz peak was best left alone

A “sub bass” filter MIGHT not remove it, but something like an 80Hz high-pass filter should greatly reduce all of that low frequency junk. (I’d consider sub-bass to be below 20Hz. Since 21Hz is in audible if loud enough I’d consider it low-bass.)

The steeper the filter the more it will reduce it but a “fairly-standard” 24dB/octave filter should be quite effective.

(Also) about the sub-bass frequencies: No doubt some of these are from the performance: guitar, vocal, even resonance as DVDdoug shared; So, if in that range are also ‘other’ noises, I assume there is no way to remove them short of also removing what should be there (?)

Even if it’s “natural” it shouldn’t be in the recording.

I’m pretty sure you’re not hearing it but it’s still “good practice” to remove (or reduce) it.

A lot of microphones have a low-cut switch to filter out low-frequency junk.

I also experimented on this one using the ZOOM’s built-in compressor with a 90 rec. level which was a mistake because it really sounded harsh (crushed),

In general, dynamic compression reduces the dynamic range (or “dynamic contrast”) by making loud parts quieter and/or making quiet parts louder. That makes the signal-to-noise ratio worse.

In practice, it usually “pushes-down” the loud parts and then usually make-up gain is used to make everything louder (including any background noise). And even without make-up gain the background noise becomes more noticeable.

[Ron-Milazzo]

Thanks DVDdoug … I do use high pass filter sometimes. So are you recommending that specifically for the first screen shot I shared, high pass filter at 80 hertz/24 dBs octave?

My guitar is the only instrument in the (already tinny) recording and wouldn’t I lose some bass/low end? Not doubting you so much as puzzled as to when this area is to be taken down, because I have viewed plot spectrums (frequency analysis) of a lot of famous songs that had substantial frequencies there below 80 hertz.

I’ve been leaving that area alone (0-80 hertz) unless it seemed to be a problem (maybe effecting the compressor or limiters’ ability) but this has me rethinking and relooking at other songs again and revisited the notch I first did (on this particular session “Heavy Blues…10/5/24,” currently working on), see pictures below. My ears favored notching over high pass filter on this particular session, all tracks. I also used EQ on the area to finish and get the slope going the right way, rising from 0 to 70 hertz and not the other way around. High pass filter (even at a minimum) seemed to do all this a bit too much for me.

BTW, most popular songs do seem to have the 0-80 hertz reduced to at least below the rest of the frequencies and usually sloping back or at the least evened out, but rarely rising the other way, higher the lower the frequency, as my raw/unedited recordings often do. I’ll include a screen shot of an anomaly too, Blind Lemon Jefferson’s ‘Match Box Blues,’ with it’s defiant sub-bass peak. - thanks, - Ron.

Hertz Range - Names

00 to 20…SUB BASS
20 to 250…LOW END
250 to 1000…LOWER MIDRANGE
1000 to 8000…UPPER MIDRANGE
8000 to 20,000…HIGH END

Blind Lemon Jefferson’s ‘Match Box Blues,’ with it’s defiant sub-bass peak.

[DVDdoug]

@Ron-Milazzo

BTW, most popular songs do seem to have the 0-80 hertz reduced to at least below the rest of the frequencies and usually sloping back or at the least evened out,

“Most songs” also have bass guitar and/or kick drum or other instruments with bass.

I would kill the subsonic noise. But of course you can keep any “regular bass” if you want. Do whatever sounds best! It might be a good idea to make a 70 or 80Hz low-pass filter to reduce everything else just to see what that lower-range sounds like.

You can also play-around with the Graphic Equalizer. Boosting in the 100-200Hz range might give some “nice” mid bass boost… or it might make it sound worse…

Blind Lemon Jefferson’s ‘Match Box Blues,’ with it’s defiant sub-bass peak.

Probably record warp.

That’s certainly not intentional. Wikipedia says it’s from 1927 and there was no bass in recordings from 1927. Certainly no sub-bass.

There were also no tape recorders so it would have been cut directly to disc. But tape really doesn’t go subsonic either.

Even “modern” vinyl usually has the bass rolled-off below abut 40Hz. Many phono preamps have “rumble filters” that usually cut-off around 30Hz (and a switch to disable it). The RIAA playback curve boosts the bass and depending on the preamp design, the boost may be extend below 20Hz and the subsonic rumble & warp will be boosted.

It may have been transferred to tape later, then possibly to vinyl. If somebody digitized their “modern” vinyl, the vinyl may have been warped… That’s my best-guess. I’m sure it came from someone digitizing a record.

…Another point-of-reference: Most subwoofers used live and in dance clubs go down to around 40Hz. It’s a compromise that allows for a more efficient-louder speaker (all speakers involve compromises) so they can fill a large venue with bass you can feel in your body. Many home theater subwoofers go lower and it’s easier to fill the smaller space with deep-strong bass.

You hearing is also about 10dB less-sensitive at 20Hz compared to 40Hz (Equal Loudness Curves). That means you need 10 times the wattage to sound equally loud at 20Hz! And that’s optimistically assuming the subwoofer is flat down to 20Hz.

[Ron-Milazzo]

REPLY TO DVDdoug about 0-80 hertz in popular songs:

DVDdoug wrote: “Most songs” also have bass guitar and/or kick drum or other instruments with bass. "

Ron: Yeah, I know, I have to re-visit and categorize my plot spectrum (frequency analysis) screen shots but many were just vocals with acoustic guitar (country blues), … old, some new.

DVDdoug: “It might be a good idea to make a 70 or 80 Hz low-pass filter to reduce everything else just to see what that lower-range sounds like.”

Ron: Ha… yeah but what am I listening for? I assume if it’s a constant noise then it’s not part of the song and not good … other then that if there’s no sound there then what, what does it matter?

TEST: I did a test cutting w/low pass full strength filter at 70 50 and 30 and could hear music except at low pass 30, which still had the 21 hertz peak. Here are the screen shots:

Thanks Doug … and good info about early recordings sub-bass and subsonic sounds … are subsonic sounds only a digital recording thing then? Am I getting that right?

DVDdoug: “Even “modern” vinyl usually has the bass rolled-off below abut 40Hz. Many phono preamps have “rumble filters” that usually cut-off around 30Hz (and a switch to disable it). The RIAA playback curve boosts the bass and depending on the preamp design, the boost may be extend below 20Hz and the subsonic rumble & warp will be boosted.”

All this still leaves me undecided. I feel my best telling are screen shots of the frequency analysis of similar (famous) songs. I can’t seem to find any books that go deep on sound engineering … and sense a lot of information is just proprietary…no one wanting to share how they really did it…and maybe a lot is lost with time as they pass away, taking secrets, tricks etc. with them. Of course a lot is in the initial recording before any editing too.

Though these subsonic frequencies are very interesting to me, ‘what is coming from where,’ and now as we move into the future with all the man-made things in the air…subsonic waves could be blasting us all the time and we wouldn’t hear it… what about animals? Thanks DVDdoug.