Frequencies below 20hz present in musical recordings.

[Stefano_Milano]

Tonight I tried to extract the tracks of some classical music CDs in my collection in wav. I realized that older recordings have a narrow frequency range. Some more recent have surprised me because the very low frequencies below 20hz are very strong. But how can you listen to them? Why insert them in the recording if they are inaudible and difficult to reproduce by the speakers.
Below the spectrum of an excerpt from the first movement of Bartok's first quartet recorded for Decca by the Tokyo Quartet in 1981.

Second movement of Job by Vaughan Williams by Chandos (2017 - downloaded 96khz/24bit file)

At last Prelude of Tristan by Wagner conducted by Solti (new remaster) for Decca 1960

 

[amirm]

DC offset is sometimes created in FFT measurements creating false results close to 0 Hz. So I am not sure if the graphs are accurate in that regard.

 

[RayDunzl]

Some more recent have surprised me because the very low frequencies below 20hz are very strong. But how can you listen to them?

If you can reproduce them, they are felt more than heard.

Einojuhani Rautavaara
Toccata for Organ, Op.59
Marcussen Organ at Tonbridge School, Kent
Nimbus 5675

16Hz, 32 foot pipe

Below 16hz is just ambient rumblings and the equipment that moves the air to and through the organ...


Hmm... How does it get down to -117dB on a 16bit CD?

 

[MRC01]

The 20-20k human hearing range is a round-off to whole numbers. Most people can't hear 20 kHz, and most people can hear a bit lower than 20 Hz.

With acoustic music, typically energy below 30 Hz is mostly noise - the room, or rumble from HVAC or other noises. But there are some exceptions, and not just giant organ pipes. A concert harp has a 25-30 Hz resonance that you can hear live if you're close enough, and that good recordings will pick up. The lowest piano note has a 28 Hz fundamental, but it's weak even in huge pianos; we hear the overtones more than the fundamental. Foot tapping and some big drums can have significant energy below 30 Hz.

Older recordings often applied a subsonic filter because there was no way to reproduce the frequencies on LP even if they weren't noise. I've noticed modern recordings are more likely to be unfiltered on the low end, with bass rumble below 40 Hz around -50 to -60 dB. Not all the time, but not uncommon either.

 

[DonH56]

Those frequencies are in the recording because they are in the music. Drums, plucked strings, piano hammers, etc. have have low-frequency percussives. Musicians playing together produce beat tones that are low in frequency. The infamous old Telarc 1812 recording had 6 Hz content from the cannons and that was in the early 1980's IIRC. The CD was a boon to LF information as they could (can) reproduce it cleanly instead of being cut off by the medium (LPs) and/or rumble filters. As for reproducing them, that is why I built my first sub ca. 1980 or so to get that bottom octave or two. To me a subwoofer was and is one of those things most notice by its absence.

IMO - Don

 

[MC_RME]

The first picture shows a constant DC value from the used ADC. That was quite common in the begining of digital audio, and also typcial in home DAT recorders. These days one often sees this constant DC in DSD recordings, again a problem of the used ADC.

 

[MediumRare]

Below 16hz is just ambient rumblings and the equipment that moves the air to and through the organ...
Hmm... How does it get down to -117dB on a 16bit CD?

Because the top of the scale is -29 dB for a net Dynamic Range of 88 dB out of 96 possible in Redbook.

OK, teachers, did I get that right?

 

[audio_tony]

I've not been 100% convinced of the spectrum analyser measurements in Audacity.

Have you tried another program?

 

[MRC01]

I've run Audacity's FFT on test signals where I knew the frequency content and they were accurate. The window function can significantly affect the results, so try different ones.

 

[Tks]

I always see this stuff on my RME's analyzer. Like you'll have tracks with nearly no bass, and then you see the 25Hz level shooting up to nearly -15dbFS with some recordings.

I don't hear it most of the time (since I listen to audio at low volume more than half the time), but if I crank up the volume, and some parts of a song are low in the mid/treble ranges, you start to hear the rumble sometimes.

 

[Juhazi]

Time span of the sample for analysis has also some effect, the algorithm makes some averaging I guess.

Yes there are big differences in recordings, also with pop/rock.

 

[Frank Dernie]

we hear the overtones more than the fundamental.

That is true of most piano notes I think.
We have a Steinway Model B and I remember setting up microphones many years ago playing a few single notes to get levels and the overtones were almost as strong as the fundamental, and on sustain the C below middle C had the second harmonic sometimes growing larger than the fundamental as their levels changed.

 

[MRC01]

That is true of most piano notes I think. ... the overtones were almost as strong as the fundamental, and on sustain the C below middle C had the second harmonic sometimes growing larger than the fundamental ...

Same with the bottom octave of most flutes. When the musician plays a big fat sound in that range, the first (2nd, depending on one's definitions) harmonic is stronger than the fundamental, which is what gives it that unique rich "purple" tone quality.

 

[waltzingbear]

one thing to remember is that if it went thru a tape recorder, it won't have much below 15-20 hz. Tape playback heads do not have response down there. Brand does not matter.

the graph of Prelude of Tristan by Wagner conducted by Solti (new remaster) for Decca 1960 shows this clearly.

Alan

 

[Blumlein 88]

I've not been 100% convinced of the spectrum analyser measurements in Audacity.

Have you tried another program?

What makes you doubt them?

Blackman-Harris window is probably better for low frequencies and you need to get the FFT size 16 k or larger to have good resolution at the lower frequencies.

I have tried other software which matches what Audacity shows.

 

[ajawamnet]

one thing to remember is that if it went thru a tape recorder, it won't have much below 15-20 hz. Tape playback heads do not have response down there. Brand does not matter.

the graph of Prelude of Tristan by Wagner conducted by Solti (new remaster) for Decca 1960 shows this clearly.

Alan

That's a fairly true statement - tho check out the curve of the A827 2" at 15IPS on jack's page:
http://endino.com/graphs/index.html

Note at 30ips it's -5dB at 20 tho... and most pro's ran these at 30ips.

BTW - Jack was the guy that did a lot of that Seattle stuff back then... up in Amir's neck of the woods.

 

[Frank Dernie]

Note at 30ipss it's -5dB at 20 tho... and most pro's ran these at 30ips.

you get higher HF limit but less bass and the bass unevenness moves higher into the audible range.
And this is one of the best...

 

[PierreV]

I have been a bit crazy with subs this year and I notice the low bass (often non-musical) content more and more. Foot taps and movement on what I assume are wooden floorboards or podiums in chamber music recordings are my favorite. I feel those environmental noises add a lot to the presence illusion.

 

[waltzingbear]

the best I've seen out of tape is one of Tim d'Paravichi's 1" 2 track Studer C37s. Custom heads running at 15ips and it extended to below 11 hz. (I would not call it flat, head bumps)

Most 2 track of pop music (that includes rock) was done at 15ips, because it did have a better low end. Jazz and classical was often done at 30ips, they needed the increase in S/N.

I wouldn't trust any of these digital spectrum displays. I trust my HP test equipment, but not something that was thrown on as an afterthought. As a general indicator they are wonderful, accurate, well maybe not so much. Especially at zero as Amir pointed out earlier.

oops, the attribution on Jack's tape curves is wrong, there is no such thing as an A827 2 track, that was only made as a multi-track, 16 or 24. The last 2 track Studer made was the A820.

Alan

 

[Blumlein 88]

DC offset is sometimes created in FFT measurements creating false results close to 0 Hz. So I am not sure if the graphs are accurate in that regard.

You can use the normalize function in Audacity and one of the check boxes is Remove DC.