Appropriate signal chain for extreme low frequency recording
 

There was once a thread where the topic of subcontra bass and associated frequencies were being discussed - it devolved as most discussions regarding this topic do, but there was valuable information interspersed that I cannot dredge up when I search for it.

So I am starting anew, and am wanting to limit discussion to the technicalities. Please leave opinions of usefulness and artistry for other discussions.

There are limitations to any tracking environment - from analog to digital - and I'd love for the details of each to land here at some point.

I ask as I am in development of an instrument and signal path within an electric bass that will provide access to a 10 Hz fundamental. I know the nature of a bass note and expect content to be predominantly 20 Hz and above, but I want as much of the signal to land as is technically possible.

I am unconcerned about playback at this juncture and would rather leave that undiscussed. I believe that to be as complex and lengthy a conversation as this one potentially is.

My knee-jerk is that a digital environment has the greater capacity, and that longer and faster (bit depth and rate) is better - that is what I recall from the original discussion - but affirmation is far better than a less complete recollection.

Please weigh in . . . .

Skip do you mind being more specific?

Analog tape has a natural frequency roll off, both on high and low end. This can vary; between machines, calibration, tape speed during recording and type of tape used.
Digital is capable of capturing frequencies much lower than analog, if the signal processing and conversion allows for those frequencies to be captured. Therefore it is likely more advantageous to use digital medium when tracking instruments that require frequencies to be represented accurately in the range of 5hz-60hz.
Of course, when it comes to the second or 3rd harmonic, the part of the sound that is more in the hearing than feeling range can sometimes be more desirable when trying to place in a mix with other sounds, and tape has it's own way of coloration and compression that can be very desirable.
Years ago before digital when I recorded my sub tuned basses to tape, (also before proper strings for these tunings), I often used synthetic process in the form of subharmonic generators, like the dbx 120xp for example, to bring back some of the missing fundamental.
I have also experimented with using frequency division (crossover) and multiple tracks with their own compression and eq to get the fundamental frequencies to more of a usable volume.

Hey Skip, after posting a few suggestions in the fEARful forum yesterday I remembered another facet of my old gig, where we did low frequency strain analysis out on the dams. These guys supplied the tools, which I'm sure are well out of your preferred price range though. You could try contacting the Applied Physics Lab at UW; they probably have the same tools (or better) and might be able to do the processing you need on a contract basis, or maybe you could interest a student in doing so for college credit.

Edit: Even my fairly humble EMU 1820M sound card is only down around .4dB at 20Hz, according to web-published testing. My quick at home testing showed -1.5dB or so at 10Hz (this was with one of my DIY preamps in front), but I'm not at all sure I trust that, given that I was using freeware. I'll dive a little deeper if no ready solution presents itself here.

Edit 2: a little research reveals that my old, no longer used Gadget Labs card is spec'ed at +/- 0.1dB at 10Hz. Got an old WinXP or Win98 PC available?

Humans can't hear 10hz
Young one, maybe can hear 20hz
There is no bass note that has a fundamental this low.

Look up "Missing Fundamental". You don't need to record or reproduce the fundamental as it is easily reconstructed in the ear/brain of humans.

There isn't a lot of information in low frequencies. It's not a challenge to record it. Reproducing it isn't too hard as long as you don't get hung up on reproducing the fundamental.

This is where I think most people just miss the point...
you can FEEL it, and that's what it's about.
Seismic activity can be fun!

It doesn't need to be a fundamental, just noise to feel some shaking.

You can't see all frequencies of the EM, light, but feel free to stick your hand in a microwave so you can feel it. :)

Well, the obvious difference is that the feeling from bass is good, the feeling from nuking your hand, not good.
And since as you stated you can't necessarily hear it, yes it is vibration that you feel shake you... in specific frequencies, controlled in performance, along with upper harmonics you can hear.
Don't like to feel bass? There's always guitar.
Anyway, enough derailing- anyone have anything to add as far as techniques to capture these subharmonics?

"Hearing at low and infrasonic frequencies"
http://www.noiseandhealth.org/articl...;aulast=Moller

It says there you can sense low frequency vibrations through your "buttock"

Buttkicker has a device


With the right impedance matching, put it next to where you feel a flutter and record. :)

Best of luck with that Skip! ;)

That's the delivery system once you've recorded the information.
Best of luck Skip, indeed...

No worries on the soundcard and software ends, as it turns out. Just have to do due diligence on the specific pieces. At least one of my DIY bass preamps only measures -1.5 dB at 10Hz, and if I bypassed the input and output transformers I'm sure it could do a little better. I've never had occasion to test any commercial ones that low.

That said, coming down to source....
proper string gage being the first hurdle, Jauqo hipped me to that with some of Gary Goodman's strings and it made a massive difference. Then the next step is going to be pickups that capture the string vibration....
something I haven't delved into but I know Jauqo has so if he's still interested in commenting i'd love to hear his experience with this and dealing with bartolini and others... or anyone else who can shed some light on what it takes to get a pickup to capture these frequencies adequately...

The same thing can be done with a buttkicker

Many audio interfaces as going to filter out the very low. It's not important anyhow. Nothing useful down there.

Sorry, the sub kick actually doesn't really go low enough to capture the frequencies we are discussing here:
" The SKRM-100 Subkick is a low-frequency capture device that picks up the low frequencies (100Hz-2000Hz)"
Although, i've used the device for recording and it delivered a nice response well below 100hz but no where near the frequencies discussed. Thanks.

There's is no audio interfaces that aren't AC coupled and limited to 20hz

You can however follow the what pipe organs have done for centuries and arrange the harmonic amplitudes as to fool human ears into thinking they're hearing the fundamental well below what the actual instrument is producing.

Try a pipe organ VSTi.
Or
The the Waves MaxxBass plugin
It works.

I will be using sensors sensitive to 3/5 Hz - very high impedance - individually buffered with a separate summing buffer. No mag pickups.

Very long instrument with my biggest string set - a 5 string set with my .254 pulling the same tension as a .105 does tuned to E, but two octaves lower. Tuning will be E A D G C - Frequencies 10.3, 13.75, 18.35, 24.5, 32.7

I am convinced that atonal noise or inharmonic noise below the second harmonic will be non-musical to the senses - much as dissonance is unsettling. Won't know til I get there.

It appears I have a good many options as A/D goes - I just need to be certain I have a 192K/24 bit environment to track in.

It'll happen by spring.

Not necessary. The samplerate only affects the upper limit of frequency response. 44.1/24 will work fine.

This made me very curious, as I know the i/o ratings on my universal audio apollo interface are 20hz-20khz, pretty much flat response-wise.
I was wondering if this was limited to these frequencies as you state, or if it was a measurement of flat response before roll off.
So, as a quick experiment I used ableton live's tone generator to produce a 5hz sine wave, ran it out of the interface and then back in and recorded it. While recording, I varied the frequency between 5 hz and 20hz.
There is definitely information going out of the interface, as well as back in, down to 5hz. There is, however, a steady roll off in volume as you lower the frequency from 20 down to 5hz.
But it definitely is there, both leaving the interface as well as being recaptured by the interface.
So, in short, you are incorrect again mr. monkey- it is actually possible to produce and capture frequencies in this range (at least with the universal audio apollo).
If you like, I can post a clip of the results.

http://www.rme-audio.de/en_products_fireface_ucx.php

"Signal to Noise ratio (SNR): 111 dB RMS unweighted, 114 dBA
Frequency response @ 44.1 kHz, -0.1 dB: 10 Hz - 20.6 kHz
Frequency response @ 96 kHz, -0.5 dB: 5 Hz - 45.3 kHz
Frequency response @ 192 kHz, -1 dB: -1 dB: 5 Hz - 90 kHz "