as promised: thread on "air flow" vs. "air column vibration"

[peterbas]

.

[bloke]

Fourier analysis is a mathematical technique so it is a math problem, has nothing to do with real life.

ok...Don't kill the messenger (ie. the person attempting to explain someone's else suggestion, but with a neutral attitude), but...

it is a math problem, has nothing to do with real life

uh...
Whether-or-not this principle has anything to do with acoustical phenomena, you might (??) wish to tighten up that generalization just a bit...
or not.

me...??
The next time I pull out the tuba (still packed from a Monday rehearsal), I'm going to run through those experiments you suggested.

[donn]

I think what he may be saying, is that while Fourier series are commonly used to arrive at a function that resembles nature, they're artificial - the validity of the analysis is only in the degree of conformance to the function, it isn't because the Fourier series analysis is related in some way to the basis of the function. The GIbbs phenomenon apparently has to do with this analysis. I sure don't understand this even to the extent I pretend to explain it, but it is something of a mystery to me as well how it enters in here. If we're talking about oscillating functions and their potential role in pushing the slide around, I'd think it would be about the main partials, not about effects derived from function discontinuities, but maybe that shows what I know.

[peterbas]

.

[matt g]

Fourier analysis is a mathematical technique so it is a math problem, has nothing to do with real life.

Thanks, I know that.

The idea is that there’s some sort of transient response when a person blows a raspberry into a piece of expensive tapered tubing.

When a person articulates the beginning of a note, especially a low/loud one, there’s going to be some broadband content that’s gotta be forced down said resonating tube. Especially when the resonating tube only wants to resonate at a certain batch of frequencies, ironically a Fourier series.

Possibly, said resonating tube, especially when it has some low effort expansive capacity, might make some sympathetic effort to accommodate this weirdness by moving a touch?

Again, I don’t think just spittooning a watermelon seed down the leadpipe would move a slide like this. There’s gotta be something else that’s at the wavefront making something fun happen.

For the record, I’ve had this happen also. When I was a kid in high school, we had King 2340s. I forget the piece, but I was able to make the lower slide on the first valve drop out (due to it being a bit loose to begin with, definitely not precision aligned) but only while playing. Armed with only algebra 2 and some very basic physics at the time, I left it to the category of unexplained phenomena.

Just something to consider.

Apologies for a tardy reply, but I was off doing something else more important, interesting, and enjoyable.

[peterbas]

.

[bloke]

You guys don't make me post that Pee-Wee Herman clip, again...

[bloke]

@peterbas, why the animosity when you post here? Sure, I was a bit flippant in my response, but prompted based on your post. I get there is also a language and cultural difference, but whatever.

To answer your questions…

Reason I mentioned it in the first place was because it reminded me of what happens when the Fourier series is used to approximate a square wave, in which a square wave approximates how @bloke is turning on/off the tuba. That’s it. Something like fodder for a “Gedankenexperiment”, if you will.

Regarding your experimental prompt, my own experience was that it was very loud (i.e., ff or louder) notes, and D below the staff, iirc, that caused this. Middle of the road dynamics and crescendos did not. No idea about Joe’s situation. Just the idea that a highly energetic start to the note would provide a sufficient transient to elicit this effect.

Regarding your last quip, no one interrupted anything of yours, specifically me. My point was that my tardy reply was due to being involved with my family, something far more important than quibbling about on a niche forum.

[bloke]

@matt g

re: "blowing the lower #1 slide out on that ancient King school tuba"

In my experience is that (ignoring damage) King tuba slides (40 years ago and prior, pretty much) were precision-aligned.

I played a (absurdly) worn-out-valves 1240, but (knowing that I know now) the slides (though snug) were quite well-aligned.

No one noticed this superb alignment, as there was no upper #1 slide at that time (thus: no one was monkeying with any slides on the fly - when playing).

I'm suspecting that (several summers) your King tuba's lower #1 slide was cleaned in a "shop" by buffing it really "pretty" with tripoli buffing compound and a spiral-sewn wheel - which defined it as both precision-aligned and loose-fitting.

Knowing NOTHING about air flow principles/theories (etc.), I can imagine that air in a tuba is under more pressure at the end where there is a c. 1/3 square inch round tube - vs. the other end (bell throat) where the tube could have expanded to 130 (etc.) square inches in area...

...with the #1 slide being in the first approx 1/8th (long mouthpipe tube, on this model) of the expansion (and not even geometrically, but only in length) and with reduced friction on its moving parts to boot, I can imagine (without even having observed it) this slide "jumping" (again: with VERY LITTLE knowledge of air flow theory/principals).

I'm now remembering (so quickly forgotten) that I ended up putting a "brake" on the SAME INSTRUMENT'S main tuning slide - which I HAD BEEN "shooting out" during similarly fortitude-packed passages (composer/arranger-marked to be exuberant/jolly xmas phrases).

bloke "duh"

[matt g]

Possibly the slide was that well aligned. The 2340s I was playing were only a few years old at the time. They were likely made somewhere between 1986 and 1988.

I think I ended up using tape or string to hold it in place. I think that’s when I learned how sticky that old Selmer brand cork/slide grease was and how liberal use of it slowed the problem.

I’m pretty sure I tried to big bad wolf the slide out without buzzing, and it never budged. It had to be done while playing.

[Rick Denney]

I scanned the thread, so probably missed this or that.

When we think of pressure and resistance, we absolutely must think in the frequency domain. When we blow through a tuba without a buzz, we are creating a constant flow, which is the one thing that does not produce more sound than a quiet hiss from whatever turbulence exists (which may well be in our lungs, throat and mouth as much as in the instrument).

We are creating an oscillation of pressure, not a flow, to produce sound. But to create that oscillation, we have to provide enough flow to fully feed the micro-flows back and forth in the oscillation. This overall airflow acts like biasing a transistor--it moves the buzz into the domain where it can function. The pulses from the buzz are pressure pulses, and a pressure pulse is a wave of denser air traveling through a medium of "average" air (this average is not atmospheric pressure, by the way--Fletcher and Rossing showed that the average pressure is a gradient through the instrument becoming similar to atmospheric pressure only at the bell). We have to store enough air quantity in the valley to completely fill out the peak flow in each pulse. That quantity of air needed to fill out both the pressure and flow of those successive pulses seems to me like power--the pressure is analogous to voltage and the flow is analogous to current, and power is the product of the two.

Resistance also lives in the frequency domain, and profoundly so. It's not about constrictions in cross section, it's about nodes and antinodes, which are in different places at different frequencies. At the nodes, flow is zero while speed (which oscillates) is at its highest. At antinodes, speed is (at that frequency) zero while flow is at its highest. Turbulence is also possible, in addition to noise (meaning: vibrations unresonant with the incoming signal) can be produced by leaks of various sorts. Resistance at a given frequency is called impedance. The bell is actually an impedance matching device--it attenuates some frequencies while amplifying other frequencies against the "wall" of still air outside the instrument. Base air flow has not much effect on that--the speed of airflow at the mouthpiece shank will be a couple of hundred times the speed of the airflow at the bell. The pulses at the mouthpiece have to have enough power in them to allow their unresonant parts to be attenuated and their resonant parts to be amplified. A buzz with too little power won't be able to fill out the pressure pulses and their associated waveforms.

So, it seems to me that air flow is an enabler more than a creator. It enables the buzz and provides the power to fill out the pressure pulses fully. Without sufficient air flow, the buzz becomes more difficult to sustain and the signal loses harmonic content sufficient to make use of the filtration that occurs in the instrument and bell. Too much pressure without sufficient airflow will make it louder (ish) but will not fill out the waveform, and lots of higher and less resonant harmonics will jump in as a result of the waveform being truncated in whatever way. Too little pressure and too much airflow might not even be possible given the impedance of the embouchure, mouthpiece and instrument. Enough airflow will create whatever pressure it needs against those constraints.

But I think the story of airflow might be done by the time the waveform has made it out of the mouthpiece, though what it takes to get out of the mouthpiece might be quite a bit different for different instruments. If all the impedance is in the embouchure, the amount of airflow needed will be similar to that of free-buzzing. As the impedance of the instrument increases, the ability of the instrument to make big waveforms might diminish. I think that is what we perceive as being "stuffy". So, the instrument needs to push back enough to resonate the buzz but not so much that it attenuates the biggest waveforms we are capable of producing at the airflow we can provide.

I struggle to provide the airflow to make big sounds because of impedances upstream of the embouchure. A bigger tuba can compensate for that by providing greater amplification of smaller signals. But when I try to push through too much upstream impedance, I am probably creating pressure at the expense of flow, leading to a truncated waveform lacking the power to be fully formed. That makes the sound edgy and bright--blatty--rather than big, full and loud. Blattiness can be piercing but not powerful.

Rick "my arm-waving on the subject" Denney

[bloke]

I really appreciate the insights, and believe (unless my ego fooled me) that I may have understood 2/3rd of Rick's contribution.

not-at-all topical, but yet (not quite) non sequitural:

My wheel covers were taken at that Memphis-located community band Superman-Main-Title-playing rehearsal (even with TWO police cars staying RIGHT THERE - at the dollar store and Walgreens across the street, which are regularly robbed)...Maybe (??) they were pissed off that my car is manual shift, and did it out of spite...
Anyway... Unless I play through that piece at home, my 1st slide probably won't be jumping like that again, anytime soon...

[peterbas]

.

[tabert]

I was able to recreate the "1st slide getting blown out" phenomenon with both all-valves-down ugly, loud playing, as well as just blowing air through the mouthpiece. In the second case, I did need to create additional resistance (ie. valves 2 and 3 down in addition to 1st valve) in order to get it to occur.

[bloke]

Thanks...
That's great !

Mine wasn't allowed to achieve an erection, due to me playing C below the staff (B-flat tuba) with 1-3, and (as outlined earlier) having reconfigured my #1 circuit so as 1st valve C (an octave higher) is in tune with #1 all the way in, and 1-3 C (an octave lower) nearly all the way out...with further saltpetering restrictions being a slide stop rod and (again) my thumb and index finger on that slide at all times.

-------------------------------------

I'm now even more convinced (perhaps, certain...??) that THE VERY SAME THING was occurring with my MAIN slide (after I shortened it's ferrules (for easier A=440 tuning achievement in the dead of winter) and (me being me) doing "too nice" of a job of aligning the main slide - after rebuilding it with shorter ferrules.

People were telling me to (simply) "mess up the slide"...but I'm not going to mess something up - when it's just about perfect...
...thus: the (considered silly by many, I'm sure) "slide brake" that I installed...which was NOT effective UNTIL I contact-cemented coarse sandpaper to the outside slide tube:

The shortening of the main slide ferrules (which defined quite a bit of disassembly) is shown here:
(luckily, at the very beginning of this LONG thread)
viewtopic.php?t=5283

and the subsequent installation of the main tuning slide "brake" is documented here:
(skipped to page 3 - since there's a PICTURE of it here...)
viewtopic.php?p=56349&hilit=brake#p56349

[peterbas]

.

[bloke]

I believe it has a great deal to do with the machinery itself. I don't think that's going to happen (neither with a frequency nor with air only) unless both pairs of slide tubes fit like pistons, in addition to being absolutely parallel and coplanar.

That looks like a pretty expensive model of tuba that they used to demonstrate it.