I like to report occasionally on instrument designs that were less than perfectly successful because, as we all know, there’s as much to be learned from the unsuccesses as from the successes. This is one of those reports.
It’s easier than you might expect to make a working version of the sort of beating reed found in organ reed pipes. In some very crude versions, reed pipes made this way can produce a sort of miserable-bovine sound that I find interesting and appealing. So I decided to make a tuned set of such pipes — in short, an organ.
[Inserted some months after the original posting of this article: The resulting instrument can now be seen and heard here.]
My thought was not to make a keyboard organ, but to try and find some non-keyboard way to deliver air to selected pipes; a way that would be simpler to make than a keyboard (making a keyboard is quite a chore). This would give the instrument a different sort of character in its gestural qualities and in the nature of the interaction between player and instrument. I spent some time brainstorming and sketching out various possibilities, and eventually I came up with the following. First, the instrument would work on human lung power. Lung power should hopefully be good enough for such an instrument as long as you don’t intend to play more than one or maybe two notes at a time. And using the lungs is far easier, construction-wise, than other options such as building a pair of bellows. It also has great benefits in terms of minimal added weight and space requirements, less mechanical noise, less required maintenance, fewer human limbs required to operate, etcetera. To direct the air from the lungs to the individual reed pipes, the plan was that the player would use a flexible blow tube having a gasket-like pad covering the far end with a hole through it to let air through. By placing this over the open end of one of the reed pipes, the player could direct the breath into the selected pipe and cause it to sound. Moving it from pipe to pipe, one could play melodies. Very simple! … though admittedly slow and clumsy: it would be difficult to move rapidly and fluidly from one reed pipe to the next, and impossible to play legato. These limitations, I figured, would be in keeping with the sort of slow and unhappy nature of the sound itself — a quality I wanted to bring out in the instrument. I did consider making a dual blow-tube, one in which the tube from a single mouthpiece forks into a pair, one for each hand, with some sort of valving system to cause air to flow through one or the other. This had the potential to make melodic playing more fluid and facile. But a bit of thought told me that the creation of the valve was going to be awfully difficult, so I decided to stick with the single tube system, at least to start with.
So that was the basic idea for this very simple pipe organ: an assemblage of tuned reed pipes positioned in a frame with their open ends exposed, plus the blow-tube system to activate them. The ergonomics of this playing method — bringing the padded end of the blow tube down over the open end of the selected pipe — works best with the pipes arranged vertically, so you can picture the instrument as a set of upright pipes held in a framework in such a way that their open upper ends are positioned at a comfortable playing height of, say, about 40”.
From the start I found myself referring to this instrument as the Sorry-Ass Organ, and that has remained its name.
But there is another important layer of complexity to the way that the blow tube works with the pipes, and to explicate this I need to first discuss how the reeds work.
In this sort of reed pipe, the reed serves as an air-gate, rapidly opening and closing under wind pressure. This allows air from the player’s lungs or other source to flow into the pipe, but not in a steady stream; rather as a series of pulses. When things work as intended, the pulsing frequency comes into agreement with the resonant frequency of the air column within the pipe. This sets up and sustains the vibration and resulting tone.
More detail: The reed lies over an opening in the pipe. Due to slight curvature in either the reed or the surface below, in its rest position the reed doesn’t perfectly cover the opening; there’s a bit of a gap there. With greater air pressure outside the pipe, air flows through the gap and into the pipe. Through a combination of factors including the springiness of the reed, Bernoulli forces associated with the air flow under the reed, and periodic pressure wave fronts arising in the pipe, the reed commences movement, alternately slapping shut against the surface below and then springing back open to let another puff of air through.
This brings us to an important consideration for such pipes. You can’t just direct an air flow at or onto a beating reed pipe the way you can with a flute pipe. To see this think of the most common use of beating reeds, in instruments like clarinets and saxophones. In these, the reeded mouthpiece of the pipe is actually placed inside the player’s mouth, creating a pressure chamber around the end of the pipe. In instruments like bagpipes, which are not directly mouth-blown, a cap is placed around the reeded portion of the pipe, and air is directed into this little pressure chamber. In organ reed pipes, a similar but larger enclosure creates the chamber around the base of the pipe. The Sorry-Ass Organ, being of the same family of reed pipes, likewise required such an enclosure to maintain air pressure and create the pulsing air movement into the pipe. As I got into design and construction, this aspect of the instrument proved to be a bugaboo, as I’ll describe shortly.
The pipes for the Sorry-Ass Organ are pvc conduit in the nominal 3/4” size, with the commonly available schedule 40 wall thickness. To allow a suitable seating for the reed, a flat area is sanded into the sidewall of the pipe at one end, flattening it over an area about 5/8” wide and a few inches long (the thickness of the pipe wall is enough to allow for this flattening without sanding through). Also here at the pipe end is an inserted stopper which extends about a half inch into the tube. At a point little beyond that a 3/8” hole is drilled in the flattened part of the tube wall.
The reed is a strip of stainless steel, a half inch wide and a few inches long. I experimented with different thicknesses for the stainless strip; got results from thicknesses ranging from .005” to .018″; and settled on .010″ as about right for the current purpose. This stainless reed is positioned in the flattened area so that one end covers the hole when the reed lies flat. It’s attached at its other end by rubber bands or equivalent. (I used 1/2” sections snipped from latex tubing of suitable thickness and diameter — essentially, extra-strong, custom-sized rubber bands.) The stainless strip is given a slight curvature so that in its natural position the end is about 3/32” above the hole. With all this in place, you can place the end of the tube, reed and all, in your mouth and blow. (Seal your lips around the rubber band; make sure nothing in your mouth touches the reed.) If all is good, the pipe will sound.
It’s pretty easy to make a reed pipe that works this way with mouth-blowing. The sound, while not beautiful, can be convincingly strong. The next design step for the current purpose, then, was to provide such pipes with reed caps in the form of air-tight enclosures over the reed end of the pipes into which I could direct the air through a blow tube. But I found that pipes that worked fine when mouth-blown often proved unsuccessful when played through the blow-tube and chamber: sometimes they wouldn’t speak; sometimes they spoke but with noticeably poorer tone than when mouth-blown, and often the resulting pitch was inconsistent. I was surprised by this: after all, many reed instruments use such a system; why were mine so uncooperative? I formulated various explanations and wasted some time trying out various fixes. Eventually I decided that the most likely explanation for my problem lay in my use of unusually flimsy material for the reeds. In those early prototypes I was using very thin stainless steel strips for the reeds, at .005” thick, while other, more successful reed-cap instruments generally use stiffer materials. When I switched to stiffer reeds, I found that they did indeed work better in the reed caps. My best guess in explaining this is that the very soft reeds are more likely to be influenced by extraneous resonances arising in the chamber and blow tube, whereas such extraneous resonances are less likely to arise within the player’s mouth. These extraneous resonances, more often than not, would interfere with what would otherwise be the natural resonance frequency of the reed. Stiffer reeds would be more immune to such interferences.
The stiffer reeds, while functioning better, had a somewhat more refined tone quality than the floppy ones. This was not good news: as I mentioned earlier, I had been enjoying the crude and wobbly sound of the original reeds, and was sorry to convert to something more respectable. But the stiffer reed’s sound wasn’t hopelessly boringly respectable; the tone was still coarse enough to be interesting, especially if I went with ones that were just a bit stiffer but not too stiff at .010”thick. The somewhat stiffer reeds also required more air pressure to sound, but not to a degree that was problematic. In fact, there was an advantage in the fact that with the stiffer reeds it was possible to create a tighter air gap under the reed, with the effect that the airflow could be made more efficient with less wasted air rushing under. This makes more sustained blowing possible on a single lungful of air.
In the usual thinking about reed instruments, one expects to couple the reed to a pipe of a certain length to get a certain note. Things are never quite this simple of course, as a variety of other factors come into play in determining how the reed interacts with the pipe’s air column and what the resulting pitch is. For this instrument’s reeds in particular the pitch situation tends to be imprecise and fluid. Once again, this is probably because although I had switched to a somewhat thicker and stiffer stainless steel strip for the reeds, they were still relatively flimsy and bendable compared to other traditional types of reeds. So for a given pipe length, you can get a wide range of pitches depending how the reed lies and how long the active portion of the reed is. This is not hard to control: if you’ve strapped your reed on with rubber bands or equivalent, then you can shorten the effective vibrating length of the reed simply be repositioning the rubber band. But while you can get a range of pitches in this way from any given pipe length, it does help to have a suitable length to allow a decent coupling between reed and pipe at the intended pitch. With these reeds, however, the ideal pipe length doesn’t correspond closely to the pipe length that In theory would resonate at the intended pitch. That’s because the reed functions in part as the stopper at one end of the pipe, and these relatively flimsy reeds make for a pretty un-solid stopper. This affects the pipe’s resonant frequency. So in choosing pipe lengths for the intended pitches of my set, I paid only minimal attention to the lengths that pipe-resonance calculations would normally suggest. Instead I proceeded empirically and by ear. Basically, I messed around with a few random reed pipes, trying various reed adjustments to see at what pitches the tone came through strongest and showed the most stability. Having established a few nice-sounding prototypes, I extrapolated lengths for the remaining pitches from those.
I found my favorite tone in the lower registers, so I fabricated a set of 21 pipes tuned chromatically over a bit less than two octaves from E2 (that’s guitar low E) to C3. The pipes range in length from about 60” to about 24”. But, a problem: to put the upper end of the vertical pipe at a practical playing height while keeping the lower end off the floor, the longest pipe can’t be much more than about 34”; yet to get the low notes that I liked, the longest pipes need to be quite a bit longer than that. To get around this I made the longest pipes in two parts: the main pipe of rigid plastic conduit, 32” long, with an extension of corrugated plastic flex tube bringing the air column to the required full length. The flex tube bends and doubles back up the main pipe, allowing the required acoustical length without requiring as much height. I wasn’t sure that these flex tube extensions would work well (I was worried that the softer plastic and the corrugated interior form would degrade the tone) but the results were good.
With that resolved, I was able to make a frame to hold the 21 pipes vertically in a suitably accessible and playable array. I put them in a 6-6 configuration, as opposed to the more common 7-5. This refers to the positioning of the twelve notes of the chromatic scale: instead echoing the standard keyboard configuration having a front row with the seven natural notes and a back row with the five sharps and flats, the notes on the Sorry-Ass Organ are arranged in two rows of six each. There are lots of pros and cons to consider for this just as for any other possible pitch layout, but lately my philosophy on this sort of question has reduced itself to a simple “why not just try it?” In any case, 6-6 is good for saving a little space.
I made the blow tube out of corrugated plastic flex pipe. Choosing a large enough diameter avoided the potential problem of whistling in the corrugated tube. As touched on earlier, the end of the blow tube needs to effortlessly seal when pressed down over the top of the chosen reed pipe, and to achieve this I used a medium-soft closed-cell sponge rubber pad. The pad is a little larger in diameter than the pipe ends it is to cover, with a hole in the center to accommodate the blow tube and allow air through. I actually made two of these blow tubes, allowing the option of having two players.
OK, so now that I’ve finished the instrument and had some time to play around with it, what do I think of it? Well, as I’ve acknowledged all along, there are a lot of not-so-great things about it, most notably: it’s impossible to play with any fluidity or legato; it takes an great deal of wind; this makes it difficult to play long lines or even long tones, and it makes prolonged playing quite exhausting; the tuning is fidgety; the tone quality is variable from note to note; the reeds often need adjustment to keep them in tune and in good speaking order. On the other hand, the tone is fun, crude and comical in a way that, in the right sort of company, can be enjoyable; the player interface is unusual and yet works intuitively; and anyone can play it, no training needed. It works well with two players, and this somewhat mitigates the slowness and awkwardness of the playing. It makes people laugh. I’ve just composed and recorded a little piece to use as an audio clip when I get the instrument posted to the instrumentarium section of this web site, and I confess that I do enjoy the unique sound in this recording. But to make the recording, I had to cheat a lot, doing many takes and lots of editing to assemble a decent finished product when the original recordings inevitably were so full of glitches and errors.
Addendum: why not use the simple reed pipe making method described here to make a mouth-blown capped reed pipe with tone holes to vary the pitch? This is problematic because the pitch is so wobbly and variable that the tone holes tend not to produce predictable or stable results. (Consider that, with the pipes described above, I was able to get at least some agreement between reed length and pipe length in order to help stabilize the coupled system for a single intended pitch, whereas in a single pipe with tone holes you’re hoping to get multiple pitches from a single reed-and-pipe configuration.) And, given the cap, the lips are not directly on the reed; this means that taming an unpredictable reed through embouchure control isn’t an option. But I must say that just writing this now makes me want to revisit this possibility once more to see if I could come up with a tone-hole reed pipe that might be, if not effective in a traditionally musical sense, then at least unique and interesting.
You can also take the idea in a more clarinet-like direction: eliminate the reed cap to allow direct embouchure control of the reed; perhaps also use a stiffer reed for a more predictable sound.