Musical strings made from rubber bands make an appealing sound. The tone has lots of fundamental and almost no overtones. The decay is rapid, and the overall feeling is sort of softly punchy, both warm and rhythmic. Almost all books of home-buildable instruments for children (there are quite a few of these) include some version of the rubber band zither, or whatever the author chooses to call it, consisting of several rubber bands stretched over the open top of a shoe box or equivalent. This instrument couldn’t be easier to make, and it actually sounds quite nice (though weak in volume). Easy to tune, too: just by pulling the rubber bands this way or that across the edge of the box you can vary the tension and resulting pitch.

Some more sophisticated elastic string instruments have been produced commercially. Most notable are a couple of string basses. It’s not surprising that they’d be basses: the elastic string tone quality is well suited for a nice plucked bass sound, and at a conveniently small scale length. The two main entrants in the field are the Ashbory Bass, apparently distributed at different times by a couple of different well known guitar firms, and the Kala U-Bass, presented as a bass ukulele. I’ll talk more about these instruments later in this article.

The elastic string idea seems ripe for further exploration, and not only in the bass range. But beware; although elastic strings work wonderfully in home-buildable children’s instruments, some obstacles appear when you set out to take them to the next level. In the following paragraphs I’ll outline some of the special challenges, as well as upsides, that come with elastic strings. After that I’ll talk about how I’ve attempted to manage these things in a couple of elastic instruments I’ve made. (At the time of this writing — April 2018 — I haven’t yet photographed, recorded, or uploaded these instruments to the Instrumentarium section of this website. So I can’t yet point you to where you can see and hear them. But hopefully soon.)   [Follow-up: I’ve now added pages for a couple of elastic string instruments in the instrumentarium section of this web site here and here.]

One problem with elastic strings is that they don’t drive a soundboard very well. Rather than forcing the soundboard into vibration when they themselves vibrate, they tend to accommodate their own motion by stretching. The soundboard, if it has any inherent mass and rigidity of its own to speak of, scarcely moves. For this reason, a rubber band string instrument with a more fully realized soundboard/soundbox is likely to be disappointingly quiet; probably not much louder than the shoebox zither. And rubber band strings don’t lend themselves easily to the usual methods of sound amplification. They don’t work with magnetic pickups (the kind used on electric guitars), because magnetic pickups require a string of steel or other magnetic metal; latex doesn’t qualify. Rubber bands aren’t great candidates for piezo pickups either. Piezos need to attached a vibrating body to gather a signal; they are usually attached to or built into a bridge or soundboard. But since the rubber band string transmits very little of its vibratory movement to its bridge and soundboard, this too is a sub-optimal situation; the piezo is given very little vibration to respond to.

However, there are methods for elastic amplification. One is to use a magnetic pickup in a particular configuration that allows it to function as a bridge and pickup in one. This system is detailed below in the description of the instrument called The Noodles. Another amplification method is to mount rubber band strings on a styrofoam body with a piezo placed somewhere on the styrofoam. A suitable styrofoam body, reinforced in appropriate ways, can have the right blend of strength and yeildingness to make this work where other materials, being heavier or more rigid, would be less effective. This idea is detailed below in the description of the instrument called Styrovessel.

Elastic strings can be made from a variety of materials, but most stretchy materials have nothing like the tensile strength of other widely used stringing materials — spring steel music wire, nylon, or sheep gut. If you try to subject elastic strings to high tension, they unceremoniously break. But they happen to possess interesting countervailing characteristics. One is that their tension requirements play out differently from other stringing materials. Normal strings need to be set at high tension because if they’re not, the harmonics tend to be out of tune, and they suffer from noticeable pitch-drop as the amplitude diminishes immediately after plucking. Neither of these are a problem for elastic strings. Harmonics, detuned or otherwise, are not an issue because whatever harmonics may be present are so weak as to be inaudible. And pitch drop doesn’t noticeably happen because — well, to jam a lot of information into a few words which you may choose to ignore, the stretchiness results in a reasonably linear restoring force over a wider amplitude range. The interesting upshot is: if you discount the problem of volume, elastic strings work well musically at tensions much lower than other types of strings. That’s one of the reasons they can work well on shoe boxes.

The comparative weakness of elastic strings limits their suitability for instruments in which you need to cover a wide range of pitches in a set of strings of the same length. The primary example here is fretted instruments such as guitars. In a guitar, all the strings, when unfretted, are the same length. The highest and lowest strings are tuned two octaves apart. With elastic strings, it’s difficult to get that range in a set of strings of the same length. Even if you set the lowest string as low in pitch as it can go without sounding and feeling ridiculously funky, it’s quite a challenge to get another string of the same length up to two octaves above that before the higher one breaks. Interestingly, making the lower strings thicker and heavier doesn’t help as much as you’d expect: in contrast to un-stretchy strings, when you have two elastic strings of the same length, making one much thinner than the other is only moderately helpful in allowing it to work for higher pitches. (The reason for this has to do with the fact that making the string thinner reduces the pull of its inherent stretchiness; thus, reducing weight has the simultaneous countervailing effect of reducing tension — a factor which is scarcely at play in other string types.) In my initial experiments with latex strings, the greatest range I was able to get in a set of strings all of the same length has been about an octave and a fourth. When I tried for anything greater than that, there was just too much tendency of the high string to break; if not immediately then inconveniently soon. But once again this gets into the question of materials — are there alternative materials that have the stretchiness and appealing tone of latex, but that are much stronger? I haven’t found any miracle solutions, but there are some helpful options. I’ll get into this question farther along in this article.    

Another advantage of elastic strings, again relating to their ability to work at low tensions, is the fact that they can work well with unusually short active string lengths. Also, they can produce pitches in the middle range or even bass range at much shorter lengths than would be possible with other stringing materials. So, very convenient: you can make mid-range or bass instruments in surprisingly compacts sizes. This is one of the selling points of the commercially made elastic string basses: they’re small.

When you set about tuning an elastic string by increasing its tension — cranking a tuning gear or turning a tuning peg — you find that it takes a lot more turning to bring the string up to pitch than is the case with conventional strings. That’s because the string accommodates some of the increased pull simply by stretching, resulting in less-than-expected increase in tension. A related effect is that pitch-bending by stretching, as guitarists seeking blue notes are wont to do, is less effective (less pitch-bend for a given amount of stretching). Another side effect for fretted instruments — this one rather convenient — is that the de-tuning associated with normal fingering is reduced: as compared to steel strings, there’s less effect on pitch when as the string stretches a bit as it’s pressed down to the fingerboard. As a result, you can have a fretted instrument with unusually high action yet minimal de-tuning. This couples nicely with the fact that elastic strings are quite un-stiff — soft and compliant under the fingers, all the more so because they’re likely set at low tensions. This means that high action doesn’t translate into more difficult playing, or at least not as much as with conventional strings. This can be useful in easily eliminating buzzes in fretted instruments, and it can also be helpful in making it easy to adjust bridge heights and, in so doing, adjusting the angle of inflection of strings over bridge.  

Related to this business of softness and low tension is the fact that elastic strings have a very different playing feel from traditional strings: very squishy. It takes some getting used to.

Elastic strings tend to gradually deteriorate over time, eventually loosing their integrity and breaking, especially when left under tension for an extended period. This problem is most pronounced with latex. Keeping the strings at low tensions helps with this. My work in this area has not yet stood the test of time, but so far this potential problem hasn’t been as bad as I had feared. However, this is a reason to try to make sure, as you design your elastic string instrument, that string changes are reasonably quick and easy to do.

Interestingly, in my experience so far the elastic strings to seem to hold their tunings reasonably well even as they age … except that there is also this slightly different aspect of the tuning-stability question: because the elastic strings are usually a bit loose, it’s easy, with vigorous plucking, to cause them to slip a bit one way or the other over the bridge or nut at either end. And because they have good traction and typically are not at very high tension, once slightly displaced in this way they don’t automatically readjust and correct themselves. This affects tension, and results in detuning. There’s a plus side to this as well: if a string is slightly out of tune you can often quickly correct it just by lifting and slightly repositioning at the bridge, thus adjusting the tension on the played side.

I mentioned a moment ago that when it comes to tuning, elastic strings require more peg-turning to achieve a desired effect on pitch. This can get to be inconvenient. Imagine you’re tuning by means of, for example, a tuning gear made for guitar. You could easily find that the winds of elastic string have quickly mounted up on the peg to the point where they’re in danger of spilling off, and you haven’t reached your intended pitch yet. For this reason, I’ve found that it’s helpful to have dual tuning mechanisms: a rough-tuning mechanism capable of quickly pulling in a lot of string to get yourself in the ballpark, to be followed by an additional fine-tuning mechanism to zero in on the desired pitch. Later in this article when I talk about some of the elastic string instruments I’ve made I’ll describe the system I’ve come up with for the rough tuning. For the fine tuning I usually use something similar to a zither pin or violin peg. The bridge-crossing-repositioning method described above also amounts to an available option for tuning, and in practice it often turns out to be most convenient for on-the-fly fine tuning. Taking all these options into account, we can say that if you care about accurate tuning on such instruments, then having multiple tuning methods is helpful and actually not so terribly difficult to build in.

One more nice thing about elastic string instruments: given the much lower string tensions involved, it’s possible to make such instruments very light. No need for very strong construction; no need to worry about soundboards collapsing, necks warping or separating from the body, or other undesirable side effects of high tension.

And now, more about materials.

Rubber bands are made of latex. They’re available in a many sizes and shapes. For most musical string purposes, narrow square-ish shapes are better than wide flat band shapes. (Round is ideal, but few rubber bands are made that way.) The size sold as “file bands,” about 8” long, is particularly useful for musical strings, being not too far out of square in shape and of suitable length at about 16” when cut so as to be no longer a loop. As a convenient trick when you need a thicker string, twist two or more file bands together and string them up as a single string.

You can also get latex cord, cylindrical in shape, in several thicknesses from outlets such as McMaster-Carr. Very useful, and a bit classier than rubber bands.

And another variation: the stretchy cord sold as shock cord (similar to bungee cords) is made of strands of latex with an over-wrap of thin synthetic cloth. It’s available from various sources in a range of diameters spanning the range that’s potentially useful for musical strings. Since the over-wrap provides UV protection, these may last longer than plain latex. In the narrower diameters I’ve found the tone not as nice as plain latex, but it can be useful when thicker diameters are called for. Possibly this larger shock cord sounded better to me than plain latex of comparable diameter because the interior of the shock cord is stranded, giving it more flex and less internal damping than a solid thick cord.

Latex is the stretchiest of the elastic materials I’ve found. Perhaps for this reason it has, to my ear, the nicest tone: very saturated fundamental (virtually no overtones), a nice plunky sort of attack, and rapid decay. It is surprisingly strong compared to some other elastic materials and can take a huge amount of stretch before breaking– but there are limits of course, since it has not nearly the tensile strength of music wire or other traditional stringing materials. One of the biggest drawbacks of latex for serious instruments is that it deteriorates over time, especially when left under tension. Another negative for some people will be the playing-feel of latex, which can seem disconcertingly soft and squishy.

One of the materials often used in the commercial elastic string basses is polyurethane, which lasts better than latex. It’s a little stronger as well — able to take somewhat higher tensions without breaking. The tone, to my ear, is a bit duller; not as rich as latex. It’s less stretchy than latex, so the playing-feel, while still soft, is not as squishy. I was able to find polyurethane cording sold as round belting in several useful diameters from a few sources online. You can also purchase a set of four polyurethane short-scale bass strings made for Ashbory Bass and similar instruments at various outlets including

The other string material sometimes used in the commercial elastic string basses is silicon rubber. I found silicon rubber cord in suitable thicknesses at McMaster-Carr in a standard version and an extra-soft version. The extra soft is comparable to latex in feel. I found the silicon rubber to be more subject to breakage than latex, and published data on tensile strength seemed to bear this out. Yet the fact that some form of silicon rubber is sometimes used in the commercial instruments makes me wonder if there’s more to this story — perhaps there’s a stronger alternative formulation for this material. As with the polyurethane, you can also purchase a set of four silicon rubber short-scale bass strings made for Kala U-Bass and similar instruments at     

Interesting to note: In the online reviews of the two best-known commercially made short scale elastic string basses, the most common complaint is that they don’t hold their tuning well. On the plus side, I haven’t seen complaints about string breakage. Instructions for installing and tuning the strings mention a few of the issues we’ve touched on here for elastic strings in general — for instance, the importance of equalizing tension on either side of the bridge and nut to minimize detuning, and the potential problem of too much bulky string piling up around tuning pegs.

There are a few other elastic materials available in the form of cords that might seem worth a try as stretchy strings, such as neoprene, buna-N and EPDM. Several of these materials are sold in cord form as material for fabricating O-rings or belts. According to the data I’ve been able to find online, these materials are not good candidates due to insufficient tensile strength. Indeed, I did try strings of buna-N and found them not up to the task. But, important to note, I’m not an expert in this field and my research wasn’t extensive; it’s quite possible that a materials scientist could point to promising candidates we haven’t considered. 

Here are descriptions of a couple of instruments I’ve made using elastic strings:

The Noodles [which has since somehow come to be called Spaghetti. Too bad; I like the name The Noodles better. But Spaghetti seems to have gained traction and it would be an uphill battle to try to change it back now.]

This is a zither made with strings strings of black 1/8” diameter round latex cord. (1/8” when relaxed that is; it stretches out thinner under tension.) The bridges are set at angles across the body so that the highest strings are very short and the lowest ones several times longer. This means that, even with all strings of the same diameter, you can get a wide range of pitches without requiring extreme variation in tension, especially since latex strings sound well at low tensions. Thus, concerns about tensile strength and breakage are mitigated.

The active string lengths are quite short, but given the low tensions, the resulting pitches aren’t so very high; the instrument functions nicely across a low-to-mid range. The strings are laid out on the zither with mid-string bridges allowing playable segments on each side. There are 14 strings, divided into 28 playable segments ranging in length from about 2.25” to 11.75”. The sounding range is just under three diatonic octaves from A1 to G4. (Some of the notes are duplicated in two string segments). The tone is just what you’d hope for from a latex zither, warm and plunky. 

The Noodles employs two magnetic pickups in a special arrangement designed to work with soft elastic strings. Magnetic pickups are usually used with steel strings because they capture the signal by responding electromagnetically to the movement of the steel. They’re normally placed below the strings of, for instance, an electric guitar, close but not touching, where they can sense the vibratory movement of the strings. In the Noodles, to allow the pickups to work with non-steel strings, I made the following arrangement: Resting directly on top of each pickup is a soft pad of foam rubber, about 1/8” thick. On top of that is a short length of flat steel bar with a segment of 1/4” steel rod glued on top. This rod-and-bar piece spans the width, forming a sort of saddle on top of the foam pad and pickup. The latex strings cross over this layered pickup assembly just as they would over a bridge, pressing down on the steel saddle. When the string segment on either side of this bridge is plucked, the vibration of the string is transmitted to the steel saddle, which is free to vibrate along with the string because it sits on the foam pad. The pickup senses the movement of the steel saddle. Two of these pickup/bridges serve as The Noodles’ middle bridges.

The instrument has several options for tuning adjustment. One is a zither tuning pin (a smaller version of a piano tuning pin) set at one end of each string. A second tuning mechanism is designed to deal with the common problem in elastic strings that to bring the string up to pitch, more string needs to be taken up than can fit on the pin without spilling over. This secondary tuning mechanism, located at the opposite end of the string, quickly and easily takes up a lot of excess string length and gets it out of the way so the tuning pin doesn’t needs to take up as much. To form this secondary tuning mechanism, the string passes through a slightly oversized hole in a wooden cross-piece at the far end. You can pull the string through to where it’s close to its intended tension and pitch, and secure it there by clamping on the far side of the hole with an alligator clip. (This only works because the tensions are low and the latex strings are soft and easily gripped by the clip; otherwise the string would slip through and not hold at the clip.) A third quick-and-easy fine-tuning method involves lifting the relevant string and slightly shifting it one way or the other across one of its bridges. This is necessary because, as mentioned above, the stretchy strings don’t always do a good job of equalizing tensions across the bridge. It’s especially important since there are playable string segments on both sides of the middle bridge, and imbalances in tension between them result in detuning.

And there’s one more tuning mechanism. The fact that the string segments on each side of the middle bridges are to be tuned and played means that it’s important to get the relative lengths on the two sides right. In theory this can be arranged by means of the geometric layout of the bridges and the distances between them, but in designing the instrument I felt that it would be useful to have fine-tuning options in this as well. So the end-bridges on each side are set up with adjustable individual saddles for each string. The expectation is that you’d fine-tune the placement of these saddles when the instrument is first strung up, and probably never bother with them again. It was a lot of work to add this feature; was it worth it? Borderline, because whatever fine tuning you do will often later be rendered irrelevant by inadvertent unequal tensions on each side of the middle bridge as described above.       

Elastic Guitar

This thing really is configured pretty much as a guitar with stretchy strings. I liked the idea of people being able to use finger-picking or classical guitar skills with the soft and round elastic string tone. But a lot of things play out differently with elastic strings, so the end result is not entirely guitar-like.

Most importantly, there was no hope of getting elastic strings in lengths suitable for a fretted instrument up to standard guitar pitch. The higher-pitched ones would break long before getting anywhere near the required tension. So although the string scale is short at 18”, the instrument is pitched an octave below standard guitar tuning. Basically, it’s tuned like a four-string bass plus two. You could ask: isn’t that just too low for guitar-like note-patterns to read well musically? Based on past experience with low-tuned guitars, I had had some optimism about this. But it turns out the answer is yes, this instrument is pretty muddy sounding when you try to get too fancy or play too many notes at once. So, as with all instruments, you gotta work with it and learn what it likes to do. It tends to sound best either when used as a bass, or when played more melodically in its upper registers (although intonation issues do kick in up there). 

Even with the very low tuning, it was quite a challenge finding elastic stringing materials to suit the need. The need was for strings such that the low note (E1, which is bass low E) could sound decent at an 18” scale, while the high E (E3 on this instrument) cold be brought up to pitch without breaking at the same 18” length. (By way of contrast, on the zither called The Noodles described above, that E3 is only 5.5”.) In the end I went with a hodge-podge of string types: the low E is of 3/16” shock cord; the 5th, 4th and 3rd strings are of latex cord; and the two highest strings are 1/16” polyurethane belting. I live in constant fear of that highest one breaking.

The pickup system is the same as described above for the Noodles: a layering of magnetic pickup, foam pad, and steel saddle assembly all doubling as the guitar’s bridge with the strings crossing over and pressing on the saddle. For tuning there are zither pins in the head stock. At the opposite end, to take up excess string length, is a version of the alligator clip configuration described above for The Noodles.


I love both the sound and the simplicity of that staple of kid-buildable instruments, the rubber bands on the shoebox. I wanted to make something that would be almost as simple, sound just as good, and, to compensate for its modest volume, be amplifiable.

To do that, I started with a smallish styrofoam picnic cooler which I cut down to a still smaller size of 9” x6” x 6”. Styrofoam, AKA polystyrene, is always a good place to start for such things, because it has an ideal blend of characteristics: the right balance of rigidity and flexibility coupled with very light weight. These qualities allow it to do an excellent job responding to vibrations even from a relatively low-impedance source, and spreading those vibrations through a large surface area for radiating sound. For strings the styrovessel uses the 8” rubber bands called file bands, simply stretched around the cooler with the segments crossing the open top available to be plucked. As with the shoebox, tuning can be done by pulling the rubber bands this way or that over the edges to vary the tension on the sounding segments. Where shorter rubber bands are called for, you can effectively shorten the file bands by knotting them at midpoint or other suitable location, so that only a portion of the band is used while the part beyond the knot is left out of play.

As suggested above, with elastic strings it’s better when possible to create a range of pitches by having strings of different sounding lengths, rather than depending on variations in tension to do the job. Given the rectangular form of the styro cooler, there was no very promising way to do this, so I reshaped the cooler. I did this by cutting a wedge out of it, bending the remaining whole together and gluing, creating a form that is wider at one end and shorter at the other. (Hot glue works well. Some other glues tend to dissolve the styrofoam on contact.) 

A styro cooler may hold up temporarily under the pressure of a few rubber bands, but as the number of rubber bands increases it will soon begin to distort drastically. It’s necessary to reinforce the rims where the rubber bands pass over. I was cautious about this, fearing that adding rigid reinforcements would mean that the structure would become too heavy and rigid to accept the vibrations from the compliant rubber bands. But I found that I was able to add pretty strong reinforcements glued to the outside of the rim that the rubber bands pass over, in the form of 1”x 1/2” redwood strips, without seriously compromising the transmission. To prevent buzzing I also added little bridges, in the form of thin bamboo shish kabab skewers, glued along the top of the rim where the strings pass over.

The styrovessel is louder than the shoebox, but still fairly quiet. For amplification I added piezo contact pickups at selected points on the cooler. Near the start of this article I mentioned that elastic strings, especially lightweight rubber bands, don’t work well with piezo pickups in many applications because the vibrating rubber bands don’t drive whatever they’re mounted on forcefully enough to give an attached piezo much of a vibration to respond to. But the just-flexible-enough quality of styrofoam comes to the rescue. The styrovessel body is light and flexible enough to take in the rubber band vibration, yet solid enough (with the benefit of reinforcement strips at the edges) to support and activate the piezo. The resulting amplified sound is nice — a pretty good recreation of the acoustic sound. A single piezo will do, but I find that the result is nicer with multiple piezos at different locations on the styrofoam body. If you don’t mind shopping overseas you can get a decently made pre-wired triple piezo pickup at low cost from

I later made a second styrovessel, similar to the first in many respects but larger, with correspondingly lower range. Its unamplified sound is louder than that of its smaller sibling, with a very nice acoustic tone.


Share This