Federation South logo

Folk Dance Federation of California, South, Inc.

How I Made a Macedonian Kaval
By Melissa Miller



In 1978, I started making authentic reproductions of kavals, traditional Macedonian village end-blown flute instruments. What I did was buy some clear plexiglass pipe at my local plastic and crafts store in the diameter closest to the Macedonian kaval I'd just obtained from Mile Kolarov. Then I measured the overall length, and the distances from the mouthpiece to the holes, and made a list of those numbers. My goal was to make as close a replica as I could, so I copied carefully, and didn't experiment. I figured the village maker had already done the designing work at his end, so I didn't need to re-invent the wheel (or the kaval). An advantage of the plexiglass kaval was that it was impervious to changes in temperature and humidity.


At that time, our band was playing on concert-pitch instruments, and I discovered the village-style kaval that I brought home from Mendocino Balkan Camp wasn't even in the same universe as the clarinet and accordion. My original idea was to make another, longer Macedonian kaval, because I wanted to start playing Macedonian kaval and stop playing Bulgarian kaval, which didn't sound or feel quite right for Macedonian music. My quick-and-dirty solution was to simply lengthen the distance from the sharpened end to the thumbhole. This threw the intonation off a bit, and it would have been better to re-tool and spread all the holes out proportionately. Fortunately, one of our band members was able to purchase an arbitrary-tuning gajda from Mile that matched my arbitrary-tuning kaval. The them-concert / me-village tuning problem was solved. The band began to develop separate sets of music for city (accordion, clarinet) and village (gajda, kaval) instruments.

Because I already had the tooling set up, I made extras and sold them to people who wouldn't otherwise be able to get their hands on a Macedonian kaval — in those days, the only way was to go to a certain village in Macedonia, and locate the maker!


Overall length692 mm - 27 1/4"
Inside diameter16 mm - 5/8"
Outside diameter20 mm - 11/16"
Distance from sharpened end to CENTER of:
        Thumb hole288mm - 11 5/16" (on "bottom")
        First finger315mm - 12 7/16" (on "top")
        Second345 mm - 13 9/16"
        Third374 mm - 14 11/16"
        Fourth401mm - 15 3/4"
        Fifth431 mm - 16 15/16"
        Sixth458 mm - 18"
        Seventh481 mm - 19 1/8"
        Devil hole #1543 mm - 21 5/16" (on "bottom")
        Devil hole #2584 mm - 23" (on "top")
        Devil holes #3         622 mm - 24 1/2" (two, one on each "side")
Diameter of holes8 mm - 5/16"

These measurements are taken from the Macedonian kaval I currently play. Tolerances are to the nearest mm or 1/16". The note we call "A" sounds at 444 MHz, A# on the piano. The all-fingers-down note, we call "C" and it sounds as C# on the piano. The intonation is not well-tempered, that is, some of the notes sound a bit sharp or flat when compared to the piano. I consider this not a serious problem, because I'm playing with an old-style Macedonian gajda, and the intervals on the gajda are also a bit off, when compared to the piano. I adjust by lipping each note up or down to match the gajda.

In 1978, I worked at a place that had a full machine shop, where I could sweet-talk the technicians into allowing me to use the lathe and drill press after hours. With my husband helping a lot, we sawed the pipe into correct lengths, then he ran the lathe to bevel the edge of the mouthpiece. (Yes, we used goggles for eye protection.)


If the length of pipe survived that process, I then inserted it into a template I'd gotten the technician to make for me, based on the numbers (see above) I'd given him. It was square-cross-section metal tubing, a bit larger than the plexiglass pipe, with the holes drilled at the correct locations. I'd wedge the butt end of the pipe into the metal template with little pieces of wood or cardboard, hold this down on the platform of the drill press, and then ever-so-carefully bring the drill down on each hole in the template.

Once the holes were all drilled, I'd use a little curved de-burring file by hand to smooth the outside edges of the holes on the pipes that survived that far. I didn't go so far as to bevel the holes, just softened the edge of the holes a bit so they'd feel nicer under the fingers. I'd push a 1/2" wooden dowel down the pipe to knock off the few curls of plexiglass that would cling to the inside edges of the holes, and clear out any shavings that remained. I sanded the beveled and butt ends a little to smooth them off, too.


In the plumbing section of the hardware store, I bought joints of sturdy white PVC pipe with (I think) inside diameter of 40 mm, 1-1/2" or so, with two end caps for each case I planned to make. I'd saw the joint into sections an inch or so longer than the kaval. I'd put a bit of foam rubber inside each end cap, and fit the end caps on. Usually I'd glue or tape one cap into place, leaving the other cap with a snug press-fit. I may have drilled a breather hole into one or both of the end caps.


Plexiglass is a difficult material to work, and requires a particular combination of correctly-angled sharp tools, slow rotational speed and gentle pressure. If you push too hard, or try to rush, it'll crack. If you stress it by swinging it hard or flexing it, it'll break in two. We ruined a lot of pipe while we learned the proper touch, and some batches of pipe were trickier than others.

The benefit of using such an unforgiving material is that it yields a very pretty, completely transparent flute with a brilliant tone. Mile Kolarov was very favorably impressed by the several kavals I gave him, and some years later, Alex Eppler also complimented my humble little plexiglass kavals.

I made and sold perhaps 50 or more kavals in the course of a decade or so. I have left that job, and don't have any of the necessary tools at my disposal anymore, so I'm out of the kaval business.

Why did I use plexiglass? (1) I didn't know how hard it would be to work with. (2) There wasn't any PVC pipe of the correct exact diameter, which I assumed was important to make a good replica. I also thought a hard, light material, much like the wood of the original kaval, would make a better tone than the relatively soft, slightly flexible light-weight PVC pipe (of the wrong diameter) that was available on the adjoining rack at the plastic and crafts store, or the bulky, thick-walled heavy-gauge PVC pipe (in several diameters, all wrong) in the plumbing department at the hardware store. (3) It was very inexpensive.

I read an article by kaval player Mike Slama that contains a great deal more technical acoustic information than I had when I started making plexiglass kavals, back in 1978. Because Mike has done a great deal more experimentation than I ever did, I defer to his assertion that the kaval doesn't really care what it's made of. If I ever need to make another Macedonian kaval, I'd take the hint, and go with PVC because messing with plexiglass is impossible without the right tools.

Used with permission of the author.