---

Jon Swayne Border Pipes in G

It is with pleasure that I respond to Jock’s request to contribute an article on my Border /Half-long style- bagpipes, but also some nervousness, since although I have never claimed that they are typical of surviving originals, I am aware that they have attracted criticism in some quarters as being ‘untraditional’ which of course they are. Much of what follows is necessarily abbreviated, but I hope enough detail is included to be informative, without it being boring through excess. First a word or two about the origins.

Origins.

For about 8 years from 1979 I made reconstructions of renaissance period Flemish and   English bagpipes. Both these instruments had a basically nine note scale, with a few crossfingered chromatic notes, and served very well, especially for the purpose for which they were originally conceived, but in continuing to explore the southern English and European repertoire, their limitations became increasingly apparent. As is well known, very little   specifically bagpipe music survives in the south of England; there is a small body of traditional music with a one octave range, but there is much more that can be played without adaption on an instrument with a slightly greater range. I was also interested in European repertoire, and in particular that of the centre of France, where some tunes span one octave and a fourth. I wanted to make an instrument with at least an English parentage, so it seemed a good idea to try to adapt the instrument to the music, rather than the music to the instrument. My main purpose was to make an instrument suited to the needs of the modem musician playing a wide ranging repertoire whether solo or in a band, but an instrument with at least one foot in the English/British bagpipe past.

My starting points were various. I had observed several sets in museums without taking any detailed measurements. In 1980 I had the opportunity, courtesy of Graham Wells, of measuring and photographing a very stylish set of Half-longs, probably dating from around 1800, which passed through Sotheby’s salerooms; this was in plumwood with brass ferrules and and ivory mounts. [see photo]. I also had a copy of the drawings in the first edition of Cocks & Bryan on Northumberland pipes. In addition I had looked at some Pastoral pipes.

Pitch.

Most originals seem to be pitched around A or B flat. I decided to go for G, since in general this is a more useful pitch for playing with other instruments, and the early English pipes I was already making were in G. D would be even better from the point of view of compatibility, but this implies either a very small or very large chanter, and was to come later. :

Chanter Reed.

When making a new pipe, do you start with the chanter or the reed? It’s a bit ‘chicken and egg’, since clearly you can’t think of them in isolation. I considered Highland pipe reeds, but although there are obvious advantages in having such things easily available over the counter, I suspected they would be too hearty for the type of sound I was after. On the other hand, because I didn’t want to have to make staples. I had decided to work with modified modem oboe staples. In making a few sketches of reed configurations a staple length of about 30mm (1 1/8") looked about right, the larger end of which measures about 4mm (5/32") internally. This decided the diameter of the chanter throat. The actual reed design is rather a simple one, compared with the reeds I had hitherto been making for the ‘historical’ bagpipes. At first I used cane, but later went over to plastic. The reasons for this were ease of manufacture (with practice a reed can be made in five minutes) and durability. Correctly treated they seem to last indefinitely, and since they need virtually no maintenance, are much easier for beginners to cope with. On my pipes, I believe a good cane reed is slightly superior in tone to a good plastic one, but there is very little in it. This does not necessarily hold true for other chanter designs; I have tried without success to make good plastic reeds for my previous chanters. For those interested, precise dimensions and instructions for   making the reeds are given below.

Chanter.

Of the original information available to me at the time, the original Cocks drawings ooked the most suitable for scaling to G, but the details are to say the least sketchy. The main points to decide are: speaking length (bottom of reed socket to foot), bore shape, throat diameter, wall-thickness, and disposition of finger- and vent-holes.

Speaking length; this would be decided by experiment, starting with a model of the Cocks drawing.

Bore shape; I already had experience of using a taper of 30:1 and felt this was about right for the quality and loudness I was after; it was also a reasonably close match to the taper shown on the drawing. I later modified it slightly to 33:1. For simplicity I decided to try for a straight cone if possible.

Throat diameter; this was already decided at 4mm.

Speaking length and wall thickness would be determined initially by the drawing. I made a version of the Cocks chanter, and fiddled with the reed and hole sizes until the scale was reasonably in tune. This gave a starting pitch, which I can’t remember, but I had to scale it down somewhat to get a 6 finger G. (A useful scaling tool derives from a method of finding pitches in equal temperament. See note below.)

The result was a chanter which played a major scale in G with reasonable stability, but overblowing and chromatic possibilities left a lot to be desired. Various prototypes were made, lay around in corners of the workshop over several years, and would be taken up and put down in odd moments as fresh ideas arose. I didn’t really make any progress until I started playing around with the part of the bore below the lowest fingerhole (which for   simplicity I will from now on call the ‘bell’).

When bell ventholes are referred to in books on wind instruments, they are usually described as affecting the harmonic spectrum,especially of the lower notes. They also have the function of equalising the tone of the lowest note, which would otherwise speak only through the open end. Normally the pitch issuing from a tonehole can be raised either by enlarging the hole, or by moving it up the bore. That the behaviour of a vented bell is not at all simple, I began to realise from the observation that if you have a chanter on which the 6finger note (tonic) is a little sharp, you can flatten it by reducing the size of the lowest     fingerhole; you can also flatten it by increasing the size of the ventholes, a result which is counter-intuitive. Conversely, reducing the size of the ventholes sharpens the tonic. (The sub-tonic will be sharpened and flattened respectively, which is what you would expect, and may not be what you want, but that is another question.) On the other hand, it remains true that moving the ventholes up tends to sharpen the lowest notes, and vice versa. The importance of the bell was confirmed when it occurred to me to insert a paper tube into the open end. A whole range of effects can be observed, from simple adjustment of the tuning of the lowest notes, to more complex ones such as the accuracy of cross-fingerings, ease of over- blowing, tuning of the second register, stability of the tonic, and so-on. Shown diagram- -matically and grossly simplified, the three different arrangements shown below may give the same basic chanter pitch, but the performance will be totally different.

In short, the length of the bell and the position and size of the ventholes appeared to be   crucial, and getting the proportions right solved many problems of overblowing and tuning.

Optimising the wall thickness took some experimentation too. Too thin a wall seemed to adversely affect stability, but an over thick wall affected good overblowing. An important point, but easy to overlook, is that unless the bore is straight i.e. the drill does not wander when boring, the wall thickness will vary around the circumference of the tube. Other     factors had also to be optimised, such as general fingerhole positions and sizes, and reed shape and dimensions.

In this way I eventually reached a point where I thought I had the basis of a useable instrument. The finger chart shows the functional performance of the chanter. I found it   impossible to get a good minor third (B flat), so borrowed from the practice of pipemakers of Central France the idea for a hole for the thumb of the lower hand.

A fingering not shown is the following, which usually gives a good G sharp.

Drones.

I offer various drone formats, including bass and two octave drones, bass octave and high fifth, and bass octave and double octave. A drone at the fifth can sound impressive with 5- or 6-finger tonic tunes, but tends to interfere with 3-finger tonic tunes, especially so if the fifth lies between the bass and octave drones. A high fifth can be made quiet enough that it does not overly interfere with 3-finger tonic tunes, but then it is hardly louder than the third harmonic of the bass drone. A switch or plug can solve the question of interference, but you still can’t mix 3- and 6-finger tunes in a sequence, in addition to which you are left with only two drones. For this reason, and because of the added richness which it gives, I prefer the double octave as the third drone.

The question of drone/chanter balance is another matter of taste. Mainly it seems to be a question of balancing the needs of the listener against those of the player. Drones which are well balanced to the listener can be tiringly loud for the player. The actual loudness of the drones is easily adjusted by choosing a suitable bore diameter/length. As far as drone reeds are concerned, provided that the diameter of the bore is not too large in relation to its length, and provided that they are properly set and adjusted, reeds with a brass or plastic tongue on a wood or plastic body are very satisfactory, plastic being much easier to set up than brass. Drone stability assumes even greater importance with an overblowing chanter, in that the drones must not change pitch when pressure is increased to go over the octave. If the drone diameter/length is correct, stability is relatively easy to achieve for a wide pressure range by reed adjustment.

A switch to turn off all the drones, an idea borrowed from the Irish Union pipes, is useful if you want to play with a harmonised accompaniment from keyboard or - accordion, for     example.

Appearance.

Rather than base the exterior design on one existing original, I decided to combine various features which I found attractive in several different instruments, including pastoral pipes, and generally to make the instrument as light and compact as possible. The sketches below show typical turning details, the top and bottom of a chanter above, and two styles of drone end below.

 

Materials.

On the whole I prefer to use indigenous woods if possible, and of these I find plumwood to be the best from the point of view of tone and appearance. Boxwood of course is superb, but I tend not to regard it as an ‘everyday’ wood for bagpipes for a variety of reasons, amongst them expense, problems of seasoning and warping and of getting in good quality in long enough lengths. I have also occasionally used ebony for a really bright sound, but it makes a heavy instrument, and you have to be careful of splitting where the tube wall is thin. I see no reason to use African Blackwood, fine wood though it is. It makes an even heavier instrument, and I feel the sound has less life than ebony.

My ‘standard’ design uses brass ferrules in structurally exposed places such as the ends of sockets, with plastic artificial ivory on the ends of the drones and top and bottom of the chanter. There are various substitutes on the market, but generally I use GPS which is a lot less fragile than it used to be. An attractive veined variety comes from France (intended for cutlery handles). Recently I used it on a set in boxwood which was to be stained with nitric acid. Not having tested it first I found that the acid attacked the plastic, causing the surface to soften and bubble up; fortunately I was able to rescue the situation before it had gone too far.

Makers beware!

I also like using black horn, both as a reinforcement and as decoration (i.e. no brass), but except for buffalo horn where you can only use the first few inches of the tip before it     begins to flatten off, good black ox horn is now very hard to get. I understand that African sources are drying up because growth hormones are increasingly used, resulting in animals coming to slaughter weight before having had time to grow decent horns. How long before hornless cattle are the rule?

Developments.

Since the original design in ‘G’ I have extended the range to include the pitches upwards of

‘A’. ‘Bb’ and ‘C’ (I have tried a ‘D’ but it doesn’t work very well yet), and downwards ‘F’, ‘D’ and ‘C’. On the two latter largest sizes, I have compressed the physical length of the bass drone by triple boring the first section, which makes the instrument more convenient to manage.

Some while ago I introduced a simplified, single (bass) drone, mouth-blown model for beginners, using exactly the same acoustic design for the chanter. More recently I added an intermediate design with two drones, also usually mouth-blown.

The vast majority of the full sets ordered are bellows-blown, but occasionally I am asked for mouth-blown versions, and see no reason not to comply with this request. Mouth-blown versions need slightly more maintenance, and moisture condensing on the reeds can cause problems in extremely cold conditions, but otherwise the performance is not affected.

General Observations.

On the whole the tone quality is clear, clean, expressive and loud enough for most purposes without being overpowering indoors. It is relatively easy to reed, and there are no real ‘problem’ notes. The extended range and chromatic possibilities widen considerably the scope of the music which can be played, and the chanter will accommodate quite a wide range of styles. An effective and tasteful vibrato can be found for most notes if required. On the other hand there are many Highland pipe gracings which it does not like, but then it was not intended to. That problem has been addressed by the development of a new Lowland Bagpipe design which uses a Highland scale of fingering, but that is another story ........

Note on ’Scaling’.

It hardly needs saying that the pitch of a woodwind instrument is directly related to the length of its bore. Frequency doubles at the octave, so broadly speaking, halving the length of the tube doubles the frequency of the note it gives. For small length changes, you can divide the length of the known tube by the frequency (Hz) of its pitch and multiply by the new frequency to get the new length. This is OK if you have a frequency counter plus microphone and probably an amplifier. On the face of it, electronic tuners are not much help directly, because they give a readout in pitch (relative to equal temperament) which is much more useful musically than frequency.

Of course you could convert the pitch reading to frequency. Equal temperament means 12 equal semi-tones per octave, each semi-tone being further divided into 100 cents. Since   frequency doubles at the octave, to find the pitch of a note one semi-tone above the pitch of a known note, the multiplier is 12/2, otherwise written as 2 1/12 (in words, “two to the power of one over twelve”) which is more easily handled on a pocket calculator. No, I don’t really understand any of it either, but it works. So A 440 Hz x 2 1/12 = A# 466Hz. Since there are 1200 cents per octave, smaller intervals can be handled by 2 1/1200. But since length is directly related to frequency, there is no need to go through the conversion. You can use the same multiplier (or divisor) on the length. Suppose you have made a prototype chanter, speaking length 350mm, and find your tuner tells you that it plays 45 cents flat of A440. 350 /2/ 45/1200 = 341 = new length. The same calculation can then be applied to the position of each tone hole and the ventholes. Strictly speaking the reed/staple length should be adjusted as well, but the amount will be small. For a reed/staple length of 45mm, the new length will be 43.8mm. This seems to work quite well for small intervals, but the bigger the interval, the more that other factors come into play which will disturb the usefulness of the result.

                                                                           SEE NEXT PAGE FOR FINGER CHART