by Walter J. Goodey, F.I.S.O.B.

Many people have asked me whether Haskell pipes have a real use in organ building and my answer to them is "Yes, very much so".

The Haskell pipe was invented by a Mr. Haskell, working in America, and when Reginald Walker visited the U.S.A. in the early 1930s, he discovered that the Patent rights had ceased, and he was therefore able to bring back a few details of how these pipes were assembled.

Haskell pipes are one of the few gimmicks, or for want of a better word, substitutes, in organ building that, with careful manufacture and voicing, give satisfactory results.

I am not going to discuss the scientific reasons for this phenomenon, but I do know that several of the Universities have borrowed Haskell pipes from me over the years to determine how and why it happens. I have seen smoke tests, but this does not convey much, except that the smoke remains static within the inner tube, and moves more or less the same as for a normal pipe from the mouth upwards. I have not heard of a satisfactory answer as to exactly how this type of pipe works.

There are, I believe, a few Haskell reeds about, but I only saw one, many years ago, and from what I can remember, it wasn't very successful, as it made the tone too smooth, and with modern tendencies to have a brighter type of reed, I decided to leave these alone, as if it is height and space that matters, then the half length reed with modern voicing is a better alternative.

In this article, therefore, I am concentrating on Haskell flue pipes and the means I have used to produce the tone of a full length bass by this method.

First, let me say that Haskelling does not halve the speaking length of the pipe. As a rough guide, you can make a 16ft CCC of a Pedal Open Wood go down to 32ft DDDD if the scale is large, but to allow a little margin, I would say 32ft EEEE for all useful purposes. It should be remembered that to Haskell an old wooden Pedal Open, even with four notes added will only give you a small sea le 32ft and you may find that the tone will not carry well in an acoustically dead church. It must be borne in mind, however, that the scale of the outer pipe is the scale that you would normally use, thus, if using an old Pedal Open Wood, you may need at least eight or nine new pipes at the bottom to increase the scale to a desirable 32ft Double Open Diapason.

The inner tube for these pipes can either be made from wood or zinc. I prefer zinc inner tubes, mainly because if the inner tube should split, then it may be a difficult job to repair when the pipe is in position in the organ. If using a wood inner tube, then I would suggest that the middle board down the centre be well glued and the seams leathered. The 32ft at Perth Cathedral, Western Australia, was made in this way, but owing to the dry climate, I decided to bed and screw the front board above the mouth, so that the inner board could be got at if it should split. All joints on this octave were also screwed. The complete wood pipe, of course, would have a stopper on the inner tube and either the outer or the inner tube can be cut if found to be too long when installed in the organ. The zinc tube can easily be fitted to the top of the pipe with a bar or metal band fitted to a hook over the end of the pipe. This, of course, would depend on the size and scale.

The inner tube should protrude above the top of the outer pipe by at least one sixty-fourth of the overall length. For example, with a 32ft pipe, the inner tube should protrude at least six inches. This distance can be varied according to your own personal likes - generally if the tube is pulled up higher above the end of the outer pipe, the tone becomes broader, but if pushed down, it will thin the tone a shade, but the pipe will tend to give trouble if you lower the inner tube to the level of the top of the outer pipe.

At this point I think it would be worthwhile to let you know the details and dimensions that are used. Very briefly, the inner tube is half the area on plan and the length of the outer and inner tubes added together is the same as the normal length of the pipe. Thus, if we are Haskelling an 8ft pipe, then the outer body can be 4ft 3ins and the inner tube 3ft 9ins. It must be borne in mind that the inner tube must not go in so far as to interfere with the column of air from the mouth, as you will soon find that the note will become unsteady. Likewise, a short piece inserted in the top of the outer body will not have any more effect than fitting a lump of wood, or what-have-you in the top. In my opinion, the inner tube works best if it goes down at least half of the length of the outer pipe. This is worth remembering when deciding on the length of the first pipe that will not stand because of height. I would see that my inner tube was a worthwhile length, going at least halfway down the pipe, and then I would step up the length of the rest of the pipes to be Haskelled so that the lowest note is about the same height as the last full length pipe.

As a guide. a good 32ft can be fitted into a height of 18ft. This could be reduced if the pipes are conveyed from the chest - there must be at least 8 inches over the top of the pipe to allow the tone to circulate. Likewise, a 16ft Salicional or Dulciana can be produced in 1Oft 6ins - again, this could be varied slightly by the length of the foot and the type of chest used. An 8ft bass can be reduced to approximately 5ft 6ins by the same method.

Most types of open flue pipes can be Haskelled, and the Voicer will find that, if anything, he can slightly increase his cut-up of the mouth as these pipes tend to take a higher cut-up than normal. I have found that it is possible to Haskell Gambas and string tone pipes quite successfully, and to help the tone, the inner tube can be slotted in the same way as you would a normal pipe - the only problem is that the method of tuning will be more costly by means of fitting a long wire to the tuning slide and cleating it to the inner tube so that it stands above tile top of the pipe, like a long reed spring.

On the larger metal basses, up to and including 16ft Dulcianas and Salicional5, a zinc slide slotted with a bolt and wing nut with wa,-,hers should be fitted to lock the slide when it is in tune. Care should be taken to see that the inner ttibe is clipped onto the slide and hangs down the side of the pipe. It is not necessary to hang the inner tube in the centre, but is desirable to keep it to one side or the other, so that it does not hang across the pipe. I used to fit two springs onto my inner tube to keep it over to one side, but you will find it is best not to hang the tube immediately over the mouth. A small piece of felt on the lower end of the inner tube will prevent any rattles.

In deciding the diameter of the inner tube, I have found that a ratio of 7 to 5 works very satisfactory. In wood pipes, half the area. Now let me at once say that this inner tube diameter should not be increased, in fact, a fraction under does not matter, and I always told my metal hands to take care not to exceed the required inner tube diameter. Some of you may have found that a slightly smaller inner tube diameter helps your voicing, and I say carry on, but I do know that if you increase and go over the diameter ratio, then the note will soon be troublesome.

Many people have asked me if it is necessary to beard these pipes and my answer it that if the full length pipes are bearded, then the Haskell pipes will need the same treatment. In fact, providing the pipes are accurately made, there should be no problem to get, say, principal basses to speak well without a beard.

A good Voicer should be able to match the tone quite easily and make it difficult for the organist to discover which is the first Haskelled pipe in a particular octave.


My first experience of having to work out a scale for a 32ft where I had only I 8th length and limited space, came in 1948, when we installed the new organ into St. Mary's, Oxford. John Webster, the organist at the time was very keen to have a full length 32ft, but this was not possible owing to lack of height and space oil the organ screen. We decided, therefore, to Haskell and lay the pipes horizontally across the screen, with the outer body in wood and the inner tube in zinc. The problem was, how to adjust the length of the pipes when the organ was built around it. The method I used was to fit a stopper into the zinc inner tube so that if the pipe was found to be too long I could move the zinc tube out, remove the stopper, cut off the desired amount, and replace the stopper, using an iron bar to secure the tube to the pipe when tuned. This worked well, and I only had one pipe on the long side, due to the fact that it was shaded by a building frame post.

No doubt you are all familiar with the methods of calculating the lengths of pipes from a given scale with a good wave length rod and a scale board. I have every confidence in this method, and if your calculations are correct, you will soon find that the Haskell lengths are the same added together as a frill length pipe, but care should be taken to see that there is a little extra length on the inner tube if you are fitting a stopper, as we all know that a stopper right oil the end of a pipe will soon become unsound - very important in a Haskell pipe.

In Dulcianas, with the outer tube length being adjusted, the inner tube can be sealed, but make sure that the end is well soldered, as a small leak at this point can cause trouble.

To sum up, Haskell pipes are one of the methods of reducing height without the necessity of fitting a stop bass or other device, and I have found them very successful. The Classical school of thought should take note that a Principal 8ft with open tipped voicing and no nicking can also be produced so that the match between the full length and Haskell pipe is difficult to detect.

In general, then, a Haskell pipe has its uses and if a little time is spent in experimenting, it will be found that it is of great value where height and space are limited.

© The ISOB

This article was first published in the ISOB Journal No. I  Vol. 5 Autumn 1980.

Copies of this issue of the ISOB Journal are available from the Secretary at the following prices:

Table of Contents:

• Foreword by the President

• Meeting at the RAC, London

• Profiles of Peter Collins and Christopher Gordon-Wells

• The Relationship between Wind Pressure, Pallet Opening Area and Pulldown Tension - by Ian Frost, FISOB

• Tenth ISO Congress

• Eleventh ISO Congress

• Alphonse Noterman - by A.A. Noterman, FISOB

• Leather?... Now just a minute! - by Robert C. Colby

• Must if be Ivory - by Herbert Norman, FISOB

• Haskell Pipes - A Practical Survey - by Walter J. Goodey, FISOB

• Some Notes on the Restoration of an Organ Case - by G.J. Jones, MISOB

• Letters