Back to Top ▲
Extract:- Journal of the Incorporated Society of Organ Builders. Volume 1
Minutes of July 1952

The Piston Action Combination Machine

By the President (Mr. Henry Willis)

    This is a vast subject, and I think it best to treat it from this historical angle.

    One point on which there cannot he controversy is that all the improvements in the mechanism of organs, including combination mechanism in all its forms, have been developed by organ builders who have taken the initiative in smoothing the way for the organist. The layman, the non-organ builder, has not made any contribution whatsoever to our improvements in such mechanisms-with the exception, a very striking one, to be referred to later.

    An historical citation may not be precisely correct, for it is a question of extracting some facts and neglecting others.

    As far as one can make out, the first form of combination mechanism was invented by Bishop in 1809, and no doubt most of you have seen very early examples of this fromt the beginning of last century -extraordinarily cumbrous in their application, and the effect being what amounted to a full organ pedal for each division. Clumsy parts were soon improved by Bishop, until a reasonable form of combination pedal was evolved.

    "Composition Pedals" - not really understanding how this term came to he used, I decided to revive my knowledge by consulting the great Audslev's work on Organ Building. I found he put in a vigorous protest against such a tern and gives the correct one as "Pedals of Combination.''

    Combination Pedals constituted a real step forward in control where hitherto none had existed. In the middle of the last century, 1847, my grandfather, the first Henry Willis, invented the thumb piston, made possible by the use of the pneumatic lever, thumb pistons were tried out in two or three small instruments (the first one of which I can find any record is that in Manor Chapel, Bermondsey), leading up to the comprehensive use in the Willis organ of the Great Exhibition of 1851 which, with all mechanisms controlled by the pneumatic lever, together with its tonal resources, represented a real revolution in English organ building and provided a basis for further process.

    The pneumatic combination mechanism in the 1851 organ were not adjustable in the modern and accepted sense, consisting of the use of mechanical double-acting fans each operated by a pneumatic lever. .Alterations of stop combinations continued to take the form of adjustment of the blocks fixed on the drawstop rods, giving '' on '' or '' off movements.

    In the organ for St. George's Hall, Liverpool, in 1854, my grandfather made further progress. Instead of using the primitive form which he did in the Exhibition organ mechanism, he chose simple pneumatic. The use of individual pneumatic motors for each stop enabled a form of adjustability to be incorporated capable of setting by an informed organist. The system was that of a pneumatically operated roller-arm working a pneumatic lever by means of tapes dependent from either the " on " or off" valve, the appropriate tape being fixed to the operating roller-arm by means of a peg. It will be noted that a "neutral" position was readily obtained by the non-connection of either tape. This somewhat primitive method of adjustment could be used by the knowledgeable organist.

    It was used in the organ in St. George's Hall, Liverpool, of which I was in charge for some four months in 1907, due to the illness of our Liverpool head tuner, where every Saturday afternoon and evening recitals were given by Dr. Peace. I was frequently called upon by him to alter the combinations for him, the afternoon performances being devoted to a programme of a classical nature-or what was termed " classical " in those days and regarded as pure organ music---and the evening recitals consisting of decidedly popular music--hence the need for re-setting the combinations to take care of essentially differing programmes.

    The form of adjustability at St. George's Hall was not repeated by my grandfather, for two reasons, its undoubted cost, and because organists indicated no desire for such facility. Now, however, such benefits are being placed before the organists, they are beginning to learn how to use them.

    I believe I am correct in saying that the first fully adjustable piston action which enabled the organist to set his combinations whilst seated at the console was that by Hilborne Roosevelt in 1882, this is the earliest form that can be traced. This design provided a series of tilting tablets placed over the drawstop jambs each side of the manual keys-one tablet for every drawstop on each piston. The tablets were set to push pins into position to be engaged by the mechanical fans either "on" or "off" as set-the tablets themselves being locked firmly in position by springs, a neutral position was not provided for.

    I was first in America some 27 years ago and I saw an old Roosevelt with this old form of adjustable combination action, and it still worked. It was noisy, and despite its apparent cumbrousness, it marked a good step forward. That design of Hilborne Roosevelt started other organ builders thinking, particularly in America, and developments more or less on those lines followed with some rapidity.

    Following Roosevelt's design in 1882, my, grandfather, in 1893, and his son Vincent, evolved a type of adjustable combination action which, following somewhat the Roosevelt application, used miniature drawstop knobs instead of tablets, and that type was used in Hereford Cathedral in 1893. These miniature knobs controlled pneumatic switches, which worked the "on " or "off" relays working the individual pneumatic motors, a central, neutral, position was provided, it was very simply done. That system at Hereford was received very well indeed by the late Dr. Allcock.

    That was another step forward.
Duval capture system
Fig. 1
Click for larger image

    Prior to that date great progress and an immense step forward was made by Salluste Duval of Canada by his introduction of the" capture " system in which the desired combination for any piston was drawn and then locked upon it by the pressing of the appropriate setter pedal. It included the use of a separate set of pedals for each department. But Duval did initiate the capture system, by means of which the player draws the stop required, and then by pressing a pedal locked it on the combination pedal or piston desired. This took the form of rockers capable of " on " and " off" settings giving pin engagement of the fan arms together with a setter mechanism to move the rockers to the "on" or "off" position. (See Fig. i.) This, the Duval capture system, is the one example to which I must call attention as being the invention of a layman, a non-organ builder, an invention of very great value to organ building and organ playing.

    Duval's invention was immediately taken up by that great firm in Canada, Messrs. Casavant Freres, and from 1891 onwards you will find in Casavant's work, mechanical combination systems-noisy, but still working. Subsequently many systems were introduced, but all based upon Duval's principle and not requiring our detailed examination, but Hutchings of U.S.A., in 1891, did initiate the utilisation of one "master" setter pedal or locking button in place of one for each piston. Developments later took the form of using Duval's principle with pneumatic, and later, electro-pneumatic, operation.

    Finding the Duval principle, together with the somewhat elaborate mechanism for operating the drawstop knobs Costly in construction, some organ builders in both England and America evolved a cheaper system of" blind "combinations the drawstops or tablets not moving and the indication of stops being "on" or "off" given by indicators taking visual forms by tablets, coloured lights and so forth-this principle was not found generally acceptable.

    In 1891, the late Mr. J. J. Binns introduced the Duval system into this country and actually obtained a patent for its use: he used the Duval system, but the form of application was different.

    About this period, although the deeper thinking organ builder was mentally progressing along the lines of combination actions, not only in America but in this country, many undesirable systems of blind combination actions were devised and brought into use.

    In this country, it is strange that despite the introduction by Binns and the use of the Duval system, it did not seem to catch on and we did not take it very seriously, and other inventive faculties concentrated on tubular pneumatic systems, with an individual capture system for each department. Adjustability was chiefly by means of pneumatic and later electric switchboards, not always readily accessible to the organist. There was provision of one or perhaps two fully adjustable pistons for each manual--engraved "A " in indication of adjustability, the setter or capture button being placed alongside, over the draw-stops, or in any convenient position. There were many examples at the beginning of the second decade of this century, up to 1910 or 1912, the most common form met with having a normal set of pistons, which were fixed and probably only capable of alteration by means of pneumatic and later electric switches, but in addition there would be one or two adjustable pistons for each department--one the more common.

    Just after the 1914-1918 war, my own firm in its rebuild at the Victoria Hall, Hanley, used a system of a separate setter button for each adjustable piston-rather clumsy, but it still works. This showed a tendency towards the use of the capture system.

    The system used was that of an all pneumatic setting of simple form individually but elaborate in complete make-up, the principle embodying a pneumatic "on" and "off" valve, worked accordingly as a drawstop, was drawn or not at the time of pressing the setter button. The drawback of this principle was the fact that any inquisitive but unknowledgeable person could, and did, by pressing the setter button "to see what happens," unknowingly alter the combinations as set, often all to the "off" position, to the chagrin of the unfortunate organist.
John Austin adjustable combination action
Fig. 2
Click for larger image

    Reference must be made to the early and exceedingly clever invention of John Austin, first used on a grand scale in the Austin organ at the San Francisco Exhibition of 1912. On a visit to his Homeland in 1911-for John Austin was an Englishman by birth-he showed me a model of his newly invented adjustable combination action which, wholly mechanical in its operation but operated by one very large solenoid unit, is well known to you all and is shown in diagrammatic form on the drawing exhibited (see Fig. 2). We should take off our hats to John Austin for an exceedingly ingenious and well carried out design of that period.

    This Austin device operated as follows-the piston on which the desired combination was to be set was pressed in and held-then the drawstop knobs and/or tablets were moved to either the "on " or "off" position, the movement forcing over the small rockers to the desired position. This Austin invention worked perfectly and with extraordinary reliability, as did all Austin mechanisms, but apart from the perhaps undue noise in operation, due to the single heavy duty solenoid and the all-mechanical connections, the chief drawback was the time taken to set up the combinations on a large instrument. It will be obvious that with one hand immobilised by holding the piston in, the other could but manipulate the draw stop knobs one at a time.

    My friend, the late Dr. Cunningham, told me that before a recital on a large Austin of some 150 stops, it took him two hours to set up his combinations (a ten-piston job). Another drawback was that the combination action was noisy. But John Austin made a memorable and useful contribution to the design of adjustable combination mechanisms.

    In passing, the German Frei-Combination system should be mentioned, but it is not a form that has made any appeal in this country, and I do not feel it necessary to devote time to its description, which is well known, and disliked, by all here.

    The "Capture " system made its inevitable conquest, and about 1922/23 we commenced to use the all-adjustable capture system in this country, and at first, at least with my own firm, it took the form of an improvement on the Duval principle-with adjustment rockers, and one setter or locking button.

    At that time also there was a system in America used by Wirsching (see Fig. 3), the only difference being that the actual valves were mechanical---they were operated by pneumatics. The system used by us subsequently at St. Paul's Cathedral, for example, and St. George's Hall re-build, was on that principle.

Fig. 3

    To ensure clarity it will be well to mention the exact operation and use of the modern capture system.

    There is one setter or locking button, usually placed to the bass of the key-slip of the lowest manual. To set any combination the operation is as follows :--
    (1) "Draw" the stops or tablets desired on the piston to be set.
    (2) Press the setter button, and while pressing it---
    (3) Press the piston on which the desired combination is to he set-this sets the combination.
    (4) Release the piston.
    (5) Release the setter-button.

    The use of the named cycle of operations is essential in the order named.

    There are several designs and methods by which the capture system can be carried out and which can be named in the order of development.
    (a) The all mechanical system as shown in Duval's original specification.
    (b) The pneumatic application, as shown in the Wirsching adaptation of Duval's principle-and the electro -pneumatic use--an evolution from it. (The use of the description title " electro-pneumatic " is due to the use of " electrics " for the piston circuits--" pneumatic," the motors to operate the rods on which the adjustable rockers are fitted and which are directly connected with the drawstops or tablets.)
    (c) The electro-pneumatic system on which the adjustable mechanism is apart and independent from the drawstop movement itself--first designed for remote control use--or with electric or electro -pneumatic mechanism to move the stops and/or tablets.
    (d) The wholly electric system capable of either incorporation in the console or remote control as desired.

    The difference between (c) and (d) is merely that of the form of power used to operate the adjustable machines, and in view of greater simplicity of the all-electric system, my further remarks will deal with that use which is, without doubt, the simplest in use and, if of correct design and perfect construction, the most reliable and hard wearing.

    The requirements can be enumerated in simple form, viz., a switch, operated by the single setter button, to cut out the normal piston circuit and bring into circuit that for piston setting-the apparatus to bring the required combination contact either "on" or "off."

    The time came when we found that if we were to supply a plentiful number of pistons, including adjustable pistons, our consoles were going to get too big. In the Willis organ in Westminster Cathedral, for instance, the size of the console is unreasonably large. It so happens that there is enough room, but nevertheless the console is very large. Then organ builders began to think of remote control. Having been to America, I did not feel too happy about it.

Fig. 4
Click for larger image

Fig. 5
Click for larger image

    In America there is the all-electric apparatus made by Reisner, the manufacturer of electrical components for the organ, utilising adaptation of the older mechanical and pneumatic systems, solenoid, to make contact with either the "on" or "off" bar. The model exhibited and now to be demonstrated embodies three drawstop knobs and two tablets-all coupled up' on the two adjustable pistons, in addition a "cancel" piston; the one tablet and one knob being also subject to reversible movements. Designed to work on 12 volts-this design was of 1928--first used in 1929 and subsequently, with minor improvements', used in all "Willis" productions on the "capture" system (Fig. 4). The model is in accordance with that first used in 1929 and normal operation is, for precision reasons, on 14 volts (Fig. 5).

    Next is a very neat and admirable system designed and used by Hill Norman & Beard Ltd. (Fig. 6). Again, it is capable of incorporation either in the console or for remote-control use, the operation of which is simplicity itself, as shown by the drawing (Fig. 7).

Fig. 6
Click for larger image

Fig. 7
Click for larger image

    Mr. HERBERT NORMAN.-This has a basically different approach from any so far illustrated. Mr. Willis's admirable action, which has been in use for a number of years, is really a form of static switch or, if you like, a setting switch with remote-control and connected through a normal relay operated by the thumb piston.

    Our systems, both this, and the one previously used, work on the principle that there is no static switch. The use of adjustable pistons is so casual that to have no switch that is not automatically switched and kept self-cleaning is in our view a disadvantage. In our system, contact is newly made every time you touch a piston and therefore every contact is self-cleaning, even after many years. In our relay the actual switched contact is moved when you press the thumb piston. The actual sequence of events when you touch the thumb piston is that instead of a relay being operated and its current routed through the adjustable mechanism, the contact in the relay is capable of being switched to the " on" or the " off" position. Besides reducing the number of contacts, it also enables an extremely compact machine to be achieved, one which very readily allows of electric operations; unlike some of the earlier ones, which were electro-pneumatic.

    Mr. WILLIS.--The next design I am going to refer to is that admirable invention of Mr. Taylor's of nearly twenty years ago, which has been used by Messrs. Compton ever since then. It represents a high development of a very compact nature. An important factor is the simplicity of the wiring.

    Mr. JAMES TAYLOR.-The model which I shall now demonstrate is in my opinion the simplest possible approach to the problem: every other piece of mechanism I have seen has to go rather a long way round to get the difference between on and off. In this case it has been done with one moving part for the setter, one moving part for the stop, and one moving part for the piston.

    Each stop has a horizontal bar, having a bakelite centre and silver on each side, the front strip being connected to the " off" movement of the stop, and the back one to the " on." Each strip is operated by a lever magnet, which is energised only when the stop is on and the setter piston in use. Each piston has a vertical strip of bakelite perforated with triangular apertures, so that when the strip is in the" off" position, the contacts can move freely without becoming displaced from the silver bars. See Figs. 8 and 9.

Fig. 8
Click for larger image

Fig. 9a
Click for larger image

Fig. 9b
Click for larger image

    To fix a combination, the setter piston is pressed, after putting down the stops required, and immediately the horizontal bars corresponding to the stops you have put on move across. When a thumb piston on which a combination is required is operated, the vertical strip with the triangular slots is pulled down. Release it, and release the setter piston, and instead of being in the left-hand slot, it has come into the right hand one and therefore come from the " off" to the "on." This mechanism has been in use since 1934- In the organ in Hull City Hall there are 48 double-touch thumb and toe pistons, adjustable on both first and second touch.

    To avoid cabling-one would otherwise have to match these contacts to a remote relay-you can see some corresponding strips which come down, they are just a continuation of this contact, which comes straight through the middle member on to some silver wires behind and connects to the contact and according to the position, either supplies the "on" or the "off" position. The same simplicity of adjustment we keep to on all capture actions.

    Mr. WILLIS.-Interesting drawings have been sent in by various members and are worth some study.

    I would like to deal with one or two other systems, viz., Hiligreen-Lane, etc.

    They are interesting but rather clumsy to-day, particularly in view of the two latest models examined. With the utmost respect to our American friends, I believe they have nothing to teach us on these combination capture systems.

    The one and only drawback of the capture system is its cost-it is and must remain more costly to produce than the simpler principle of hand-operated setter switches as used for the smaller instruments.

    The alternatives to a fully adjustable combination system are systems which involve the use of setter-boards or setter-switches. They are quite reasonable for use with small organs where the possible permutation of combinations is very limited. Messrs. Walker have sent a small model of a type of setter switch, the switches being either over the drawstops, behind a movable music desk, in drawers, or similar, on either side of the console. The actual application varies a great deal, but the system is very simple, the switch going to the " on " or the "off" position.

    In conclusion, we all know that in many cases in presenting this beautiful form of combination, it is just not appreciated in some quarters, but now a younger school of organists is beginning to appreciate the enormous convenience of what is placed at its disposal and learning to use the adjustable combinations in the correct manner, altering the setting not only for services and recitals but in between pieces.

    In this short review I have omitted reference to various alternative selective electro -pneumatic types with which we have experimented in this country but discarded on account of complexity and easily getting out of regulation. In all these systems, however, the basic principle remains, and must remain, the same.


    CHAIRMAN: I think you will agree that the talk to which we have just listened is one of the very best that has been given to the Society. The subject has interested organ builders for very many years, and it shows the way to do things which modern organists are definitely demanding. But some older organists still prefer to have their pistons absolutely fixed.

    I agree with Mr. Willis that in an important job, particularly where recitals are going to be given by various organists, it is a very great advantage to have the capture system.

    MR. H. NORMAN: I think Mr. Willis's Paper is characteristic of the research which he always puts into a matter like this. At first it looked a very simple subject, but when you look round these walls and see the types and systems in the many diagrams there and then remember those you have read about and others you have not read about, you realise what a tremendous amount of thought has been put into this one part of organ mechanism. To cover the whole field in a short review is an impossibility.

    The problem of the use or non-use of the adjustable piston mechanism I think has had a very big influence on the design, and perhaps Mr. Willis has not said enough about the importance of the quality of the materials. I think that is particularly interesting, because if you look at the successful British designs and then you look at the presumably successful-because they exist in large numbers-American designs, you cannot help noticing that we use superior contacting metals. They seem to get away with phosphor bronze and so on under conditions which would be quite hopeless in our climate, and probably they can do this because of the greater use and therefore self-cleaning of the mechanism. The use of adjustable pistons on the other side of the Atlantic is quite staggering, the idea of even a small organ with even fixed pistons is looked upon as n trograde and even as unacceptable. Their use of thumb pistons is quite amazing, and the organ sounds very different from player to player because of the difference in the setting of the combinations. They tend to use far too frequent tonal changes, changes in the middle of a phrase even, things which no orchestral writing or scoring would permit, then facility of mechanism intrudes upon good musical taste. How far can we be held responsible for that sort of thing ? Evidently from what we see here, we have stood up to the problem of how to do it.

    MR. H. WILLIS: This point has not been touched on, but it is rather obvious that these mechanisms have to be very carefully made and the components must be dead accurate, e.g., Messrs. Hill Norman & Beard's model, where machine-like accuracy too many thousandths of an inch is achieved. But one takes that for granted, and that is why I omitted to touch on it in my opening talk.

    No Church or Cathedral organist wishes to have his piston combination set-up messed up by a deputy. We supply a lock beside the setter button, and whoever has the key can use it. This is good practice and very advisable.

    MR. JAMES TAYLOR : Another answer to our President's remarks is, of course, to put in two sets of pistons, and that is what was done at the B.B.C. organ-put the key in and combinations can be altered. The mechanism can be enlarged by putting in some extra bars worked by the same magnets. If he wishes to adjust the combination, the organist puts on his stops, presses the setter button, and the tuner works the setter by hand, and once altered they cannot be altered again until the tuner comes in again. There are 36 actions along a bar. For other pistons you allow another inch it is not worthwhile for a few pistons, but very worthwhile for many.