Notes on the Diaphone

By JOHN COMPTON

None of the Hope-Jones tonal innovations has been more misjudged than the Diaphone. This ponderous stop, however, is usually much more highly esteemed by the organist who uses it than by the organ-builder, whose business it is to tune it; and the reason for this is not, I think, a very subtle one. To the organ player the diaphone is a useful and impressive pedal bass for the louder combinations, because he rarely or never uses it alone, and he does not worry very much about its precise tone character. The skilled voicer and tuner, on the other hand, is accustomed to judge every organ stop with his voicing-shop ears; he cares little about its effect in combination, but must needs try it up and down, note by note, deliberately searching for the technical defects which will no doubt offend his ears, but which are probably quite unnoticed even by the trained listener when the stop is used in the normal manner. I do not by any means underrate the importance of minute regularity and perfection of speech and tone in organ stops of all kinds. It should always be demanded, if only as evidence of good design and skilful workmanship. But the diaphone is not primarily or essentially a solo stop, and its musical value is not totally dependent upon the individual perfection of its notes. It may frankly be admitted that the best diaphone to be found in any Hope-Jones organ is more or less irregular and faulty in tone, and can seldom be used with really good effect except in combination with other stops.

None the less, I consider the diaphone a most valuable and desirable voice, comparable in importance with the tympani of the orchestra. It usually has a harmonic development equal or superior to a robust violone or Schulze contrabass (but with immensely greater power), and therefore its pitch is very much more definite and its stride more aggressive than that of the usual "pedal open," wood or metal. In rare instances, as, e.g., at St.Clement's, Ilford, the manual diaphonic horn can be used with excellent effect for horn passages in orchestral transcriptions ; and some specimens blend very satisfactorily with the diapasons. But in the majority of his British organs Hope-Jones did not continue the diaphonic construction above middle C or F; from this point to the upper end of the manual compass, he usually substituted high-mouthed flue pipes of indefinite tone quality, - often windy and unsteady. The example at St. Michael's, Chester Square has no treble at all. It plays in 16ft. and 8ft. pitch on the pedal, but on the great (forty-two notes) in 16ft. pitch only, - a very singular arrangement! In some early instruments the diaphones were playable only on the pedals ; in 32ft. and 16ft. pitch at Worcester and Edinburgh, and in 16ft. and 8ft., or 16ft. only, in other places.

Our present purpose is to consider how to avoid the defects which have prevented the more popular employment of this valuable means of tone production. I will not describe in detail any of the earlier and more complex constructions, which are sufficiently illustrated and explained in Wedgwood's Dictionary of Organ Stops, and other standard works; but will confine my remarks chiefly to the type of diaphone which has proved more serviceable.

It will be convenient to discuss the subject in three parts: (I) the Windchest; (2) the Vibrator . (3) the Resonator.

I have put the wind-chest down for first consideration because its influence on the speech and tone of diaphonic pipes, though highly important, has apparently been neglected not only by Hope-Jones himself, but also by the few organ-builders who, since his day, have employed organ pipes of this class. Every experienced voicer knows that some reed pipes which have been adjusted to speak quite satisfactorily on the voicing machine are nevertheless apt to be faulty and troublesome when removed to their permanent position in the organ. This difficulty occurs most frequently with broad-tongued, non-harmonic reeds, or in stops such as the vox humana which have relatively short resonators; and seldom or never in the narrow-tongued French bombarde, trompette and hautbois. A full explanation of this phenomenon and its incidence would involve a long digression and an intimate examination of the acoustical conditions attending the speech of reed pipes in general, which is beyond the purpose of this article. I will therefore say briefly that in a broad-tongued reed or diaphone (which is but a variant of the reed form) a relatively large area of the vibrator is exposed to the influence of the air column in the resonator. Within certain limits the frequency of the vibrator (and therefore of the note produced) is controlled by the natural frequency of its associated air column. Thus the vibrator (be it a reed tongue or a diaphone spring) having a frequency of, say, one hundred and twenty-eight vibrations per second, may, when a resonator of suitable length is applied, have its actual frequency reduced (i.e., its note will be flattened) to one hundred vibrations, or even less, per second. This is a matter of common knowledge and experience.

But it is not so commonly known that the note of the vibrator is also liable to be affected (and sometimes controlled very powerfully) by the resonance of the body of air in the windchest, or in the groove immediately beneath the pipe, or even by the short air column in the socket of the pipe, often with very mischievous results. This fortuitous influence is most commonly encountered in the voicing and tuning of vox humanas, which have only a short and relatively impotent air column in their resonators, and are therefore very much at the mercy of the resonant cavities beneath them; and in diaphones, which have a large vibrator area exposed both to their proper resonators above and to the windchest below. When the windchest has a natural resonance of a frequency not far removed from that of the pipe the two are apt to disagree, sometimes with horribly discordant effect. The vibrator seems to hesitate as though it did not know whether to follow the lead of its resonator or to yield to the compulsion of the windchest resonance. Under these circumstances the pipe will sometimes settle down to a weak, rather characterless note a little below its proper pitch. At other times, particularly if the temperature has fallen since the last tuning, the vibrator apparently tires of compromise and coercion and decides to sing a note, harsh and dissonant, at its own natural pitch, i.e., at an indefinite interval perhaps a third or a fourth, or even more, above the proper note of the pipe. In its less violent manifestations windchest resonance can still be very troublesome. It is one of the commonest causes of irregularity in the power and tone quality of reed and diaphone pipes, often affecting them in such manner as to make accurate tuning and regulating very difficult and to cause them to go out of tune again very soon after adjustment. Every Hope-Jones diaphone has some note or notes with an unduly heavy fundamental tone, and other notes with insufficient fundamental. This is no reflection upon the voicer's skill, for the most accomplished voicer in the world cannot overcome the difficulty by manipulation of the pipe alone, nor until the malefic influence of the unwanted resonator is destroyed. Many expedients have been tried in the attempt to overcome this objectionable interference. Harmonic (double, triple, or quadruple length) tubes are helpful, as also are short sockets, for reed trebles; but diaphone basses cannot very conveniently be made of double length, nor can their sockets or windways be very materially reduced. The whole trouble, in so far as it affects diaphones, can be avoided by using a windchest whose "note" is at least half an octave higher than that of any pipe standing upon it, or a whole octave lower than that of the largest pipe. Or better, by contriving that the windchest shall not resound appreciably to any note at whatever pitch. When I speak of the windchest I intend to include also the windtrunk together with which it sometimes forms a long and very powerful resonator. To ensure prompt speech it is of course necessary that the diaphone windchest shall have ample pallets and windways and a really rapid action, none of which were ever lacking in the Hope-Jones organs, whatever other shortcomings they may have presented.

We will next consider the vibrator mechanism. In its simplest form it consists of "a valve which is not necessarily a spring, combined with a spring which is not necessarily a valve," as its inventor described it. This is known as the valvular reed, and is the most common and useful type of diaphone. The valve, usually a disc of metal or wood (sometimes cork) faced with felt and leather, is carried on the free end of a vibrator spring of flat steel, brass or aluminium. The spring is usually forked at the lower end to facilitate adjustment or removal, and is gripped between two cast iron blocks. This disposition is clearly shown in the annexed drawing of an example by Hope-Jones. The vibration rate of the spring and valve combination varies according to the length, width and thickness of the spring and the weight of the valve. One notable difference between a diaphone and a reed pipe is that the diaphone is not provided with a tuning wire, the vibrator spring being too thick and inflexible to be conveniently adjusted by means of a tuning wire or similar device. Moreover, it is not at all important that the vibrator should be exactly tuned; usually it may be flattened or sharpened a quarter-tone or more without noticeably affecting the tone quality or power. All close tuning is done at the top of the pipe, as with flue pipes.

The tone character of a diaphone is of course dependent to a certain extent upon the form of its resonator, of which we will speak in due course; but it is also affected in an important measure by the treatment of the vibrator. To produce a very free, reedy tone we may use a long vibrator and a fairly light valve; or, if we want an intensely smooth, fundamental tone, we shall probably keep the spring short and stiff, and load the valve heavily. If we set the valve too far away from the shallot or block face, it will either be slow of speech or quite dumb. If it is set too close, it may again be dumb or it may sound a weak or a rough note. In fact, the setting of the diaphone spring and valve is in this respect almost as critical as that of a reed tongue. The Hope-Jones method of mounting and adjusting the vibrators (as illustrated) is a very crude one. The voicer had no means of altering the setting of the valve except by the laborious method of bending or twisting the spring. A valuable modern aid is the fine screw adjustment whereby the vibrator can very quickly and accurately be set in the best position for good speech and tone. The relative inaccessibility of the vibrators - especially those of the larger pipes - is another palpable defect of the old method. The lower ends of all large diaphone and reed pipes should preferably be so designed that the vibrators can be inspected and adjusted without lifting or disturbing the heavy resonator tubes. This construction, which I have always used, together with the above mentioned screw adjustment, enables the voicer to obtain the best results with minimum labour and profanity. Diaphone voicing was formerly a formidable task, requiring endless patience and involving a lot of heavy manual labour.

Other forms of diaphone mechanisms were used in a few Hope-Jones organs both in this country and in America. At Worcester Cathedral and at the McEwan Hall, Edinburgh, may be found a double valve device which operates by the agency of a pneumatic motor. In the 32ft. octave this type is fairly satisfactory. The Edinburgh specimen was in remarkably good form in this part of the compass when I tried it last month. But it does not work so well in the upper octaves, where irregularities of various disagreeable kinds very quickly develop. One or two other types were tentatively tried, but not with great success; and I am of opinion that for all-round satisfaction there is nothing to equal the simple and not too expensive "valvular reed." I have said that the form of the resonator to some extent determines the tone character. For fairly obvious reasons there cannot be so much useful variety in true diaphones as in reed pipes; but, nevertheless, they furnish quite a reasonable range of tone colours. A resonator of conical form, as of a trumpet pipe, gives a tone with the harmonic series I, 2, 3, 4, 5, &c.; while a cylindrical or parallel-sided one will suppress the even numbered partials and will give a tone approximately similar to that of a gedackt or clarinet. An exceedingly intense tone, almost purely fundamental, can be got from a large scaled diaphonic pipe of almost any shape by very "close" voicing,- i. e., with a short vibrator tuned as sharp as possible without over-blowing. But if any noticeable harmonic development is desired, the precise form of resonator becomes more important. Resonators of several distinct forms are employed in the new organ at Shepherd's Bush Pavilion. The 16ft. diaphonic salicional has tubes of small scale, partly conical and partly cylindrical. The I6ft. diaphonic diapason has large scaled conical tubes, with cylindrical additions in the lowest octave. The 32ft. diaphonic subbass has double cylindrical tubes, and in addition there are I6ft. cylindrical basses of large and small scales. It should be noted that the walls of diaphone pipes need to be much thicker and heavier than is customary with light wind basses, and that they must be very firmly stayed to prevent the jarring sounds which so often accompany and spoil their tones.

Diaphones are very accommodating in several ways. They may be mitred with impunity, a fact which makes them especially useful for situations where there is not sufficient height for long flue pipes. Also, they can be enclosed in swell boxes with perfectly good results, as their pitch and speech are not appreciably affected by enclosure.

I have said very little about diaphones as manual stops, because there is some difficulty in making reliable and satisfactory trebles. With heavy wind pressures it is quite possible to continue them as high as treble G, or possibly to the C above the staff. breaking at this point into large-scaled flue pipes or very wide-tongued harmonic reed pipes. But it is doubtful whether the musical value of diaphone trebles is enough to counterbalance their cost of manufacture.