As the hammer-idea evolved itself in the mind of Cristofori and the other experimenters who were contemporaneously bending their efforts towards the improvement of the dynamic possibilities of keyed instruments, we may be certain that much perplexity was caused them by the problem of providing some mechanism that should form the connecting link between the hammer and the key. We may understand how the ingenious Italian must have labored, with the picture of the dulcimer and its player continually in his mind, to obtain an efficient mechanical substitute for the uncertain stroke of the hand operated dulcimer-hammer, as well as the dynamically incapable harpsichord jack and quill. It is surprising to note, indeed, how early the present ruling principles of action-mechanism were elaborated by these pioneer workers. Cristofori, as we have already mentioned, obtained, ere he died, a complete check to the rebound of the hammer and a form of escapement that may be seen unaltered in essentials in surviving square pianofortes of the “English action” variety. We may further reflect that the invention of Backers, as improved by John Broadwood the First, remains today the approved mechanism of many English and other European grands. Not only is this so, but until the general adoption of the Steinway overstrung model, there existed German makers who were content to fit their moderate priced grands with a slightly modified form of the Viennese action invented by that remarkable woman Nanette Stein, afterwards Madame Streicher, the conspicuous feature of which was the mobility of the hammer-butt and the stationary condition of the jack. The fact that these early mechanisms remained satisfactory to performers until recent years is the best proof of the thorough and sure mechanical basis upon which they were designed.

It is not necessary to go into elaborate detail in describing these early actions. Reference to the accompanying cuts will be sufficient to lay bare their actuating principles. We may, however, observe that the radical difference of touch between the Viennese (Streicher) models and those of Broadwood (known as the English grand action) continued to be a source of annoyance to performers until the increasing technical demands of the modern virtuoso school and the improvements in wire-drawing and iron casting made inevitable the supersession of the very light Streicher action by the heavier, more durable and more efficient Broadwood English model, which continued in favor until within recent years.

Shortly before the general adoption of the Broadwood model, however, there was brought forward a new type of grand action, which was destined in its turn entirely to supersede the then favorite English type. This was the invention of Sebastian Erard, of Paris, founder of the still existing house of S. & P. Erard. His new action was termed the “double repetition action,” and it fully deserves the name. By the use of this device the leverage that exists between the hammer and the key is so manipulated that the very slightest touch upon the keys is sufficient to cause the hammer to deliver a blow. It will be observed by reference to the cut that the Erard action differs very largely from that of Broadwood. In the first place, we observe that the hammer heel is no longer provided with a notch in which the jack works, but that this heel is reduced to the smallest dimensions and acts merely as a pivot. Upon the shank of the hammer is fastened a felt-covered roller. The jack acts upon this roller. Between the jack and the roller, however, we observe a long lever, one end of which has a slot through which the jack passes, while the other end is pivoted on to a rail projecting from the wippen. This is the “escapement” or repetition lever, and forms the main improvement of Erard. It will be observed that the repetition lever bears against the roller, and indeed lifts it before the jack can come into engagement with it. The jack, working through the slotted portion of the lever, is not brought into play until the lever has raised the hammer a little way. The long, double spring that acts both upon the lever and the jack gives them their motions, and the result of the depression of the key is that the repetition lever is always in engagement, and when the angular key-motion is not sufficient to bring the jack into play, the work will be done by aid of the lever. Furthermore, the lever operates to maintain the hammer-shank in precisely the proper position for striking, at all times, and without reference to the angular position of the key.

It can easily be understood that such a vital change in action mechanism did not become popular at once. Many pianists objected to double escapement. Notable among these were Chopin and Kalkbrenner; these two being the greatest pianists of their day in Paris. Chopin preferred pianofortes fitted with the Broadwood action.

Nevertheless, the merits of Erard’s invention continued to impress themselves upon the musical world, although Pierre Erard, who had patented his uncle’s invention in 1821, was obliged to obtain an extension of the English patent in 1835, on the ground of the loss occasioned in working it. It did not become generally adopted, even after the expiration of the extended patent, until modified and simplified by Henri Herz, the famous pianist and pianoforte manufacturer. Then its merits, in the modified form, were so generally recognized that it is now in use by Steinway, Chickering, Knabe and all manufacturers of grand pianofortes in the United States; by Broadwood, Collard & Collard, Brinsmead, and others in England; by Bechstein & Bluthner in Germany, and by most French makers of eminence with the sole exception of the house of Erard itself. This eminent firm continues to employ the pure and original form of action as patented in 1821, with improvements only in details of workmanship and material.

Whatever we may say as to the general adoption of the Erard action only after it had been considerably modified, we must not withhold admiration from the Erards, whose genius and courage gave to us the enduring double-escapement principle that has never yet been superseded.

The main features of the Herz-Erard mechanism may be seen at a glance. The roller is replaced by the notch, once more restored to favor, and the repetition lever is made more effective by being pivoted centrally with two free ends, and by having means provided to limit its up and down motion. Moreover, the wippen is shortened and made straight, the single spring is replaced by two, and the damper is once more permitted to fall down on the strings by its own weight without recourse to artificial springs.

The Herz-Erard action has been taken up by American makers, but we find that they have modified it again, after their own ideas. For example, if we look at the accompanying cut showing a modern American grand action of the highest class, we shall observe that the original roller of Erard is restored, the damper mechanism simplified and additional means of controlling the movements of the jack introduced. These two latter improvements combine to make the present American grand action most perfect, and the achievements of the specialist action makers must ever command our respectful admiration.

The reader has now been able to take a comparatively complete survey of the progress of invention in grand pianoforte action mechanism. We shall therefore turn to a critical examination of the modern standard grand action here illustrated, in order that the adjustment of it within the instrument may be understood.

It will be observed that the key (1) is supported upon a key-frame, part of which is clearly shown in the drawing. This frame is provided with a cloth strip at its rear end, also shown, and with a “balance rail” upon which the key is pivoted by means of a pin. The position of this balance rail must be such that the divisions of the key thus made are to one another as 3:2. The part nearest to the player’s finger is the longer. But such a proportion as this holds good only for the actual operative lengths of the keys from the front to the point where the sticker, or capstan screw is placed. For otherwise we have a distinct change in the mechanical relations of the parts of the lever-system, and this, of course, entails a corresponding change in the forces that are brought into operation. It is therefore necessary to have a standard length of key, as between the points mentioned, and this length is placed at 15³/₄ inches. But, in order to maintain the mechanical relations between the dimensions of the lever we must also consider the depth to which the key sinks in front, and also the height to which it rises at the back. If, then, we arrange that the depth in front shall be ³/₈-inch full and the rise in the rear ¹/₄-inch full, we shall be able to maintain the position of the balance pin as already calculated. Such a proportion supplies the very best arrangement. In order to fix the proportions in the reader’s mind we repeat them in tabular form, as follows:

The front portion of the key is mounted upon a rail of the key-frame which is provided with a pin. This pin works in a mortise cut in the bottom of the front portion of the key and bushed with cloth. The pin is oval in form and adapted to be turned so as partially to increase the amount of space occupied by it whenever wear or age tend unduly to enlarge the size of the mortise.

The exact size of all the various action parts that are mounted above and in connection with the keys depends largely upon the exact interior dimensions of the grand pianoforte case. So that it becomes necessary carefully to measure the height from the key-bed to the level of the strings, and also the distance from the front of the keys to the line of damper lifter wires, and from there to the extreme rear end of the key-bed. Likewise the exact position of the middle string of each three-string group, the middle point of each two-string group and the position of each single string must be marked upon a stick for the guidance of the action-maker. The latter will then be able to effect the correct alignment of the hammers, the rise of the escapement lever and jack, and the exact position, in reference to these, of the other parts of the action. It will, of course, be understood that the proportionate dimensions of all the action parts depend upon the measurements that we mentioned just now, as well as upon the dimensions of the keys. The greatest care is therefore necessary in making these measurements.

The junction of these parts with the damper mechanism lies within the province of the action finisher and regulator and will therefore be treated in the chapter on action regulation.

It will be observed then, to return to the consideration of the action in its actual movements, that the depression of the key causes the wippen to rise at its forward end. This brings the escapement lever (4) to bear upon the roller (6), with which it is in continual contact, and raises the hammer-shank and hammer-head (5) and (7). At the same time the jack (3) is raised and its tail end is brought nearer to the button (8), which it finally touches. As soon as this happens the jack is tripped up and its head comes out of contact with the roller. The latter in the meantime, having been raised away from the escapement lever by the action of the jack, falls back into contact with the former and is then borne up on the lever so that the hammer is maintained in position near to the strings. The result of this is that so long as the finger is held on the front of the key, the whole action is continually in proper position to actuate the hammer. The jack is not permitted to fall away, for the roller is maintained in the right place by the escapement lever and it thus happens that a very small angular motion of the key obtained by slightly raising and depressing the finger, is sufficient to cause the whole action to be set in motion, and the operation of actuating the hammer to be gone through as often and as rapidly as required.

The pedal motions of the grand action are simple. The whole action and keys, as we know, are mounted upon the key-frame together, and thus form a homogeneous structure. The “piano” pedal is caused to shift this frame so that the hammers each strike only two of the members of each three-string group and only one of each two-string group. This is effected by the use of a heavy spring that is set in the side of the interior case and bears against the key-frame, and by a lever that operates from a hole cut in the bottom of the key-bed and engages with one of the bars of the key-frame. The pedal is connected with this iron lever so as to shift the key-frame, and the spring operates to push the key-frame back into place whenever the foot-pressure is removed from the pedal.

The “forte” pedal simply pushes up the whole line of damper levers (16) each of which is pivoted on flanges (21) for that purpose. The lifting is effected by means of a rod connecting with the pedal which is projected through the bottom of the key-bed and engages with a rail set under the line of damper levers and adapted to raise them when operated by the pedal-rod.

There are, however, two variations to the pedal mechanism of grand pianofortes that require some attention on account of their mechanical interest. These are clearly illustrated and described in the cut and specifications shown on adjoining page.

The working of the sostenuto pedal and of the new soft pedal device will be readily comprehended by reference to the cut.

It will be noted, in the first place, that the metallic action brackets which were omitted in the previous drawing for the sake of exhibiting the working parts more clearly, are now shown.

The “sostenuto pedal” is devised to permit the sustaining of a chord while the fingers of the performer, for the purpose of continuing the melody, are withdrawn from the keys. It will be observed that the action is provided with a rod (22) which is connected with a pedal and is adapted to be brought into contact with an additional damper-lever (23). This additional lever is provided with a tongue of felt, and when the rod is turned on its axis, this tongue engages with a similar metallic device on the rod, thus holding the damper up as long as the rod is kept in the position of contact. When the key is pressed down, the damper rises to clear the string, and brings the additional lever up with it, so that, if the rod is now caused to revolve, the tongues on each engage one with another and the damper is held up as long as the pedal is pressed down and irrespective of the position of the key. Provided that the keys are first depressed, the pedal will always hold up as many of the dampers as are thus raised. The result is to sustain the sound of the strings after the hand has been removed from the key. This gives a more permanent character to the harmonization of a melody, without the dissonance that comes from the releasing of all the dampers by the ordinary “forte” pedal. The “sostenuto pedal” is generally situated between the others in the lyre or pedal case at the level of the player’s feet.

The second device is also most interesting. It consists of a method for softening the sound of the string, by a method similar to that which is employed in the upright. It is claimed, and with some justice, that the use of the shifting key-frame is attended with various complications, and that the motion tends to hinder the adjustment of the action to the strings. Moreover the hammers become unevenly worn by being continually caused to strike two, instead of three, strings and this operates to cause loss of directness in the hammer-strokes.

In this new device, the hammer is provided with a swing-rail (25), having a cushion for the hammer-rebound similar to the ordinary cushion that is supported upon the wippen in most actions, and a rod (26) for pivoting it to the metallic action frame (24). The swing-rail also is fitted with a lifter (27), which is operated by the soft pedal through a lifter-rod (28). So far the process is simple. The effect of the motion of the swing-rail is to bring the hammer closer to the string and to soften the tone. But this is only accomplished at the cost of lost motion between the capstan-screw and the key. To avoid the loss of touch thus brought about, the lifter-rod is connected with two compensating levers (29) and (30), which are so adjusted that they form a continuous link between the parts where the lost motion would occur, and thus preserve the continuity of touch and the weight that is lost by the different actuating positions of the hammer. The ingenuity of this device commends it to the student of pianoforte mechanism, and the makers are entitled to great praise for the successful achievement of this important modification.

We may conclude our survey by giving a list of the materials of which grand actions are constructed.

We are now able to turn our attention to the action mechanism of the upright pianoforte. The historical chapter of this work contains a short description of the early history of the upright form. Of course, it will at once be seen that the only real mechanical difficulty in designing a vertical type of pianoforte was to be found in the department of touch-mechanism. The problem was indeed primarily of the action. As such it is interesting to note the efforts of Hawkins, Southwell and Loud, to produce an action that would approach in efficiency that of the grand. It was Robert Wornum, however, who found the solution, though not until nearly the end of the first quarter of the nineteenth century. The principle applied by Wornum possessed such value that it has never been superseded, and remains today the distinguishing feature of the upright action.

We refer, of course, to the tape. As applied to the action of uprights, this device has become universal, and all upright actions in consequence have for years been everywhere similar as to general design. Thus we do not find those radical differences of actuating principle that to this day distinguish American from certain European grand pianofortes. This is partly due to the fact that the great battles had already been fought out when the first successful upright actions were made. At present, the only remaining types of the ancient “sticker” leather-hinged action are to be found in old-fashioned and obsolete European pianofortes of the cheaper grade. None are made now, however, except in the case of a very few worthless commercial English instruments. In the United States the “sticker” action has hardly been heard of.

Thus it seems unnecessary to give detailed drawings of any but the accepted type of upright action. The reader’s attention is immediately called, therefore, to the drawing herewith given.

The method of operation may be seen very clearly. Depression of the key, to which is screwed the key-rocker (1), causes the Abstract (2) and the Wippen (6) to rise. This brings the Jack (7) to bear against the butt (18), which is raised, forcing the hammer (20) and (21) through the shank (19) against the string (32). As the hammer strikes, the rise of the jack has brought its tail against the regulating-button (16), which trips it up so that it falls towards the back-stop (14). The butt, being thus left free, drops back, assisted by the pull of the tape (13), until the back-stop is caught and held by the check (9), thus leaving time for the jack to get back under the hammer butt through the aid of the spring (8). Repetition of the hammer is thus assured, and while the rapidity of stroke cannot be so great as that of the grand, it is sufficiently so to satisfy ordinary users. Sometimes a spring is fastened to the front of the jack in connection with a silk loop and cord which stretches back and engages with the hammer spring (23). The latter is then fastened to the hammer butt instead of to a rail (24). This arrangement gives a far more rapid repetition to the action and causes it to approach the delicacy of the double escapement of the grand.

The pedal mechanism is the same as in the grand. The “piano” effect is obtained by pushing forward the hammer-rail (22), so that the hammers approach nearer to the strings. The “forte” pedal effect is obtained by use of the damper lifting rod (26) which is turned rearwards and forces back the whole line of damper heads so that they leave the strings.

The lost motion that occurs in the action when the “piano” pedal is used has been overcome in a most ingenious manner, and by means similar to those already described in the discussion of the grand action. The trouble to be overcome is the same, namely the lifting up of the abstract from the rocker or capstan screw caused by the forcing of the hammers towards the strings. This causes loss of touch and is very irritating to the performer.

The drawing given herewith also shows the metallic action bracket for supporting the action and the capstan screw that is used as an alternate to the key-rocker.

The lost motion attachment operates through the rod (36) which lifts the hammer rail and at the same time brings into play the compensation lever (38), which takes up the space between the abstract and the key that would otherwise intervene. Thus the touch remains true irrespective of the position of the hammer.

The materials that are required for the manufacture of the upright action do not differ from those, which we have already classified as pertaining to the grand.

The adjustment of the upright action requires that the same precautions be taken in the measurement of the key-bed, of the height of the action between the keys and striking-point of hammers, the distance of hammer from strings when at rest and so on. All these assist the action maker to adjust the alignment of the hammers and the height of the abstracts. The size of the pianoforte sometimes requires the omission of the abstracts, in which case the bottom of the wippen is felted and brought into contact with the capstan or rocker direct. All these details are attended to by the action maker when the proper measurements are given. The marking on a stick of the run of the strings is done just as was suggested for the grand scale.

In conclusion, it may be mentioned that much similarity exists between certain parts of the action of both types of instrument. It has become customary to make the length of the hammer shank and hammer butt, when fitted together, about five inches, counting from the center pin of the butt to the middle of the hammer molding. This measurement will be found correct for both grands and uprights.

The length of the hammer blow is always as nearly as possible 1²⁴/₂₅ inches for the bass end, graduating to 1⁹/₁₀ scant in the treble. These dimensions should be maintained at all costs. The key lengths that we gave before are applicable to all forms of uprights. When provided with these data the action maker will be able to put out a mechanism that will fulfill the individual requirements of each kind of instrument.