Figure 2 - ¾-view of the spine register-regulation plate in situ on the instrument.  Scale:  approx. 1:1.

Double-manual harpsichord by Pascal Taskin, Paris, 1769

Russell Collection, University of Edinburgh

          Figure 3 below shows a plan view of the register-regulation plate removed from the instrument.  Unfortunately 2 of the 3 machine-screw bolts are now missing and the wood screw at the left-hand end of the plate is a replacement 19th-century screw, probably dating from the restoration of the instrument by Tomasini in Paris in 1882.  The presence of an original eighteenth-century screw (on the right in Figure 2 above and Figure 3 below), leads me to believe that the plate and the system of regulation it provides is original to the instrument.

Figure 3 - Plan view of the spine registers regulation plate with two wood screws.  Scale 1:1.

Double-manual harpsichord by Pascal Taskin, Paris, 1769

Russell Collection, University of Edinburgh

          The idea of the register-regulation plate is quite simple.  When the registers are pressed against them and the machine-screw bolts are screwed in and out, the position of the register in the instrument can be regulated very accurately.  The screw pitch of the one remaining machine-screw bolt in the plate was measured and found to be very close to 1mm.  Therefore screwing the machine-screw bolt in by one full revolution (360º) moves the end of the bolt by 1mm.  A half a revolution (180º) moves it by 0.50mm, a quarter of a revolution (90º) by 0.25mm, etc.  Hence a rotation of the machine-screw bolt by 36º moves the end of the bolt by only 0.10mm.  Rotating the screw bolt by less than 36º moves the end of the register by even less than 0.10mm!  Clearly this is a very accurate method of positioning the register, and is completely reliable and reproducible:  every time the register is pushed against the end of the machine-screw bolt it returns to exactly the same position.  Therefore the amount that the quills in the jacks project under the strings is the same each time the register of jacks is moved against the machine-screw bolt, and the regulation and voicing of the instrument is maintained with precision and complete reliability.

          Like virtually all French eighteenth-century double-manual harpsichords that I know, the 1769 Taskin in the Russell Collection has the registers in the following order (looking at the registers from above and from the keyboard end of the instrument).  The arrows indicate the direction of plucking of the quills:

                               ←  -  Rear lower-manual 8'

                                            ←  -  4' sitting on the lower manual

                                                          →  -  Upper-manual 8' permanently engaged

The two rear registers both pluck to the left (bass side) and are controlled by register levers that are attached to the registers and project through the nameboard with a screw pivot placed in the wrestplank.  Using these levers the registers can be moved from side to side so that that the register of jacks can be engaged and disengaged by the player.  For the two rear registers, moving the register lever to the right at the player end moves the register to the left at the register end and engages the jacks for that register.  Doing so therefore pushes the register against the machine-screw bolt in the register-regulation plate where its position is determined by the amount the machine-screw bolt projects through the plate.  Therefore the principle of the use of the register-regulation plate is the same for the two rear registers (or for the three rear registers if the instrument is endowed with a peau-de-buffle row of jacks which always plucks the same string as the main lower-manual 8' row of jacks).

          The principle of operation of the near upper-manual 8' set of jacks which plucks toward the treble side is clearly different from that of the lower-manual jacks all of which pluck to the left (bass side).  The position of the register for the upper-manual row of jacks is fixed and always remains engaged.  It cannot be removed and disengaged, and so it has no register lever associated with it.  Since it plucks to the right, the machine-screw bolt for this register cannot control the engaged position of the register and the quills as happens with the rear registers.  Instead the upper-manual register is equipped with a small, strong spring hidden inside the treble end of the register.  This spring can be seen in Figure 4 below with the upper-manual register removed from the instrument and again in Figure 5 where the register and spring can be seen in the register gap.

Figure 4 - The treble end of the upper-manual register with the spring.

Double-manual harpsichord by Pascal Taskin, Paris, 1769

Russell Collection, University of Edinburgh

This spring acts to push the register and its jacks towards the bass side of the instrument and against the end of the corresponding machine-screw bolt in the register-regulation plate.  Therefore the machine-screw bolt can be used to move the register with its jacks to the left and right in order to control the amount that the quills project underneath and past the upper-manual strings.  The near upper-manual register is therefore always held in the same position and can be positioned with the same precision and reliability as the rear registers.

Figure 5 - The treble ends of the three registers in situ showing the near upper-manual register with its spring. 

Double-manual harpsichord by Pascal Taskin, Paris, 1769

Russell Collection, University of Edinburgh

          The treble ends of all three of the registers in the 1769 Taskin double-manual harpsichord can be seen in Figure 5 above.  The register lever plate had been temporarily removed when this photograph was taken, and the spring can be seen expanded to its full extent.  Here the screw hole in the 4' register for the 4' register lever can be seen in the middle register (the rear 8' register is operated by a register lever on the bass side of the instrument).  When the regulation plate is screwed in position the spring at the treble end of the upper-manual register is compressed leaving about only 1mm of spring exposed and 1mm of space between the end of the upper-manual register and the inside surface of the cheek.

          Each of the three machine-screw bolts on the register-regulation plate can therefore be used to position its corresponding register very accurately and reliably so that the regulation of the instrument is always maintained with the same precision and constancy.

          It is somewhat surprising that this register plate has not been reproduced by makers of copies of this instrument.  The 1769 Taskin double-manual harpsichord is perhaps the most copied French double-manual harpsichord in the world, but I know of no one who has made a register-regulation plate for their copy.  The drawing[1] that I made of the instrument in 1974 shows the register-regulation plate clearly and with 3 machine-screw bolts.  It is clear from a photograph taken by the author in 1979[2] when the instrument was being restored by John Barnes (see Figure 6 below), that there were the original 3 machine-screw bolts present at that time.

Figure 6 - Photograph of part of the spine, the bass ends of the registers and the bass end of the wrestplank with the nuts and plugged tuning-pin holes.

Double-manual harpsichord by Pascal Taskin, Paris, 1769

Russell Collection, University of Edinburgh

          The photograph shown in Figure 6 was taken during the restoration by John Barnes of the instrument in 1979.  Before 1979 the instrument had large modern zither-pins instead of traditional flat-sided tuning pins as made in the historical period.  Barnes removed the zither pins, plugged the holes in the wrestplank with hardwood plugs and then drilled these out to take new reproduction tuning pins of a traditional design and smaller diameter as was used in other instruments by Taskin.  The photograph shows that the tuning-pin holes have been plugged with hardwood plugs, but these have not yet been drilled out for the new reproduction tuning pins.  This fact therefore dates the photograph precisely to the year of the restoration of the instrument which was 1979.  The important feature of this photograph is that it shows the heads of the 3 machine-screw bolts of the register-regulation plate at the spine side opposite the bass ends of the registers.  This means that in 1979 all three of the machine-screw bolts existed and were being used.  As mentioned above there now remains only one of the machine-screw bolts indicating that in the intervening period two of the three bolts have gone missing at the hands of the restoration staff at the Russell Collection.  Hopefully new machine-screw bolts will be made for the register-regulation plate with the same screw pitch as the original remaining bolt so that the original function of the plate can be re-instated.

          As mentioned above I discovered the original register-regulation plate and its original function in the process of the restoration of a stunning Franco-Flemish double-manual harpsichord that I have been restoring in the past years.  Naturally for an instrument of its significance and importance, I wanted to utilise this system for regulating the registers in the restoration of this instrument.  The Franco-Flemish harpsichord has the usual French disposition with a French keyboard coupler, coupler dogs, and four registers including a row of peau de buffle jacks at the rear plucking the same string as the main lower-manual 8' set of strings.

          In this case, therefore, the plate had to have 4 machine-screw bolt regulators including one for the rear peau de buffle register.  The dimensions of the  register-regulation plate and the positions of the holes for the retaining wood screws and those of the regulating screw bolts was adapted to suit the Franco-Flemish instrument.  The result of the new design is shown in the drawing of Figure 7 below.  Here the dimensions of the machine-screw regulation bolts, the thickness, width and beveling of the plate, the upper-manual register spring and the original screw are exactly the same as those of the 1769 Taskin double-manual harpsichord in the Russell Collection.  The machine-screw bolts have a length of 8.2mm of threaded lower section.  The thickness of the register plate itself is 5.2mm so that the machine-screw bolts can project a maximum of 3.0mm towards the bass ends of the registers.  This corresponds to the maximum amount that the ends of the registers can be positioned away from the outside spine surface of the instrument.  In practice it is relatively easy to cut the length of the registers so that the gap between the end of the register and the end of the machine-screw bolt is about 1mm or less.  Therefore this 3.0mm of possible movement is more than sufficient in practice given the small amount of sideways movement that the jacks can have given the small amount of space available between the 3 sets of strings. The position of the holes for the retaining wood screws was measured out to be the same as the existing wood-screw holes beside the register gap-cut-out (perhaps for a similar regulation plate) on the spine side of the Franco-Flemish harpsichord being restored.  Using these dimensions a new 4-machine-screw bolt register plate was made.

Figure 7 - A drawing of the spine register-regulation plate made for the c.1630/1750 Franco-Flemish harpsichord which has 4 registers, unlike the 1769 Pascal Taskin harpsichord in the Russell Collection which only has 3.  Scale 1:1.

          The new register-regulation plate made for the Franco-Flemish harpsichord is shown below in Figure 8Figure 8.  The wood screws used to hold the plate in place are 19^th-century narrow-slot screws.  Otherwise the plate and the regulation bolts are all new.