1863 E. & G. G. Hook Opus 322 Church of the Immaculate Conception Boston, Massachusetts Part 1

 A virtually complete documentation and tonal analysis derived from the data, drawings, and photographs from the restoration of 1976

by Michael McNeil and David Sedlak

Michael McNeil has designed, constructed, and researched pipe organs since 1973. He was also a research engineer in the disk drive industry with 27 patents. He has authored four hardbound books, among them The Sound of Pipe Organs, several e-publications, and many journal articles.

Editor’s note: Part 1 of this article was published in the August 2018 issue of The Diapason, pages 16–19. Part 2 was published in the September 2018 issue, pages 20–25. See the October 2018 issue, pages 26–28 for Part 3.

A graphical analysis of William Johnson’s scaling and voicing

The graphical models used in this section provide a visual means of understanding the scaling and voicing of an organ. More importantly, they serve as a means of comparing other styles of scaling and voicing. From these models we can understand how the tonal structure of an organ can be designed to suit any desired outcome.

The graphical data of Johnson’s 1864 Opus 161 at Piru Community United Methodist Church are presented side-by-side with data from E. & G. G. Hook’s Opus 322, built for the Church of the Immaculate Conception in Boston, Massachusetts, in 1863. Unfortunately, no recordings of Johnson’s Opus 161 are known, but the Samuel Green organ originally built for Litchfield Cathedral and now located at the Church of St. John the Baptist, Armitage, Staffordshire, bears a striking resemblance to the scaling, voicing, and tonal quality of the Johnson. The Green organ can be heard in the CD listed in the section on Recordings at the end of this article.

Normal Scale tables were copied into a spreadsheet and restructured in a manner that would allow an Excel matching function to find the Normal Scale values of the Johnson data. The spreadsheet for the Johnson data calculates mouth width fractions of the pipe circumferences, C values of the pipe toes, toe areas, flueway areas, and ratios of toe areas to flueway areas. The spreadsheet generates graphs for Normal Scale pipe diameters, Normal Scale mouth widths, Normal Scale mouth heights, toe C values, flueway depths, and ratios of toe areas to flueway areas. Spreadsheets of both the Johnson and Hook organs may be obtained at no charge from the author; see References at the end of this article.

Scot L. Huntington carefully documented the restoration of Johnson’s Opus 16 on pages 163–207 in the book, Johnson Organs 1844–1898, co-authored with Barbara Owen, Stephen L. Pinel, and Martin Walsh (see References). The only important data missing from this work are pipe flueway depths, but the flueway data on Opus 161 likely provides a guide for Opus 16 as well. The basis for this assumption is the similarity in scaling and voicing of the two Johnson organs, both of which show a significant reduction in power as the pitch of the stops in the principal chorus rises, and both are very unlike the Hook. The reader can find Huntington’s tabulated data in the new book.

Pipe diameters

The Normal Scale of pipe diameters is a way to visualize relative power, where a flat line from bass to treble will produce relatively constant power. Pipes extending higher in the graph will produce more power. Each half tone on the vertical scale is worth 0.5 dB of power. Interested readers can refer to The Sound of Pipe Organs for a discussion of the underlying theory and principles of all of the graphical models of the Johnson data. 

The scales of the principal chorus of the Hook in Figure 15 are relatively constant, i.e., they are the same for all of the pipes of a given pitch—only the Hook V Mixture is significantly narrower than the foundations. By contrast in Figure 14, the scales of the Johnson upperwork descend as the pitch of their stops rises, i.e., the scales of the 4′ Principal are narrower than the 8′ Open Diapason, and the 2′ Fifteenth is narrower than the 4′ Principal. The narrowest stop in the Hook chorus, the VII Cymbal represented by the blue line, was built by Johnson in 1870.

Note the extremes of Johnson scaling in the wide 8′ Clarabella and the narrow 8′ Keraulophon. These stops share a common narrow bass, which was scaled to match the modest power of the Keraulophon.

Like Samuel Green, Johnson greatly widens his deepest foundation pipes; note the scale of the 16′ Pedal Double Open Diapason at +9 half tones, the single blue data point in the upper left of Figure 14. This stop produces a strong tactile effect even in the dry acoustic of the Piru church, whose walls are too thin to reinforce bass tone.

Mouth widths

The Normal Scale of mouth widths operates just like the pipe diameters, where a flat line from bass to treble will produce relatively constant power. Pipes extending higher in the graph will produce more power. Each half tone on the vertical scale is worth 0.5 dB of power. 

Mouth widths are nearly always a better indicator than pipe diameters of power balances; this is because mouth widths can be designed to vary considerably within the same diameters of pipes. Narrower mouths will produce less power. 

The Johnson principal chorus in Figure 16 remains mostly unchanged, but the 8′ Clarabella mouths are now slightly narrower than the 8′ Open Diapason. 

Note that the bass of Johnson’s 8′ Open Diapason is as wide as the Hook 8′ Open Diapason in Figure 17—this is remarkable when we consider that the Johnson was built for a much smaller acoustic. Again note that the Johnson VII Cymbal in the Hook chorus in Figure 17 is the narrowest stop in that chorus, even though it was designed for the large and vibrant acoustics of the Church of the Immaculate Conception in Boston, the original home of the Hook. We can see how Johnson compensated for the larger acoustic at Immaculate Conception by observing that he scaled this VII Cymbal slightly wider than the 2′ Fifteenth in his Opus 161 in Figure 16.

The wind pressure of the Johnson was probably about the same pressure as the Hook (76 mm) in its original state. The reduction of the scales of Johnson’s upperwork stops shows that he wanted a very refined chorus, and indeed the Johnson chorus is never overbearing. Like Samuel Green, Johnson provides grandeur to his chorus by making his basses extremely wide and powerful; note how the mouth widths of the Great Open Diapason increase rapidly from the tenor to the bass, and also note how the mouth width of the Pedal Double Open Diapason (the single data point in the upper left of Figure 16) extends this trend linearly to 16′ low C.

Mouth heights and toe diameters

Mouth height, or “cutup,” as it is commonly called by voicers, is the primary means of adjusting the timbre of a pipe. Low cutups will create a bright tone with many higher harmonics, while high cutups will produce smoother tone. It is not uncommon to find flute pipes cut as much as 12 half tones higher than principal pipes in many classical pipe organs.

In the Normal Scale of mouth heights, a higher cutup value on the vertical scale will result in smoother tone. Cutups may be adjusted higher for one or both of two reasons: the voicer wants a smoother timbre, or the voicer wants more power at the same timbre. More power means more wind, and this means a larger toe opening (or deeper flueway) to admit more wind and raise the pressure at the mouth. More pressure at the mouth will always produce a brighter tone, so the voicer can make a pipe louder and preserve a certain timbre by opening the toe and raising the cutup until the timbre is restored.

Pipe toe diameters can be normalized to the diameter of the pipe, the width of the mouth, and the depth of the flueway. Higher values of C indicate larger toes with more flow of wind and higher pressures in the pipe foot. 

Now we can understand the graphs. In the Hook graph of mouth heights (Figure 19) and toe C numbers (Figure 21) we see very high values. Hook was after power, and these graphs show how you get it, even on a modest pressure of 76 mm.

In the graphs for Johnson’s mouth heights (Figure 18) we see that the mouth heights are cut lower as the pitches of the stops rise. This would normally make the upperwork brighter, but Johnson also reduces the toe C numbers (Figure 20) as the pitches of his stops rise, and this keeps the timbre constant. The net effect is that Johnson’s upperwork is significantly reduced in power; the 4′ Principal is quieter than the 8′ Open Diapason and the 2′ Fifteenth is quieter than the 4′ Principal. This fits the description of Samuel Green’s work by Stephen Bicknell perfectly.

We can see how Johnson compensated for the larger acoustic at Immaculate Conception by observing that the toe C numbers of his VII Cymbal in Figure 21 are as wide as the Hook voicing and as fully winded as the Hook pipes—these VII Cymbal toe C numbers are much larger than the toe C numbers of the Opus 161 2′ Fifteenth in Figure 20. In contrast, Johnson’s VII Cymbal in the Hook chorus in Figure 19 has lower mouth height than the Hook voicing, and it is indeed brighter than the Hook mixtures.

Note the mouth height of the low C of the Johnson Pedal 16′ Double Open Diapason in the extreme upper left of Figure 18. At a value of +11 half tones, this stop produces copious power on full wind without harmonic stridency, a further extension of the balances sought by Green.

Flueway depths

Like the pipe toe, the flueway depth controls the flow of wind and strongly correlates to the power and the speed of the speech of the pipe. Both organs, the Johnson in Figure 22 and the Hook in Figure 23, exhibit flueways that are deep enough to claim that power is not at all regulated by the flueway. These flueways are characteristic of Romantic and Classical French voicing. Power in these pipes is regulated at the toes. The flueways of the Johnson 4′ Principal and 2′ Fifteenth in Figure 22 are wide even by normal Romantic standards. Classical Germanic voicing typically maintains an open toe and controls power at the flueway. Gottfried Silbermann is the famous exception to the Germanic custom; he learned organbuilding in France.

In his book, The Johnson Organs, John Elsworth noted that the Johnson pipemakers would set the flueway and the Johnson voicers would adjust the toe and mouth height. It is probably safe to assume that a Johnson voicer would adjust the flueway depth if it were needed, but Elsworth’s description of this process is interesting—it is the exact opposite of Germanic practice.

Ratio of toe and flueway areas

Once the scaling is set, the flow of wind and the available range of power are controlled by the voicer at the toe and flueway of a pipe. The ratio of the area of the toe to the area of the flueway is important. If the area of the toe is less than the area of the flueway, which is a ratio less than 1:1, the speech will be slower. “Slowness” in this instance does not refer to the voicer’s term (which reflects how the voicer adjusts the relative position of the languid and upper lip) but rather to the effect of resistances (the toe and flueway areas) and capacitance (the volume of air in the pipe foot). These resistances affect the rise time of the buildup of sound to full power. The ratio is exactly 1:1 when the area of the toe and flueway are equal, and this is the normal lower limit for pipes with prompt speech. For example, the vast majority of the principal chorus pipes in the range of 4′ to 1′ pitch in the Isnard organ at St. Maximin, France, exhibit a value of almost exactly 1:1, with higher pitches approaching a value of 3:1. This gives the Isnard foundation pipes a lovely “bloom” to their speech.

The speed of pipe speech is important. A well-knit chorus of pipes may have slower pipes or faster pipes, but never both. The ear is very sensitive to the speed of pipe speech—it can sense changes in milliseconds.

With this background in mind, we can see that the speech of the Johnson chorus is slower than the Hook chorus. Indeed, the voicing of the lovely Johnson chorus works well with the relatively low resonant frequency of its wind system to impart what the author noted in 1976 as a “light ‘give’ on full organ, a relatively fast buildup to full flow.” The Johnson 8′ Open Diapason is a bit faster with toe-to-flueway ratios above 1:1, but the upperwork is slower with ratios well below 1:1.

The Hook speech is very fast. The Hook chorus develops a lovely surge on full organ; this is not due to the voicing but rather the lower resonant frequency of its wind system. 

Reflections

The William A. Johnson tonal design is eminently suited to the dry acoustics of most American churches. Johnson’s VII Cymbal in the Hook organ provides us with a window into Johnson’s thinking on the scaling and voicing for a much larger and reverberant acoustical setting.

Whatever the reader’s opinion of the aesthetic value of the Johnson chorus, its documentation has proved to be quite valuable. The Piru church disregarded the advice of the author to engage the services of Manuel Rosales for the maintenance of the organ when the author departed California in 1993. Some time after the Northridge, California, 6.7-magnitude earthquake in 1994, the church contracted a different Los Angeles firm to rebuild Johnson Opus 161; Piru was approximately 30 miles distant from the epicenter. The work had not been completed by the summer of 2017 when Manuel Rosales was contracted to perform an inventory of the organ and assess its condition; the organ had been dismantled leaving only the windchests stripped of their topboards and sliders in the frame. The pipes, topboards, sliders, and most of the mechanical parts, stored in trays in the parish hall, were suffering damage from constant handling. In an effort to keep the pipework intact, Kevin R. Cartwright has been engaged recently by the church to reinstall the pipework in the organ; there is no funding to make it playable. The documentation in this essay has provided a useful reference during the reassembly. Mr. Cartwright has twenty-one years’ experience in organbuilding, the last three of them working as a contractor to Manuel Rosales.³²

This essay on Johnson Opus 161 was a considerable effort. The goal was to provide a template for the documentation of important and historically valuable organs. Such documentation is often the only insurance we have against well-intentioned modernizations. It is the author’s hope that this essay will inspire more thorough documentation of the world’s priceless gems.

Although drawings crafted on computers are visually pleasing, most organbuilders do not have the time or funding to make such graphics. If we want to see good documentation in print, we must also be willing to accept the lack of polish in hand drawings. The editorial staff of The Diapason has shown courage in their willingness to publish such drawings.

There is some evidence that the need for more thorough documentation is gaining traction in the organbuilding community. Pierre Chéron and Yves Cabourdin published complete scaling and voicing data on the Isnard organ at St. Maximin in 1991; Frank-Harald Greß published similar data on the organs of Gottfried Silbermann in 1989, although neither work addressed the documentation and analysis of wind systems. Of great importance is the work of William Drake, Ltd., in the United Kingdom. Their recent restoration of a 1755 Snetzler organ included documentation that has the depth of the data found in this essay on Johnson. This gold-standard level of documentation can be found on their website: www.williamdrake.co.uk/portfolio-items/clare-college-cambridge/. If more organbuilders follow the lead of Drake we will begin to really understand how the sounds that inspire us are achieved.

Notes and Credits

All photos, drawings, tables, and illustrations are courtesy of the author’s collection, if not otherwise noted. Most of the color photos were unfortunately taken by the author with an inferior camera in low resolution. David Sedlak used a high quality camera, lenses, and film to produce the high-resolution color photos of the church and its architectural details; these are all attributed to Sedlak.

32. see www.cartwrightpipeorgan.com/
recent-projects/.

Recordings

Preston, Simon. 5 Organ Concertos, The English Concert, Simon Preston, 1984, Archiv D 150066. The organ concertos of George Frederick Handel are played on the Samuel Green organ, 1789–1791, Church of St. John the Baptist, Armitage, Staffordshire, England. Although this organ was built for Litchfield Cathedral and was later moved to its present location in a smaller acoustic (lending more force to the impact of the organ), its sound bears a striking resemblance to that of William A. Johnson’s Opus 161. The Green organ on this CD is tuned in meantone, resulting in a gravity that enhances the already rich timbre of Green’s scaling and voicing. The Johnson organ would sound equally at home in this temperament. There are, unfortunately, no known recordings of Johnson Opus 161.

Murray, Thomas. The E. & G. G. Hook Organ, Immaculate Conception Church, Boston, Sheffield Town Hall Records, Album S-11 (ACM149STA-B), Santa Barbara, CA.

Murray, Thomas. An American Masterpiece, CD, AFKA SK-507. (E. & G. G. Hook Opus 322)

References

Bicknell, Stephen. The History of the English Organ, Cambridge University Press, Cambridge, 1996, 407 pp.

Cabourdin, Yves, and Pierre Chéron. L’Orgue de Jean-Esprit et Joseph Isnard dans la Basilique de la Madeleine à Saint-Maximin, ARCAM, Nice, 1991, 208 pp.

Elsworth, John Van Varick. The Johnson Organs: The Story of One of Our Famous American Organ Builders, The Boston Organ Club, 1984, Harrisville, 160 pp.

Greß, Frank-Harald. Die Klanggestalt der Orgeln Gottfried Silbermanns, VEB Deutscher Verlag für Musik, Leipzig, 1989, 176 pp.

Huntington, Scot L., Barbara Owen, Stephen L. Pinel, Martin R. Walsh. Johnson Organs 1844–1898, The Princeton Academy of the Arts, Culture, and Society, 2015, Cranbury, 239 pp. 

McNeil, Michael. The Sound of Pipe Organs, CC&A, Mead, 2012, 191 pp., Amazon.com.

McNeil, Michael. Johnson_161_170807, an Excel file containing all of the raw data and the models used to analyze the Johnson Opus 161, 2017, available at no charge by emailing the author at [email protected].

McNeil, Michael. Hook_322 Scales Voicing_170228, an Excel file containing all of the raw data and the models used to analyze the Hook Opus 322, 2017, available at no charge by emailing the author at [email protected].

Nolte, John M. Scaling Pipes in Wood, ISO Journal, No. 36, December 2010, pp. 8–19.

Owen, Barbara. The Organ in New England, The Sunbury Press, Raleigh, 1979, 629 pp. 

Michael McNeil has designed, constructed, and researched pipe organs since 1973. He was also a research engineer in the disk drive industry with 27 patents. He has authored four hardbound books, among them The Sound of Pipe Organs, several e-publications, and many journal articles.

Michael McNeil was awarded twenty-seven patents over a period of forty years as a research and development engineer, and in a parallel career he designed and constructed four mechanical action pipe organs. He has written five books and two technical papers, three of which are e-publications.a

We are all familiar with organs that have an overbearing harshness. There is now a backlash under way towards an organ sound better suited to the typically dry acoustics of American churches and concert halls. The organ reform movement of the twentieth century was itself a reaction to the sounds of great builders such as Ernest M. Skinner, who knew very well how to deal with dry acoustics, but Skinner’s sound was not at all convincing when playing the compositions of Baroque masters. The twentieth century reform movement looked to the home of those Baroque masters, especially that of J. S. Bach, for guidance.

While the pipe scaling and voicing of many organs in northern Germany works extremely well in the interpretation of Bach’s music, it is not suited to the smaller, dry acoustics of American churches. An examination of two famous and successful organs will suggest ways in which we might deal with this, while maintaining a sound that does justice to the older masters.

The monumental work Gli Organi della Basilica di San Petronio in Bologna by Oscar Mischiati and Luigi Ferdinando Tagliavini (2013) contains data detailed enough to allow an analysis of the scaling and voicing of the remarkable Gothic organ constructed by Lorenzo da Prato in 1475. This book provides evidence of the originality of much of what we see and hear in the present organ. The publication in 1991 of the book by Yves Cabourdin and Pierre Cheron, L’Orgue de Jean-Esprit et Joseph Isnard dans la Basilique de la Madeleine à Saint-Maximin, gives us the most complete documentation of any organ ever attempted and the opportunity to analyze a famous French Classical design constructed in 1774 by Jean-Esprit and Joseph Isnard. These organs are very different in their design, but are so successful that they have survived centuries virtually unmolested. They can teach us something of value.

The principals are the main voice of an organ. The smoothness or stridency of a principal chorus is mostly the result of two factors: 1) the size of the acoustic, i.e., the distance of the organ from your ears, and 2) the scaling and voicing of the organ’s pipes. 

The acoustic 

Large rooms make a sound less strident and less powerful. There are two things going on here. First, sound power falls off with distance, so much so that if you listen to an organ from twice the distance, the organ will sound four times less powerful. But for higher frequencies, the sounds we perceive as “harsh” or “strident,” the sound falls off much faster with distance. This is because air has mass (you feel the force of this mass on your face in the wind, and it powers the sails of your boat), and only the lower frequencies (the alto down to the deep bass) travel easily through this mass of air. The higher frequencies (the soprano up through those high mixture pitches) are absorbed by air and travel much shorter distances. This type of loss starts to noticeably affect pipes at 1⁄2′ pitch and it gets vastly more pronounced as the pitch rises. To show just how vast these losses are at very high pitches, a 1⁄16′ pipe loses 98% of its original power when heard 500 feet away; this is the “high C” of the 2′ stop on your 61-note keyboard. It is virtually inaudible at this distance, even if it screams at the console.

Pipe scaling

So this is all very interesting, but what’s the point? The point is that many northern German organbuilders relied on large distances to refine their choruses. They used what is termed a constant pipe scale, which simply means that all of the pipes at a single pitch have the same “scale,” or width, regardless if they are used in the foundations, the octaves, or the mixtures. For example, in such a scale the 2′ pipe in the Principal at middle C will have the same width and power as the pipes at this pitch in the Octave, Superoctave, and mixtures. A pleasing sound will contain harmonics that fall off in power relative to the fundamental, but a constant scale has no fall off in power for higher-pitched stops like the mixtures. So how does a constant scale work? It works by the absorption of the high frequencies in the air over great distances. Transplant this constant scale into a small, dry acoustic, and you have a recipe for overbearing harshness. 

Some of my colleagues would now point out that we can reduce the power of mixtures by reducing their toes (which admits less wind) or reducing the flues at the mouth (which admits less wind). I would counter that there is limited leeway in the reduction of toes and flues, and a steep price is paid for such measures—the speech of the pipe is slowed—and a chorus that is well-knit and cohesive will have pipes that speak at the same speed. A good chorus can have pipes that are all relatively slow or all relatively fast, but a mix of the two will not produce a cohesive chorus, and our ears are extremely sensitive to this. 

So if we don’t want to use large reductions in toes and flues, how can we obtain a well-balanced chorus? The wisdom of da Prato in 1475 and the Isnards in 1774 shows us two alternate paths. While those builders didn’t have the technical tools we use today to analyze acoustics and organ pipes, they were obviously superb experimental engineers who used their ears to great advantage in the practice of their art.

The alternate paths

The comparison of organ sounds requires a specific set of data and its analysis requires a model. In 2012 the author published a detailed description of such a model in The Sound of Pipe Organs. The model is based on the physics of sound, its perception by humans, scaling and voicing parameters, and the effects of distance and atmospheric losses. A fully worked example of the model was applied in this book to the Isnard organ. In this article we will look at just the scaling of pipes and the balances of power.

Take a look at the figures. These are scaling charts used by many organbuilders. They are based on the “normal scale” devised by Töpfer in the nineteeth century. Töpfer assigned an arbitrary width to each pipe in the normal scale, and all other things being equal, pipes made to this scale produce a relatively constant power from the deep bass to the treble. The normal scale is represented by the line that runs from left to right at a value of zero (0) in the middle of the graphs. Pipes wider than the normal scale have more power, and narrower pipes have less power. 

At the bottom of the graphs we see the pitch of the pipes increasing, from 16′ bass pipes to 1⁄8′ treble pipes. Look at the left-hand columns in the graphs; the pipes widen in units of “half tones” of scale, with very wide scales at the top and very narrow scales at the bottom. The different colored lines represent the relative widths of the pipes in the different stops in the chorus; these stops are identified in the tables at the right of each graph. 

For example, look at the middle C pipe at 2′ length in Isnard’s 8′ Montre in Figure 1. The pink arrow points to it in the pink line, and at 2′ pitch it is -2 half tones in width. The yellow arrow in Figure 1 points to the Isnard 4′ Prestant, which has a 2′ pipe at tenor C, scaled three half tones narrow. 

Many trends leap off these graphs. The da Prato principal chorus in Figure 2 is very tightly clustered around a constant scale at about -7 half tones, rising to about -5 half tones at the high a″ of the compass. The two flutes of the da Prato organ in Figure 2 are remarkably wide-scaled relative to the principal chorus. The only member of the principal chorus in Figure 2 that is widely scaled is the 8′ Principale, and it has wider pipes to adjust for its highly unusual position in the back of the organ—this organ has two façades, one in the front with the 24′ (16′) Principale, and another in the back with the 12′ (8′) Principale (the compass of this organ extends to low F). The pipes of the 8′ Principale need to be wider and louder to compensate for their unusual placement—they speak not towards the front, but face backwards at the rear of the organ. In stark contrast, the Isnard chorus in Figure 1 is very spread out, with wide foundations and increasingly narrow upperwork.

Now having shown you the more traditional pipe diameters of the principal choruses of the da Prato and Isnard organs, I have to confess that there is a more accurate way to compare power. Pipe diameters, or widths, are the most common descriptions of organ pipes, but the power of a pipe is much better related to the width of its mouth, not the width of its resonator. Those wishing to understand this in more depth can find detailed explanations in the author’s book The Sound of Pipe Organs. The power relationships of the da Prato and Isnard choruses are much better described by the mouth widths of their pipes, seen in Figures 3 and 4. 

In Figures 3 and 4 we see something similar and very striking. The pipes noted in group 1 in both figures start to widen at 1⁄2′ pitch and increase dramatically by nine to eleven half tones at 1⁄8′ pitch. Both builders were compensating for distance losses due to the atmosphere. A pipe at 1⁄8′ pitch will drop to about 28% of its power when heard from 500 feet away. As the author has shown, such losses can be completely compensated for at this distance by widening the pipes at 1⁄8′ pitch by twelve half tones; both da Prato and Isnard were scaling their pipes to be heard correctly in the large acoustical spaces in which they built these organs.

In Figures 3 and 4 we also see something very dissimilar and very striking. The pipes in the principal chorus by da Prato in Figure 3 in group 2 follow a constant scale where the upperwork is scaled roughly the same as the foundations, while Isnard in Figure 4 in group 2 chose to drastically reduce the scale of the stops in the principal chorus as they ascended in pitch; the mixtures are eight half tones narrower than the 16′ foundations.

The exciting but not overbearing sonority of the Isnard chorus is clearly explained in the graph at the left by the reductions in the scales of the higher-pitched stops. But the well-balanced chorus by da Prato is explained very differently. Italian organs used a device rarely seen outside of Italy: the rack board that holds the pipes in place on the Italian wind chest is placed above the mouths of the pipes, and the pipes that are placed far to the back on the windchest are greatly muted by the effect of this rack board. The da Prato upperwork is placed towards the back, and the sound of those pipes must find its way under a rack board and around the feet of hundreds of pipes before they can project into the room. This is the secret of the Italian organ, its constant scales, and its refined chorus.

With the examples of the da Prato and the Isnard organs we see a purposeful effort to reduce the power of the high-pitched pipes relative to their foundations. This is the evidence for the assertion that a constant scale will sound overbearing in most American churches and concert halls. While the Italians used a rack board above the pipe mouths to mitigate their stridency, and Isnard used narrower scaling of upperwork, there are German traditions where neither compensation is made, and these traditions depend on vast acoustical spaces to succeed. The use of constant scales in smaller, dry American acoustics does much to feed the current backlash in organbuilding.

There are fine examples of classical organbuilding in the United States that are well scaled to their rooms. One such example is the universally liked Fisk organ at Old West Church in Boston, a building neither vast in scale nor highly reverberant. Its scaling is reportedly based on the design of J. A. Silbermann at Marmoutier, France, whose pipe scales look remarkably like those of the Isnard at St. Maximin. Designing for American acoustics is always a difficult challenge, but we can also learn from good examples. ν

References

Cabourdin, Yves, and Pierre Cheron. L’Orgue de Jean-Esprit et Joseph Isnard dans la Basilique de la Madeleine à Saint-Maximin. Nice: ARCAM, 1991, 208 pp, ISBN 2-906700-12-6.

McNeil, Michael. A Comparative Analysis of the Scaling and Voicing of Gothic and Baroque Organs from Bologna and St. Maximin. CC&A, Mead, 2016, 8pp. ISBN 978-0-9720386-3-8, e-book on Lulu.com. The current article is a shortened version of this publication. The e-book includes a voicing analysis and tables of measurement data.

———. The Sound of an Italian Organ. CC&A, Mead, 2014, 78 pp. ISBN 978-0-9720386-6-9, e-book on Lulu.com.

———. The Sound of Pipe Organs, CC&A, Mead, 2012, 191 pp. ISBN 978-0-9720386-5-2.

Mischiati, Oscar, and Luigi Ferdinando Tagliavini. Gli Organi della Basilica di San Petronio in Bologna. Bologna: Pàtron Editore, 2013, 577 pp.

Michael McNeil has designed, constructed, and researched pipe organs since 1973. He was also a research engineer in the disk drive industry with 27 patents. He has authored four hardbound books, among them The Sound of Pipe Organs, several e-publications, and many journal articles.

Editor’s note: Part I of this article was published in the July issue of The Diapason, pages 17–19.

Mouth heights 

Mouth height, or “cutup,” as it is more commonly called by voicers, is the primary means of adjusting the timbre of a pipe. Low cutups will create a brighter tone with many harmonics, while high cutups will produce smoother tone with fewer harmonic overtones. For interested readers, see The Sound of Pipe Organs, pp. 68–80. In older organs, it is not uncommon to find flute pipe mouths cut twelve half tones higher than principal chorus pipes.

In the Normal Scale of mouth heights, a higher cutup value on the vertical scale will result in smoother tone. Cutups may be adjusted higher for one or both of two reasons: 1) the voicer wants a smoother timbre, or 2) the voicer wants more power at the same timbre. More power means more wind, and this means a larger toe and/or flueway to admit more wind at the mouth. More wind at the mouth will always produce a brighter tone, so the voicer can make a pipe louder and preserve its original timbre by opening the toe and raising the cutup until the timbre is restored.

Now we can understand the graphs. In Figure 6 we see that the Hook principal chorus has high cutups and that they do not significantly vary from bass to treble. Hook pushes pipes to higher power with much more open toes (Figure 8), and the voicer raises the cutups to avoid a strident timbre at the increased power. The timbres are relatively constant from bass to treble. Note the lower mouth heights of the William A. Johnson Cymbal VII, which makes its timbre brighter than the Hook voicing (also note that the Cymbal’s toes are winded as robustly as the Hook pipes in Figure 8).

In contrast, the Isnard chorus in Figure 7 shows much lower cutups in the bass and mid-range, and much higher cutups in the highest treble. We will see in the data for toe diameters in Figure 9 that Isnard is restraining his pipes for less power and voicing for an ascending treble.

Pipe toe “C” values

Pipe toe diameters can be normalized to the diameter of the pipe, the width of the mouth, and a normalized depth of the flueway. For interested readers, the derivation of this normalization is explained in detail in The Sound of Pipe Organs, pp. 43–47. Higher “C” values mean the toe is larger and flows more wind relative to its mouth width and flueway depth. This is a primary voicing tool for regulating power.

The contrast in the toe diameters of these two organs is striking in many ways. The Hook toes in Figure 8 are much wider overall than the Isnard toes in Figure 9, demonstrating the primary source of the power of the Hook. This power would normally encourage chiff in the pipe speech, but this is suppressed in the Hook by the use of very deep and regular nicking of the languids of the pipes. Ninety percent of the Isnard pipes are free of nicks, and when nicks are found, there are typically only two or three very fine, shallow nicks on a languid. Contrast this with the treatment of the 16′ Open Diapason of the Hook: 22 fine nicks at C1, 20 nicks at c13, 29 nicks at c25, 24 medium nicks at c37, and 19 medium nicks at c49, all of the pipes having their nicks cut very deeply into the languid. There is no discernible “chiff” to the speech, but this is desirable for the interpretation of Romantic music. Interested readers can refer to The Sound of Pipe Organs, pp. 94–96, for a graphic illustration of the effects of such nicking on speech transients.

Figure 8 demonstrates another key element of the Romantic tradition—large toes supplying more wind and power to the bass and mid-range. In contrast, the toe constants of the Isnard are much smaller, more constant across the compass, more constant for all stops of the chorus, and exhibit a subtle rise to support an ascending treble. 

Flueway depths 

Like the pipe toe, the flueway depth also controls the flow of wind and strongly correlates to the power of the pipe. Interested readers can refer to The Sound of Pipe Organs, pp. 50–63 and 77–82. 

In Figure 10 we see another essential characteristic of Romantic voicing—a very deep flueway. Much of the Romantic voicing tradition grew out of the French Classical voicing style, which maintained deep flueways and controlled the power of a pipe by restricting its toe, much as we see in Figure 9. The restorer of the Isnard organ, Yves Cabourdin, noted that the flueways of the Isnard organ seen in Figure 11 are “closed up” relative to normal French Classic practice, yet the flueways of the Isnard are very deep relative to the common North German practice of regulating power by closing down the flueways while maintaining open toes. For interested readers, some examples of historical practice in flueway depths may be found in The Sound of Pipe Organs, pp. 50–51.

The extremely deep flueways of the Hook organ are consistent with Romantic voicing in general, along with more generous toe diameters and the nicking required to suppress chiffing in the pipe speech at the greatly increased power levels of this style. 

The flueways of the Hook organ appeared in general to be very well preserved and were very consistent. The anomalous lower value of the flueway in Figure 10 for the Hook 16′ Open Diapason at c25 (4′ pitch) may have been the result of handling damage to that pipe or modifications when the pitch was changed. The robust flueway depth of the Hook 16′ low C pipe is literally off the chart at 4.8 mm.

Ratios of toe and flueway areas

The flow of wind and power balances are controlled by the voicer at the toe and flueway of a pipe. The ratio of the area of the toe to the area of the flueway is important. If the area of the toe is less than the area of the flueway, which is a ratio less than 1:1, it will cause a significant drop in the pressure at the mouth, and what is more important, the speech will be noticeably slower. When the area of the toe and flueway are equal, the ratio is exactly 1:1, and this is the lower limit for pipes with faster speech. Interested readers can refer to The Sound of Pipe Organs, pp. 56–63 and 114–116 for a discussion of this very important musical characteristic and its effect on the cohesion of a chorus. (A well-knit chorus may contain slower pipes or faster pipes, but never both.)

The Hook ratios in Figure 12 never descend below a ratio of 1:1 and typically ascend to extremely high values in the treble. It is this technique with which Hook obtains an ascending treble.

The Isnard ratios in Figure 13 reside at a value of 1:1 for the bass and mid-range and ascend to much higher values at the highest pitches. Like the Hooks, the Isnards achieved an ascending treble with this technique, but unlike the Hooks, the Isnards crafted the bass and mid-range ratios to values of almost exactly 1:1. The Isnard pipe speech has a lovely “bloom,” which is a direct result of these very carefully crafted ratios; the term “bloom” refers to a slower buildup of power in the initial speech of a pipe. The Hook organ also exhibits a distinct bloom, but this bloom has no speech transients, and it derives from the low resonant frequency of the wind system when it is working hard to supply wind.

The wind system 

The design of the wind system plays a large role in the dynamics of the wind and the musicality of the organ. Dry acoustics favor faster wind systems, which support faster tempos; live acoustics fill dramatic pauses with a halo of reverberation and encourage slower tempos. Wind systems can be designed to enhance the grand cadences of historic literature written for live acoustics, and such wind systems will have a slower response. For interested readers, this response can be described as the resonant frequency of the wind system, and it is fully described in The Sound of Pipe Organs, pp. 99–113, using the Isnard organ as a worked example.

Documentation of the wind system is probably the most overlooked feature in descriptions of pipe organs. The Hook’s wind system was measured in some detail, but not completely due to the constraints of time.

The wind of the Hook organ has no perceptible shake. The tutti does not noticeably sag in pitch. The speech onset of the full Hook plenum is characterized by a dramatic surge, the result of weighted bellows and large system capacitances. The current wind system shows some modification of the 1863 design, largely as a result of the 1902 addition of the Solo Division. 

The static wind pressure of the Great was measured to be 75 mm water column at c′(25) of the 4′ Clarion, the last stop on the back of the chest. The static wind pressure at a′′′(58) of the 16′ Open Diapason was measured to be 76 mm; drawing all of the stops reduced that pressure to 67 mm.

All divisions in the organ are fed with ducts that have cross sections many times what is necessary to wind the tutti. These ducts are also very long, with the result that they are calculated to have Helmholtz resonances in the very low range of about 4 Hz; this frequency is not audible when the organ is played, suggesting that the damping of the wind system is considerable (some concussion bellows are present). The main ducts have about 0.56 m³ of volume.

The two bellows that together feed the Great and Choir (and originally also the Swell), are massive with 8.4 m³ of volume, having two inward folds and one outward fold. The resonant frequency of the two bellows, two pallet boxes of the Great division, and wind ducts as a function of their mass and volume is calculated to be 1.23 Hz. Such a low resonant frequency is the primary source of the grand surge in the tutti of this instrument. It is a musical wind with grand drama, exhibiting none of the nervousness of organs with sprung bellows. Both the mass and volume of this wind system compare favorably with the Isnard organ. And although the Hook organ features double-rise bellows and the Isnard features wedge bellows, they have very similar and low resonant frequencies at 1.23 Hz and 1.20 Hz, respectively. Figure 14 is a table showing the measurements of the wind system and its calculated resonance. 

Another important characteristic of a wind system is its wind flow and damping. The total demand on a wind system is equal to the areas of all of the toes of all of the pipes that can be played at the same time on full organ. We then look to see if the key channels can flow sufficient wind to those toes, if the pallets can flow sufficient wind to the key channels, and if the wind ducts can flow sufficient wind to all of the pallets. This analysis was performed on the Isnard organ (see The Sound of Pipe Organs, pp. 120–127), with the interesting result that the Isnard wind trunk just barely flows adequate wind for the coupled principal choruses of the Grand Orgue and Positif, but it is wholly inadequate for any form of tutti. This sort of restriction is not uncommon in older organs, and it performs the function of adding significant resistance to the wind flow, which in turn dampens Helmholtz resonances in the cavities of the wind system, e.g., wind shake from the wind trunks and pallet boxes. We do not have enough data for all of the stops of the Hook to perform this analysis, but the very large cross-section of the wind trunk suggests that it has much more winding than the Isnard, and that would be consistent with a Romantic organ and the requirement that it support a full tutti. The table in Figure 15 shows the data for wind flow in the windchests of the Great division.

The Great division 

There are two windchests for the Great division, split diatonically C and C# with the bass notes at the outer ends and a walkboard in the middle. Figure 16 shows the pipes on the C side windchest from the 8′ Open Diapason Forte at the left (front of the chest) to the treble end of the III Mixture at the right. The order of stops is:

8′ Open Diapason Forte

8′ Clarabella

16′ Open Diapason

8′ Viola da Gamba

8′ Open Diapason Mezzo

4′ Octave

4′ Flute Harmonique

3′ Twelfth

2′ Fifteenth

III Mixture

V Mixture

VII Cymbal (Johnson, 1870)

16′ Trumpet

8′ Trumpet

4′ Clarion

Figure 17 shows the treble end of the mixtures on the C side. The toeboard on the left contains both the III Mixture and V Mixture. From left to right, we see the III Mixture, V Mixture, and on the right toe board, the later addition of the VII Cymbal (red arrow).

Most of the treble pipes are cone tuned and exhibit almost no damage. This is a tribute to the tuning skills of the Lahaise family. Few organs of this age have survived with such intact mixture pipes. The pre-restoration photos of the Isnard organ at St. Maximin show the more typical fate of such pipes.

All of the tin-lead pipes in this organ are constructed of spotted metal, with the notable exception of the Cymbal (added by Johnson in 1870), which is planed metal. This accounts for the obvious difference in the construction of the rackboard for this stop. The VII Cymbal (red arrows) includes a third-sounding rank, and in the style of Johnson it is silvery (lower cutups) and restrained in power (very narrow pipe diameters and mouth scales). Although no records exist, there must have been a fascinating story behind the inclusion of a competitor’s mixture in this organ.

Figure 18 shows the back of the C side Great chest. The order of reed stops, from left to right, is: 16′ Trumpet, 8′ Trumpet, and 4′ Clarion. Note that the 4′ Clarion is cut dead length in all pipes except the newer, slotted low C pipe added at the time of repitching the organ. Trebles of the 16′ and 8′ ranks are also cut dead length without slots. The intent here is obvious: don’t tune these reeds on the scrolls, tune them on the wire.

General observations

16′ Open Diapason

All of the pipes of the 16′ Open Diapason from the mid-range downward into the deep bass exhibit very bright harmonic content. The reason for this becomes apparent with a close examination of the middle D pipe. When the organ was repitched from A=450 to A=435 Hz, a new low C pipe was made for many of the stops and the original pipes were moved up one half step. The tuning distance between 435 Hz and 450 Hz is less than a half step, with the result that the pipes were now much too flat. The scrolls were then rolled down to bring the pipes into tune at 435 Hz. We can see from Figure 19 that to achieve correct tuning on the middle D pipe, the tin-lead scroll was completely removed and the zinc resonator was crudely cut and broken to make the slot deeper.

This was apparently not sufficient to bring this pipe into tune. Figure 20 shows that the toe of this pipe was crudely opened and flared outward without the benefit of a normal toe reamer or toe chamfering tool. This is very informative because it explains the much brighter timbre of this pipe relative to its treble or other foundations. The opening of the toe increased the pitch and brought the pipe into tune, but at the expense of more power and a much brighter timbre relative to the original voicing. Even with this increased power it would have been possible to have preserved the original timbre by slightly raising the cutup. Inspection of the upper lips indicates that this was not done; the upper lips of all pipes are slightly skived to about one half of the metal thickness, and this was still intact on all pipes. Note that the crudely damaged toe shows bright metal; there was no bright metal on the upper lips, indicating original cutups but modified toes. This voicing damage is typical throughout the bass of this stop. 

Figure 21 shows the back of the low D façade pipe. Note that the tin-lead scroll is completely missing, the zinc is rolled back at the bottom of the slot, and the tin-lead adjacent to the top of the slot is bent outwards on both sides. The author verified that the wind to the toe was likely altered as well; the wooden slides in the toeboard that regulate wind flow were completely open. The façade pipes were all speaking on maximum wind. Figure 22 illustrates the condition of the scrolls in the back of the façade for 16′ c, 16′ G#, 8′ C, 8′ D, 8′ E, and 8′ F#, going from left to right in the figure.

8′ Open Diapason Forte

The cutups appear original, the toes were crudely opened, and this stop indeed sounds too loud and too bright relative to any other 8′ stop. In fact, this stop obliterates the sense of chorus when using it in the traditional French fonds. One would normally expect the 8′ Forte to be slightly more powerful, but less bright, than the 8′ Open Diapason Mezzo; they would then combine as a fine chorus. In fact, this stop is much more powerful than the Open Diapason Mezzo and also brighter. This rank shows the same tuning modification seen in Figure 19, and the toes of this rank were opened in the same crude manner seen in Figure 20. 

While there is some evidence of selective toe adjustment in other stops, no other ranks show such crude treatment and excessive opening of the toes. They have normal chamfers and round bores. Lending further evidence to the hypothesis that this was damage inflicted at the time of repitching the organ, it was seen that the same crude method of opening the toes was applied to all of the new low C pipes in all of the ranks.

We are fortunate in at least one respect. The workmanship during the repitching was very crude, and this allows us to better understand the order of events and the anomalous tonal balances.

III Mixture

The mixture pipes were all moved up one half step when the organ was repitched, widening the scales by a half step and moving the breaks up by the same amount. The new pipes added at low C were crudely matched in diameters, mouth widths, and toes. The width scales of the fifths are about two half tones narrower than the 4′ Octave, similar to the scaling of the Twelfth. The octaves are as wide as the foundations. The current breaks are:

C1 2′ 11⁄3′ 1′

c#26 4′ 22⁄3′ 2′

V Mixture

Although not measured, the flueways were visually consistent with other Hook stops. This mixture is scaled about 3 to 5 half tones narrower than the foundations. The current breaks are:

C1 2′ 11⁄3′ 1′ 2⁄3′ 1⁄2′

c#14 22⁄3′ 2′ 11⁄3′ 1′ 2⁄3′

c#26 4′ 22⁄3′ 2′ 11⁄3′ 1′

c#38 8′ 4′ 22⁄3′ 2′ 11⁄3′

c#50 8′ 51⁄3′ 4′ 22⁄3′ 2′

VII Cymbal (Johnson, 1870)⁴

Although not measured, the flueways were visually consistent with the other Hook pipework. This mixture, designed and built by William A. Johnson and installed in 1870, is 6 to 7 half tones narrower than the foundations. It has similar robust winding in its toes and flueways to the Hook work, but it is cut up relatively lower than the Hook mixtures, giving the Johnson mixture a more silvery timbre. It is a magnificent sound and provides a scintillating crown to the principal chorus of the Hook. Unlike the spotted metal of the Hook pipework, these Johnson pipes are all made of planed metal, probably containing Johnson’s typical alloy of 33% tin.⁵ This stop includes a third-sounding rank; its current breaks are:

C1 13⁄5′ 11⁄3′ 1′ 2⁄3′ 1⁄2′ 1⁄3′ 1⁄4′

c#14 2′ 13⁄5′ 11⁄3′ 1′ 2⁄3′ 1⁄2′ 1⁄3′

g20 22⁄3′ 2′ 13⁄5′ 11⁄3′ 1′ 2⁄3′ 1⁄2′

c#26 4′ 22⁄3′ 2′ 13⁄5′ 11⁄3′ 1′ 2⁄3′

g32 51⁄3′ 4′ 22⁄3′ 2′ 13⁄5′ 11⁄3′ 1′

d#40 8′ 51⁄3′ 4′ 22⁄3′ 2′ 13⁄5′ 11⁄3′

c#50 16′ 8′ 51⁄3′ 4′ 31⁄5′ 22⁄3′ 2′ ν

Notes and Credits

All photographs, tables, graphs, and data are by the author except as noted.

4. Huntington, Scot L., Barbara Owen, Stephen L. Pinel, Martin R. Walsh, Johnson Organs 1844–1898, OHS Press, Richmond, Virginia, pp. 17–18.

5. Elsworth, John Van Varick. The Johnson Organs, The Boston Organ Club Chapter of the Organ Historical Society, Harrisville, New Hampshire, 1984, p. 45.

To be continued.

Michael McNeil has designed, constructed, and researched pipe organs since 1973. He was also a research engineer in the disk drive industry with 27 patents. He has authored four hardbound books, among them The Sound of Pipe Organs, several e-publications, and many journal articles.

Editor’s note: Part 1 of this article was published in the July issue of The Diapason, pages 17–19. Part 2 was published in the August issue, pages 18–21.

Re-pitching of the Pedal 

In Figure 23 we see the C side of the Pedal 16′ Trombone in the front row, and the Pedal 16′ Open Diapason in the back row. Both stops have their pipes in the original position. Note the crude addition of boards to the top of the Trombone pipes as the means of lowering the pitch from A450 to A435 Hz. Relative to its original voicing, this stop is choked off in power and brilliance. Also note the more professional lengthening of the resonators of the Pedal 16′ Open Diapason pipes.

Impact of the Solo division 

The Solo division was added in 1902 as Opus 1959 of E. & G. G. Hook & Hastings, placing the windchest over the C# side of the Pedal and Great divisions. Figure 24 is a view from below up into the bottom of the Solo chest. The Pedal wood Trombone pipe in the center is speaking directly into the bottom of the Solo chest, muffling its tone. The Trombone pipe on the left has been mitered to clear the Solo chest.

In Figure 25 one can see that the low C# pipe of the Great 16′ Trumpet speaks directly into the bottom of the Solo chest. In an effort to restore the tuning and power to the pipe, the entire scroll has been crudely forced open. In Figure 26 one can see the more normal scroll of the unobstructed low C pipe of the Great 16′ Trumpet. The diatonic differences heard in the voicing of many bass pipes are entirely due to the unfortunate placement of the Solo division. The craftsmanship and engineering skills of 1902 were clearly inferior to those of 1863.

The change of pitch

The organ was originally pitched at A=450 Hz. Sometime before 1902 the organ was repitched to A=435 Hz.⁶ The current pitch of the organ, 435.3 Hz at 74 degrees F, was measured in June 2000 with a Widener electronic tuner using the 4′ Octave of the Great as the reference pitch, while confirming that this stop was in good tune with itself and the rest of the chorus. The tuning of the organ is quite stable as a result of the use of scrolls in the bass pipes, cone tuning for the trebles, and generous pipe flueways, which do not easily become choked with dust. 

Resonator lengths of the reeds

How did this change of pitch affect the timbre of the reed chorus? Raising the pitch of a reed pipe by pushing down on its tuning wire will eventually force it to overblow to its octave. As an overblowing reed pipe’s tuning wire is slowly raised and the pitch flattened, the pipe will at some point flip back to its fundamental pitch. This is called the “flip point,” and it represents the pitch with the warmest fundamental power. As the wire is raised further, tuning to yet lower pitches, the fundamental will weaken and the harmonics will strengthen in power. The same effect will occur if the resonator is shortened at the flip point. Most reed pipe resonators are adjusted to a length where the flip point is just slightly sharp of the desired pitch—the speech is faster and the harmonic balances are more pleasing with good fundamental warmth and some fire in the harmonics. A good resonator length is not so close to the flip point that it “flips” to the octave when it is tuned on the wire to the flue pipes on the hottest summer days, but it is close to that condition.

With this in mind, the author saw an opportunity to explore the flip points of the Hook chorus reeds. With the exception of the low C pipe, which was added when the organ was repitched to 435 Hz, the resonators of the 4′ Clarion were cut dead length with no scrolls and no evidence of having been shortened. This afforded the opportunity to explore the timbre of these stops relative to what they might have been in 1863. 

The reeds were tested for flip points at 70 degrees Fahrenheit when the tuning of the 4′ Octave was 434 Hz. The pipes were tuned on the wire sharp to their overblowing octaves, then tuned down carefully to their flip points, and the pitch of the pipe relative to A was measured on a Widener electronic tuner. The table below (Figure 27) shows the flip point frequencies for the Great reed chorus and Pedal Trombone.

16′ 8′ 4′ 2′ 1′

Gt 16′ 434.2 441.4 434.3 434.5 445.2

Gt 8′ 435 444.2 435.8 434.5

Gt 4′ 444.1 439.2 449

Pd 16′ 437 434.6 432.6

Pitch @ 70° 434 434 434 434 434

Figure 27

When looking at this table we need to bear in mind that the flip point frequencies need to be higher than the relative pitch of A to which we want to tune the chorus, i.e., these flip points should be significantly higher than 434 Hz. What we find are values ranging from 432.6 Hz to 449 Hz. The direct inference, assuming that the pipes have not been otherwise modified, is that the original chorus was significantly brighter than what we now hear. The dead length reed resonators were apparently not shortened and their tuning wires were used to achieve A=435 Hz, pushing many of the pipes very close to, or even beyond, their flip points. This is a significant offset in the flip point from the original voicing. It is clear that as beautiful and inspiring as it is, we hear a darker approximation of the original 1863 reed chorus in the present organ.

The magnitude of the deficit

The issue of pitch is complicated. Figure 28 shows a graphic depiction of the problem. The shift in pitch at middle A from 450 to 435 Hz is a change of 15 Hz. The distance between a half step at this pitch is about 25 Hz, and when the pipes were moved up a half step, middle A was then repitched to about 425 Hz. The 10 Hz deficit between 425 and 435 Hz was corrected by retuning the pipes. In the case of the dead length reeds, the tuning wires were simply pushed down to raise the pitch, so we know that the original Hook pipes in the table in Figure 27 would have “flipped” at frequencies about 10 Hz higher (at middle A) than what we measured in the table. To bring the pipes back to their original timbre at the current 435 Hz, the resonators would need to be shortened on all reed pipes by an amount that would produce about a 10 Hz increase in pitch at middle A. This may be inadvisable as it would reduce the scale of the resonators.

The Pedal Trombone was not moved up a half step, but large flaps of wood were added to drop its pitch from 450 to 435 Hz, covering the tops of its resonators and reducing its power and brilliance (Figure 23). The correction would entail the removal of the flaps and a lengthening of the resonators, which may be also inadvisable, as it would increase the scale of the pipes, an effect opposite to the correction needed for the reed chorus pipes of the Great division. 

The flue pipes suffered a similar fate and were retuned 10 Hz higher by one or both of two methods: making the pipes shorter and/or opening their toes. Of the two methods, the opening of the toes had a major effect on the timbre and power of the pipes. The impact of such changes is described in the notes on the 16′ Open Diapason and the 8′ Open Diapason Forte, with the result that the current balances deviate markedly from the original intentions of the Hooks. The correction would entail a reduction of the toes where they were opened, and a further shortening of the pipes. Since nearly all façade pipes have had their scrolls rolled down to the maximum extent, or even removed, the correction would require deeper cutouts and new scrolls on all pipes, not a simple or necessarily desirable proposition.

Raising the pitch from 435 to 440 Hz would push some reeds beyond the flip point, further darkening the sound, and it would increase the tuning deficit to 15 Hz. Such an increase in pitch would require further deepening of the façade pipe scroll openings, most of which are already at their limit. Further opening of the toes of the façade pipes would make their timbre and power even more imbalanced than their current state. All of these reasons suggest why the organ was never repitched to 440 Hz. 

Reflections

The Hook organ was put back into regular service use during the tenure of Fr. Thomas Carroll, SJ, as the director of the Jesuit Urban Center at the Church of the Immaculate Conception. Many notable organists at that time visited the church and played the instrument in concerts that were warmly and appreciatively received. 

It is hoped that the research presented in this study will inform those who restore this organ at a future date. Virtually all of the tonal modifications made to this organ resulted from the change to its pitch and the addition of the Solo division; the rest is vintage and very well preserved E. & G. G. Hook. 

Serious consideration should be given to the relocation of the Solo division in a manner that does not encroach upon the tuning of the original Hook pipes or limit the sound egress of the original Hook layout. The raw data indicate that the 1902 installation of the Solo division had a major impact on both counts. If the decision is made to remove the 1902 Solo division from the organ, and that conclusion should not be reached lightly, it should be carefully crated and stored, not discarded. It is a part of the Romantic tapestry and history of this organ.

Three possibilities now suggest themselves: 

1) Leave the organ at 435 Hz and reposition the Solo division to allow sufficient clearance to the Great and Pedal bass pipes. This preserves the current sound but corrects for the tonal and mechanical damage inflicted by the Solo division installation. It does not address the darker character of the reed chorus or the tonal imbalances of the 16′ and 8′ Open Diapasons.

2) Same as Option 1, but shorten the manual reed resonators to their original flip points, i.e., about 10 Hz shorter at middle A. Lengthen the wooden resonators of the Pedal Trombone and remove the obstructing boards. Restore the toes of the Diapasons to their original values and further deepen the tuning slots of all façade pipes. This involves significant expense in pipework restoration, it comes closer to the original Hook sound and power balances, but it permanently and perhaps inadvisedly changes the diameter scales of the many reeds that are cut to length.

Note that most of the scrolls on the reed pipes in Figure 29 (see page 22) are excessively rolled down in an effort to achieve 435 Hz; restoring the original pitch would correct this, so . . .

3) Repitch the organ to its original 450 Hz and move the pipes back to their original positions and voicing, restore the toes of the two Diapasons back to their original values, and restore the tuning scrolls of all pipes back to their original positions. This restores the original sound of the Hook. Repositioning of the Solo division is still essential.

Option 3 would not be the exact sound familiar to those of us who have heard the organ at Immaculate Conception, but it would be faithful to the original intent of the Hooks. The reed chorus would come alive. The author strongly recommends Options 1 or 3 over Option 2. Repitched to 450 Hz, the organ will not be compatible with orchestral instruments tuned to 440 Hz, but neither is the present organ compatible at 435 Hz, and the pipework will clearly not support 440 Hz. The argument can be made that we have a great many organs tuned to 440 Hz in our concert halls, while we have very few large Hook organs in their original state designed for superb acoustics like those of Immaculate Conception. Hook Opus 322 presents us with a unique challenge: it has been passed down to us in superb condition by the careful attention of the Lahaise family, and it may be the best opportunity we have to hear a large, well-preserved Hook chorus of Civil War vintage designed for a stunning acoustic.

The importance of the choice we make of the restoration options pales in comparison to the decision of the site of the organ’s new home. Much of this organ’s fame was the result of its placement in the stunning acoustics of the Church of the Immaculate Conception. When selecting or building a new acoustic for this organ it is important to realize that architects are not accustomed to the requirements of pipe organs. Be especially aware that definitions of reverberation by architects will not even remotely correlate with your musical perception of those acoustics. See The Sound of Pipe Organs, p. 32, for a detailed discussion of this ubiquitous problem. If the Church of the Immaculate Conception still exists in its original acoustical form, an unlikely event, take the architects there and make the accurate replication of those acoustics a requirement. If that acoustic doesn’t exist, take the architects to the Duke University Chapel in Durham, North Carolina. Architects will know how to measure it, but they will be stunned by the request to replicate it. The fame of the Hook organ and its original acoustical environment are inseparable. As any organbuilder will tell you, the best stop in any organ is the room in which it is placed, or to put it more bluntly, a wonderful organ placed in a mediocre room will sound­—mediocre.

Professor Thomas Murray, Yale University organist, has been deeply involved with this Hook organ, has made recordings of it (listed in the discography), and possesses a deep knowledge of the Romantic literature. Future restorers of this organ could benefit from his advice. 

We are incredibly fortunate to have at least some detailed data on the Hook organ, and we owe the Jesuit community and especially Fr. Thomas Carroll, SJ, a great debt for the opportunity to acquire it. Fr. Carroll now resides at the Collegio Bellarmino in Rome, Italy, a home to a community of more than 70 Jesuits representing more than 35 countries. He is the spiritual director for many of the Jesuits pursuing advanced theological degrees, conversing with about half in English and half in Italian. He provides guidance for young Jesuit scholars in the preparation of theses written in English, and for whom English may be a second, third, or fourth language.

Notes and Credits

All photographs, tables, graphs, and data are by the author except as noted.

1. Owen, Barbara. “A Landmark within a Landmark: The 1863 Hook Organ,” undated typescript.

2. Excel files with all raw data taken on the Hook and the spreadsheets that produced the graphs and tables may be obtained at no charge by e-mailing the author at: [email protected].

3. McNeil, Michael. The Sound of Pipe Organs, CC&A, Mead, 2012, 191 pp., Amazon.com.

4. Huntington, Scot L., Barbara Owen, Stephen L. Pinel, Martin R. Walsh, Johnson Organs 1844–1898, OHS Press, Richmond, Virginia, pp. 17–18.

5. Elsworth, John Van Varick. The Johnson Organs, The Boston Organ Club Chapter of the Organ Historical Society, Harrisville, New Hampshire, 1984, p. 45.

6. Noack, Fritz. Preliminary Report about the Pipework of the 1863 E. & G. G. Hook Organ, July 9, 1999.

Discography

Murray, Thomas. The E. & G. G. Hook Organ, Immaculate Conception Church, Boston, Sheffield Town Hall Records, Album S-11 (ACM149STA-B), Santa Barbara, CA.

Murray, Thomas. An American Masterpiece, CD, AFKA SK-507.

Useful References

Cabourdin, Yves, and Pierre Chéron. L’Orgue de Jean-Esprit et Joseph Isnard dans la Basilique de la Madeleine à Saint-Maximin, ARCAM, Nice, France, 1991, 208 pp.

Huntington, Scot L., Barbara Owen, Stephen L. Pinel, Martin R. Walsh, Johnson Organs 1844–1898, The Princeton Academy of the Arts, Culture, and Society, Cranbury, New Jersey, 2015, 239 pp.

McNeil, Michael. The Sound of Pipe Organs, CC&A, Mead, 2012, 191pp, Amazon.com.

Owen, Barbara. The Organ in New England, The Sunbury Press, Raleigh, North Carolina, 1979, 629 pp.

Michael McNeil has designed, constructed, and researched pipe organs since 1973. He was also a research engineer in the disk drive industry with 27 patents. He has authored four hardbound books, among them The Sound of Pipe Organs, several e-publications, and many journal articles.

Preface

Good documentation of organs with enough pipe measurements to permit an analysis of both scaling and voicing is extremely rare. Pipe diameters, mouth widths, and mouth heights (cutups) may be sometimes found, but toe diameters and especially flueway depths are rare. Rarer still are wind system data, allowing a full analysis of wind flow and wind dynamics, parameters that have an enormous impact on the sound of an organ. The reader will find all of this in the following essay on William A. Johnson’s Opus 161.

Good documentation is important for several reasons. We can make useful comparisons with other organs to learn how a specific sound is achieved. And perhaps most importantly, we can document the organ for posterity; while organs are consumed in wars and fires, they are most often replaced or modified with the changing tastes of time. They never survive restorations without changes. Comprehensive documentation may also serve to deter future interventions that intend to “modernize” an organ. Lastly, future restorations of important organs will be more historically accurate if they are based on good documentation.

The mid-nineteenth-century scaling and voicing of William A. Johnson is very similar to the late-eighteenth-century work of the English organbuilder Samuel Green, as evidenced by the data from Johnson’s Opus 16 and Opus 161. Stephen Bicknell provides us with detailed descriptions of Green’s work.¹ Johnson’s scaling is utterly unlike the work of E. & G. G. Hook, whose 1843 Opus 50 for the Methodist Church of Westfield, Massachusetts, set Johnson on a career in organbuilding when he helped the Hooks with its installation.² In this essay we will explore Johnson’s Opus 161 in detail and contrast it with the Opus 322 of the Hooks, both of which were constructed within a year of each other.³ While the Hooks used a Germanic constant scale in their pipe construction, Johnson significantly reduced the scale of his upperwork stops, much in the manner of Samuel Green and classical French builders.

The question arises as to whether Johnson came to his design theory by way of a process of convergent evolution (i.e., independently), or whether he was exposed to the organ Samuel Green shipped to the Battle Square Church in Boston in 1792, and which “was played virtually unaltered for a century,” according to Barbara Owen.⁴ The author suggested to Owen that the Green organ may have had a strong influence on Johnson, but she thought it unlikely that Johnson would have made the long trip from Westfield, far to the west of Boston. 

Travel would indeed have been much more difficult in 1843 when Johnson was exposed to the Hook organ at Westfield. But of some significance was the extension of the Western Railroad from Boston to Westfield in 1843. This new railroad may have been the means by which the Hook organ was shipped to Westfield. Elsworth (see endnote 2) clearly makes the case that Johnson was intoxicated by organbuilding with his exposure to the Hook organ. It is easy to imagine that he would have made a pilgrimage to Boston, at the time a mecca of American organbuilding, perhaps invited by the Hooks to accompany them after finishing their installation in Westfield.⁵

The author was engaged in 1976 by Mrs. Gene Davis, the organist of the Piru Community United Methodist Church, to evaluate the organ at that church. The identity of the organ was in question as no nameplate was in evidence on the console, the organ was barely playable, and its sound was greatly muted by the crude placement of panels in front of the Great division to make it expressive by forcing its sound through the shades of the Swell division above it. An inspection showed that nearly all of the pipework was intact, and a contract was signed to restore the organ to playable condition. The organ was cleaned, the pipes repaired, the few missing pipes replaced, and much of the action repaired by Michael McNeil and David Sedlak.

The church office files produced an undated, typed document that stated: 

The pipe organ in the Methodist Church of Piru was built by William Johnson, of Westfield, Mass., in the early 1860s, making it probably the oldest operating pipe organ in California. It was a second-hand organ when transported by sailing ship 17,000 miles around Cape Horn before 1900, and installed in a Roman Catholic Church in San Francisco. After the earthquake and fire of 1906, the organ was moved to another church and probably at this time parts damaged in the quake were replaced. After many more years of service it was retired and put into storage until, in 1935, Mr. Hugh Warring was persuaded to purchase it for the Piru church. It was purchased for the storage cost of $280.

Evidence of a different and more likely provenance was discovered during the removal of pipework and the cleaning of the organ. Three labels were found glued to the bottom of the reservoir (perhaps as patches for leaks). Two labels read: “Geo. Putnam ‘Janitor’ Stockton California July 1 ’99.” A third label read: “From the Periodical Department, Presbyterian Board of Publication, and Sabbath = Schoolwork, Witherspoon Bldg, 1319 Walnut St., Phila. PA.” At a much later time Reverend Thomas Carroll, SJ, noticed that the clues of Stockton, California, and the Presbyterian church correlated to an entry in the opus list of Johnson organs, compiled in Elsworth’s 1984 book, The Johnson Organs. Opus 161 was shipped in 1864 to the “Presbyterian Church, Stockton, Cal. The church is Eastside Presbyterian.” The organ was listed as having two manuals and 22 stops.⁶ At this time such features as couplers and tremulants were counted as “stops,” and this roughly fit the description of the Piru organ. The façade of the Piru organ is also consistent with the architecture of organs built by Johnson in the 1864 time frame. Elsworth’s illustrations include a console layout of Opus 200 (1866) virtually identical to the Piru organ layout; Opus 134 (1862) exhibits the impost, stiles, and Gothic ornamentation of the Piru organ; Opus 183 (1865) has similar pipe flats and also the console layout of the Piru organ.⁷ Many other details verified the Johnson pedigree, among them the inscription “H. T. Levi” on the reed pipes. Barbara Owen pointed out that Levi was Johnson’s reed voicer during the time of manufacture of Opus 161.⁸ The pieces of evidence fell together when Jim Lewis discovered a newspaper photo of Opus 161 in the Eastside Presbyterian Church of Stockton that matched the façade of the Piru organ. The most likely scenario is that Johnson shipped Opus 161 directly to that church. The Gothic architecture of the Johnson façade also reflects the architecture of the Eastside Presbyterian Church façade. A handwritten note on the Piru church document stated: “Pipe organ and art glass memorial windows dedication June 2, 1935 per Fillmore Herald May 31, 1935, a gift of Hugh Warring.”

It is possible that the organ went from the Presbyterian church into storage, and was later moved to its present location in the 1934–1935 time frame. Even so, we can say with nearly absolute certainty that this organ is William A. Johnson’s Opus 161.

Tonal design overview

It is obvious from even a casual glance at Elsworth’s study of Johnson organs that the Johnson tonal style was based on a classical principal chorus that included mixtures in all but the more modest instruments. But the voicing style is gentle and refined, and bears great similarity to the late-eighteenth-century English work of Samuel Green, whose meantone organ at Armitage in Staffordshire is an excellent surviving example.⁹ Tuned in meantone, Johnson Opus 161 would easily pass muster as the work of Green. The tonal contrast between Green and Hook is stark, and the Hook data serve as an excellent counterpoint to the data from the Johnson organ. Green was the organbuilder favored by the organizers of the Handel Commemoration Festival of 1784, who went so far as to have one of Green’s organs temporarily installed in Westminster Abbey for that occasion. King George III paid Samuel Green to build an organ for Saint George’s Chapel at Windsor.

Stephen Bicknell’s The History of the English Organ relates important details of Samuel Green’s work that we find in Johnson’s Opus 161. “. . . Green’s voicing broke new ground . . . . Delicacy was achieved partly by reducing the size of the pipe foot and by increasing the amount of nicking. The loss of grandeur in the chorus was made up for by increasing the scales of the extreme basses.”¹⁰ And “Where Snetzler provided a chorus of startling boldness and with all the open metal ranks of equal power, Green introduced refinement and delicacy and modified the power of the off-unison ranks to secure a new kind of blend.”¹¹ The Hooks, like Snetzler, used a constant scale where all of the pipes in the principal chorus at a given pitch had about the same scale and power.

The most basic data set for describing power balances and voicing must include, at a minimum, pipe diameters, widths of mouths, heights of mouths (“cutup”), diameters of foot toe holes, and depths of mouth flueways. The data in this essay are presented in normalized scales for inside pipe diameters, mouth widths, and mouth heights. Tables showing how raw data are converted into normalized scales may be found in the article on the E. & G. G. Hook Opus 322 published in The Diapason, July 2017. The full set of Johnson data and the Excel spreadsheet used to analyze them may be obtained at no charge by emailing the author.¹² Also available is the book The Sound of Pipe Organs, which describes in detail the theory and derivation of the models used in this essay.¹³

Pitch, wind pressure, and general notes

The current pitch of the Johnson and Hook organs is dissimilar and should be taken into consideration when observing the scaling charts. The Hook organ is now pitched at A=435.3 Hz at 74 degrees Fahrenheit, while the Johnson organ is now pitched at 440 Hz. The original pitch of the Hook organ was 450 Hz; new low C pipes were added when the pitch was changed to 435 Hz, and the original pipework was moved up a halftone, widening its scales by a halftone. The original pitch of the Johnson organ was approximately 450 Hz; the pipes were lengthened to achieve a lower pitch.¹⁴ The Hook and Johnson organs are both tuned in equal temperament. The wind pressure, water column, of the Hook is 76 mm (3 inches); the Johnson organ was measured at 76 mm static and 70 mm under full flow on the Great division. The pressure was reduced during the restoration to 63 mm static. This allowed the pitch of the pipes to drop, making the adjustment to 440 Hz with fewer changes to the pipe lengths; most of the pipes that were originally cut to length had been crudely pinched at the top to lower their pitch. With the reduction in pressure the ears of the 4′ Flute à Cheminée, with its soldered tops, achieved a more normal position. 

The Piru room acoustic was reasonably efficient, and while the Johnson voicing is very restrained, it was adequate to fill this room on the reduced pressure. The Piru church seats 109, has plastered walls, wood and carpet flooring, and a peaked ceiling about 30 feet high; the reverberation, empty, as heard with normal ears, is well under one second (this is not the measurement used by architects that erroneously reports much longer reverberation). Elsworth relates that “the wind pressure which Johnson used during this period was generally between 21⁄2 and 23⁄4 inches [63.5 and 70 mm], and, in rare examples, nearly 3 inches [76 mm].”¹⁵ The photograph of the original Eastside Presbyterian Church for which the Johnson was designed implies a larger acoustical space than that of the Piru church.

The compass of the Johnson organ is 56 notes in the manuals, C to g′′′, and 27 notes in the pedal, C to d′.

Stoplist

The Johnson console was found in poor condition, missing the builder’s nameplate and many of its stop knob faces. Correct stop names were derived from the markings on the pipes and the missing faces were replaced. The original stoplist is reconstructed as follows (Johnson did not use pitch designations):

GREAT

8′ Open Diapason

8′ Keraulophon

8′ Clarabella

4′ Principal

4′ Flute à Cheminée (TC)

22⁄3′ Twelfth

2′ Fifteenth

8′ Trumpet

SWELL

16′ Bourdon (TC)

8′ Open Diapason

8′ Stopped Diapason

8′ Viol d’Amour (TF)

4′ Principal

8′ Hautboy (TF)

Tremolo

PEDAL

16′ Double Open Diapason

Couplers

Great to Pedal

Swell to Pedal

Swell to Great

Blower signal

The above list adds up to 20 controls. The Johnson company opus list describes Opus 161 as having 22 “stops.” This may have reflected the original intention to supply the organ with stops having split basses, which are commonly found in Johnson specifications. The sliders for the Keraulophon and the Trumpet were found with separate bass sections from C to B, professionally screwed together with the sections from tenor C to d′′′. The two additional bass stops would account for a total of 22 “stops.” There are no extra holes in the stop jambs to indicate the deleted split bass stop actions. The extant stopjambs are apparently a later modification from the time of the installation at Piru or before. Elsworth noted that all Johnson organs of this period were constructed with square stop shanks.¹⁶ The current shanks are round where they pass through the stopjambs and are square where they connect to the stop action.

Several stop knobs were switched during the 1935 installation at Piru; e. g., the Viole d’Amour in the pre-restoration photo of the right jamb belongs in the position noted on the left jamb with the black plastic label “Bell Gamba,” which indeed is how this stop was constructed. The Swell Stopped Diapason was operated by a knob labeled “Principal” [sic]. The illustrations of the left stopjamb and right stopjamb diagrams provide the correct nomenclature as restored in the correct positions, with the incorrect 1935 nomenclature in parentheses ( ) and the correct pitches in brackets [ ].

The wind system

The wind system can be modeled from two viewpoints: the restriction of flow from the wind trunks, pallets, channels, and pipe toes; and the dynamics of the wind. Wind dynamics are fully explained in The Sound of Pipe Organs and are a very important aspect of an organ’s ability to sustain a fast tempo with stability or conversely to enhance the grand cadences of historic literature. The data set on the Johnson allows us to model all of these characteristics. Figure 1 shows the Johnson wind flow model.

In Figure 1 we see a table of the pipe toe diameters and their calculated areas; values in red font are calculations or interpolations from the data (e.g., wood pipe toes are difficult to measure when they have wooden wedges to restrict flow). These areas are measured for a single note in each octave of the compass.

A model for the total required wind flow of the full plenum of the organ assumes a maximum of ten pallets (a ten-fingered chord), as described in the table, and the flow is multiplied by the number of the pallets played for each octave in the compass. The sum of the toe areas of all ten manual pallets in the tutti is 5,057 mm². The total area of the manual wind trunks is 38,872 mm², and we see that the wind trunks afford 7.7 times more wind than the tutti requires, so much in fact that the trunks do not at all function as an effective resistance in the system.

Interestingly, the Isnard organ at St. Maximin, France, used the main wind trunk as a strong resistor to dampen Helmholtz resonances in the wind system, and that organ has ratios of wind trunk area to a plenum toe area of only 1.07 for the coupled principal chorus of the Grand-Orgue and Positif, but with no reeds, flutes, or mutations. Helmholtz resonances are the source of what is normally called wind shake, and we would expect some mild wind shake with the Johnson’s large wind ducts and low damping resistance. The author’s notes from 1976 state: “Very little sustained shake . . . a considerable fluctuation in pitch when playing moderately fast legato scales, which stabilizes very rapidly . . . this imparts a shimmer . . . .”

In Figure 1 we also see dimensions of the key channels, pallet openings, and the pallet pull length (estimated from the ratios in the action). These allow us to calculate the relative wind flow of the channels and pallets. We find that there are robust margins in wind flow from the channels to the pipe toes (244% at low C to 737% at high C on the Great). This accounts for the small drop in static pressure at 76 mm to a full flow pressure of 70 mm with all stops drawn. Pallet openings are less robust and flow about 100% of the channel area for the first three octaves and 190% in the high treble.

The underlying dynamics of a wind system are the result of the mass of its bellows plate and the volume of air in the system. These factors produce a natural resonance that can enhance the grand cadences of literature with a long surge in the wind, or it can produce a nervous shake if it is too fast. A grand surge in the wind is characterized by a resonant frequency of less than 2 Hz (cycles per second), and it is most often produced by a weighted bellows. A nervous shake results from a sprung bellows. We correct the latter condition with small concussion bellows in modern organs, but the Johnson organ does not have such devices; instead, it features only a large, weighted, double-rise bellows. 

We can model the dynamic response of an organ by using its wind pressure, the area of the bellows plates, and the combined internal volume of its bellows, wind trunks, and pallet boxes. The model in Figure 2 shows the dynamic response of the current Johnson wind system at a relaxed 1.61 Hz. This low resonant frequency drops further to 1.47 Hz when the pressure is raised to its original value of 76 mm. The author’s notes from 1976 state: “Light ‘give’ on full organ; relatively fast buildup to full flow.” That “light give” is the result of the low resonant frequency of the system. The resonant frequency of the Hook organ was modeled at 1.23 Hz, a value lower than the Johnson, and the Hook chorus does indeed exhibit a slower and grander surge on full organ. Figure 3 shows the modeled resonant frequency at the original pressure of 76 mm for the Johnson organ. The equation for modeling the resonant frequency of a wind system along with a worked example on the 1774 Isnard organ at St. Maximin may be found in The Sound of Pipe Organs, pages 99–113.

The wind system in pictures

See the accompanying pictures: Notebook sketch 1, Great windchest, Toeboard, Notebook sketch 2, Notebook sketch 3, Notebook sketch 4, Great pallet box, Pallet springs, Notebook sketch 5.

The layout in pictures

“Green’s organs stand on an independent building frame with the case erected around it, rather than being supported by the structure of the case itself.”¹⁷ Bicknell’s description of a Samuel Green organ applies equally well to this Johnson organ. The casework is built entirely of black walnut, a wood mentioned by Elsworth in reference to Johnson cases. The organ is situated within the front wall of the church. The original black walnut side panels (typical of early Johnson organs) were found crudely cut up and nailed behind the façade in an effort to make the whole organ expressive through the Swell shades. This had the effect of making the Great division sound like a diminutive Echo division. The typical layout of a Johnson organ is well described by Elsworth: “The framework was arranged to carry the chests of the Great organ and the supporting framework for the Swell, which was usually above the Great organ and slightly to the rear.”¹⁸ Such layouts, shown in Figure 4, are common in nineteenth-century American organbuilding. The walkway behind the Great allowed access to the pipes and pallets placed at the rear of that chest, and the rollerboard to the Swell division was normally placed just behind this walkway, allowing access to the Swell pallets that were placed at the front of the Swell windchest. Opus 161 was installed in an opening in the Piru church that was far too shallow to allow the depth of a rearward placement of the Swell division. 

As a result, there is evidence that the Swell windchest may have been reversed, placing its pallets to the back of the windchest, and the chest brought forward over the Great division. Note the lack of clearance between the 4′ Principal pipe and the bottom of the Swell chest in Figure 5. The internal framework shows signs of crude saw cuts; the order of the notes on the Swell chest is the same as the Great, but it is reversed; the Swell rollerboard appears to have been likewise reversed and now faces toward the walkway where the action and rollers are exposed to damage. 

To say that the Piru layout was cramped would be an understatement; no one weighing over 150 pounds would gain access to the pipes for tuning or to the action for adjustment without damaging the pipework or the key action. The author weighed less (at the time) and was barely able to navigate inside the organ. The current layout is shown in Figure 6. 

It is also possible that the current layout reflects the original layout by Johnson, but that the Swell was simply lowered to fit the height of the Piru church and brought forward to fit the limited depth available, reducing the depth of the walkway.

Notes and credits

All photos, drawings, tables, and illustrations are courtesy of the author’s collection if not otherwise noted. Most of the color photos were unfortunately taken by the author with an inferior camera in low resolution. David Sedlak used a high quality camera, lenses, and film to produce the high-resolution color photos of the church and its architectural details; these are all attributed to Sedlak.

1. Stephen Bicknell, The History of the English Organ, Cambridge University Press, 1996, Cambridge, pp. 185–187, 190–191, 207.

2. John Van Varick Elsworth, The Johnson Organs, The Boston Organ Club, 1984, Harrisville, p. 18.

3. A detailed study of the E. & G. G. Hook Opus 322 may be found in The Diapason, July, August, and September issues, 2017.

4. Barbara Owen, The Organ in New England, The Sunbury Press, 1979, Raleigh, pp. 18–19.

5. see: en.wikipedia.org/wiki/Boston_and_Albany_Railroad.

6. The Johnson Organs, p. 100.

7. Ibid, pp. 23, 50, 57, respectively.

8. The Organ in New England, p. 275.

9. 5 Organ Concertos, 1984, Archiv D 150066, Simon Preston, Trevor Pinnock, The English Concert.

10. The History of the English Organ, p. 185.

11. Ibid, p. 207.

12. McNeil, Michael. Johnson_161_170807, an Excel file containing all of the raw data and the models used to analyze the Johnson Opus 161, 2017, available by emailing the author at [email protected].

13. McNeil, Michael. The Sound of Pipe Organs, CC&A, Mead, 2012, 191 pp., Amazon.com.

14. The Organ in New England, p. 75.

15. The Johnson Organs, p. 25.

16. Ibid, p. 23.

17. The History of the English Organ, p. 187.

18. The Johnson Organs, p. 23.

To be continued.

Michael McNeil has designed, constructed, and researched pipe organs since 1973. He was also a research engineer in the disk drive industry with 27 patents. He has authored four hardbound books, among them The Sound of Pipe Organs, several e-publications, and many journal articles.

Editor’s note: Part 1 of this article was published in the August 2018 issue of The Diapason, pages 16–20.

The casework in pictures

The entire casework of Opus 161 is executed in solid black walnut, and in the author’s opinion is among the best of Johnson’s cases with its elegant proportions and understated Gothic ornamentation. The window above the entrance of Eastside Presbyterian Church, its original home, displayed similar, restrained Gothic form and ornamentation. Elsworth’s book illustrates a great many of Johnson’s organs, among them Opus 134, built in 1862 for St. Luke’s Episcopal Church in Lanesborough, Massachusetts.¹⁷ Opus 134 has nearly identical stiles and ornamentation, but its proportions do not soar in the elegant manner of Opus 161, perhaps the result of limitations in height. It is ironic that one of Johnson’s best aesthetic creations has languished in anonymity for decades. Many American churches built in the early nineteenth century did not have a provision for a pipe organ, and as a consequence Elsworth noted that most of Johnson’s earlier organs were furnished with sides to the cases of the free-standing organs produced for such churches.¹⁸ As previously noted, Opus 161 originally had such side panels to its casework, and these were found crudely sawn and nailed behind the façade. The Piru church elected to place the façade casework flush with the wall of the church, necessitating the removal of the side panels.

As was typical of nearly all nineteenth century organs, the façade contains no smaller pipes. The side flats contain pipes of the Open Diapason with considerable overlengths. This is the only architectural flaw in this otherwise stunningly designed case. The use of pipes of very different lengths is an important architectural device—it gives a sense of scale, making the larger pipes appear more imposing in contrast. But façades with pipes of extremely different size are more complex and more expensive to make. Compared to the vast majority of nineteenth-century façades, Opus 161 is one of the finest aesthetic designs.

The keydesk in pictures

The reader should refer to Part 1 of this series for photographs of the keydesk and stop jambs (August 2018, pages 17–18). Elsworth described the keydesks of Johnson organs from the period of Opus 43, 1855, to Opus 268, 1868:

The manual compass was invariably fifty-six notes, from CC to G3. The stop knobs were disposed in vertical rows on each side of the manual keyboards, and always had square shanks with round knobs that had flat faces. Into these faces were set the ivory labels with the stop names. The labels were always engraved in Spencerian script with no pitch indication. The nameplates up to about 1867 or 1868 were of silver, engraved “Wm. A. Johnson, Westfield, Mass.”¹⁹

This description provides some evidence that the organ was modified during its installation at Piru. The stop action does indeed have square shanks leading to the bellcranks, but the shafts connecting to the square shanks and leading through the stop jambs are round. The author had initially believed that the stop jambs were original, observing well-worn and professionally installed felt bushings in the openings of the stop jambs. But a more likely explanation is that the round shafts and extant jambs were added at a later date, and this goes a long way to explain the disappearance of the split bass stops, all of which were screwed together to make continuous stops with no splits. And this nicely explains the current specification with 20 controls instead of the 22 controls indicated in the opus list of the Johnson factory.

The organ was initially supplied with a hook-down Swell shoe, normal fare for Johnson’s work of this time. This feature was deleted, and a balanced Swell shoe was installed by crudely re-routing the action of the Great to Pedal coupler rollerboard. Note the added Swell pedal in Figure 7, the missing hook-down pedal in Figure 8, and the damage to the action in Figure 9 and Figure 10. All of this damage was repaired in the 1976 restoration and the original hook-down mechanism refabricated. The figures show the condition of the console prior to the restoration.

The key action in pictures

The basic layout of the key action can be seen in Figure 6 in Part 1 of this series (August 2018, page 20). With the exception of the repositioning of the Swell chest and the addition of the balanced Swell pedal, the key and stop action of Opus 161 was well worn but virtually unaltered in 1976. The damage to the trackers on the Pedal couplers from the installation of the balanced Swell pedal was repaired in 1976 with new trackers, wires, felts, and buttons, and basic repairs to the stickers on the Swell to Great coupler were made, but this was a stopgap solution. At this time the console was in need of a complete disassembly and refurbishment of the leather on the couplers, the felts, and the leather buttons. The action was well designed, had served for a period of more than a hundred years, and had survived a move from Stockton to Piru. But the leather facings of the key tails where the coupler stickers made contact and the felts and leather buttons were showing their age. There were no funds for such work in 1976. 

In Johnson’s action we see similarities to Samuel Green. Bicknell writes: 

Green introduced or developed numerous refinements to the mechanism. He often arranged pipes from f# up in chromatic order on the soundboards, even in large organs. This reduced the extent to which rollerboards were required. . . . To make the key action readily adjustable the ends of the trackers were fitted with tapped wires and leather buttons. The appearance of Green’s consoles was enhanced by the use of ivory inserts screwed into the heads of the stop knobs, engraved with the name of the stop. . . . Green also usually made keyboards with white naturals and black sharps. . . .²⁰

All of these features are found on Opus 161. The photographs of the action were all taken in 1976 prior to the restoration work.

The stop action in pictures

The stop action of Opus 161 is conventional, with metal squares and square wooden shanks. The stop action to the Pedal 16′ Double Open Diapason is a ventil valve to the three windchests of that stop, which are placed at the sides (largest pipes, diatonic) and the treble pipes at the back (chromatic). The photographs show the details of the stop action construction.

A description of the stops and general notes on the scaling and voicing

This section provides a detailed description of the stops; two of the Swell stops were not measured (16′ Bourdon and 8′ Stopped Diapason). For the stops which were measured, a table of data in millimeters is shown. The photographs show some details of the construction, although the poor resolution of the camera is regrettable.

As earlier noted, there is a close resemblance between the organs of Samuel Green in late eighteenth century England and the organs of William A. Johnson in nineteenth-century America. Bicknell writes:

On the tonal side Green seems to have adopted the trend towards delicacy and developed it still further. . . . Green’s first line of development in securing the effect he desired was to experiment . . . with the scales of the chorus . . . . in 1778 the Open Diapason is larger than the rest of the chorus. . . . The appearance of extra pipes in some ranks, definitely by Green and contemporary with the instruments themselves, together with re-marking of the pipes, suggests that Green took spare pipes with him to the site and rescaled stops during the tonal finishing in the building. This is considerably removed from the standardised scaling and voicing adopted by, for example, Snetzler. The reasons for this become clearer when one understands that Green’s voicing broke new ground in other aspects as well. Delicacy was achieved partly by reduction of the size of the pipe foot and by increasing the amount of nicking. The loss of grandeur in the chorus was made up for by increasing the scales of the extreme basses. . . .²¹

As we will see in the graphical analysis of the data, all of the features mentioned by Bicknell about Samuel Green would apply equally well to Johnson’s Opus 161. Bicknell observes, “Where Snetzler provided a chorus of startling boldness and with all the open metal ranks of equal power, Green introduced refinement and delicacy and modified the power of the off-unison ranks to secure a new kind of blend.”²²

As earlier noted by Elsworth, Johnson’s wind pressure during the period of 1855 to 1868 “was generally between 21⁄2 and 23⁄4 inches (63 and 70 mm), and in rare examples, nearly 3 inches.”²³ The lower wind pressures, narrower scales of the upperwork, and reduced toes produced a sound with restrained brilliance. 

Referring to his conversations with Edwin B. Hedges (1872–1967), a voicer for Johnson organs, Elsworth made some telling observations. In the process of making the pipework, “ . . . the languids were carefully soldered in place, and the flues were properly adjusted.”²⁴ This is a very important comment, because today the flueway is considered a variable for adjusting power in some voicing styles, especially North Germanic voicing. Johnson’s flueways are very open, often the maximum that would produce good speech, even with Johnson’s bold nicking. Power balances, for Johnson as well as Green, were designed into the scales and further adjusted by the voicer at the toe. “The voicing of flue pipes, such as Diapason, Dulcianas, and strings, consists of nicking the languid, cutting up the upper lips to the proper mouth height, and adjusting the positions of the languid and the upper and lower lips. The amount of wind entering the pipe foot must be carefully adjusted by opening or closing the orifice in the pipe toe.”²⁵ There is no direct evidence that William A. Johnson had first-hand knowledge of the 1792 Samuel Green organ delivered to Boston, but the legacy of Green is obvious in Johnson’s work.

A few comments are in order on the nicking and languid treatment. The languids contain a counterface with a negative angle; the more usual angle is vertical, or 90 degrees. The Isnards made a positive-angled counterface at about 75 degrees with a normal bevel at about 45 to 55 degrees. The negative counterface of the Johnson languid is unusual. This languid is nicked at an angle with a knife, cutting a fine nick as deep as halfway into the languid bevel. Long knife cuts were also in evidence inside the lower lip. As a general rule there are the same number of nicks on a languid, regardless of pitch. These languids work well and produce fast speech even when the lower, negative languid bevel shows above the top edge of the lower lip; the upper lip is not pulled out to compensate for this languid position. Ears are generally found up to 1′ in pitch in the principal chorus, but they are very narrow, not extending far in front of the mouth.

Many of the pipes were found in 1976 to be crudely pinched at the top, part of an effort to reduce the pitch to the modern standard. All of this damage was repaired on mandrels, and tuning slides were fitted.

Great division

8′ Open Diapason 

This is the first stop on the front of the Great windchest. It has zinc resonators from low C to tenor B and planed common metal feet from about tenor E. All pipes from middle C are planed common metal (30% tin, 70% lead). Zinc wind conductors to the façade pipes supply copious wind; the conductor diameters are 38 mm at low C and 25 mm at tenor C. If memory serves, at least one or two of the pipes in the side flats were dummy pipes, implying that the speaking façade pipes extended to tenor D. The façade pipes were tuned with scrolls at the back, which were entirely rolled up as a consequence of the drop in pitch to 440 Hz, where the original pitch was probably closer to 450 Hz. See the earlier notes on the pitch and wind pressure. As with all of the stops in the principal chorus, the ears are very narrow. 

The author feels obligated to point out a grave error he made in the restoration by removing the heavy nicking on the languids of the Open Diapason, and only on this stop. To make the record clear, David Sedlak advised against doing this, and the author regrets that he did not take Sedlak’s advice. These nicks should be renewed in the manner used by Johnson.

8′ Keraulophon

The second stop on the chest, the Keraulophon pipes were found badly pinched at the top along with crudely reduced toe bores in an effort to reduce the pitch. All of the pipes were straightened on mandrels and tuning slides added. Toes that were not damaged were used as a guide for readjusting damaged toes. This stop is voiced with tuning slots and ears, but no beards of any kind. The bass octave is common with the Clarabella, five pipes from tenor C to E have zinc resonators, and the rest have planed common metal resonators. The nicking is bold and often crossed to keep the speech stable. Flueways were often more closed on one side. This is a bolder string than a Dulciana. 

8′ Clarabella

This is the third stop on the chest. Bass pipes C to tenor E are stopped wood; the remainder are open wood with lead plates covering the tops for tuning. These lead plates are somewhat closed down to accommodate the lowered pitch. The internal blocks forming the languids are lower than the front plates by 2.0 mm at tenor E, and 1.5 mm at tenor F. The bevel of the upper lip is internal for the open pipes and external for the stopped pipes. The stopped pipes have narrow, slanted strips at the sides of the mouth to form narrow ears; the open pipes have no extra strips functioning as ears. The nicking is deeper and heavier than the pipes of the principal chorus. The scales and voicing of this stop place its power on the same level as the principal chorus foundations. The only concession to power is a greatly reduced mouth width in the bass octave, a concession to its function as a common bass to the Keraulophon. 

The effective inside diameter of a wooden pipe is a calculation of its diagonal, a method proposed by Nolte.²⁶ The potential power of a round pipe is related to the amplitude of the standing wave in the pipe, which is in turn related to its diameter. Following this logic, Nolte has pointed out that the amplitude of a standing wave in a rectangular pipe is related to its widest point, i.e., its diagonal. We often see modern conversions of wood pipe scales by relating their rectangular areas to those of round metal pipes with equivalent areas, but this does not produce balanced power. The consequence is that conventional modern wisdom decrees that wood pipes should be scaled a few half tones narrower than round pipes of equivalent area. This disconnect disappears with Nolte’s observation of the relevance of the diagonal, not equivalent areas. This is not a new idea. Many older organs, e.g., J. A. Silbermann’s organ of 1746 at Marmoutier, show very disjointed scales between the rectangular wood bass of the 16′ Montre and its metal pipes when plotting by equivalent areas. Convert the Silbermann wood bass scales to diagonals and those scales merge seamlessly into the scales of the metal pipes. Diagonal computations of the effective diameters for the Johnson Clarabella can be found in the table, and those calculations are used in the graphical analysis. 

4′ Principal

The fourth stop on the chest, the Principal has five zinc resonators from C to E; the rest are all planed common metal. These pipes showed very little damage. The flueway depths are remarkably wide, especially in the treble, and demonstrate that Johnson regulated power entirely at the toe, not the flueway. Such flueway depths are often found in classical French voicing. This data set can be taken as reasonably accurate evidence of Johnson’s unmolested voicing.

4′ Flute И CheminОe

 The fifth stop on the chest from tenor C, this is a classically constructed flute in planed common metal with soldered domed tops, chimneys with no tuning mechanism, and very large ears for tuning. Those large ears had been pushed in far enough to virtually touch each other when found in 1976, another effort to reduce the pitch. The cutups were lightly arched. There was considerable handling damage to the flueways. The toes were reasonably intact. The reduction in pressure from 76 mm to 63 mm allowed these pipes to speak much more freely with the ears much more opened (but not completely straightened). The pipe construction becomes open at g#′′, i.e., the last twelve pipes, and they are noticeably wider across the break. The table above shows a calculation of the total resonator length, i.e., the body length plus the chimney, and the percentage of the chimney length to the total length. This gives an idea of the harmonics that Johnson was trying to emphasize with the chimney. At tenor C the chimney is 25% of the total length, emphasizing the fourth harmonic, while at middle C the chimney is 30% of the total length, roughly emphasizing the third harmonic. The chimney progresses to larger percentages of the total length as the pitch rises. The chimney is not a constant percentage of the total length.  The photograph shows the classical construction of this stop. 

22Џ3′ Twelfth

The sixth stop on the chest, this stop consists entirely of planed common metal pipes that had minimal damage.

2′ Fifteenth

The seventh and last flue stop on the chest, the 2′ Fifteenth continues the trend of extremely deep flueways and closed toes. The flueway depths of this stop are perhaps the largest the author has measured on any organ. Remarkably, this planed, common metal stop has no ears on any pipe, and its sound is exquisite. The toes are very restrained and represent the means of controlling power. The diameter and mouth width scales are considerably narrower than the Open Diapason, continuing the trend of narrower scaling with higher stop pitches, a characteristic introduced by Samuel Green. This progression can be clearly seen in the graphical analysis, in stark contrast to the Hook’s constant scaling of  the principal chorus. By this means Johnson and Green achieved a chorus with more refinement and less impact, but they compensated with very wide scaling of the extreme basses.

8′ Trumpet

The extant pipework of this eighth and last stop on the chest was constructed of planed common metal with zinc bottom sections from tenor C to tenor B. The Trumpet has an obscure history. In 1976 only two octaves of pipes were found from tenor C 13 to C 37. These were all in fairly good condition without obvious modifications; some crude slotting of the tops was repaired and the pipes spoke well on 63 mm wind. All of the original pipes were cut to exact length with no tuning slots or scrolls. The bass octave of the Trumpet was originally separated on the slider, but found screwed together in 1976. Interestingly, while the bass topboards were bored and chamfered to receive pipes, the chamfers were not burned in like all other borings on both windchests. With the repositioning of the Swell chest over the Great chest, it was now impossible to reconstruct a full-length bass set of pipes, and a half-length set was fabricated with limited tonal success (a few of the half-length pipes needed mitering to clear the Swell chest). The missing treble pipes were recreated by the firm of Stinkens to scales extrapolated from the original pipework. These were quite successful and a good tonal match. The high treble from c#′′′ to g′′′ were obviously flue pipes, and the rackboard borings provided guidance for their scales. All shallots are brass and are marked “H. T. Levi,” one of the reed voicers for William A. Johnson, according to both Barbara Owen²⁷ and Elsworth.²⁸ This stop bears a strong resemblance to the Trumpet heard in the recording of the Samuel Green organ at Armitage, Staffordshire, England (see the section on Recordings).

The Trumpet was carefully disassembled during the restoration and its measurements carefully tabulated; see the drawings and tables below. Measurements unfortunately omitted were the height of the block and the length and width at the top of the main taper on the tongues.

II Mixture

The author added a two-rank mixture in planed common metal to the Great during the 1976 restoration. While the merits of this can be debated, it was added in a manner that did not affect the other stops. A thick oak board was mounted at the back of the key channels, extending backwards and upwards, making this the ninth stop on the Great. The pipework was narrowly scaled in the manner of Johnson, roughly -7 half tones from 2⁄3′ pitch to 1⁄4′ pitch, then widening to about -3 half tones at 1⁄8′ pitch. A great many Johnson organs of this size had mixtures. It should be noted that Johnson mixtures of the time period during which Opus 161 was created were called Sesquialtera, and they included third-sounding ranks. Elsworth states, “ . . . these were composed of 17th, 19th, and 22nd ranks [i.e., 13⁄5′, 11⁄3′, and 1′, the same pitches observed in Samuel Green’s Sesquialteras] with two or three breaks.”²⁹ The mixture added by the author is more typical of later Johnson work in its composition without thirds.

The voicing of the cutups was a fortunate accident, where the pipes were mouth-voiced before realizing that they were left many half tones overlength by the pipemaker. When the cone-tuned pipes were cut to length, it was obvious that the cutups were very high. But this was fortuitous, because it taught the lesson that high cutups can have a superb blend, and this mixture provided a fine sparkling glitter in the plenum with no hint of harshness. There are no ears on any pipes. The toes are relatively more open than what Johnson would have done and the cutups are higher. The mixture composition is as follows:

C 2⁄3′ 1⁄2′

c 1′ 2⁄3′

c′ 11⁄3′ 1′

c′′ 2′ 11⁄3′

c′′′ 4′ 2′

Barbara Owen noted that William A. Johnson was hired to add a VII Cymbal to the Hook organ.³⁰ This mixture was installed in 1870, and no records indicate how this happened. The political implications invite much speculation, of course. The differences in scaling and voicing of the Johnson mixture relative to the Hook chorus illuminates the different approach to chorus design between Johnson and Hook. We will look at this in detail in the graphical analysis. The Johnson VII Cymbal provides a scintillating crown to the Hook chorus and contains a third-sounding rank. In 1871 William H. Johnson, the son of William A. Johnson, joined his father as a partner in the firm and the mixtures built from that time deleted the third-sounding rank.³¹ ν

Notes and Credits

All photos, drawings, tables, and illustrations are courtesy of the author’s collection if not otherwise noted. Most of the color photos were unfortunately taken by the author with an inferior camera in low resolution. David Sedlak used a high quality camera, lenses, and film to produce the high-resolution color photos of the church and its architectural details; these are all attributed to Sedlak.

17. The Johnson Organs, p. 50.

18. Ibid, p. 22.

19. Ibid, p. 23.

20. The History of the English Organ, p. 186.

21. The History of the English Organ, p. 185.

22. Ibid, p. 207.

23. The Johnson Organs, p. 25.

24. Ibid, p. 45.

25. Ibid, p. 47.

26. John M. Nolte, “Scaling Pipes in Wood,” ISO Journal, No. 36, December 2010, pp. 8–19.

27. Scot L. Huntington, Barbara Owen, Stephen L. Pinel, Martin R. Walsh. Johnson Organs 1844–1898, The Princeton Academy of the Arts, Culture, and Society, 2015, Cranbury, pp. 11, 13, 14, 16.

28. The Johnson Organs, p. 36.

29. Ibid, p. 48.

30. Johnson Organs 1844–1898, pp. 17-18.

31. The Johnson Organs, p. 48.

To be continued.