July 7, 2006

Dear David [Boe] and Jim [James David Christie]:

Michael Kraft suggested I write to you about my ideas for practice organs, as you informed him
that Oberlin is beginning a long-term process of replacing its practice machines. Following are a
letter and an email to Robert Poovey and an email to John Schreiner espousing conclusions of
my decades-long ruminations on the subject. Also, a picture and the specification of the organ I
am in the process of building for Robert.

3 Currier's Court
Newburyport, MA 01950-2303
June 22, 1996

Dear Robert,

Enclosed are renderings and specifications of my dreamed-of "line" of residence and table-
positive organs.
Residence Organ
 Manual I
1. Double Open Diapason 16' 63 pipes
2. Prestant 8' 63 pipes
3. Quint 6' 63 pipes
4. Octave 4' CC-d2 from c1-d4 of 1. +24 pipes

Manual II
5. Double Stopt Diapason 16' 63 pipes
6. Traverse Flute 8' CC-BB from c0-b0 of 1. 51 pipes
7. Aeolian Harp 8' CC-FF# from c0-f#0 of 5.,
GG-d4 from CC-g3 of 3. -
8. Pandean Flute 4' CC-d2 from c1-d4 of 5. + 24 pipes
351 pipes

Pedal
9. Double Open Diapason 16' All from 1. -
10. Double Stopt Diapason 16' All from 5. -
11. Tenor Prestant 8' All from 2. -
12. Soprano Prestant 4' All from 2. -

Couplers
I to I Octaves
II to I
I to Pedal
II to Pedal

63-note manuals
32-note pedalboard
Bebung tremulant
Serpentone Projector
Swell Box enclosing all pipes except Prestant and lowest 12 notes of
Double Open and Double Stopt Diapasons
Stands under an 8' ceiling

I have been experimenting with different case designs of the residence organ. One has a "wasp-
waisted" silhouette, a most impractical design because the pedalboard is wider than the lower
case! I really would prefer to include round towers; the cul-de-lamp brackets would require the
lower case to be the same width as the upper case. Since the casework would all be made of
plywood or paper or plastic or Styrofoam(!), it would have to be painted rather than natural
wood. I'd love a dark olive (drab) green with gold pipes and accents, but here the rendering is of
the case painted white, reminiscent of Tannenburg’s cases. The two pedal towers are inspired by
English telephone booths and Japanese shoji; because the pipes inside would be enormously fat
Sonotubes (a cardboard tubing used by building contractors for concrete forms), I figure it is
pointless to try to have reasonable-looking front pipes in them.

I would like to hear your opinions of the specification. This is a 16' organ; variety is obtained by
playing combinations up an octave as well as in situ, and this is the reason for the 63-note
manuals. Do you think that is enough notes for playing up an octave or should there be still
more, especially for the sake of the super coupler? Actually, 63 9-mm slider seals side by side
push the limits of the case width just as it is. I arbitrarily chose the case height as 7' 6"; the
golden mean then determined the case's width, and I have stuck with those two dimensions
unquestioningly for ten years, now. Another reason I settled on the 63-note manual compass is
that it allows playing the top note of English organ music (Händel, Stanley, Wallond) played up
an octave.

My reasoning behind the 16' pitch basis is the observation that the old Estey two-manual-and-
pedal reed organs always had 16' manual stops, and that this contributed to the grand cathedral
growl they produced, even in a dead room. The 5 1/3' might seem insane according to the
conventional wisdom, but an experience years ago with Tom Murray convinces me otherwise:
When I was singing for him in the Cathedral Choir, and he was still living in Newbury, he would
come by Sunday mornings to pick me up. I had set up in my living room a chest and keyboard
with an 8' Gedeckt and an 8' Viol with a Haskell bass on it. I had also gotten from Fisk another
old set of Viol pipes, from tenor f, and for lack of a better storage place had set them on a spare
slider—at 5 1/3' pitch because that is where they fit most conveniently into the existing rack
holes. I had not bothered to open up the toe holes on them to accommodate the 1" wind pressure,
so they barely whimpered. Its effect with the two 8's and all the difference tones produced was
undeniably grand! Tom would play on these stops while I finished getting dressed to go to
church; he raved about the sound and suggested I really should build a whole instrument with
these stops as its basis. The Quint 6' pipes also serve as the Aeolian Harp—the céleste!—on the
second manual. Being tuned pure as the 5 1/3' makes them slightly sharp for the céleste.

At first blush it might seem strange to have a harmonic flute as the parent rank for a céleste, but I
would not have the harmonic pipes start till 1' c, and the normal non-harmonic pipes of a
harmonic flute are really more salicionals than flutes. So, I think the two might actually work
well together in the range that is most important for the céleste effect.

Further contributions toward the lush, warm tone I believe essential to a pleasing-sounding
residence organ in its non-acoustic:
The multiple unison combinations possible--two complete 16' stops and three almost
complete open 8' stops
The "Serpentone Projector" which is a takeoff on the Baldwin "Choratone Projector" and
the Allen "Gyrophonic Projector", two devices for mushing up sterile electronic sound. I
envision a slowly revolving, wobbling 4' disk behind the pipes that imparts a very slow Doppler
effect, at least to the pipes enclosed in the swell box. (The "serpent" part comes from the fact I
was born under what some astrologers believe is the 13th sign of the zodiac—Serpentaires the
Snake Handler—describing those who get where they are going twisting and turning like a
snake!)
The pipes in the two pedal towers would be played by tubular pneumatic action that
would allow some flexibility of placement in cramped quarters. Rather than being "CC side" and
"CC# side" towers, each tower holds all twelve pipes of the Open 16' and of the Stopt 16',
respectively. Having two spatially separated sources of 16' tone would contribute to warmth
since they would not be physically close enough to pull each other into tune and into phase.
Trying to simplify the construction of the pedalboard as well as to reduce the overall height of
the organ, I propose to make the pedal keys as "broomsticks" like those of electronic spinet
organs. Do you think they would feel annoyingly dissimilar to the usual pedals considering that
they would be pivoted at the back of the organ instead of at the back of the organist as normal? I
am trying to devise a lever arrangement that would provide a "virtual" pivot in the normal
position on the key, but this would complicate the internal construction substantially. The pedal
keys are so low that they would thunk on the floor (on a rug, I imagine) when they are played!
The dimensions and placement relative to the manuals would be identical to those of Fisk's
Meyerson and SMU organs with their 32-note flat pedalboards.

Another simplification of construction involves the Swell box. I want to avoid the labor of
making and fitting up the usual sort of shade, so I am considering using an idea I saw in an old
Italian organ in which the regal was enclosed in a box with two disc-shaped dampers with pie-
shaped openings. You can just see this in the rendering. I wouldn't dream of doing this in a
"real" organ where you want maximum tonal egress when the box is open, but it might work just
fine for a residence organ.

Table Positive
 1. Open Diapason [Prestant] 8' Treble 31 pipes
2. Stopt Diapason 8' Treble
Stopt Diapason 8' Bass 55 pipes
3. Traverse Flute 4' Treble
Traverse Flute 4' Bass 55 pipes
4. Fifteenth 2' Treble
Fifteenth 2' Bass 55 pipes
196 pipes
Keyboard compass: CC-f3, 54 keys
Keyboard slides into case when not in use.
Transposable one-half step: A=440 or A=415
When A=440, the Bass/Treble break is between b0/c1 as in early English organs.
When A=415, the Bass/Treble break is between c1/c#1 as in early Spanish organs.
Electric blower and jelly bag-style bellows within the case
Winker beneath keyboard to enable a finger tremolo like the "bebung" of the clavichord
The whole to be of extremely light (balsa wood and cardboard) but sturdy construction to
allow transport by one person, alone
Pedestal table included
Width: 33"
Height: 42"
Depth: 18"
I submitted this design and a proposal to Edith Ho at the Advent, though nothing will come of it
there. This has come out larger in physical size than I would like, because I really want it to be
light enough and small enough to be carried in one hand and set on a table. I so dislike the idea
of the "trunk organ" that sits on the floor with its keyboard on top. When I was organist at St.
Paul's Episcopal here in Newburyport, I sent for Casavant's brochure of their continuo organ,
hoping I might interest parishioners in obtaining one for use in other areas of the church plant as
well as in the sanctuary. One woman took a single glance at the picture and the $9500 price tag
and said, "I wouldn't pay that much for something that doesn't even look like a pipe organ!" I
had to admit I entirely agreed with her. If I could give up the transposing option, I could use a
short bass octave and get the whole thing substantially smaller. My line of organs would also
include a table positive with subsemitone keys and 1/4-comma meantone tuning, externally
identical to the renderings. One thing I am counting on to enormously simplify construction is
the Stopt Bass idea of Haskell, something he took out a patent on but which I have never seen in
an organ. I have done some experiments in this regard which have been less than promising.
Again, as far as the specification is concerned, I want to avoid any screeching upperwork, hence
the open 8' from middle c rather than a 1 1/3' or a cymbal. Granted, the Traverse Flute is weird,
but it would have an open Haskell bass, and the bottom 24 or 25 pipes would be voiced so that
they combine so perfectly with the Stopt Diapason that they would sound just like the
continuation downward of the Open Diapason.

Organ-in-a-Book Organ

This is the project I should really get busy on since it is so small and therefore doable. The idea
is to provide everything needed to build a 37-note portative out of the paper, cardboard, and skin
of leather that could be bound together in book form. It needs to be simple enough, and more
particularly require little time enough in its construction, to enable a youngster to see his
instrument through to completion. I presently have the organ drawn with a tenor f 0 to f3
compass, acknowledging that this is the tessitura of the recorder consort and that it would allow
playing hymns, albeit at 4' pitch. [The scale of the pipes is taken from the reconstruction of the
Gothic organ depicted in the 1432 van Eyck altarpiece at Ghent.] Probably I should reduce the
compass to the 25 notes between middle c1 and c3, sufficient for Händel’s and Haydn's clock
pieces.
I want to get this to the post office, at long last, so I'll quit now. Hope you'll find things here of
interest!

Sincerely yours,

Steve

Amidst conversations with Robert when he was getting serious about commissioning a
practice/house organ for him and Gordon, I wrote:
April 05, 2003
Dear Robert,
Your call the other night got me thinking again about practice organs.
Just as an exercise, what single stop would come to mind if you were to have:
1. 61 flue pipes representing a Manual III Swell/Récit/Schwellwerk division
2. 61 flue pipes representing a Manual II Positive/Chaire/Positif/Rückpositiv division
3. 61 flue pipes representing a Manual I Great/Grand Orgue/Hauptwerk division
4. 58 flue pipes representing every Pedal division ever devised????
My off-the-top thoughts:
For manual III, enclosed in a box with my pie shutter idea, facing back towards the wall:
Gambe 8', complete to the bottom in mitered or Haskell pipes—representing Trompette or
Hautbois or 8+4+Tinkle Mixture or 8+4+2 harmonic flutes.
For Manual II, enclosed in a second box facing forward: Quintadena 8', representing
Cromorne 8' or Cornet Décomposé or Piquant RP 8+4+2+Mixture.
For Manual I, in the façade: Montre 8', representing Trompette or Grand Jeu or Harmonic
Flute solo or 8+4+2+Mixture.
For the Pedal, on the sides and maybe the treble in the façade: Double Dulciana 16', Haskell,
representing Bombarde 16' or Bourdon 16' or Flûte 8' or Choral Bass 4' or 16+8+4+Mixture.
Pull couplers would be III-II and II-I (III+II-I when both are on; no III-I only). I sense you
want to be able to have the Great on the middle manual sometimes—maybe a manual
transfer would be possible by moving Manual I (!?) Maybe no manual to pedal couplers.
I'm trying here to keep the one-pipe-to-a-key idea, for maximum independence of all musical
lines, avoiding tying the organ up in knots.
Steve
When I assumed John might be building Robert’s house organ, I wrote this: 

May 13-19, 2005

 
Dear John,
 
It is only by accident that I've been drawn into the present conversation.  At Fisk an email from
Dermadog/George was interqueued with a multi-page document I had sent to the printer—
couldn't help but notice it was something about a 3-manual practice organ and immediately
thought of Robert Poovey.
 
Robert and I have been tossing house organ ideas back and forth for years and years and years.  I
met him fall of 1977 while he was still a high school student, turning pages and pulling stops for
me for one of the dedication concerts I played for the Detlef Kleuker organ at Our Lady of
Grace, Greensboro.  My Peabody Conservatory organ teacher, Arthur Howes, had been
responsible for getting that instrument installed there.
 
Attached is an interesting file that Robert found on PIPEORG_L about small three manual
organs.  Sort of puts one in the necessary frame of mind to be able to consider this 3-manual
business!  A link to an article by the same author, Julian Rhodes,
is http://www.ondamar.demon.co.uk/essays/kilk.htm.
 
I have long thought that open pipe tone is most essential for clarity in a practice organ.  I
remember so well trying to practice the Brahms "Herzlich tut mich verlangen" (the heart-throb
setting) on the Hammerburg 8/8/P8--a chimney flute and two gedeckts--in one of the Peabody
practice rooms.  The effort was absolutely pointless.  Covered stops in the 8' and 4' octaves have
no pitch definition because the first upper partial, if it exists at all, is the twelfth, not the octave. 
At best, those pipes threaten to transpose everything up a fifth.
 
This goes double or triple for 16' stops.  You can't use a 16' bourdon alone. [I will barely admit I
was surprised that the Pedal 16' "Pommer," really a Quintadena, in the harp studio Beckerath at
SMU (Robert must know it) did sort of work alone--but otherwise, how awful!]  Charles Fisk
seemed aware more than any other neo-classical organbuilder the value of a quiet 16' Prestant in
the pedal (usually borrowed from the Great).  Because its first upper partial is the octave, the
rank can be used alone without anything coupled to it.  Being able to use every rank alone, not
dependent on anything else being added to it, is obviously of first importance in an absolute
minimum house/practice organ.  So, I shout in the dark to anybody and everybody to do
whatever is necessary to provide that tone.  Remember that the pedal piano had the pedalboard
permanently coupled at 16' pitch, viable because of the strong octave first upper partial and the
otherwise full and orderly harmonic development of those bass piano strings.  We need that same
piano-string-timbre tone in the 16', 8', and 4' octaves of every practice organ.
 
I probably am guilty of believing the Haskell bass to be the great panacea—the answer to every
problem.  I will in fact grant that it doesn't provide quite the same or quite as good a tone as that
of a true open pipe—it may be that the tierce or even the septième is unduly strong.  But at least
the first upper partial is the octave, not the twelfth.  So it seems to me still clearly superior to
the stopped basses organbuilders have employed for centuries to fill out, downward, their various
open stops.  Do keep as many true length pipes as possible, but don't resort to the classical
builders' only available alternative to pipes that are too long, the god damned gedeckt!
 
Here are some dimensions:
 
My Estey Labial Saxophone 16' is not maximally Haskellized, probably because it is being
forced so much to get the reedy tone.  But its scale might be good, i.e. CC 145mm x 182 inside,
182 x 218 outside.  The length is 150 + 2701 + 56 = 2907mm (114 1/2"). When I remove the
beard and blow it softly it makes a perfectly fine quiet Double Open Diapason tone.
 
According to William Haskell's patent applications, a pipe can be shortened a maximum of 10
semitones, a minor 7th.  So, count on taking the length of AA# for the body length to which must
be added the canister protrusion and the block height.  Fisk Opus 93, Niantic, has a fully
Haskellized bottom 6 notes for the Prestant 16'.  We used old pipes from 16' AA# and inserted
metal canisters.  The scale of the new CC was 197 x 228, wall thickness 1".  The body speaking
length was 2554mm to which add 97mm for canister protrusion and perhaps 150mm for block
height = 2801mm (110 1/4").
 
THE ESTEY MINUETTE!!!...
It is interesting to note that no outside dimension for either of the Estey Minuette versions, which
included a 16' Haskell Violone, was as much as 110 1/4" (9' 2 1/4"). The dimensions for the
Estey Minuette "Grand" were 3' 3" high by 5' 5" wide by 8' 4" deep, 3600 lbs. (And this was
tapered like a grand piano toward the rear!)  The dimensions for the Estey Minuette "Upright"
were 5' 7 1/2" high by 7' wide by 4' 8" deep, 3600 lbs.
There were 231 pipes, which I judge to have been distributed thus:
    Violone 16' 85 pipes
    Gedeckt 8' 85 pipes
    Diapason 4' 61 pipes
Among the 2 manuals and pedal the Violone was available at 16', 8', and 4', the Diapason at
8'(TC) and 4', and the Gedeckt at 8', 4', 2-2/3', 2', and 1-3/5'.  May I suggest that all we really
need to do today is update the Estey Minuette to play direct mechanically rather than direct
electrically?!  I so wish I had purchased one of the two Minuettes that had become available
over the years—for about $3000 each.  The only reason I wasn't interested was because I learned
they had electric action.  But restoring one of them would have given me all kinds of insight into
its problems and virtues.  Seems to me one could get those 3 Minuette ranks or preferred
equivalents on 3 manuals into the space Robert has available by embracing the Minuette concept.
 
The above photographs of the "Grand" Minuette in Vol. 1 of David Junchen's Encyclopedia of
the American Theater Organ clearly show a metal Haskell pipe that has also been mitered not far
above the mouth.  So, I guess it can be done.  I might point out that the Haskell set we still have
in storage at the Fisk shop that was deemed too large for your Rice instrument was the all-wood
version (and for a church organ Double Open Diapason to boot).  I'm quite certain the equivalent
wood-with-metal canister version would have a smaller footprint, and probably the all-metal
version would be smaller still.
 
Here are some of the Haskell patents, available for viewing and printing from the Patent Office:
 
The Haskell short-resonator length pipe patents:
965,896  - both the all-wood and all-metal versions of open voices
967,911 - a Haskellized stopped pipe!  [I did some experiments with this idea but could get only
the faintest sound out of it.  Still, I want to believe!]
971,502 - Haskellized conical reed
 
The Labial Reed patents:
871,272 - labial oboe
965,897  - what became the labial clarinet
965,898 - another kind of labial clarinet? or, improved labial oboe
1,327,996 - labial tuba
 
Patent Office link:
http://www.uspto.gov/
 
This is the TIFF image viewer you need to view the patents:
http://www.alternatiff.com/
 
As for free reeds!...
 
I have long pondered the possibility of using free reeds to save space in continuo and house
organs.  If free reeds were the panacea, we would all be perfectly satisfied with the two and even
three manual Estey reed organs that appeared over the years.  What's the problem?:  the slow,
oozing, lugubrious speech innate to the devices.  (Ever notice how a recording of a piano played
backwards sounds exactly like a harmonium? —reverse percussion, reverse chiff!)  Summer of
'67 when I was tuning organs in the D. C. area with Lewis & Hitchcock, I played one of the
Wicks with free reed 16' bass octave like what Robert mentioned.  My impression was that the
free reeds would have been more successful if they had been carried through the whole pedal
compass.  The 12 notes did not at all match the flue stop they were supposed to.  A similar sort
of thing that was even less successful was Mel Butler's "Kilgen Petit Ensemble" which had a
beating reed Vox Humana 16' 12 pipes to serve as the bass to the Bourdon!  Again, if the Vox
had just been carried through the whole pedal compass, it would have been slightly more
acceptable.  But those 20 extra beating reed pipes would have been expensive.
 
The one possible salvation for the free reed is the percussion stop idea found on some French
harmoniums.  The only one I have ever experienced was in the chapel on a big estate, a country
home, in King's Lynn, G. B.  I was startled, totally amazed!  With the "percussion" stop engaged
the reeds spoke instantly!  ...nothing more to be desired.  Tiny piano-like hammers with an
escapement action struck the reeds to get them going just as the key also opened the pallet.  Ever
since, I've imagined making little pneumatics with hammers that would play just the bottom 12
notes of a continuo organ 8' Gedackt.
 
In addition to the poor speech another disadvantage of combining free reeds with flue pipes is
that they do not change pitch with the flues as the temperature changes.  Perhaps in our present-
day (theoretically) climate-controlled homes this temperature question is not a problem (though
the typical residential thermostat doesn't expect to keep the temperature steadier than + or - one
degree Fahrenheit--a two degree variation.)  The one experiment I've done involved attaching a
2' cylindrical resonator to an American Organ (suction) reed playing 4' C.  The resonator not
only promoted a strong fundamental; it also allowed the reed to be tuned by moving the tuning
slide on the resonator up and down.  The resonator was so influential that it just pulled the reed
tongue along.  This probably would help it follow the flues with temperature changes.  The idea
came from the free reed clarinets in Aeolian organs I've seen.  So, perhaps the 16' octave of the
practice organ could be a free reed Bass Clarinet--the 8' long cylindrical resonators mitered as
necessary.  A Bass Clarinet would be more appropriate for the bottom of a Gedeckt; the best
solution for an open flue stop would be a mitered 16' long conical resonator—so we have all the
harmonics, not just the odd-numbered ones.
 
I think it would be worthwhile to check out first hand Sebastian Gluck's free reed bass with
resonating cavity—to hear how successful it actually is.  Perhaps he's done the R & D to make it
work really well and would be willing to share the information.
 
"The Beehive Reed Organ Studios", where you can almost certainly purchase sets of harmonium
reeds:
http://alfred-beehive.org/index.cfm.  See especially: 
http://alfred-beehive.org/ReplacementReeds.cfm.
  
Considering the question in a practice organ of pluck, the feel of the keys, the tactile aesthetic—
the importance of which it seems you and Robert agree with me upon...
 
Ludwig Hoffmann in ISO Information, June 1986, has attempted to quantify pluck versus pallet
spring tension precisely.  The thing to aim for is a 1:1 ratio between pluck and hold down, that is
to say, that the pluck provide half the resistance and the spring provide half the resistance at the
top of the key travel.  The absolute values are 120-160 grams at the top and 60-80 grams to hold
the key down.  Another clue is E. M. Skinner's take on what was necessary, even in an
electric keyboard, for a user-friendly, ergonomic feel.  He suggests a pressure of 3 1/2 oz. (99.2
gr.) to cause the key to fall, and 1 1/2 oz. (42.5 gr.) to hold it down.  Willis suggests 4 oz. (113.4
gr.) at the top, 2 3/4 oz. (78 gr.) at the bottom.  I think one really should grant each key its own
pallet so the feel can be optimized for each.  Three keyboards x 61 keys = 183 pallets.  Ideally,
for maximum musical independence each of those 183 keys should also have it’s own pipe. 
Perhaps, yet also, adjustable bleed holes for all the pallets should be provided, their wasted wind
directed back to the blower intake.  I was hoping one could put all 183 (+12 for the pedal
extension) pallets-side-by side across the back of the organ, but even at manual keyscale it would
require more than 3 x 33" = 99" (8' 3"), and certainly the pallets need to be wider than that for
pluck's sake.
 
 This is what I put forward as a minimal 3-manual practice organ for your consideration at this
moment:
 
III:  Violin Diapason 8' 61 pipes, CC-BB Haskellized in Estey's traditional way, rest of rank
slotted
II:  Harmonic Flute 8' 49 pipes, CC-BB common with Violin Diapason, Co-E1 natural length,
not slotted, F1-C4 harmonic. The Violin Diapason, or else both the Violin Diapason and the
Harmonic Flute, are enclosed (4' Swell box).
I:  Spire Flute 8' 61 pipes, the Fisk Opus 76/76a prestant 8', (the one about which Joan
Lippincott, when she was helping Charles Fisk finish the Westminster Choir College
instrument, said something like it's the only set of pipes you'd ever need for keyboard practice. 
This 5/6 taper, 2/9-mouth stop is really more a quiet Diapason, the virtue of its taper being its
poppy, precise speech.  Our often-used Pedal "Octave" or "Baarpijp" 8' is the same idea—really
a principal, not a flute, 5/6 taper to promote quicker speech—for trio sonatas and such.)  The
Spire Flute is the Prestant, maybe even should be called that instead, chromatic in the treble so as
not to jump around distractingly, voiced to be in balance for the organist even though it be too
soft several feet away from the organ.  The 8' octave would not be Haskell, rather, be exactly
what you did for the [Doug and Helen] Reed's instrument [tapered, mitered wooden open pipes].
Pedal:  Double Spire Flute 16' (!?) 12 pipes, the continuation downward of the I: Spire Flute (not
tapered, though!!).  The Pedal plays this extended rank at 16', 8', and 4'.  Being prestant, the 4'
range would let a cantus sing out presently and nicely (for the organist)—the major point of a
pedal 4'.
 
The total number of pipes is 183.
 
It ought to be possible for the sake of warmth and acoustical expansiveness, as you have implied,
to combine at least two of the manuals, i.e., two of the 8' stops.  Maybe III-II and II-I—or just
III-I, considering how effectively the slotted Violin Diapason would come and go against the
Spire Flute.  I think being able to play an unenclosed rank off against an enclosed one is a
valuable extra for bothering to have a swell enclosure at all.  But I absolutely do appreciate that
an entirely swell-enclosed organ accomplishes most everything that could otherwise be desired. 
(Shouldn't there be some much cheaper way than the traditional one of making a Swell box, at
least for a house organ?)  Probably there should be III-Pedal and II-Pedal couplers.  Have you
heard the hullabaloo at Fisk caused by Opus 123's Richard Hoskins?  We now think we should
make our couplers so they will not couple through!  This does allow the organist more flexibility,
especially in a small organ.  I was aware of this problem and its possible solutions long before I
came to Fisk, but I could never imagine daring to bring up the subject here myself!
 
John, I've lost sleep over trying to prepare this email; I've got to quit and get it off to you!  For
most of the years I've been pondering the problems of house organs I've only thought in terms of
2-manual instruments.  It is only this past year that Robert has prodded me into thinking
THREE!  So, now my house organ that I'll never build will have three manuals, also!  I still sort
of think it's a bit beyond the pale, but I certainly am fascinated with Julian Rhodes' examination
of the idea, and I do think the minimal 3-manual practice organ notion is eminently worth
pursuing.

Sincerely yours,

Stephen Paul

And what Robert has actually contracted with me to build for him at the Fisk shop:
SPECIFICATION FOR A THREE-MANUAL AND PEDAL PIPE ORGAN

For the residence of Dr. Robert L. Poovey of Rochester, New York

By Stephen Paul Kowalyshyn

GREAT (Manual I) SWELL (Manual III ─ expressive)

1. Flauto Traverso 16’ 11. Violin Diapason 8’ 73 pipes

2. Prestant 8’ 61 pipes 12. Céleste (Serpentone Projector II)
3. Quint 6’
4. Violin Diapason 4’
5. Swell to Great PEDAL
6. Positive to Great
7. Great to Great 8ves 13. Flauto Traverso 16’ 12 pipes
14. Prestant 8’
POSITIVE (Manual II ─ expressive) 15. Prestant 4’
16. Prestant 2’
8. Flauto Traverso 8’ 61 pipes 17. Great to Pedal
9. Céleste (Serpentone Projector I) 18. Positive to Pedal
10. Swell to Positive 19. Swell to Pedal
207 pipes
20. General Tremulant
All pipes played by direct mechanical action. 21. Great/Positive Transfer
I hope these materials will prove to be of some interest to you.

Sincerely yours,

Steve Kowalyshyn