Because the Ampico spinet piano was only 36.5" tall, the American
Piano Company (APC) was faced with severely limited space within which
to incorporate a conventional box-type vacuum pump (16" X 16" X 8"
without distributor block).  Consequently, APC utilized a smaller pump
(15" X 9" [max, min 3.5"] X 8" without 12" diameter offset sheave)
which became known as a 'grasshopper' pump.  This followed because
the pump somewhat resembled the overall shape of the jumping insect.

Production of the Ampico spinet with J. & C. Fischer, Steck, and
Marshall & Wendell brand names (maybe others) appears to span only the
years from 1937 through at least 1938.  Does anyone know how many of
these Ampico spinets were produced by the American Piano Company and/or
any other piano brands in which the grasshopper pump was installed?
When was the first one introduced and when was the last one sold?

Curiously, the so-called grasshopper pump itself, which is commonly
associated with the Ampico spinet, was in production well before 1937.
Moreover, it was incorporated into certain smaller-size Ampico grand
pianos.  These instruments (presumably less costly than larger grands)
included a modified version of the Ampico B drawer and some other
related B components, but clearly were implemented to function with
pressure-regulating Ampico A expression devices.

Because the construction of the reproducing mechanism was a blend of
components from both Ampico A and B systems, it is sometimes denoted
by Ampico A/B.  This author recently acquired all the components
(thankfully original and untouched) from such an instrument.  The
piano was a derelict and unrestored 4'6" Haines Bros. Ampico grand
(serial number 78566) manufactured in 1931.

The motivation to acquire these components was to obtain a 'spare'
grasshopper pump, with which to compare faintly audible emanations from
the marginally 'thumpy' grasshopper pump in my J. & C. Fischer Ampico
spinet (serial number 171908) from 1937.  I have not yet progressed to
a rebuilding point where a direct comparison can be made, but at least
the raw materials are now at hand.

Orphan grasshopper pumps do not 'show up' very often, especially in
original, unrestored condition.  The spinet grasshopper and the grand
grasshopper differ only slightly, with the 1+1/8" spinet vacuum port at
the very rear of the pump (straight nipple) and the grand vacuum port
on the side (toward the rear) with a right-angle elbow.

After removing the Ampico A/B components from the small Haines Bros.
piano, it was only natural to examine the intricacies of the relatively
uncommon design of the player system.  The intent of this article is to
briefly describe some of the unusual characteristics of the Ampico A/B
implementation.  Though this description reflects only one example of a
small grand piano in which a grasshopper pump was originally installed,
it should help Ampico piano aficionados recognize similarly constructed
instruments in the field.

The characteristics listed below reflect many (probably not all)
of the differences between the Ampico A/B design and the ordinary
Ampico A and Ampico B grands that were installed in pianos generally
larger than 4'6".

1. Grasshopper pump mounted on a carefully configured vertical support
board parallel to the fall board and directly under the soundboard.
Thin felt (3/16" thick) tacked on the support board beneath the pump,
ostensibly used to muffle the external flap valves from only two of the
four lobes of the pump.

2. Ampico B drawer with four plastic radio control knobs
(Stromberg-Carlson?) replacing the usual fancy cast brass indicators
found in normal Ampico B drawers.

3. Ampico B reservoir assembly with integral spill valve consisting of
box-covered (along with 3' long coiled silencing tube 3/8" diameter)
spring-loaded trap door positioned on reservoir movable rear board.
This valve is activated by a long screw adjusted from the front
stationary distributor block board when the reservoir collapses
sufficiently, e.g., during the rewind operation.  Only one compression
spring (wishbone-type) is used to store energy in the reservoir.

4. Ampico B valves in the A stack (with constrained in-line poppet rod
construction with wooden guide rails).

5. Stack vacuum feed lines (4) connect to 5/8" diameter nipples each
held in place with internally positioned single piercing rods
(through-the-deck nails?) ostensibly to prevent unintended pullout of
the nipples from stack decks.  The presence of these small diameter
retainers is not a significant air flow blocker.  Has anyone ever
noticed this fabrication refinement?

6. Pump flap valves 'backed up' with rubber striker pneumatic cloth
to prevent stretching of valve leather under tensile load from
ever-present tensioning springs.  This was not done with the pump in
the Ampico spinet (only one example).

7. Stack vacuum supply (from step intensity unit) elbows consisted of
a pair of 5/8" diameter side-by-side elbows soldered to a common base
plate.  The plate was mounted to the intensity regulator with three
screws instead of the usual four (when 1+1/8" diameter tubing was
used).

8. No 5/32" diameter elbows were used to connect the tubing from the
rear of the B-drawer to the 83-note stack.  Tracker bar tubing was
bundled in bunches (of 10 tubes or so) and loosely draped between
drawer and stack.  The underside of the keybed was thinner than normal
(lightly built) so as to provide enough room to accommodate the
associated tubing bulges.

9. The valve controlling the soft pedal pneumatic was mounted on
the movable board of the pneumatic.  The sustain pedal pneumatic was
conventional with its controlling valve not mounted on the pneumatic.

10. Bare-bones construction of bass and treble step intensity units.

11. Only one crescendo pneumatic (A-type) was used to simultaneously
activate both the bass and treble step intensity units (also A-type).

The above list is probably not complete because other structural
differences were not apparent from only an exterior examination of
the Ampico A/B components.  Those MMD readers who have more extensive
experience with the rebuilding of the Ampico A/B system in general,
and the grasshopper pump in particular are welcome to contribute their
observations.

Bill Koenigsberg