Pipe-Organ Reservoirs, Regulators, and Schwimmers Explained

The regulator function and the reservoir function are usually carried out by the same folded unit, but they are separate jobs that come into play at different moments. The Garden State Theatre Organ Society draws the line this way: the regulator function holds steady pressure from the blower before wind enters the chests, while the reservoir function comes into play when many pipes play at once and demand is heavy. The distinction is functional rather than physical. One folded reservoir does both, which is why a unit proportioned only for steady-state pressure can be short of stored volume when a sudden chord draws far more wind than the moment before.

The balance between the two functions has shifted over time, at least in emphasis. Writing in 1957 from a theatre-organ standpoint, Roy Gorish characterized the modern organ reservoir as not much more than a regulator, in which the static or raw air from the blower is steadied and regulated, and contrasted it with the earlier accordion-folded reservoirs weighted with lead that both stored and regulated wind. That sits in some tension with the Garden State description of one unit doing both jobs, but the difference is one of emphasis rather than of fact.

The same 1957 account in The Tibia went on to locate unsteady wind, as distinct from insufficient wind, primarily in the regulating stage. That placement is a period view rather than a statement of present-day design practice, and unsteady wind can also arise elsewhere in the system, so it describes where regulation happens rather than fixing where trouble must originate.

The regulating apparatus exists to keep the working pressure constant while the volume of wind being drawn changes. James H. Cook, in an educational statement published by the Organ Historical Society, puts the reason plainly: without some means of regulating the air pressure so that it remains constant, playing a large number of pipes simultaneously would deplete the air in a windchest and cause a drop in pressure. Every speaking pipe draws wind, and a chest holds a finite volume at any instant, so a full chord draws down the supply behind the pipes unless something replaces what is taken at the same pressure. Cook states the principle only; how much pressure would drop on a particular instrument, or how quickly, is not something the principle settles.

That is why regulation and wind demand have to be kept apart in thought. Demand is a volume question, the quantity of wind the playing instrument is drawing; pressure is a force question, the push behind each pipe. The regulator’s task is to let the first vary widely while holding the second nearly still. Steadiness of that kind is deliberate and continuous, which makes it the opposite of a tremulant: a tremulant is a controlled disturbance introduced into the same wind on purpose, and it is explained on its own page on the pipe-organ tremulant.

The regulating valve is attached to the moving top of the reservoir, so the position of the top itself decides how far the inlet is open. The AIO guidelines for new organbuilding work direct that each reservoir be provided with a valve of appropriate type, such as curtain, poppet or pallet, attached to the moving top to regulate the flow of air from the main supply trunk. That attachment is the essential regulating link. No separate gauge or sensor is needed, because the reservoir’s own height is the sensing element and the valve simply follows it.

The principle is the same whichever inlet valve a builder uses. Johan Liljencrants describes it as common to curtain, cone, poppet and vane inlet valves, and his description is of the principle only; it does not describe any one builder’s linkage geometry, and the way a particular shop connects top to valve is not something a general account can supply.

A regulator holds pressure by letting its own movement open and close its supply, which is a closed loop. In Liljencrants’s account, the sensing bellows collapses somewhat when output pressure tends to be too low, opening the valve disk so that flow from the unregulated input increases; as the bellows refills and rises, the valve closes again. The reservoir therefore keeps finding the height at which inflow matches outflow, and that height corresponds to the working pressure it was built to hold.

What sets that pressure can be written down. For a spring-loaded regulator Liljencrants gives the regulated output pressure as P = F/(A – B), where F is the spring force, A the effective bellows area and B the area of the valve hole. In words, regulated pressure is spring force divided by net area, fixed by the spring and the top rather than by anything the blower is doing. He adds that a balancing bellows on the upstream side, whose effective area is about the same as the valve hole area, prevents input pressure from affecting output; without that balance the output tends to track the input. The relation was derived for his own regulator design, published in the Mechanical Music Digest technical pages in 2001, and it explains the behavior rather than prescribing any dimension for an organ.

The loop has a resting state. John Bishop describes it: when the reservoir filled, the regulating valve closed, stopping the flow of air from the blower. With the organ silent and the reservoir full, the blower runs against a closed or nearly closed inlet, which is the condition under which it develops its static pressure rather than its working pressure, and it is why the two figures for the same instrument differ. A full reservoir closes its own inlet; a reservoir being drawn down opens it.

A reservoir’s working pressure comes from weights, springs or a combination of the two pressing on its moving top. The AIO guidelines contemplate reservoirs of the folded or diaphragm (schwimmer) type fitted with weights, springs or a combination to maintain a stable working pressure, and they treat electric blowing as the normal wind source with manual feeders as an optional adjunct. Those guidelines address new work and do not purport to govern the restoration of historic instruments.

Published catalog material gives a concrete sense of scale. In its 2006 catalog Organ Supply Industries offered standard reservoirs from 18 by 24 inches at 57 pounds up to 48 by 60 inches at 325 pounds, with built-in curtain valves mounted on nylon bearings. The same section listed coil springs from number 4 to number 40, in wire diameters from 0.120 to 0.192 inches, and compass springs from number 6 to number 30, in thicknesses from 0.042 to 0.080 inches, with springs included for a specified wind pressure. That spring-per-pressure listing shows something basic about sprung reservoirs: the pressure is set at manufacture by selecting the spring against the area of the top. The figures establish what one supplier offered in 2006; they do not establish what any particular organ contains, and current catalog contents would have to be re-verified.

Weights and springs give the moving top very different amounts of mass, and that mass governs how the reservoir moves. A. Taylor’s research at the University of Salford, published in 2019, characterizes the organ reservoir as a mass-spring system, with the mechanical mass of a weighted reservoir exceeding 100 kg and dominating its behavior, while a sprung reservoir uses approximately 25 percent of that mass. Those figures are characteristic magnitudes within the author’s test framework rather than universal values.

The same research reports a way in which the blower itself, not only the playing, can unsettle the wind: under certain conditions the blower excites the reservoir at its natural resonant frequency with sufficient amplitude to cause unwanted amplitude modulation on a sounding organ pipe. Taylor’s survey of 83 instruments in the United Kingdom found that 57 percent had organ pipe flutter, a figure that belongs to that survey alone and is not a general prevalence.

One consequence follows from the two findings together, although as an inference rather than a published result. Because a weighted reservoir carries far more moving mass than a sprung one, the two types have different natural frequencies and therefore different susceptibility to being driven into resonance by blower blade-passing excitation. The published abstract states the mass difference and the resonance mechanism but does not publish comparative natural frequencies for the two types, so the inference says only that the choice changes the behavior, not which type is the safer one. None of this diagnoses any instrument; it explains why two reservoirs holding the same working pressure can respond differently to the same blower.

A schwimmer is a pressure regulator mounted on the chest rather than on the wind line. Wicks Organ Company’s glossary defines it as a pressure regulator usually attached to the bottom of a chest, whose movable lid moves inward under spring tension as air is drawn from the chest and thereby opens an inlet valve admitting air from the blower. That is a builder’s own definition of the component and does not establish how the schwimmer on any particular instrument is built.

Suppliers describe it in the same functional terms. Organ Supply Industries describes a schwimmer as a chest-mounted regulator that takes the place of a reservoir, quoting in its 2006 catalog a 24-inch minimum width and 57-inch standard length, with 4 inches of clearance required inside and 8 inches outside, and suggesting schwimmers for wind pressures over 5 inches. The clearance figures are a practical constraint on fitting a schwimmer into an existing chest layout, and the parameters are one supplier’s own and would need re-verifying against a current catalog.

A schwimmer differs from a conventional reservoir in where it sits and which way it moves. It is attached directly to the underside of a windchest rather than sitting on the wind line between blower and chest, and its movement is downward instead of upward. That description covers geometry and direction only; it does not compare regulation quality between the two arrangements.

The consequence of that position is that the chest does the storing. Parkey OrganBuilders describes schwimmer lids as wind-regulating devices mounted directly to the bottom of the chest that use the windchest itself as the reservoir. Regulation can also be built into the chest as manufactured: Organ Supply Industries cataloged slider chests with built-in schwimmers using pantograph springs, and unit chests with a schwimmer bottom. That is a supplier’s own description of its products and says nothing about how common integral schwimmers are among installed instruments.

The argument made for the arrangement is one of response. The specification of United States Patent 3,209,636 asserts that mounting the regulator directly on the chest eliminates the resistance and impedance of the relatively large diameter tubing that earlier separately mounted regulators required, and it claims recovery of a pressure cut to the level of 0.03 second. That is the applicant’s own performance assertion, stated relative to unspecified prior designs and not an independent measurement.

A regulator sets the working pressure, a reservoir stores a working volume, and the small devices near the chests do neither; they damp and compensate locally. Concussion bellows, known in the trade as winkers, are described by James H. Cook as small wedge-shaped bellows that respond to pressure changes, and Organ Supply Industries catalogs them as shock absorbers for wind systems in an 8 by 16 inch type, a 12 by 18 inch type and a tunable variant. The tunable variant implies that the response can be adjusted, but neither source states what parameter is adjusted or what frequency range it addresses.

Two accounts, a century apart, describe what a winker does. Roy Gorish, writing in 1957, states that concussion bellows help eliminate wind-line rumble conveyed through wind trunks, which identifies the role as damping transients traveling in the trunk rather than setting steady pressure. Grove’s first edition credits Bishop with a concussion bellows using springs to compensate automatically for fluctuations in demand and so maintain steady pressure near the pipes, which shows that the device was understood in the nineteenth century as local compensation placed close to the point of use; a priority claim of that kind is not settled by one reference work. Either way, a winker smooths what the regulator has already set and does not set the working pressure itself.

The word winker carries a second meaning in theatre-organ practice that must not be confused with the first. As recorded by the Garden State Theatre Organ Society, a winker or equalizer in that world is a small regulator usually used for tuned percussions such as xylophone and chimes, or for air-driven swell shade motors, rather than for pipes, and a dice box is a small regulator without a reservoir placed between the regulator and the chest. In classical-organ usage a winker is the concussion bellows on the wind line; in theatre-organ usage it is a small regulator, and the two senses name different devices.

Regulation is also commonly staged. Organ Piper Pizza describes its own instrument as sending wind from the blower through large ducts to regulators near the chamber floor that reduce static pressure to what the pipes require, with larger regulators feeding several ranks and smaller ones generally a single rank. One instrument cannot stand for all, but it documents the general pattern: a single static supply is stepped down to several working pressures by separate regulators, some of which serve one rank only. Where that wind comes from upstream and how it is carried onward belong to the rest of the wind system.

Practitioners do not hold one view of how much the working pressure should be allowed to move while the organ is played, and this page records the positions rather than choosing among them. The AIO guidelines for new work require that there be no adverse pressure drop in the course of normal playing when additional keys are depressed on the same manual, and the reason given is that one must be able to tune in a satisfactory manner. The rationale is tuning practicality rather than musical effect, and the guideline sets no numeric limit on acceptable drop.

Charles Fisk argued the case for movement on musical grounds. He held that wind pulses contribute to clarity in counterpoint by making the motion of inner voices audible through sustained notes, and that at the Steinkirchen organ the entire effect of the organ is colored by its wind supply. That is an aesthetic argument by a named organbuilder, not a measured or perceptual study.

Parkey OrganBuilders states a position between the two. The firm holds that a slight flex in pressure is desirable for an artistic touch, but that huge dips or wind pressure drops under heavy demand are not desirable, and that wind flex should not create bouncy, unsteady or nervous undulations resulting in unstable tone. That is a builder’s stated design philosophy, expressed without numbers, and the three positions are left standing side by side.

Four named parties have recorded different judgments of how a schwimmer-winded organ sounds, and they are set out here as they stand. Wicks Organ Company states that when properly designed with a large lid and appropriate springs, the schwimmer produces the most stable wind pressure, which is a company’s claim about a design it uses.

Charles Fisk, in a 1969 essay arguing for flexible winding, criticized the schwimmer regulator on the ground that it precludes any unsteadiness of wind except for vibrations in the range of the vibrations produced by the pipes themselves. The criticism is an aesthetic argument rather than a measurement, and the quality one party offers as the schwimmer’s merit is the quality the other names as its fault.

Parkey OrganBuilders, having described schwimmer lids as using the windchest itself as the reservoir, states that they can sometimes seem sterile in the management of wind supply and pressure, which is one firm’s aesthetic judgment stated on its own website. Alan Jackson, writing in the Journal of American Organbuilding, states that he personally likes the sound of schwimmer-winded organs, comparing them to a singer with good breath control and pressure, and presents that as an explicitly personal preference. The four positions remain four; this page neither ranks them nor turns them into a recommendation.

When more than a few pipes are played and both pitch and pressure fall away, the storage-and-regulation stage is one candidate among several, and the symptom itself settles nothing about which condition is present. The documented possibilities include schwimmer maladjustment or table drift, an undersized or obstructed supply, and leakage, and no cited source treats the symptom as proof of any single cause. Those conditions can sit at different places in the system, and not all of them lie in the storage-and-regulation stage. Telling them apart is measurement work carried out on the instrument by a qualified technician, and nothing about the symptom alone says which is present or which is more likely.