Appl. sci. Res. Section B, Vol.

ANALOGY OF HYDRAULIC, MECHANICAL,

ACOUSTIC AND ELECTRIC SYSTEMS
bij J. c. SCHONFELD
Rijkswaterstaat, Central Research l)ivisiou, The Hague

Summary
A clearer conception of the analogies is pursued by t r e a t m e n t of hydraulic
systems as distinct from mechanical systems, interpretation of acoustic as
mixed mechanic-hydraulic systems, analysis of the operation of piston
mechanisms, and consideration of gyro-elements.

§ 1. Introduction. Many mechanical and electric systems show

analogous properties, as has been demonstrated clearly in particular
for systems of solid bodies or compressible fluids 1), by treating such
systems by the impedance concept 2) 3). However, systems in which
an incompressible fluid (e.g. water) circulates have never, as far as
we know, received a similar systematic treatment on this basis, al-
though the analogy of such systems with electric systems has often
been recognized and even forms a well-know,ha didactic means to
explain the properties of a flow of electricity.
By considering the effects of gravity on an incompressible fluid,
which in fact form an essential feature of many hydraulic problems,
we can arrive at a more complete conception of fluid-transmitting
systems than when the elasticity of the fluid is considered, but the
effects of gravity are disregarded. Such fluid-transmitting systems
we shall call hydraulic systems.
The hydraulic systems appear as distinct from what we shall call
mechanical systems. The most outstanding difference lies in the oppo-
site character of fluid and rigid joints. It is basic for the dinstinction
between series and parallel coupling of elements and also decisive
for the analogy with electric systems. H e c h t's 1) conception must
be confined, we think, to systems with rigid joints ("mechanical"
systems).

--417--
418 J.C. SCH6NFELD

Hydraulic and mechanical systems m a y be coupled b y means of

piston mechanisms, and an analysis of such mechanisms further
elucidates both the analogy and the difference in character of h y d r a u -
lic mM mechanical systems. Acoustic systems m a y be considered in a
dualistic way as partly hydraulic, p a r t l y mechanical systems.
The analogies are of practical importance in the first place for
didactic purposes, in the second place for the establishment of a n
elegant, clear and relatively simple t r e a t m e n t of compound systems,
such as electro-acoustic transducers, and in the third place in the
form of models, such as electric models of hydraulic systems for the
investigation of tidal p h e n o m e n a ~), groundwaterflow ~) or water-
h a m m e r ~1).

List of symbols.
.q cross-sectional area Q hydraulic discharge (current)
B hydraulic capacitance Qe electric charge
C electric capacitance R hydraulic resistance
c mechanical compliance Rc electric resistance
D hydraulic deliverance r mechanical resistance
E energy S area of watersurface
F force Sk hydraulic gyro-susceptance
G electric conductance Se electric gyro-susceptance
g gradient of gravity s mechanical gyro-susceptance
~,,,j mechanical glidance t time
h height watersurfaceabove datum U electric potential
H total head above datum 1.' volume
I electric current v velocity
L electric inductance II: power
31 hydraulic inertance X hydraulic gyro-reactance
m mass; mechanical inertance Xe electric gyro-reactance
P hydraulic potential .r coordinate
p pressure a',,, nlechanicat gyro-reactance

§ 2. Hydraulic systems. By hydraulic systems we shall for the

m o m e n t u n d e r s t a n d systems in which an incompressible fluid (e.g.
water) circulates between rigid walls.
R e s i s t a n c e. Consider l a m i n a r flow in a tube (fig. 1). Let Q
be the dischaJ~e (currant, volume velocity) through the tube, i.e. the
volume of liquid t r a n s m i t t e d per u n i t time. Let H denote the total
head at some point, that is the sum of the geometric height z of the
point above some zero level (see fig. 1), the pressure head p/~g
(~o = density) and the velocity head v2/2g.
 ANALOGY OF HYDRAULIC AND OTHER SYSTEMS 419

H e a d is t h e q u a n t i t y c o m m o n l y used in hydraulics, a n d therefore

it is most p r a c t i c a l in applications to define h y d r a u l i c impedances
as, relations between head a n d discharge. Presently, however, we
intend to give a more general physical discussion of h y d r a u l i c
systems, a n d then it is preferable to use the q u a n t i t y

P = og H = eg: + P + -1.-'v2,
_,
which we shall call the hydraulic potential.
F r i c t i o n in the conduit of fig. I causes a loss of p o t e n t i a l

A P = og A H = RQ,
where R = 8o31/~a 4, and R = 3[n,l/4ab a for a round tube a n d a
crevice respectively (v = kinematic viscosity). We shall call the
factor R the hydraulic resistance (for r e c t a n g u l a r and elliptic cross-
sections cf. ")).

...... zero - t e v e t

Fig, 1. Hydraulic circuit of a pump and a tube with resistance.

A p o t e n t i a l difference A P t h r o u g h a b o d y of porous material (e.g.

soil) Obeying D a r c v ' s law produces a discharge
() = D A P ,
where the factor D = kA/ogl (k = permeabili@ of the material,
I ---- length a n d A = cross-sectional area of tim body) will be called
the hydraMic deliverance; this is the reciprocal of the resistance.
W h e n we use heads i n s t e a d of potentials, the formulae for tile
resistances m u s t be divided b y og and the fornmlae for the deliver-
ances m u s t be multiplied b y ~g.
Most other h y d r a u l i c resistors are non-linear and often q u a d r a t i c
in c h a r a c t e r 7). Their c o m m o n p r o p e r t y is t h a t t h e y obey a functional
relation between the discharge a n d the p o t e n t i a l difference required
to m a i n t a i n t h a t discharge. Ill the present discussion it is i r r e l e v a n t
420 j . c . SCHONFELD

whether the functional relation is linear or not; for b r e v i t y we con-

fine ourselves to linear systems.
The energy dissipated per unit time in a linear h y d r a u l i c resistor
a m o u n t s to tV = Q A P = RO 2 = D(AP) 2.
I n e r t i a. Consider a variable discharge through a tube or an
open conduit. Ignoring friction for the moment, a p o t e n t i a l differ-
ence will be required to accelerate or to decelerate the flow,
a m o u n t i n g to
A P -= M dQ/dt, where M ---- ol/A ---- m/.42.
This factor M will be called the hydra.ulic i~erla.lzce of the conduit.
As there is also friction, the p o t e n t i a l difference is in fact a func-
tion of the discharge Q a n d its t e m p o r a l variations. In case of a
prevailing or at least not subordinate resistance we m a y p u t
A P = RQ + MdQ/dt, where M now must be given a g r e a t e r value
t h a n ~I/A (cf. 6)). W h e n heads instead of potentials are used, the
formulae for the i n t e r t a n c e m u s t be divided b y t)g.
Both R and M are directly p r o p o r t i o n a l to the length I of the
c o n d u i t ; M is inversely p r o p o r t i o n a l to the cross-sectional area A,
and R s u b s t a n t i a l l y inversely p r o p o r t i o n a l to .42. A nearly pure
resistance of a definite value can therefore be oLtained in principle
b y m a k i n g A 2 and 1 small in a definite n m t u a t ratio, and a more or
less pure inertance can be o b t a i n e d b y making A and 1 large in a
definite m u t u a l ratio.
A h y d r a u l i c intertance conserves the kinetic energy Ek = '.,~MQ2.
Storage. W h e n the potential P a t the entrance of the open
vessel of fig. 2a increases, the liquid in the vessel will rise so t h a t a
volume A V = SAh must enter the vessel. Neglecting inertia and
friction and ignoring the barometric pressure (the l a t t e r is equiva-
lent to a d o p t i n g a n o t h e r zero level) we have P = egh and hence
A V -~ B AP, where B : S/~g is the storage or hydraulic capacita~ce
of the vessel. We can deduce
Q -- B dP/dt

for the discharge entering. W h e n using heads instead of p o t e n t i a l s

we m a y identify the storage with the surface area of the liquid.
Now consider two non-mixing liquids with different densities
(o > e') in a b i - t e r m i n a l system like t h a t of fig. 2b. Let h be the
height of the interface between the liquids a n d p the pressure in the
 A N A L O G Y Oi: H Y D R A U L I C A N D O T H E R S Y S T E M S ~t2l

interface, then the potentials of the two liquids are P = p + egh

and P' = p + (/gh and hence P - - P' = (o~ --- o~') gh, if the velocities
of the liquids are zero or at least negligibly small.
W h e n a volume A V of the lower liquid enters at the left, an equal
volume A V of the upper liquid is expelled at the right, and the inter-
face is raised Ah -= A V/S, where S is the area of the interface. This
results in an increase of the potential difference between the liquids
given b y A V --= / 3 / I ( P - P'), where B = S ( L , - ~)') g is the di//er-
e~zlial storage, or briefly storage of the system.
The uni-terminaI system (fig. 2a) m a y be considered as a bi-
t e r m i n a l system of the kind of fig. 2b in which the u p p e r liquid has
a negligible density.

l ; i g . 2. H y d r a u l i c storage elements; a. U n i - t e r m i n a l ; b-d. B i - t e r m i n a l .

A combination of two systems of the t y p e of fig. 2b is represented

in fig. 2c. I n e r t i a being supposed negligible, all places in the u p p e r
liquid have the same potential, a n d hence t h e p o t e n t i a l difference
between the entrance a n d the exit is PI - - P2 = (e - - {") g(hl - - h2)
Let now- V denote the volume t h a t has been introduced at the left
a n d delivered at the right when we suppose t h a t originally ]h a n d h 2
were equal. Then we have 1" = B ( P I - - P 2 ) , where I//3 = 1/B 1 -F
-~ 1/B2 =- ( ~ ' - 0') g ( 1 / S i - F 1/.52). Here /3 is the resulting storage of
the system. A n o t h e r combined storage element is represented in
fig. 2d (here ¢, < ¢/).
The uni-terminal system (fig. 2a) operates according to the same
law as would do the hi-terminal systems of fig. 2c and d if the right-
h a n d vessels of these systems were infinitely wide a n d the interface
in them at zero level.
The storage elements conserve potential energy in the g r a v i t a -
tional field. This energy is E~ = ½/3P, or E~ = . ~ / 3 ( P - - p , ) 2 , or
ER ~ ..]3(t
1 ) ~ - - P2) 2, p r o v i d e d the zero level of p o t e n t i a l energy is
defined b y the g r a v i t a t i o n a l energy of the system when all liquid
elements have been carried to the zero level of height.
S e a. We consider a fictitious basin of infinite extent, so t h a t its
422 J.C. SCHONFELD

p o t e n t i a l remains constant, irrespective of w h a t quantities of water

are a d d e d or withdrawn. A sea constitutes a basin of so large an
e x t e n t t h a t it will not a p p r e c i a b l y a l t e r its p o t e n t i a l as long as the
discharges are not too excessive (ahstraction is m a d e from a n y t i d a l
motion in the sea). We shall therefore call our fictitious basin of
c o n s t a n t potential seh. I t is convenient to identify the zero level,
which is in principle a r b i t r a r y , with the level of sea.
A uni-termina[ storage elenlent (fig. 2a) m a y be considered as a
simplification of a b i - t e r m i n a l element (fig. 2c or d) of which the
second entrance is connected to sea.
G e n e r a t o r (p u m p ) , A p u m p (fig. I) produces a difference of
head and hence a potential difference P 2 - P~. Meanwhile it m a y
t r a n s m i t a discharge Q from the low- to the h i g h - p o t e n t i a l side. I t
therefore delivers h y d r a u l i c power IV = (P2 ~ Pi) 0- P r a c t i c a l l y a
p u m p has always internal energy losses; hence the p o t e n t i a l differ-
ence still depends on the discharge delivered.
An oscillating m e m b r a n e or piston may, produce sinusoidally
v a r y i n g discharges and heads, in a similar w a y as an o r d i n a r y p u m p
produces c o n s t a n t discharges a n d heads.

_- c. r' ,&

Fig. 3. a. Fluid joint; b. Electric joint; c. Rigid joint.

Fluid joint. W e consider a joint of two or more h y d r a u l i c

systems (fig. 3a). We idealize the joint b y neglecting the effects of
inertia a n d friction in the liquid and b y supposing the solid walls to
be rigid a n d not leaking, Then all the systems have the same poten-
tial at the joint, and the algebraic sum of the discharges t o w a r d s
the joint is zero:

PI = P2 = G . . . . . ; G + c22 + G + . . . . . o.

(Hence at least one of the discharges is negative, i.e. opposed to the

sense indicated in fig. 3a).
Series and parallel coupling. W h e n two systems A B
and CD are coupled in such a w a y t h a t the terminals A a n d C form
 A N A L O G Y OF H Y D R A U L I C AND O T H E R SYSTF.MS 423

one entrance, and likewise the terminals BC, the systems AB and
CD are said to be parallel. Then the potential difference on AB is the
same as on CD and the same as on the whole system, and the dis-
charge t r a n s m i t t e d through the whole system is the sum of t h e
discharge through A B and t h a t through CD,
W h e n the two systems are coupled in such a way t h a t B is con-
nected to C, the systems are said to be in series. Then the discharge
t h r o u g h AB is the same as through CD and the same as t h r o u g h the
whole system, and the p o t e n t i a l difference on the whole system is
the sum of the potential difference on AB and t h a t on CD.
Two e l e m e n t a r y impedances are composed in ,series or parallel
according to the following rules:

Hydraulic In series Parallel

resistances R = R 1 + R2 D=D 1 +D 2
inertances M = M I + 3.t 2 I / M = I / M t + 1/M u
capacitances 1/B = 1/I511 + 1/B 2 B=B 1 + B2

§ 3. Eleclric syslems. \Ve presume t h a t the properties of electric

systems are so well-known t h a t a brief exhibition m a y suffice.
R e s i s t a n c e. An electric current I t h r o u g h a linear resistor
sets up an electric potential difference A U, such that

A U = Ro.I or I = G AU,

where R. is the declric resislan.ce and G the conduclance. We have in

particular G = ?,.-1/1 for a uniform resistor with cross-sectional area
A , length l, and co,nduclivi@ y.
The resistor dissipates the energy per unit time W = I A U =
= R~I 2 = G ( A U ) 2.
I n d u c t a n c e. The formula for a self-inductance (inductor) is

U = L dI/dt,

where L is the inductance. The inductor stores energy E,,, = -~-LI2

in its magnetic field.
C a p a c i t a n c e. The formulae for a condensor (capacitor) are

Qc = C A U or I = Cd(AU)/dt,

where Q,. is the electric charge of the condensor and C its capaci-
424 J . c . SCHONFELD

:a~we. The condensor stores energy E~ = ½ C ( A U ) 2 in its electric

field.
E a r t h. In electrical engineering this denotes a fictitious con-
d u c t o r of infinite extent so t h a t its p o t e n t i a l remains c o n s t a n t
irrespective of the quantities of electricity with which the conductor
is charged or dischSrged. I t is called earth, since the e a r t h consti-
tutes a conductor of so large an e x t e n t t h a t it will not a p p r e c i a b l y
alter its p o t e n t i a l b y a n y currents of not too excessive magnitudes.
I t is convenient to identify the zero potential, in principle a r b i t r a r y ,
with the p o t e n t i a l of earth. In theoretical problems of electric fields,
the point at infinity m a y be considered as earth.
G e n e r a t 0 r (e 1 e m e n t). An electric element or other generator
produces a tension, while it m a y t r a n s m i t a current from the low- to
the high-potential side.
Ideal transformer. W h e n the coupling of a t r a n s f o r m e r
is perfect a n d the inductance so great t h a t it m a y be idealized as
infinite, the p r i m a r y voltage and current are in a definite p r o p o r t i o n
t r a n s f o r m e d into a secondary voltage a n d current in such a w a y t h a t
p r i m a r y a n d s e c o n d a r y power are equal.
E 1 e c t r i c i o i n t. W h e n we idealize the j o i n t of a set of
electric systems (see fig. 3b) b y neglecting inductance, resistance,
c a p a c i t y and leakage at the joint, the formulae for the joint are

U l = U2= U3 ..... ; I 1 +I 2 +I 3 + ..... 0.

(The l a t t e r formula shows t h a t at least one of the currents is negative,

i.e. opposed to the sense indicated in fig. 3b).
Series and parallel schemes. The rules for elemen-
t a r y electric impedances in series or parallel schemes are:

Electric In series Parallel

resistances R. = R,l + R. 2 G = Gt + G 2
inductances L = L 1+ L2 1/L = 1/L l + 1/L 2
capacitances I / C = 1/C I + 1/C 2 C = C I + C2

§ 4 . Hydra~dic-eleclric a~alogy. Complete and incom-

plete analogy. W h e n we compare the formulae of § 3 for
electric systems to those for h y d r a u l i c systems as deduced in § 2, it
will be clear t h a t the following analogy exists:
 A N A L O G Y OF H Y D R A U L I C AND O T t l E R SYSTEMS 425

Complete analogy
Hydraulic Electric
vo'lume charge
discharge current
potential difference potential difference
resistance resistance
deliverance conductance
inertance inductance
capacitance capacitance
kinetic energy magnetic energy
gravitational energy electric energy
sea earth
generator (pump) generator (element)
If we consider only the formulae for the impedances, we might
also arrive at the following analogy:

Incomplete analogy
Hydraulic Electric
discharge potential difference
potential difference current
resistance conductance
deliverance resistance
inertance capacitance
capacitance inductance
kinetic energy electric energy
gravitational energy magnetic energy
The rules for joints, however, have then been disregarded. Now
we may overcome this objection sometimes by introducing mesh
currents (see the arrows in fig. 4a) so that the current in every branch
of an electric network m a y be considered as the difference of two
mesh currents s). Then the rules for a mesh (fig. 4b) are

I 1 2 = I 2 3 =134 . . . . . ; U12+ U23+ U34+ . . . . . 0.

Here Ii2 is the mesh current of the mesh considered (fig. 4b) as it
flows through branch 12, I2a as it flows through branch 23, etc.;
furthermore U12 is the drop of potential along branch 12, U23 along
branch 23, etc. In what we here called the incomplete analogy, these
rules for a mesh in the electric network must provide the counterpart
Appl sci. Res. B 3
426 J.C. SCHONFELD

of the rules for a joint in the h y d r a u l i c network and conversely.

Then the networks are d u a l l y opposed (like e.g. the two n e t w o r k s
of fig. 4c and d).

Fig+ 4. Examples of networks.

A network, however, can only be t r e a t e d b y mesh currents when

it is topologically plane, i.e. when it can be drawn on a plane w i t h o u t
branches crossing 8). This means t h a t not every hydraulic n e t w o r k
has an electric analogue according to the incomplete analog3;. This
is why we characterized it as incomplete.

HydrauLic resistance inert.ante caPaCitance impedance sea generator(pump]

_a. ~ --"m.P- - c a - m +---I + . _ . ~ - . - ~ _

ELectric resistance inductance capaci~nce +rnpedance earth tjenerato~"

Mechanical resistance compl.iance inertance impedance immobile generator [motor)

Fig+ 5. Schematic symbols; a. Hydraulic; b. Electric; c. Mechanical.

S y m b o 1 s. The complete analogy is illustrated b y fig. 5a and b

where the well-known electric symbols have been used, whereas
analogous hydraulic symbols are introduced on which we r e m a r k :
The symbols Ior resistance and inertance were suggested to us b y
the relative narrowness and length of a tube, the s y m b o l for the
storage b y fig. 2c and d, and the symbol for sea b y a usual s y m b o l

Fig, 6. Damped series oscillators; a. Hydraulic; b. Electric; c. Mechanical.

Fig, 7. Damped parallel oscillators ; a. Hydraulic ; b, Electric ; c. Mechanical.

 ANALOGY OF HYDRAULIC AND OTHER SYSTEMS 427

for a water surface. The symbol for a pump, deduced from fig. I b,
denotes in particular a constant-head pump. An oscillating-head
p!~mp (membrane, piston) may be represented b y the other symbol
for a generator.
According to the complete analogy, series and parallel connections
in a hydraulic system correspond to series and parallel connections
in the analogous electric system (figs. 6a and b, and 7a and b).
Hydraulic oscillators. A U-tube as represented in fig.
8a is an example of a hydraulic oscillator. The two legs are uni-
terminal storage elements. The tube itself acts moreover as an iner-
tance. Hence we can represent the system by the scheme of fig. 8b.

e. 9 . ~

Fig. 8. U-tube.

By considering the'free space above the tube as the upper connection

filled with air in a system like that of fig. 2c, we can also treat the
two legs as one bi-terminal storage member and then arrive at the
scheme of fig. 8c, or, if the resistance in the tube is taken into
account, at the scheme of fig. 8d. (For a theoretical treatment of
the system cf. 6)). The system is analogous to an electric oscillator
as indicated by one of the schemes of fig. 8e, / or g, which correspond
to the hydraulic schemes of fig. 8b, c, or d.

~ _ p n
t~, I@I

a............................... _b. $ a
Fig. 9. P i t o t velocity meter.

In fig. 9a we have represented a pitot velocity meter. The nose

entrance of the pitot tube is acted upon by the total head and the
side entrance by the static head (height plus pressure head). Thus it
428 j . c . SCHONI:ELI)

is the velocity head (v2/2g) that activates the system, and the mano-
meter in the form of the Fl-tube will indicate this head as long as the
velocity remains constant. When the velocity is varying, the iner-
tance and resistance ill the tubes cause deviations of the indication
of the manometer from the actual value of the velocity head. Since
the manometer works as a hi-terminal storage member (fig. 3b),
the system can be represented by the scheme of fig. 9b, where the
generator denotes the action of the varying velocity head. The
resistances and inertances in the scheme of fig. 9b m a y be composed
according to the rules of series coupling, and then we obtain the
simplified scheme of fig. 9c.
The pitot meter is analogous to an electric system as represented
schematically in fig. 9d. The electromotive force of the generator
represents the velocity head to be measured, and the tension on the
condensor represents the reading of the manometer.

I I I I I I l
,~- tI t2 13 14 15 16 i? 18 =g rio ~11
-" / / / / / / / / / / / / / / 1 2 k / / / / / / / / / / / k / / / / 2 / / / / 2 / / / / 2 / / / / 2 k 2 / / 2 / / /

b_. ~-~r -~ M~ ~-Ir-~ M~ ~-Tr-J MR ~ r - ~ ~R

Fig. 10. \Vave motion in a canal.

Hydraulic wave-systems. Besides the more simple

oscillators discussed above, in which storage, inertia and resistance
are discrete properties, we encounter also hydraulic systems in which
storage, inertia and resistance are continuously distributed. Such
systems are capable of propagating waves 9).
Fig. 10a represents an open conduit (canal, river, estuary etc.) with
storage on one hand, and resistance and inertia in the stream bed on
the other hand. In order to reduce the system, we separate schema-
tically these two functions by supposing the storage of the sections
2-4, 4-6 etc. (fig. 10a) as concentrated in open basins (fig. 10b}, and
the inertia and resistance of the sections 1-3, 3-5 etc. as concen-
 A N A L O G Y OF H Y D R A U L I C A N D O T H E R S Y S T E M S 429

trated in closed conduits (tubes). Next we consider the open vessels

as bi-terminal storage elements connected to sea (fig. 10c). The
electric analogue is a free conductor with resistance, inductance
and capacity against earth (fig. t0d).
Another example of a wave-system is formed by two layers of
liquids (e.g. salt and fresh water) with different densities. In that
arrangement so-called internal waves are possible at the interface
between the two liquids (fig. 11 a). We suppose that the upper liquid
is covered by a rigid b o u n d a r y (practically the same effect is ob-
tained by making the difference in density small compared with the
densities themselves, because the surface disturbances attending
the internal waves are then very small). The system m a y be schema-
tized by sections of parallel branches formed by bi-terminal storage
elements of the type of fig. 2b, and two lines of inertances in series,
one line formed by the inertances of the upper liquid (Ms) an the
other line by those of the lower liquid (21I,) (fig. 1 lb).

fresh water P~
a_.~
salt water • /o~
"/lil>'llillllll~llllllllllllllllll/.

Fig. I 1. Internal wave motion.

The electric analogue is a set of two conductors, each with series

inductance, and parallel capacity between them (fig. 1 l c).

§ 5. MechaJzical systems. By mechanical systems we shall for the

moment understand systems of solid bodies moving relative to one
another.

-- i i -- i I,, i - -

~'. I., I I,. I,, ; I. ,, .. ,. ~,

]Zig. 12. a. Mechanical resistance; b. Mechanical compliance: c-e.

Mechanical inertances.

R e s i s t a n c e. Consider two bodies t h a t can glide along each

other (fig. 12a). Let v I be the velocity.of the first and v2 of the second
430 ] . C. SCHONFELD

body. Then the force F exerted b y each b o d y upon the other depends
on the relative velocity v ~ - v2. I n case of lubricated surfaces we
may put
F = r(v I - - v 2 ) , where r = e~,S/6,
(e = density, 7, = kifiematic viscosity of the lubricating liquid,
S = area of the lubricated surfaces, and 6 = distance between
them) m a y be called the mechanical resistance of the system. Neglect-
ing inertia, the system as a whole transmits the force F. We can also
put
v1 - v 2 = g , , F , where g , , = 1/r
m a y be called the glidance of the system. The resistance element
dissipates energy, per u n i t time a m o u n t i n g to W = F(v t - - v 2 ) =
= g m F 2 = r(v! - V2)2.
-

S p r i n g. Let to be the length of the spring of fig. t2b, then a

compression or expansion requires a force F = {10 - - (x2 - - xl)}/c,
where c is the compliance of the spring. Neglecting the mass of the
spring, it transmits the force F. The spring stores elastic energy,
E, = ½cF 2, which is a kind of potential energy.
I n e r t i a. If a force F works on a mass m (fig. !2c), the mass is
accelerated according to Newton's law

F = m dv/dt, where v -= dx/dt.

Now consider a mass at the end of a lever (fig. 12d). Then we have

I;1 =mo(lll2/120)(Ol--v2)--mo(12/ll) O" V2=mo(121x/12)(vl--O2)--mo(l,/lo) ~)2.

Let now m,0 decrease a n d lj(/2) increase in such a way t h a t molll2/t20
tends to a limit m. Then we approach a system obeying the formula

F = F t = F 2 = n~(~) 1 - ~ ) 2 ) ,

where m is the mechanical inertance of the system.

Another hi-terminal system is represented b y fig. 12e. If x~ a n d x 2
v a r y similarly, the two masses will not be accelerated provided the
m o v e m e n t s are small, a n d then the t r a n s m i t t e d force will be zero.
A difference in variation of x~ and x2, however, requires forces to
accelerate the masses a n d thus force can be t r a n s m i t t e d .
A mechanical inertance stores kinetic energy, E k = ½my2, or
E k = ½~n(v I - - v2) 2, when i t is a single mass or a b i - t e r m i n a l element
respectively.
 A N A L O G Y OF H Y D R A U L I C A N D O T H E R SYSTEMS 431

I m m o b i 1 e. I n treating machines we frequently consider

some m e m b e r of the machine as immobile although forces m a y be
working on such a member. We m a y do so if the m e m b e r considered
is of such a great mass or if it is connected to a body of such a great
mass (often the earth) that, as long as only forces of not too excessive
m a g n i t u d e are working, these cannot set t h a t m e m b e r or t h a t body
into perceptible motion. A body of such a great mass will here be
denoted as the immobile.
According to the classical relativity principle, the state of rest is
undefined as long as only uniform translatory movements are in-
volved. It is c o n v e n i e n t to fix our coordinate system, supposed to be
at rest, to the body considered as immobile.
A single mass m as represented in fig. 12c is in principle equivalent
to a bi-terminal system with inertance m of which the second m e m b e r
is connected to the immobile.
IX'[o t o r. A motor in general is some machine t h a t moves one
body with respect to another body. Often one of these bodies m a y be
considered as immobile or connected to the immobile, b u t this is not
strictly necessary. In fig. 13a we have represented a motor t h a t moves
a rod b y means of a pinion. The motor itself m a y glide, not rotate,
while friction a n d inertia are supposed to be negligible. Then the
system m a y t r a n s m i t a force, meanwhile giving a differential move-
m e n t x 2 - xl. The power delivered is W = F(v 2 - - v l ) . There are
also motors that produce oscillatory motions.

Fig. 13. a. Motor; b. Pair of levers.

Pair of levers. A pair of levers as indicated in fig. 13b

operates according to the formulae F t / F 2 = (v~2- - ½2)/(vii --- ½1) =
=12/l ~. This means t h a t a small force and a great velocity difference
m a y be transformed into a great force and a small velocity difference
or conversely, the power remaining the same.
Rigid joint. At a r i g i d and massless joint of two or more
solid bodies all the bodies have the same m o v e m e n t and the sum of
the forces working on the joint is zero (see fig. 3c):
vl = v 2 = v a ..... ; Ft + F2 + Fa + . . . . . O.
432 j.C. SCHONFELD

These rules may be interpreted both vectorially for the total veloci-
ties and forces and algebraically for the components of all velocities
and forces in any particular direction. (At least one of the forces is
negative, i.e. opposed to the sense indicated in fig. 3c).
Series and parallel coupling. Two mechanical
systems AB and CD are coupled parallel to each other if A is joined
to C and B to D. Then the movement of B relative to A is the same
as that of D relative to C, and the force transmitted by the whole
system is the sum of the forces transmitted by AB and CD separately.
Two mechanical systems AB and CD are coupled i n series if B is
joined to C. Then the force transmitted by AB is the same as that
transmitted by CD, and the movement of D relative to A is the sum
of the movement of D relative to C and that of B relative to A.
The composition rules for the elementary impedances are:

Mechanical In series Parallel

resistances g,. = g,,,i + g.a r = r I + r2
compliances c = cI + c2 l/c = 1/c 1%- 1/c 2
inertances 1/m = 1i m t + 1/m 2 m = m 1%- m 2

§ 6 . Mechanic-eleclric analogy. Complete and incom-

plete analogy. By comparing the formulae for electric
systems (§ 3) to those for mechanical systems (§ 5) the following
analogy establishes itself clearly:

Complele analogy

Mechanical Electric
force current
velocity difference potential difference
resistance conductance
glidance resistance
compliance inductance
inertance capacitance
kinetic energy electric energy
elastic energy magnetic energy
immobile earth
generator (motor) generator (element)
pair of levers ideal transformer.
 A N A L O G Y OF H Y D R A U L I C A N D O T H E R S Y S T E M S 433

By considering only the formulae for the impedances we m a y also

arrive at the following analogy:
Incomplete analogy
Mechanical Electric
force potential difference
velocity difference current
resistance resistance
glidance conductance
compIiance capacitance
inertance inductance
kinetic energy magnetic energy
elastic energy electric energy.
There is a controversy among different authors 2) as to which of
the analogies should be preferred. By virtue of our considerations on
hydraulic-mechanic systems (§ 7) we have arrived at the following
standpoint :
The second analogy which is often, although incorrectly we think,
called the direct analogy, does yet not account satisfactorily for the
rules of electric and mechanical joints. Therefore we agree with
H e c h t ~) (cf. ~0)) in his interpretation of mechanical systems as
force-transmitting systems, and we believe the first analogy, which
we called the complele analogy, to be more veracious. We think,
however, that H e c h t's proposal should be confined to what we have
called mechanical systems (§ 5), but that it then gains in persuasive
strength since all the counter-arguments based on properties of
hydraulic systems are then irrelevant.
It m a y appear as an unsatisfactory consequence of the point of
view propagated above that the intuitively felt analogy of forces and
electromotive forces is disregarded. A stronger, more rational argu-
ment of this kind would be that according to our point of view a
mechanical inertance is considered as the analogue of an electric
capacitance whereas a hydraulic inertance is considered as the
analogue of an electric inductance; thus a contrast is created be-
tween mechanical and hydraulic inertances, although they both
express the same inertia law. And yet we think that this rather
paradoxical contrast must be accepted because it is characteristic
for an essential difference between mechanical and hydraulic
systems, as we shall discuss further in § 7.
434 J . c . SCHONFELD

S y m b o 1 s. An illustration of the complete analogy is given in

fig. 5b and c where we have also introduced schematic symbols for
the mechanical impedances. As to the latter we m a y remark: The
symbol for a mechanical resistance was introduced b y H e c h t ~)
and the symbol for a spring is selfevident. We have not adopted
Hecht's symbol for inertance as it rather represents a mechanically
working hydraulic inertance than a pure mechanical inertance; so
we adopted a symbol suggested by fig. 12d. The symbol for the m o t o r
was derived from fig. 13a.

7///~

_c. d_. /-~2/~ ' _e. r.

Fig. 14. M e c h a n i c a l oscillators.

As a consequence of the view adopted, a system of parallel

mechanical impedances corresponds to an electric system with the
analogous electric impedances likewise parallel, and a mechanical
series system corresponds to an electric series system. This is repre-
sented in fig. 6b and c, and fig. 7b and c.
Mechanical oscillators. Fig. 1 4 a r e p r e s e n t s a n o s c i l -
lator formed by a mass m suspended by a spring with compliance c,
and fig. 14b-e represent schemes of it.

_a, I I I I I I I I I t ......... I
g ~

Fig. 15. W a v e motion in a r o d .

In the pendulum of fig. 14/, the horizontal component of the ten-

sion is F = Tx/l, so that the suspension by the chord works as a gra-
vitational compliance c = l / T = l/mg, and hence we arrive likewise
at the schemes of fig. 14b-d (the scheme represents the way in which
elements are interconnected, not the direction in which they move),
 ANALOGY OF HYDRAULIC AND OTHER SYSTEMS 435

Mechanical wave systems. The elastic rod of fig.

t 5a, vibrating longitudinally, works as a combination of springs and
masses (cf. fig. 15b). The scheme of the system is given in fig. 15c,
and the analogous electric scheme in fig. 15d. We arrive likewise at
these schemes when we consider a transversally vibrating string.

§ 7. Mechanic-hydraulic syslems. P i s t o n-m e c h a n i s m s. In

the double-action piston with cylinder of fig. t6a let friction, inertia
and elasticity be negligible; then the forces F must balance the
potential difference and the discharge Q is proportional to the velo-
city difference :
F = (P2 - - P~) A ; (2 = (v~ - - v2) A.
When the piston moves to the left by the action of pushing forces F
or to the right by the action of pulling forces - - F , the mechanical

P" F

i •

Fig. 16. P i s t o n m e c h a n i s n l s ; a. D o u b l e - a c t i o n ; b. S i n g l e - a c t i o n ,

forces deliver power and the hydraulic potentials consume that

power: l,V,,,i, = F ( v l - - v 2 ) = ( P 2 - - P 1 ) Q . When the piston moves
to the right by the action of a potential difference P2 - - P~ or to the
left by the action of a potential difference P I - P2, the hydraulic
potentials deliver power which is consumed by the mechanical for-
ces: Wh,, = F ( v 2 - - v l ) = ( P I - - P 2 ) Q . In this way the piston
operates as a mechanic-hydraulic convertor. The single-action
piston of fig. 16b operates likewise as a convertor. It m a y be con-
sidered as a double-action piston of which one side is connected
to sea.
Two pistons with different areas A m a y be hydraulically coupled
as in fig. 17a, or mechanically coupled as in fig. 17b. The system of
fig. 17a operates as a mechanical transformer b y hydraulic inter-
mediary (hydraulic lever), and the system of fig. 17b operates as a
hydraulic transformer by mechanical intermediary ; hence the latter
transforms a large discharge of low potential difference into a small
discharge of large potential difference, or conversely. One connection
436 j . C . SCH(}NFELD

between the cylinders of fig. 17a m a y be b y sea (dotted lines) and

then we can also introduce single-action pistons. The cylinders of fig.
16b m a y both be connected to the immobile (dotted lines) instead of
d i r e c t l y to each other.

a_. -.-~- b_. ~J ~.

Fig. 17. Coupling of piston mechanisms ; m Hydraulically ; b. Mechanicalh'.

Mechanic-hydraulic translation of impe-

d a n c e s. A piston max; also be used to t r a n s l a t e a mechanical into
a h y d r a u l i c impedance, or conversely. If we p u t e.g. a hydraulic
inertance 21/ between the two sides of the cylinder of fig. 16a, we
have P 2 - - P 1 = M d Q / d t , and hence F = . q ( P 2 - P I ) = - I M
dQ/dt = A 2 M d ( v l - - v 2 ) / d / . This is a p p a r e n t l y equivalent to a
mechanical inertance m = A2M.
If, on the o t h e r hand, we p u t a mechanical inertance m between
piston and cylinder, we have F = m.d(v I - v 2 ) / d t , and hence
P 2 - PI = (m/A2) dQ/dl, which is equivalent to a h y d r a u l i c iner-
tance M = m/A 2.
Treating in the same way o t h e r impedances, we arrive at the
following diclionary :

Mechanical Hydraulic
force p o t e u t i a l difference
velocity difference discharge
inertance inertance
compliance capacitance
resistance resistance
glidance deliverance
This is represented in schemes in fig. I8.

Fig. 18. Translation of impedances; a-c. Mechanical into hydraulic; d-/.

Hydraulic into mechanical.
 ANALOGY OF HYDRAULIC AND OTHER SYSTEMS 437

Mechanic-hydraulic translation of series

and parallel coupling. Let us now t r a n s l a t e two mechan-
ical impedances in series (fig. 19a) from the mechamcal into the
h y d r a u l i c picture. Thus we replace piston mechanisms b y corre-
sponding h y d r a u l i c impedances (fig. 19b). Let the area of the two
pistons be the same; since t h e y b o t h t r a n s m i t the same force, t h e y
will then produce also the same potential difference. Hence we can
interconnect the two cylinders (fig. 19c) without affecting the opera-
tion of the system. The two piston mechanisms now form an un-
necessary complication since one such mechanism can s u b s t a n t i a l l y
do the same work (fig. 19d). Hence it appears t h a t the series coupling
of the two mechanical elements m u s t be t r a n s l a t e d into a parallel
coupling of the analogous h y d r a u l i c elements. This can also be in-
ferred immediately, although more a b s t r a c t l y , from the fact t h a t
two mechanical elements in series t r a n s m i t the same force and yield
the sum of their relative displacements. This must correspond to
two h y d r a u l i c elements working on the same p o t e n t i a l difference
and delivering the sum of their discharges, i.e. two elements parallel
to each other.

mechanical series
hydraulic
_a. d_. paraLLel.

Fig. 19. Translation of a mechanical scheme into a hydraulic one.

By similar a r g u m e n t s we can show t h a t we generally must trans-

late series into parallel and conversely when t r a n s l a t i n g mechanical
into h y d r a u l i c systems or vice-versa. Hence, when a mechanical and
a h y d r a u l i c system are equivalent, t h e y are also considered as net-
works, d u a l l y opposed. Not every mechanical or h y d r a u l i c system
can be t r a n s l a t e d into an equivalent h y d r a u l i c or mechanical system,
because not every network has a d u a l l y opposed network s).
The above s t a t e m e n t is a consequence of the essential difference
between fluid and rigid joints. Consequently hydraulic and mechan-
ical systems are opposite in c h a r a c t e r when considered from the
point of view of network analysis.
Mechanical fluid-inertance. Hecht 1) t r e a t s all
kinds of h y d r a u l i c a n d mechanical impedances only from w h a t we
m a y here call the mechanical point o f view. As an example of an
438 j . c . SCHONFELD

elementary fluid inertance he considers the system of fig. 20a. When

the plates move relatively to each other, the fluid moves in and out
the space between the plates. This has the effect of a mechanical
inertance.

Fig. 20. Mechanical fluid inertance.

From our point of view the system is rather a complicated than an

elementary one. For the two plates should be considered as pistons
converting the mechanical movement into a fluid movement, while
the space between the plates moreover works as a hydraulic iner-
tance ; one side of this inertance is put under pressure by the pistons
and the other side is connected to the wide space outside which
practically is not subject to pressure fluctuations and therefore
should be considered as sea. Thus we arrive at the scheme of fig. 20b
or its modification of fig. 20c. This is indeed equivalent to a mechan-
ical inertance (cf. fig. l gb). Since, however, parallel and series
schemes are interchanged when hydraulic impedances are translated
into mechanical impedances, the obscuring of the essential differ-
ence between the two kinds of impedances must lead to great con-
fusion with regard to the notions parallel and in series. We shall
demonstrate this in § 8.
Hydraulic elastic storage. As a kind of counterpart
of the mechanically working fluid inertance, we m a y consider a
membrane as a hydraulically working spring. The membrane is
simultaneously an elastic element with a mecha,-mal compliance,
and a piston. The doub~'e-action membrane of fig. 21 a therefore can
be represented by the scheme of fig. 21b; comparison with fig. 18/
shows that it works Iike a bi-terminal hydraulic capacitance. In the
same way the single-action membrane of fig. 21c can be represented
by the scheme of fig. 21d or e; it works like a uni-terminal hydraulic
capacitance.

Fig. 2t. Membranes; a-b. Double-action; c-e. Single-action.

 A N A L O G Y OF H Y D R A U L I C A N D O T H E R SYSTEMS 439

§8. Acoustic systems. Dualistic conception of

acoustic systems, By an acoustic s y s t e m we shall here
u n d e r s t a n d a fluid system of which the b e h a v i o u r is p e r c e p t i b l y
influenced b y the elasticity of the fluid (e.g. air). I t is clear t h a t such
a system, as far as the elasticity of the fluid is not involved, is essen-
tially the same as a h y d r a u l i c system. Thus resistance and inertia
in an acoustic system can be t r e a t e d as h y d r a u l i c resistance and
inertia. W h e n the elasticity comes into action, however, an acoustic
system shows an analogy with a mechanical system. An air-cushion
e.g. closely resembles a mechanical spring. We believe t h a t the
clearest conception (although perhaps not always the simplest) of
acoustic systems will usually be o b t a i n e d b y considering them as
mechanical-hydraulic combinations in which hydraulic resistances
and inertances are coupled to mechanical springs b y means of piston
mechanisms. We shall d e m o n s t r a t e this b y discussing various
acoustic systems.
Fluid cushions. F r o m the mechanical point of view the
fluid in the c h a m b e r of fig. 22a can be t r e a t e d as an elastic b o d y
with length l and cross-section A having a mechanical compliance
c = 1/KA, where K denotes the compression-modulus of the fluid.
In the case of a gas we m a y p u t K = P0 or K = (cp/c~) Po, del)ending
on whether the compressions and expansions are isothermic or
isentropic (Po = average pressure of the gas, cp, c,, = specific heat
at constant pressure, volume respectively).

', . . . . . . . .

F i g . 22. A i r c h a m b e r s ; a-b. W i t h o n e e n t r a n c e ; c-d. W i t h t w o e n t r a n c e s .

F r o m the hydraulic point of view the fluid c h a m b e r can be

considered as a uni-terminal storage element since a v a r i a t i o n of the
pressure at the entrance of the c h a m b e r is a t t e n d e d b y a volume
displacement through the entrance. The storage of the c h a m b e r is
B = Vo/K (V 0 = AI = volume of the chamber). This can be dedu-
ced i m m e d i a t e l y from the compression law of the fluid, but also b y
considering the c h a m b e r as a mechanical spring o p e r a t e d hydraulic-
ally b y means of a fictitious piston of area A (fig. 22b).
The air cushion in the c h a m b e r of fig. 22c can likewise be con-
440 j . c . SCHONFELD

sidered as a mechanical spring. W h e n considerh~g it from the hy-

draulic point of view, we can d e p a r t from the scheme of fig. 22d. Now
the two pistons b o t h exert the same force on the spring between
t h e m and so produce the same pressure. Hence we m a y m a k e the
h y d r a u l i c connection indicated b y the d o t t e d line. Then, however,
the two piston mechanisms can be replaced b y a single such mecha-
nism so t h a t we arrive again at the scheme of fig. 22b.
This deduction makes clear t h a t the system of fig. 22c, although
it has two entrances, m u s t still be considered from a h y d r a u l i c point
of view as a u n i - t e r m i n a l storage element (like t h a t of fig. 2a). Hence
an air cushion a p p e a r s on one h a n d as a b i - t e r m i n a l mechanical
compliance element, and on the o t h e r h a n d as a u n i - t e r m i n a l
h y d r a u l i c storage element.
I t is clear t h a t a t r e a t m e n t of w h a t we have called h y d r a u l i c
systems, must remain incomplete when the b i - t e r m i n a l g r a v i t y
storage elements of fig. 2b-d are ignored and only the uni-terminal
compressibility storage elements of fig. 22 are taken into considera-
tion. We t h i n k t h a t this m a y have been the reason for H e c h t ' s not
always satisfactory e n d e a v o u r to t r e a t air impedances from the
mechanical point of view.

a.

Fig. 23. Simple acoustic resonator.

Acoustic oscillators. The resonator of fig. 23 can be

t r e a t e d as an air c h a m b e r of which the m o u t h works as an inertance.
W h e n a sound wave excites the resonator, the fluid is set into an
oscillating motion t h r o u g h the mouth. The s y s t e m can be schema-
tically t r e a t e d as a spring coupled b y a piston to the inertance of the
mouth, which is p u t in series with a g e n e r a t o r representing the
action of the sound waves (fig. 23b). In this scheme we can either
t r a n s l a t e the spring into a h y d r a u l i c storage element (fig. 23c) or
the h y d r a u l i c inertance a n d the g e n e r a t o r in series into a similar
mechanical c o m b i n a t i o n in parallel (fig. 23d). B o t h the electric series
oscillator of Iig. 23e (see fig. 23c) and the parallel oscillator of fig. 23[
(see fig. 23d) m a y be considered as analogues of the resonator of fig.
 A N A L O G Y OF H Y D R A U L I C AND O T H E R SYSTEMS 44!

23a. Whether we arrive at the one or at the other, depends on

whether we a_dopt a hydraulic or a mechanical point of view.
In fig. 24 we have a system of two air chambers with two mouths,
which can be represented by the scheme of fig. 24b where the air
chambers are coupled hydraulically (viz. by equating pressures and
adding volume displacements) with each other and with the two
parallel inertances of the two mouths. The two fictitious pistons
between the chambers work as a mechanical lever, and hence we
can replace the back-chamber by a spring with the compliance
ClA~/A ~ coupled mechanically in series with the spring of the first
chamber (fig. 24c). By translating the springs into storages we obtain
the hydraulic scheme of fig. 24d. By translating the hydraulic in-
ertances into mechanical inertances on the other hand, the mechan-
ical scheme of fig. 24e is derived.

e x ~_"

Fig. 24. Compound acoustic resonator (lst example).

Fig. 24e clearly illustrates the drawback of the purely mechanical

interpretation of acoustic systems. For, as long as the mechanic-
hydraulic translation is ignored, it must appear as rather confusing
that the two evidently parallel mouths are represented in fig. 24e as
inertances in series. On the other hand we feel that a purely hydraulic
conception would not make quite clear why the two air chambers of
fig. 24a must be treated as parallel storages, since they appear as
elements in series.
The electric schemes of fig. 24/and g both represent analogues of
the resonator of fig. 24a. The first one is derived by the hydraulic,
the second one by the mechanical conception.
When comparing the system of fig. 25a to that of fig. 24a, one might
be tempted to treat the air chambers as two parallel springs. This
would be wrong, however, as the way of coupling of the chambers is
Appl, sci, Res. B 3
442 j . c . SCHONFELD

of a hydraulic nature. A true scheme of the system therefore is that

of fig. 25b, which can be reduced either to the scheme of fig. 24d or
to that of fig. 24e.

Fig, 25. Compound acoustic resonator (2nd example).

Mechanic-acoustic combinations. The dualistic

interpretation of acoustic systems here proposed will offer no princi-
pal difficulties in treating mechanic-acoustic combinations. We
shM1 demonstrate this b y discussing the system of fig. 26a, which
m a y e.g. represent a microphone.

Fig. 26. Microphone.

One of the two air chambers, which are separated by a membrane.

has an entrance with inertance through which sound waves m a y be
received. The membrane is supposed to have elasticity and mass.
Hence we arrive at the dualistic scheme of fig. 26b. This m a y be
reduced either to the hydraulic scheme of fig. 26c or to the mechan-
icalscheme of fig. 26d. From the hydraulic scheme of the microphone
we m a y derive the electric scheme of fig. 26e, and from the mechan-
ical scheme the electric scheme of fig. 26/.
A c o u s t i c w a v e-s y s t e m s. In an organ pipe or other tube
filled with a compressible fluid (fig. 27a) the propagation of waves is
defined by the interaction of the inertia and the elasticity of the
fluid. We separate these two functions section-wise and arrive at the
dualistic scheme of fig. 27b. Then we can make either of the following
simplifications:
1. The two pistons a and b exert the same force upon the spring
between them (cf. fig. 22b) and hence produce the same pressure
 A N A L O G Y OF H Y D R A U L I C A N D O T H E R SYSTEMS 443

difference, provided the areas of the pistons are equal. We m a y

therefore make a hydraulic coupling between the piston mechanisms
a~and b as indicated by the lower dotted line. Then the two mecha-
nisms a and b m a y be replaced b y a single piston mechanism, so t h a t
we obtain the scheme of fig. 27c. Translation of the springs into stor-
ages finally yields the scheme of fig. 27d, of which fig. 27e represents
the electric analogue.
2. Since the two piston mechanisms b and c (fig. 27b) are connect-
ed by the inertance between them, they have equal discharges, and
so the two pistons b and c must have the same motion. We m a y
therefore make the mechanical connection indicated b y the upper
dotted line, and then the two piston mechanisms b and c m a y be
replaced b y one such mechanism (fig. 27]). Then we can translate
the hydraulic inertance into a mechanical one and thus arrive at the
scheme of fig. 27g, of which fig. 27h is the electric analogue.

I...... 17
- a V-b-------~--l_..J'-d--7---~-7__I-f------g'~

- - c ' c ¢

Fig. 27. Wave motion in a column of air.

The scheme of fig. 27d can also be obtained immediately b y the

consideration t h a t an air cushion works as a hydraulic storage, so
that the organ pipe is an analogue of a canal (fig. 10). The scheme of
444 J.C. SCHONFELD

fig. 27g, on the other hand, can be obtained by the consideration

that the inertances are bodies of air (cf. fig. 15b), so that the organ
pipe is an analogue of a vibrating rod. Our treatment of the system
as a combination of springs and hydraulic inertances makes clear
that the purely mechanica.1 and the purely hydraulic interpretation
are formally equivalent. Physically however, the hydraulic nature
of acoustic systems prevails over the mechanical nature, we think, as
appears e.g. from the fact that couplings of acoustic elements are
usually hydraulic, not mechanical in character.
The LC-schemes of fig. 27e and h are mutually dual schemes, as
the inductances of fig. 27e and the capacitances of fig. 27h both
represent the inertances M, and the capitances of fig. 27e and the
inductances of fig. 27h both represent the compliances c. Both
schemes can equally well be justified as electric substitutes for the
organ pipe.
When we have an incompressible fluid in an elastic tube (water in
a rubber-hose, blood in an artery) the tube works like a spring
according to the scheme of fig. 27c. Such a system is clearly an analo-
gue of an organ pipe, a canal, a rod, or a string.
The propagation of shock waves in a tube filled with liquid (so-
called waterhammer) is defined by the interaction of the inertia and
the elasticity of the liquid, whereas often the elasticity of the walls
of the tube (mostly iron) supports the elasticity of the liquid (mostly
water). We then arrive either at a scheme in which the compliance
of the liquid cushions and the compliance of the tube sections are in
series, or at a scheme in which the equivalent storages are parallel,
the mathematical performance of the coupling of these elements
remaining the same in both interpretations.

§ 9. Gyro-elemeuts. The problems arising in clarifying the analogy

between hydraulic, mechanical and electric impedances as discussed
above, are similar to problems relating to what we m a y call gyro-
dements. Therefore it is interesting to consider these gyro-eIements
briefly in this connection.
Hydraulic gyro-element. In fig. 28a we have repre-
sented a hydraulic gyro-element consisting of two crossing conduits
SN and W E which are rotating with the angutar frequency o~. When
e.g. a flow Qs = Q,v is running from S to N, there will be set up a
potential difference between E and W as a result of the Coriolis
 ANALOGY OF HYDRAULIC AND OTHER SYSTEMS 445

forces. In the same way a flow from W tot E sets up a potential

difference between S and N. If we neglect all other effects of inertia,
storage and resistance, the system operates according to the for-
mulae
z-,~. - - P,,. = ~x((2:,~ + (2~), P~ + P,,. = P~,, + P.~,
P . ~ - PN = ,.,:x(Q,,, + O,-), O,,. + O~ = O~ + Q:,..
Here X , which we shall call the hydraulic gyro-reactance of the
system, is defined by X: = 2Coo~a, where a denotes the height of the
crossing conduits. The reciprocal Sh of X is called the hydraulic gyro-
susceptance. A schematic symbol for the gyro-element is proposed
by fig. 28b.
~,,I~ N o~- N S E

o~/ S I _a. _b. S c_. N W

Fig, 28. H y d r a u l i c gyro-element.

When the gyro-element is operated in such a way that Q:v=Qs=

= Q,, and QE = Qw = Qi as represented in fig. 28c. the fourpole
equations are
Pi = XQ,,, Qi = ShP,,.
The f0urpole is conservative. We shall call a gyro-element operating
in this way, a hydraulic gyralor.
Hydraulic gyro-reactances and -susceptances can be introduced
advantageously e.g. in computations of the tidal motion in wide
channels, such as the Dutch Bight and the English Channel which
form the connection between the North Sea and the Atlantic.
Mechanical gyro-element. The mechanical gyro-
element of fig. 29a consists of a top rotating with the angular fre-
quency ,, in a rhombic frame WSEN, the axis of rotation being
normal to the plane ot the frame. When a couple by equal forces F~v
and F s is applied, a rotation of the gyro-element about the axis SN
is set up by the gyro-effect. In the same way forces F w and FE set
up a rotation about WE. When we neglect all other effects of inertia,
elasticity, gravity and resistance, the system obeys the formulae
{(Fx + Fs) = x,,,(Vr - - vw), vE + %. = % , + Vs,
{-(F,,. + F;..) = x , , , ( v s - vA.), F,v + Fs = FE + Fx.
446 J . C . SCHONFELD

Here x,,, characterizing the reaction of the top, is called the mechan-
ical gyro-reaclance. Its value is Xm = O ) ( I p - I,)/2b=b 2, where tp
denotes the m o m e n t of inertia of the top about the main axis and I,
that about any axis normal to the main axis. The reciprocal s of x,,,
m a y be called the mechanical gyro-susceptance. Fig. 29b represents a
schematic symbol.

,W v~ vu
, E W E
N
,~a. _b. S C.

Fig. 29. Mechanical gyro-element.

When the element is operated in such a way that F u = F s = F , ,

and FE = F , , = Fi, it forms a mechanical gyrator obeying the
equations
v i = sF,, F i ~. xmv,,
where v; is the velocity difference %. - - vE and v, = v x - - Vs. The
scheme of the gyrator is fig. 29c.
Electric gyro-etement. The Hall effect m a y be used to
construct an electric gyro-element. This is shown in fig. 30a, repre-
senting a rectangular conductor put in a magnetic field B and pro-
vided with lateral extensions W, S, E and N for the input or output
of currents through the rectangular conductor. When ordinary in-
ductance, capacitance and resistance are neglected, the system
works according to the equations
U j . : - Uw = ½X~(Iu + Is), UF- + U,. = U x + Us,
Us- U x = ½X.(Iw + IF.), I . - + I s = IF- + I x .
Here X o the electric gyro-reactance, can be expressed as X = B/aa
where a is the thickness of the conductor and ~ a material coefficient
with the dimension of an electric density (C/m a in Giorgi-units). The
reciprocal S~ of X, is the dectric gyro-smceptance. Fig. 30b represents
a schematic symbol.
a[ N...-#~ .I N S g

/SS I a_. _b. £. N~,~.,~"',....~,L~

' W

Fig. 30. Electric gyro-element.

 ANALOGY OF HYDRAULIC AND OTHER SYSTEMS 447

The fourpole of fig. 30c, in which the gyro-element is operating as

an electric gyrator, obeys the equations
Ui = X J,,, I i = SeU,.
I t is a conservative fourpole as X~ is a real impedance. (It should be
noticed that the electric gyrator is not a kind of invertor, that is a
fourpole obeying equations U,. = - - Z I , , , U,, = ZIi. An invertor is
conservative when the inversion impedance Z is imaginary).
G y r o-i n v e r s i o n. Consider a hydraulic gyro-element of
which two opposite orifices, e.g. W and E (see fig. 28a) are shunted
by a resistance R (fig. 31a). Then it operates as a gyrator and it is
easily verified that we get the relation P~ = (X2/R) Qi at the en-
trance. Hence the whole system is equivalent to a resistance X2/R
between S and N (fig. 31b). When the gyrator is shunted by an
inertance M (fig. 31c) between W and E, it can be deduced that the
whole system works as a storage with the value M I X 2 (rig. 31d);
when the shunt is formed by a storage B (fig. 31e), the whole system
works as an inertance with the value BX2 (fig. 31/).

Fig. 31. H y d r a u l i c g y r o - i n v e r s i o n .

In fig. 31g the gyrator is shunted by a resistance R and an in-

ertance M in series between W and E. In this instance the system is
equivalent to a parallel-combination of a deliverance R / X 2 and a
storage M / X 2 between S and N. Hence we see that a gyrator inverts
inpedances and parallel and series schemes as follows:

Hydraulic gyro-inversion
resistance deliverance
deliverance resistance
inertance capacitance
capacitance inertance
series scheme parallel scheme
parallel scheme series scheme.

In a similar way we can consider mechanical and electric gyrators,

and then we arrive at the following inversions:
448 .1. c. SCH()NFELD

Electric gyro-inver- Mechanical gyro-inver-

sion sion
resistance conductance resistance glidance
conductance resistance glidance resistance
inductance capacitance inertance compliance
capacitance inductance compliance inertance
series scheme parallel scheme series scheme parallel scheme
parallel scheme series scheme parallel scheme series scheme
It is clear that the gyrator is a fourpole that realises the transition
from any scheme to the dually opposed scheme.
Gyrator and piston mechanism. In fig. 32a we
have represented a mechanic-hydraulic system in which a mechan-
ical resistance r and a spring with compliance c are put in series
with a piston mechanism with piston area A which translates the
mechanical impedance into a hydraulic impedance. This hydraulic
impedance has the character of a hydraulic deliverance A2/r parallel
with a storage A2c. The hydraulic performance of the system of fig.
32a is therefore equivalent to that of a system like the one of fig. 31h.
Hence we infer that the system of fig. 32a is also hydraulically
equivalent to the system of fig. 32b where we have a hydraulic
resistance and inertance put in series with a gyrator proportioned
according to A2: S~-----R: (I/r) = M : c . This illustrates the analo-
gy between a gyrator and a piston mechanism.

Fig. 32, Translation by gyro-element; a-b. Mechanic into "hydraulic;

c-d. Hydraulic into mechanic.

This analogy also becomes clear by comparing the formulae for a

hydraulic gyrator and for a piston mechanism. Then we see that the
hydraulic potential difference and discharge in the exit conduit of
the gyrator are analogous to the velocity difference and force
respectively involved in the mechanical performance of the piston
mechanism. In a similar way it can be demonstrated that a piston
mechanism is an analogue of a mechanical gyrator. This is illustrated
by the two systems of fig. 32c and d which are equivalent in their
performance.
 ANALOGY OF HYDRAULIC AND OTHER SYSTEMS 449

Universal mechanical, hydraulic, and elec-

tric schemes. The analogy between gyrators and piston
mechanisms enables us to substitute purely hydraulic or purely
mechanical schemes for hydraulic-mechanic systems in a way differ-
ent from what we have discussed in preceding paragraphs. As an
example we consider the system of fig. 24c of which the scheme of
fig. 33a is derived by some slight modifications. By using the analogy
of the piston mechanism with a hydraulic gyrator we arrive at the
purely hydraulic scheme of fig. 33b where the mechanical compli-
ances in series have been replaced by hydraulic inertances in series.
The hydraulic scheme of fig. 24d can then be deduced from the
hydraulic scheme of fig. 33b by applying the ruies of the hydraulic
gyro-inversion (v. above).
By using the analogy with a mechanical gyrator, we can like-
wise deduce the mechanicaI scheme of fig. 33c from fig. 33a. Appli-
cation of the rules of the mechanical gyro-inversion (v. above) then
yields the mechanical scheme of fig. 24e.

Fig. 33. C o m p o u n d a c o u s t i c oscillator; a. D u a l i s t i c s c h e m e ; b. H y d r a u l i c

g y r o - s c h e m e ; c. M e c h a n i c a l g y r o - s c h e m e ; d. Electric g y r o - s c h e m e .

There is also an analogy between electric gyrators and hydraulic

or mechanical gyrators. By using the analogy with a hydraulic gyra-
tor we can derive the electric scheme of fig. 33d from the hydraulic
scheme of fig. 54b. This electric scheme can also be derived from the
mechanical scheme of fig. 33c by using the analogy of the electric
gyrator with the mechanical gyrator. The electric schemes of fig. 24/
and g can be deduced by applying the rules of the electric gyro-in-
version (v. above) either to the left-hand part or to the right-hand
part of the scheme of fig. 33d.
It will be clear now that the analogy of piston mechanisms with
gyrators enables us to draw up a new mechanic-hydraulic analogy
that can be combined with the complete hydraulic-electric and the
complete mechanic-electric analogy as follows:
4~0 ANALOGY OF HYDRAULIC AND OTHER SYSTEMS

Universal analogy
Hydraulic Electric Mechanical
potential potential velocity
discharge current force
resistance resistance glidance
deliverance conductance resistance
inertance inductance compliance
capacitance capacitance inertance
gyro-reactance gyro-reactance gyro-susceptance
gyro-susceptance gyro-susceptance gyro-reactance
pump generator motor
sea earth immobile
series scheme series scheme series scheme
parallel scheme parallel scheme parallel scheme.
The mechanic-hydraulic translation presented in § 7 can then be
interpreted as the result of combining the rules of the mechanical
or hydraulic gyro-inversion with the above universal analogy.
We think that the mechanic-hydraulic-electric analogies are most
elegantly treated bY means of the above universal scheme, and by
then considering piston mechanisms as the dualistic mechanic-
hydraulic analogue of the gyrators. In this way ambiguity is avoided,
and all the more simple, customary schemes can be derived easily
from the universal schemes by eliminating gyrators by means of the
rules of the gyro-inversions. Of course there is no principal objection
against deriving any of the customary schemes directly, but especial-
13, in case of complicated systems the longer way by means of the
universal schemes may serve to avoid mistakes.
Received 17th December, 1953.
REFERENCES
1) H e c h t, H . , S c h a l t s e h e m a t a u n d D i f f e r e n t i a I - g l e i c h u n g e n e l e k t r i s c h e r u n d m e c h a -
niseher Schwingungsgebilde, Leipzig (1939).
2) G e h 1 s I1 6 j, B., Eleetromechanieal a n d electroaeoustieal analogies. Tl~esis Copen-
hagen (1947).
3) Symposlon: Eleetrisehe, aeoustisehe en optisehe impedantie. Ned. T. N a t u u r k u n d e
14 (194B) 289.
4) V e e n , j . v a n , De lngenieur 5 8 (1946) B t7.
5) V r e e d e n b u r g h , C . G . J . , a n d O . Stevens, De Ingenieur 411 (1933) B 187.
6) S e h 6 n f e l d , J . C . , A p p l . sei. Res. 2t- I (1948) 169.
7) S c 1 1 6 n r e i d , J . C . , D e Ingenieur 6 5 (1953) B 2 1 9 , 2 4 4 .
8) T e l l e g e n , B . D . H . , T . Ned. R a d i o G e n . tl (1941) 37.
9) S c h 6 n f e 1 d, J. C., P r o p a g a t i o n of tides and similar waves. Thesis Delft (1951).
10) P i k e . t, J . , T . Ned. R a d i o G e n . 10(1943) 227.
11} P a y n t e r , t-L lv[., Trans. A.S.C.E. 1111 (1953) 962.