2.

5 INPUT-OUTPUT CONFIGURATION OF INSTRUMENTS AND MEASUREMENT

SYSTEMS.
Before we discuss instrument performance characteristics, it is desirable to develop a generalized
configuration that brings out the significant input-output relationships present in all measurine
apparatus. Ascheme suggested by Draper, McKay, and Lees' is presented in somewhat modified form
inFig. 2.9. Input quantities are classified into three categories: desired inputs, interfering inputs, and
modifying inputs.
Output component due to
Interfering input interfering input and iM
F
A

FM
Modifying input
Output
FMD

Desired input
Fp
Output component due to
desired input and iM

Fig. 2.9 Generalized input-output configuration.

Desired inputs represent quantities that the instrument is specifically intended to measure. Interfering
inputs represent quantities to which the instrument is unintentionally sensitive. Adesired input produces
a component of output according to an input-output relation symbolized by F,, where F, denotes the
mathematical operations necessary to obtain the output from the input. The symbol F, may represent
diferent concepts, depending on the particular input-output characteristic being desceribed. Thus, F,
might be aconstant number Kthat gives the proportionality constant, relating a constant static input to
the correspondingstatic output for alinear instrument. For anonlinear instrument, asimple constant is
not adequate to relate static inputs and outputs; a mathematical function is required. To relate dynamie
inputs and outputs, differential equations are necessary. Many times, for repeated equal static inputs
there can be a scatter or dispersion of the output, in which casea statistical function is required to detine
the possible ranges of this scater. The symbol F, encompasses allsuch concepts. Similarly. the symbu
F, serves a similar function for an interfering input.
The third class of inputs are the modifying inputs, which appear to be similar to interfering inpu
but are distinctly different from them. Modifying inputs are the quantities that cause a change in
input-output relations for both the desired and interfering inputs; that is, they causea change inwhich
f, anuan
FM orepresent (in the appropriate form) the specific manner in
and F,
or F, The symbols Fu, and

C.S. Draper, W. McKay, and S. Lees, "Instrument Engineering," vol. 3,MeGraw-Hill, New York, 1995, p. 0
 Generalized Configurations ond Functianal Descriptions of Measuring Instruments 21

interfering input i affects F, and Fg Tespectively. These symbols. F, , and F,, nare interpreted in the
same general way as F, and F
The block diagram of Fig. 2.9 illustrates the above concepts. The circle with across in it is aconventional
symbol for asum1ming device. The two plus signs as shown indicate that the output of the summing device
is the instantaneous algebraic sum of the desired and interfering inputs. Since an instrument system may
have several inputs of each of the three types as well as several outputs, it may be necessary to draw more
complex block diagrams than in Fig. 2.9. This extension is, however, straightforward.
Examples The above concepts can be clarified by means of specific examples.
(|) Consider the mercury manometer used for differential-pressure measurement as shown in
Fig 2.10a). The desired inputs are the pressuresp, and p, whose difference causes the output displacement
x. which can be read off the calibrated scale. Figures 2.10(b) and (c) show the action of two possible
interfering inputs. In Fig. 2.10(b), the manometer is mounted on some vehicle that is accelerating. A
simple analysis shows that there will be an output xeven though the differential pressure might be zero.
Thus, if you are trying to measure pressures under such circumstances, an error will be engendered result
because of the interfering acceleration input. Similarly, in Fig. 2.10(c), if the manometer is not properly
aligned with the gravity vector, it may give an output signal xeven though no pressure diference exists.
Thus. the tilt angle Ois an interfering input. (t is also amodifying input in this case.)
Tilt
Gravity angle
Vector

Pz

Acceleration

(a (b (c)

Fig. 2:10 Spurious inputs for manomeler

Modifying inputs for the manometer include ambient temperature and gravitational force. Ambient
temperature manifests its influence in a number of ways. First, the calibrated scale changes length with
temperature; thus, the proportionality factor relating p, - P, to x is modified, whenever temperalure
varies from its basic calibration value. Also, the density of mercury varies with temperature, which again
leads to achange in the proportionality factor. Achange in gravitational force resulting lrom changes in
location of the manometer, such as moving it to another country or putting it aboard aspaceship, leads
to asimilar modification in the scale factor. Note that the effects of both the desired and the interlering
inputs may be altered by the modifying inputs.
(2) Apressure transducer exhibits a temperature sensitivity of 0.l units per degree celsius and a
pressure sensitivity of 2.5 units per MPa. If the temperature changes 20°C during a measurement ofl a
pressure of 120 MPa, deternine the error due to the dual sensiuvity of the transducer.
=[Link]/°C temp. =0.| x 20 = 2 units
= 2.5 units/MPa pressure = 2.5 x 120 = 300.0 units
 22 Doebelin 's Measurement Systems

True value = 300units

Indicator = 302units
Error = 2 units

% Error
300
(3) As another example. consider the electric-resistance strain-gage setup shown in Fig. 2.11. The
gage consists of a fine-wire grid of resistance R, firmly cemented to the specimen whose unit strain gat
a certain point is to be measured. (Detailed presentation in Chapter 9.)
The resistance change is proportional to thestrain. Thus, if we could measure the resistance, we could
compute the strain. The resistance is measured by using the Wheatstone-bridge arrangement shown
in Fig. 2.11. When no load Fis present, the F 60-Hz a-c field
bridge is balanced (e, set to zero) by adjusting
R. Application of load causes a strain, a AR
Test
and thus unbalances the bridge, causing an
specimen
output voltage e, which is proportional to Strain
E. and can be mcasured on a meter or an goge
oscilloscope. The voltage e, is given by R
e

K,
Rb
e,= (GF)R eE, (2.1)
(R, + R,)
The desired input here is clearly the strain
E. which causes a proportional output voltage Fig. 2.11 Interfering input for strain-gage circuit.
e,.One interfering input, which often causes
trouble in such apparalus, is the 60-Hz magnetic field caused by nearby power lines, electric motors,
etc. This field induces voltages in the strain-gage circuit, causing output voltages e, even when the
strain is zero. Another interfering input is the gage temperature. If this varies, it causes a change in gage
resistance that willcause a voltage output even if there is nostrain. Temperature has another interfering
effect since it causes a differential expansion of the gage and the specimen, which gives rise to astrain
e and a voltage e even though no force F has been applied. Temperature also acts as a modity1ng npu
since the proportionality factor is sensitive to temperature. The battery voltage E, is another modity1ng
input. Both these are modifying inputs, since they tend to change the proportionality factor between the
desired input [, and the output e, or between an interfering input (gage temperature) and output e,.

2.51 Methods of Correction for Interfering and Modifying Inputs

nullifying or reducing the
In the design and/or useof mcasuring instruments, a number of methods for widely uscd.
effects of spurious inputs are available. We briefly describe some of the most
suggests, isthe sound
Method of Inherent Insensitivity The method of inherent insensitivity, as the name
inherently be sensitive to only the desireU
design plhilosophy that the elements of the instrunent should simpliciy
inputs. Although it is not always possible to design instruments for inherent insensitivity. the
terms of the general contiguration ot y
of this approach is worth considering wherever feasible. In as necarly equal to zero as possibie:
made
2.9, this approach requires that somehow E, and/or E,,M, Dbe may exist, they cannot attect the our
inputs)
Thus, even though i, and/or i, (lnterfering and modifying
to the strain gage of Fig. 2.1l. we might try to
put. As an example of the application of this concepttemperature coefficient of resistance, while reta
some gage material that exhibits an extremely low
 Generalized Configurations and Functional Descriptions of Measuring Instruments 23
material can be found, the problem of interfering
its sensitivity to srain. If sUch a temperature inputs
0N at least partially solved. In fact. 1n recent tmes such materials are commonly used in many strain
amilarty. in mechanical apparatus that must maintain accurate dimensions in the face of ambient:
temnerature changes, the use of a materialo very small temperature coefficient of expansion may be
blniul Twosuch materials are the metal alloy lnvar and the glass/ceramic Zerodur.
C
If stiffhess (such as
in aspring) must be temperature insensitive, cons1der the metal alloy Ni-Span
Method of High-goin Feedback Feedback Systems are generally used in control systems to minimize
the error between the reference input and the actual output of adynamic system. The same principle can
be used in instruments to minimize errors resulting in them by modifying them in such a way that they
coTect their errors automatically. The ifference between an open-loop system and afeedback or closed
joop svstenm, as applied to an instrument, is explained by means of the following example.
Suppose we wish to measure a voltagee by applying it to a motor, whose torque acts on a spring,
causing a displacement X, which may be measured on a calibrated scale. By proper design. the
displacement x, might be made proportional to the voltage e, according to
, = (KyK)e (2.2)
where K, and K, are appropriate constants. This arrangement, shown in Fig. 2.12(a). is called an open
loop system. If modifying inputs and i,n exist, they cause changes in K,, and K that lead to errors
in the relation between e andx. These errors are in direct
proportion to the changes in K,.
we can correct the errors if we have means of changing the value of these proportionality and Ksp
and
constants. But
that would be a difficult process,. if not impossible.
Suppose. instead. we construct a system as in Fig. 2.12(b). Here, the output .x, is measured by the
feedback device, which produces a voltage eo proportional to x,, This voltage is subtracted from the
voltage e, and the diference is applied to an amplifier, which drives the motor and thereby the input
produce x, We may write spring to

(e,-e)k KM. Ksy =(e,-K,)Kay K. Ks =X, (2.3)

e, Ky Ky K, = (1 + AM K. Kg Kex, (2.4)

1+ KaK MoKspKFB (2.5)

Suppose, now, that we design K,, to be very large (a "high-gain" system), so that
Then KKKK.
(2.6)

The significance of Ey. (2.6) is that the effect of variations in K,., K, and K, (is a result of noditying
nputs s and i,), on the relation between input e and output, has been made negligible. We
Ow reguire only thut K, stuy constant (unaffected by i,) in order to maintainonstant input-oitpul
calibration as shown by Eq. (2.6).
Tou may question whether much really has been gaincd by this somewhat elaborate schene, since
we merely transferred the requirements for stability from K, and K, oK, n pratiee, hwweer, this
[Link], "Selection of StrucluralMalerials lor Precision Devices," Prcivion tng.. vol.9, o,Januay
1987, pp. 3-6.
 24
Doebein's Meosurement Systems

since the amplifier sunn:

method often leads to geat improvements in accuraey. One reason is that,
most of the power needed. the feedback device can be designed with low power- handling capacity.
general. this leads to geater accuracy and linearity in the feedback-device characteristics A
In
thus, the fecdback system extracts less ener
the input signal c need carry oniy ncgligible power: system. This, of course, results in
from the measurd mediunm than the corresponding open-loop measuring instrument. Finallv :t
distortion of the measured quantity because of the presence of the
devices, each susceptible to its own spur
the open-loop chain consists of several (perhaps many) of high amplification and a stable, accurat
inputs, then all these bad effects can be negated by the use
feedback device.
'M2
|M1

FMD1 FMD.2

Voltage Torque |Displacement

Kmo K_p
Xo

Motor Spring
(a) Open-loop system

|'M3

FMD.3 FM.D.1 FM.D2

e
FAM Kmo Ksp Xo
eo
Amplifier Motor Spring

Feedback device

KFB

FMD4

|M4
(b) Closed-loop or feedback system

Fig. 2.12 Use of feedback to reduce effect of spurious inputs.

Before we pass on to other methods, we should mention that application of the feedback principle
is not without its own peculiar problems. The main one is dynamic instability, wherein excessively
high amplification leads to destructive oscillations. But we can design feedback systems that can avod
instability and give the desired response within the range of measurement of our interest. The study of
the design of feedback systems is awhole field in itself,and many texts treating this subject are availabl:
however, a brief discussion is presented in section 3.15.2.
 Generalized Configurations and Functional Descriptions of Measuring Instruments 25)
Method of Calculoted Output Corrections This method requires one to measure or estimate the
maenitudes of the interfering and/or modifying inputs and to know quantitatively how they af
fect the output. With this information, it is possible to calculate corrections, which may be added
Io or subtracted from the indicated output, so as to leave (ideally) only that component associat
ed with the desired input. Thus, in the manometer of Fig. 2.10, the effects of temperature on both
the calibrated scale's length and the density of mercury may be quite accurately computed if the
tomnerature is known. The local gravitational acceleration is also known for a given elevation and
latitude, so that this effect may be corrected by calculation. Since many measurement systems today
can afford to include a microcomputer to carry out various functions, if we also provide sensors
for the spurious inputs, the microcomputer can implement the method of calculated output correc
tions on an automatic basis, giving a so-called smart sensor. But in recent times, even a tiny-sized
micro-controller can be used to build smart sensors.

Method of SignalFiltering The method of signal filtering is based on the possibility of introducing certain
elements ("filters") into the instrument which in some fashion block the spurious signals, so that their effects
on the output are removed or reduced. The filter may be applied to any suitable signal in the instrument, be
it input. output, or intermediate signal. The concept of signal filtering is shown schematicaly in Fig. 2.13
for the cases of input and output filtering. The application to intermediate signals should be obvious. In
Fig. 2.13(a). the inputs i,and i, are caused to pass through filters, whose input-output relation is (ideally)
zero. Thus, i and i, are zero, even if i, and i,, are not zero. The concept of output filtering is illustrated in
Fig. 2.13(b). Here, the output o, though really one signal, is thought of as a
superposition of o, (output
due to interfering input), o, (output due to desired input), and o, (output due to modifying input). If it
is possible to construct filters that selectively blocko, and o,, but allow o, to pass through, this may be
symbolized as in Fig. 2.13(b) and results in o consisting entirely of o,. You may notice that output filter
ing is not very easy to understand since the output signal contaminated by unwanted signals cannot be
easily separated from them. But it is just a symbolic notation. In actual practice, for example in case of
signals occurring at different frequencies, the output signal when passed through a series of appropriate
frequency filters can remove the modifying and interfering outputs.
Filter
Zero F
Filter

Filter Fu! -Zero +

o'M
OD
iMZero
FM.0
output FMD |OM
Filter t
-Zerow
output
Fp Fp

(a) Input filtering (b) Output filering

Fig, 2.13 General principle of filtering.

The filters necessary in the application of this method may take several forms; they are best illustrated
by examples. If put directly in the path of a spurious input, a filter can be designed (ideally) to block
completely the passage of the signal. If, however, it is inserted at a point where the signal contains both
desired and spurious components, the filter must be designed to be selective. That is, it must pass the
desired components essentially unaltered, while effectively suppressing all others.
 26 Doebelin's Measurement Systems

Often it is necessary to attach delicate instruments to structures that vibrate. Electromechanical

devices for navigation and control of aircraft or missiles are outstanding examples. Figure 2.14(a) shows
how the interfering vibration input may be filtered out by use of suitable spring mounts. The mass-spring
system is actually a nmechanical filter, which passes on to the instrument only a negligible fraction of the
motion of the vibrating structure.
The interfering tilt-angle input to the manometer in Fig. 2.10(c) may be effectively filtered out by
means of the gimbal-mounting scheme of Fig. 2.14(b). If the gimbal bearings are essentially frictionless,
the rotations 0, and 6, cannot be communicated to the manometer; thus, it always hangs vertical.
Gimbal
SUspension

Vibration-sensitive
instrument

Vibration
isolation
mount

Vibrating
structure Interfering tilt-angle inputs
(a) (b)

-60-Hz field
Measuring junction

Thermal
Varying insulation
ambient
tenperature Reference
junction Magnetic
shield
(c) (d)
R

e
Strain plus
interference

o
Filler
characteristic
Strain only

2 Hz 60 Hz
Frequency
(e)

Fig. 2.14 (a)-fe) Examples of filtering.

 Generatized Configurations and Functional Descriptions of Measuring Instruments
27
,E c). the thermocouple referencc junction is shielded from
, n s of thermal insulation. Such an arrangement acls as a tlter forambient-temperature
temperature
fuctuations
or heat-flow inputs.
The strain-gage ciruit ot Fig. 2.14(d) is shielded from the
tal box of some sort. This solution corresponds to filteringinterfering 60-Hz field by enclosing it in
the interfering input. Another p0ssible
salution. which corresponds to selective filtering of the output, is shown in Fig. 2.14(c), For this Bnnro
be efective. it is essential that the frequencies in the desired signal
norate£ from thOse in the undesired component of the signal. In the occupy a range considerably
Strains to be measured are mainly steady and never vary more rapidly thanpresent example, suppose the
2Hz. Then, it is possible to
insert a simple RC filter, as shown, that will pass the desired signals, but almost
60-Hz interference. completely block the
Figure 2. 14(f) shows the pressure gage of Fig. 2.2, modified by the insertion of a
berween the source of pressure and the piston chamber. Such an arrangement is useful, flow restriction
for example, if
vou wish to measure only the average pressure in a large air tank that is
being supplied a reciprocating
by
compressor. The pulsations in the air pressure may be smoothed by the pneumatic filtering effect of the
Aow restriction and associated volume. The variation of the
output-input amplitude ratio lp /pl with
freauency is similar to that for the electrical RC filter of Fig. 2.14(e).
pressures are accurately measured. while rapid variations are strongly Thus, steady or slowly varying input
mav be in the form of a needle valve, which allows easy adjustment of the attenuated. The flow restriction
A"chopped" radiometer is shown in simplified form in filtering effect.
T of some body in terms of the infrared radiant energy
Fig. 2.14(g).This device senses the temperature
emitted. The emitted energy is focused on a
detector of some sort and causes the temperature T, of the detector, and
thus its output voltage e, to vary.
D: AAverage pressure
plus fluctuationsy

Pot Average pressure

only

Frequency
()
Infrared radiant source
"Chopper" Ts plus Tambient
Tambient

Rotating
shutter e
("chopper") ampl
eo4 Tambient frequencies Chopper
frequency Chopper T, only.

Frequency
(9)

Fig. 2.14 () and (g) examples of filtering (continued).

 28 Doebetins Measurement Sys tems
The difthculty with such devices is that the ambient temperature, as well as 7, affects T, This effect js
serious since the radiant energy to be measured causes very small changes in 7; thus, small ambjent
dnifts can completely mask the desired input. An ingenious solution to this problem interposes arotating
shuter hefween the radiant source and the detector, so that the desired input is "chopped." or modulated,
at aknown frequency. This frequency is chosen to be much higher than the frequencies at which ambient
drifts may occur. The output signal e of the detector thus is a superposition of slow ambient fluctuations
and ahigh-frequency wave whose anplitude varies in proportion to variations in T, Since the desired and
interfering components are thus widely separated in frequency, they may be selectively filtered. In this
case, we desire a filter that rejects constant and slowly varying signals, but faithfully reproduces rapid
variations. Such acharacteristic is typical of an ordinary ac amplifie, and since amplification is necessary
in such instruments in any case. the use of an ac amplifier as shown solves two problems at once.
In summing up the method of signal filtering, it may be said that, in general, it is usually possible to
design filters of mechanical, electrical, thermal, pneumatic, etc., nature, which separate signals according
to their frequency content in some specific manner. Figure 2.15 summarizes the most common useful
forms of such devices.

Outouta Method of Opposing Inputs The method ofopposing inputs consists

Input
No-pass of intentionally introducing into the instrument interfering and/or
modifying inputs that tend to cancel the bad effects of the unavoid
Frequency able spurious inputs. Figure 2.16 shows schematically the concept
Low-pass
for interfering inputs. The extension to modifying inputs should be
obvious. The intentionally introduced input is designed so that the
High-pass
Signal o,, and o, are essentially equal but act in the opposite sense;
thus the net contribution o, - o, to the output is nearly zero. This
method actually might be considered as a variation on the method
Band-pass
of calculated output corrections. However, the 'calculation" and ap
plication of the correction are achievèd automatically owing to the
Band-rejection
structure of the system, rather than by numerical calculation by a
((notch") human operator. Thus, the two methods are similar; however, the
distinction between them is a worthwhile one since it helps to orga
Fig. 215 Basic frequency nize your thinking in inventing new applications of these generalized
filter types. correction concepts.
Unavoidable Some examples of the method of opposing
Inilertering
input
inputs are shown in Fig. 2. 17. A millivoltmeter.
Fn shown in Fig. 2.17(a), is basically a current
Essentially
zero
sensitive device. However, as long as the total
Intentionally circuit resistance is constant, its scale can be
inroduCed
Oj1.0)
intertering calibrated in voltage, since voltage and current
are proportional. A moditying input here is
input

the ambient temperature, since it causes the

coilresistance R.,to change, thereby altering
Desired
Output the proportionality factor between current
nput and voltage. To correct tor this error, the
Compensating resistance R . is introduced
into the circuit, and its material is caretully
Fig, 2.16 Methods of opposing inputs. chosen to have a tenperature coetticient o
 Generalized Configurations and Functional Descriptions of Measuring Instruments 29

resistance opposite to that of R Thus, when the temperature changes, the total resistance of the circuit
is unafected and the calibration of the meter remains accurate.
Figure 2.17(6) shows a static-pressure -probe design due to L. Prandtl. As the fluid flows over the
surface of the probe, the velocity of the fluid must increase since these streamlines are longer than those
in the undisturbed flow. This velocity increase causes a drop in static pressure, so that a tap in the surface
of the probe gives an incorrect reading. This underpressure error varies with the distance d, of the tap
from the probe tip. Prandil recognized that the probe support will have a stagnation point (line) along
its front edge and that this overpressure will be felt upstream-the effect decreases as the distance d,
increases. By properly choosing distances d, and d, (by experimental test). these two effects can be made
exactBy to cancel. giving a true static-pressure value at the tap.
A device for the measurement of the mass fow rate of gases is shown in Fig. 2.17(c). The mass flow
rate of gas through an orifice may be found by measuring the pressure drop across the orifice, perhaps by
means of a U-tube manometer. Unfortunately, the mass flow rate also depends on the density of the gas,
which varies with pressure and temperature. Thus, the pressure-drop measuring device usually cannot
be calibrated to give the mass flow rate, since variations in gas temperature and pressure yield
different
mass flow rates for the same orifice pressure drop. The instrument shown in Fig. 2.17(c) overcomes this
problem in an ingenious fashion. The flow rate through the orifice also depends on its flow area. Thus,
if the fiow area could be varied in just the right way, this variation could compensate for
pressure and
temperature changes so that a given orifice pressure drop would always correspond to the same mass flow
rate. This is accomplished by attaching the specially shaped metering pin to a gas-filled bellows as shown.
When the temperature drops (causing an increase in density and therefore in mass flow rate), the gas in
the bellows contracts. which moves the metering pin into the orifice and thereby
reduces the flow area.

+0
Rcomp

N S Rool

Rcoil
Rcoil

Rcornp Temperature

Riolal
8,*Real Reame)
Iambnt
(Modtying input) Effect of Rconp

Etfecd of Ru

Fp
(Desired input)

Fig. 2.17 (o) Exanples of method of opposing Inputs tor u milivoltnete.

30 Doebelin's Measurement Systems

Free stream Stagnation pont

static pressure Pstream Static pressure tap Pprobe
mass density P
velocity V
d d Slagnation
point

2
Pstream
Probe
support
Overpressure due to
Underpressure due to stagnation point (|ine)
increased velocity of
flow over surface of on probe support
probe

V
(Intertering
input) Manometer

Pstrea Pprobe
Fp
(Desired input)

Fig. 2.17 (b) Example of method of opposing inputs for a static-pressure probe.

Effect of
Fuo density
Metering Gas-filled Gas temperature change
pin bellows
Orfice -Effect of area
(Modifying input)
change

Fiow D0000000

Manometer
Mass flow rate reading
Fp
(Desired input)

Manometer
Effect of density
change
Gas pressure
Etfect of
(Moditying input) area
change

Fig. 2.17 (c) Method of opposing inputs for a mass flow meter.

This returns the mass How rale to its proper value. Similarly, should the pressure of the tlowing gas
increase, causing an increase in density and mass flow rale, the gas-filled bellows would be compressed
again, reducing the flow area and correcling the nmass flow rale. The proper shape for the netering pin is
revealed by a detailed analysis of the system.
Afinal example of themethod of opposing inputs is the rate gyrcUscope of Pig, 2.17). Such devices
are widely used in aerospace vehicles lor the generalion of stabilizationsiguals i the control system
when they undergo rotational movenents. The action of the device s that a vchicle rotalion at angulu
 Generalized Configurations and Functional Descriptions of Measuring Instruments 31

Silicone fluid fill Damping gap

Nylon
cylinder
(Ospin
Case
Torsion
bar

Gimbal

aspin

Ambient -Effect of viscosity

FM1 D1 change
temperature
(Modifying input) -Effect of gap thickness
FM D2 change

e, Angular viscosity
(Desired input) Fo

Fig. 2.17 (d) Method of opposing inputs for a rate gyrOScope.

velocity 6,causes a proportional displacement , of the gimbal relative to the case. This rotation is
measured by some motion pickup (not shown in Fig. 2.17(d). Thus, a signal proportional to vehicle
angular velocity is available, and this is useful in stabilizing the vehicle. When the vehicle undergoes
rapid motion changes. however, the angle @, tends to oscillate. giving an incorrect angular-velocity
signal. To control these oscillations. the gimbal rotation is damped by the shearing action of a viscous
silicone ffuid in a narrow damping gap. The damping effect varies with the viscosity of the fluid and the
thickness of the damping gap. Although the viscosity of the silicone fluid is fairly constant, it does vary
with ambient temperature, causing an undesirable change in damping characteristics. To compensate
for this, a nylon cylinder is used in the gyro of Fig. 2.17(d). When the temperature increases, viscosity
drops, causing a loss of damping. Simultaneously, however, the nylon cylinder expands, narrowing the
damping gap and thus restoring the damping to its proper value. By proper choice of materials and
geometry. the two effects can be made to very nearly cancel over the operating temperature range of
the equipment.