Lecture 4

Signal Generators
Multi-vibrators
 Outline
• Basic Principles of Sinusoidal Oscillators.
• OP Amp-RC Oscillator Circuits.
• LC and Crystal Oscillators.
• Bistable Multivibrators.
• Generation of Square and Triangular Waveforms Using
Astable Multivibrators.
• Generation of a Standarized Pulse. The Monostable
Multivibrator.
• Integrated-Circuit Timers.
• Nonlinear Waveform-Shaping Circuits.
• Precision Rectifier Circuits.
 13.4 Bistable Multivibrators
• In this section we begin the study of waveform-generating circuits of the other
type-nonlinear oscillators or function generators. These devices make use of a
special class of circuits known as multivibrators. As mentioned earlier, there are
three types of multivibrator: bistable, monostable, and astable. This section is
concerned with the first, the bistable multivibrator.
• As its name indicates, the bistable multivibrator has two stable states. The circuit
can remain in either stable state indefinitely and moves to the other stable state
only when appropriately triggered.
 13.4.1 The Feedback Loop
• Bistability can be obtained by connecting a dc amplifier in a positive-feedback loop
having a loop gain greater than unity. Such a feedback loop is shown in Fig.
13.17; it consists of an op amp and a resistive voltage divider in the positive-
feedback path. To see how bistability is obtained, consider operation with the
positive input terminal of the op amp near ground potential. This is a reasonable
starting point, since the circuit has no external excitation. Assume that the
electrical noise that is inevitably present in every electronic circuit causes a small
positive increment in the voltage v+. This incremental signal will be amplified by
the large open-loop gain A of the op amp, with the result that a much greater signal
will appear in the op amp's output voltage Va.
• The voltage divider (R1, R2) will feed a fraction  = R1/(R1 + R2) of the output signal
back to the positive input terminal of the op amp. If A is greater than unity, as is
usually the case, the fed-back signal will be greater than the original increment in
v+. This regenerative process continues until eventually the op amp sat­urates with
its output voltage at the positive saturation level, L+. When this happens, the
voltage at the positive input terminal, v+, becomes L+R1/(R1 + R2), which is positive
and thus keeps the op amp in positive saturation. This is one of the two stable
states of the circuit.
• In the description above we assumed that when v+ was near zero volts, a positive
incre­ment occurred in v+. Had we assumed the equally probable situation of a
negative incre­ment, the op amp would have ended up saturated in the negative
direction with Vo = L-_ and v+ = L-R1/(R1 + R2). This is the other stable state.
• We thus conclude that the circuit of Fig. 13.17 has two stable states, one with the
op amp in positive saturation and the other with the op amp in negative saturation.
The circuit can exist in either of these two states indefinitely. We also note that the
circuit cannot exist in the state for which v+ = 0 and Vo = 0 for any length of time.
• This is a state of unstable equilibrium (also known as a metastable state); any
disturbance, such as that caused by electrical noise, causes the bistable circuit to
switch to one of its two stable states. This is in sharp contrast to the case when
the feedback is negative, causing a virtual short circuit to appear between the op
amp's input terminals and maintaining this virtual short-circuit in the face of
disturbances. A physical analogy for the operation of the bistable circuit is
depicted in Fig. 13.18.
• FIGURE 13.17 A positive-feedback loop capable of bistable operation.
• FIGURE 13.18 A physical analogy for the operation of the bistable circuit. The ball cannot remain
at the top of the hill for any length of time (a state of unstable equilibrium or metastability); the
inevitably present disturbance will cause the ball to fall to one side or the other, where it can remain
indefinitely (the two stable states).
 13.4.2 Transfer Characteristics of the Bistable
Circuit
• The question naturally arises as to how we can make the bistable circuit of Fig.
13.17 change state. To help answer this crucial question, we derive the transfer
characteristics of the bistable. Reference to Fig. 13.1 indicates that either of the
two circuit nodes that are connected to ground can serve as an input terminal.
We investigate both possibilities.
• Figure 13.19(a) shows the bistable circuit with a voltage VI applied to the inverting
input terminal of the op amp. To derive the transfer characteristic Vo-VI assume
that Vo is at one of its two possible levels, say L+, and thus v+= L+.
• Now as VI is increased from 0 V we can see from the Circuit that nothing happens
until VI reaches a value equal to V+ (i.e., L+). As VI begins to exceed this value, a
net negative voltage develops between the input terminals of the op amp. This
voltage is amplified by the open-loop gain of the op amp, and thus Vo goes
negative. The voltage divider in turn causes V+ to go negative, thus increasing the
net negative input to the op amp and keeping the regenerative process going.
This process culminates in the op amp saturating in the negative direction: that is,
with Vo = L- and, correspondingly, V+ = L-. It is easy to see that increasing VI
further has no effect on the required state of the bistable circuit. Figure 13.19(b)
shows the transfer characteristic for increasing VI. Observe that the characteristic
is that of a comparator with a threshold voltage denoted VTH, where VTH = L+.
• Next consider what happens as VI is decreased. Since now, v+ = L-, we see that
the cir­cuit remains in the negative-saturation state until VI goes negative to the
point that it equals L-. As VI goes below this value, a net positive voltage appears
between the op amp's input terminals. This voltage is amplified by the op-amp
gain and thus gives rise to a positive volt­age at the op amp's output. The
regenerative action of the positive-feedback loop then sets in and causes the
circuit eventually to go to its positive-saturation state, in which Vo = L+ and v+ = L+.
The transfer characteristic for decreasing VI is shown in Fig, B.19(c}. Here again
we observe that the characteristic is that of a comparator, but with a threshold
voltage VTH = L-.
• The complete transfer characteristics, Vo-VI, of the circuit in Fig. 13.19(a) can be
obtained by combining the characteristics in Fig. 13.l9(b) and (c), as shown in Fig.
13.19(d). As indi­cated, the circuit changes state at different values of VI depending
on whether VI is increasing or decreasing. Thus the circuit is said to exhibit
hysteresis; the width of the hysteresis is the dif­ference between the high threshold
VTH and the low threshold VTL. Also note that the bistable circuit is in effect a
comparator with hysteresis. As will be shown shortly, adding hysteresis to a
comparator's characteristics can be very beneficial in certain applications. Finally,
observe that because the bistable circuit of Fig. 13.19 switches from the positive
state (vo = L+) to the negative state (vo = L-) as VI is increased past the positive
threshold VTH, the circuit is said to be inverting. A bistable circuit with a
noninverting transfer characteristic will be presented shortly.
• FIGURE 13.19 (a) The bistable circuit of Fig. 13.17 with the negative input terminal of the op amp
dis­connected from ground and connected to an input signal V I. (b) The transfer characteristic of
the circuit in (a) for increasing VI. (c) The transfer characteristic for decreasing VI. (d) The
complete transfer characteristics.
 13.4.3 Triggering the Bistable Circuit
• Returning now to the question of how to make the bistable circuit change state, we
observe from the transfer characteristics of Fig. 13.l9(d) that if the circuit is in the
L+ state it can be switched to the L- state by applying an input VI of value greater
than VTH = L+. Such an input causes a net negative voltage to appear between
the input terminals of the op amp, which ini­tiates the regenerative cycle that
culminates in the circuit switching to the L- stable state. Here it is important to note
that the input VI merely initiates or triggers regeneration. Thus we can remove VI
with no effect on the regeneration process. In other words, VI can be simply a
pulse of short duration.
• The input signal VI is thus referred to as a trigger signal,
or simply a trigger.
• The characteristics of Fig. 13 .19(d) indicate also that the
bistable circuit can be switched to the positive state (vo =
L+) by applying a negative trigger signal VI of magnitude
greater than that of the negative threshold VTL.
 13.4.4 The Bistable Circuit as a Memory
Element
• We observe from Fig. l3.l9(d) that for input voltages in the range VTL < VI < VTH, the
output can be either L+ or L-, depending on the state that the circuit is already in.
Thus, for this input range, the output is determined by the previous value of the
trigger signal (the trigger signal that caused the circuit to be in its current state).
Thus the circuit exhibits memory. Indeed, the bistable multivibrator is the basic
memory element of digital systems, as we have seen in Chapter 11. Finally, note
that in analog circuit applications, such as the ones of concern to us in this chapter,
the bistable circuit is also known as a Schmitt trigger.
 13.4.5 A Bistable Circuit with Non-inverting Transfer
Characteristics
• The basic bistable feedback loop of Fig. 13.17 can be used to derive a circuit with
non-invert­ing transfer characteristics by applying the input signal VI (the trigger
signal) to the terminal of R1 that is connected to ground. The resulting circuit is
shown in Fig. 13.20(a). To obtain the transfer characteristics we first employ
superposition to the linear circuit formed by R1 and R2, thus expressing V+ in
terms of VI and Vo as

 R2   R1 
v  vI    vo   (13.28)
 R1  R2   R1  R2 
• From this equation we see that if the circuit is in the positive stable state with vo =
L+, posi­tive values for VI will have no effect. To trigger the circuit into the L- state,
VI must be made negative and of such a value as to make V+ decrease below zero.
Thus the low threshold VTL can be found by substituting in Eq. (13.28) Vo = L+, V+
= 0, and VI = VTL. The result is

 R 
VTL   L  1  (13.29)
 R2 
• Similarly, Eq. (13.28) indicates that when the circuit is in the negative-output state
(vO = L-), negative values of VI will make V+ more negative with no effect on
operation. To initiate the regeneration process that causes the circuit to switch to
the positive state, V+ must be made to go slightly positive. The value of VI that
causes this to happen is the high threshold volt­age VTH, which can be found by
substituting in Eq. (13.28) Vo = L- and v+ = 0. The result is

 R 
VTH   L  1  (13.30)
 R2 
• The complete transfer characteristic of the circuit of Fig. 13.20(a) is displayed in
Fig. 13.20(b). Observe that a positive triggering signal VI (of value greater than
VTH) causes the circuit to switch to the positive state (vo goes from L- to L+). Thus
the transfer characteristic of this circuit is non-inverting.
• FIGURE 13.20 (a) A bistable circuit derived from the positive-feedback loop of Fig. 13.17 by
applying VI through R1. (b) The transfer characteristic of the circuit in (a) is noninverting.
(Compare it to the inverting characteristic in Fig. 13.19d.)
 13.4.6 Application of the Bistable Circuit as a
Comparator
• The comparator is an analog-circuit building block that is used in a variety of
applications ranging from detecting the level of an input signal relative to a preset
threshold value, to the design of analog-to-digital (A/D) converters (see Section
9.1). Although one normally thinks of the comparator as having a single threshold
value (see Fig. 13.21a), it is useful in many applications to add hysteresis to the
comparator characteristics. If this is done, the comparator exhibits two threshold
values, VTL and VTH, symmetrically placed about the desired reference level, as
indicated in Fig. 13.21(b).
• Usually VTH and VTL are separated by a small amount, say 100 mV.
• To demonstrate the need for hysteresis we consider a common application of
compara­tors. It is required to design a circuit that detects and counts the zero
crossings of an arbi­trary waveform. Such a function can be implemented using a
comparator whose threshold is set to 0 V. The comparator provides a step change
at its output every time a zero crossing occurs. Each step change can be used to
generate a pulse, and the pulses are fed to a counter circuit.
• Imagine now what happens if the signal being processed has-as it usually does
have ­interference superimposed on it, say of a frequency much higher than that of
the signal.
• It follows that the signal might cross the zero axis a number of times around each
of the zero­-crossing points we are trying to detect, as shown in Fig: 13.22. The
comparator would thus change state a number of times at each of the zero
crossings, and our count would obviously be in error. However, if we have an idea
of the expected peak-to-peak amplitude of the interfer­ence, the problem can be
solved by introducing hysteresis of appropriate width in the com­parator
characteristics. Then, if the input signal is increasing in magnitude, the
comparator with hysteresis will remain in the low state until the input level
exceeds the high threshold VTH.
• Subsequently the comparator will remain in the high state even if, owing to interfer­
ence, the signal decreases below VTH. The comparator will switch to the low state
only if the input signal is decreased below the low threshold VTL. The situation is
illustrated in Fig. 13.22, from which we see that including hysteresis in the
comparator characteristics provides an effective means for rejecting interference
(thus providing another form of filtering).
• FIGURE 13.21 (a) Block diagram representation and Transfer characteristic for a comparator
having a reference, or threshold, voltage VR, (b) Comparator characteristic with hysteresis.
• FIGURE 13.22 Illustrating the use of hysteresis in the comparator characteristics as a means of
rejecting interference.
• FIGURE 13.23 Limiter circuits are used to obtain more precise output levels for the bistable circuit.
In both circuits the value of R should be chosen to yield the current required for the proper
operation of the zener diodes. (a) For this circuit L+ = VZ1 + VD and L- = -( VZ2 + VD), where VD is
the forward diode drop. (b) For this circuit L+ = VZ+VD1+VD2 and L-= - (VZ+VD1 +VD2).
 13.4.7 Making the Output levels More
Precise
• The output levels of the bistable circuit can be made more
precise than the saturation volt­ages of the op amp are by
cascading the op amp with a limiter circuit (see Section
3.6 for a discussion of limiter circuits). Two such
arrangements are shown in Fig. 13.23.
 13.5 Generation Of Square And Triangular Waveforms Using
Astable Multivibrators
• A square waveform can be generated by arranging for a bistable multivibrator to
switch states periodically. This can be done by connecting the bistable
multivibrator with an RC circuit in a feedback loop, as shown in Fig. 13.24(a).
Observe that the bistable multivibrator has an inverting transfer characteristic and
can thus be realized using the circuit of Fig. 13.19(a). This results in the circuit of
Fig. 13.24(b). We shall show shortly that this circuit has no sta­ble states and thus
is appropriately named an astable multivibrator.
 13.5.1 Operation of the Astable
Multivibrator
• To see how the astable multivibrator operates, refer to Fig. 13.24(b) and let the
output of the bistable multivibrator be at one of its two possible levels, say L+.
Capacitor C will charge toward this level through resistor R. Thus the voltage
across C, which is applied to the nega­tive input terminal of the op amp and thus is
denoted v-, will rise exponentially toward L+ with a time constant  = CR.
Meanwhile, the voltage at the positive input terminal of the op amp is v+ = L+.
This situation will continue until the capacitor voltage reaches the positive
threshold VTH = L+ at which point the bistable multivibrator will switch to the other
stable state in which Vo = L- and v+ = L-. The capacitor will then start discharging,
and its volt­age, v-, will decrease exponentially toward L-.
• This new state will prevail until v- reaches the negative threshold VTL =
L-, at which time the bistable multivibrator switches to the positive-
output state, the capacitor begins to charge, and the cycle repeats
itself.
• From the preceding description we see that the astable circuit
oscillates and produces a square waveform at the output of the op
amp. This waveform, and the waveforms at the two input terminals of
the op amp, are displayed in Fig. 13.24(c).
• The period T of the square wave can be found as follows: During the charging
interval T1 the voltage v- across the capacitor at any time t, with t = 0 at the
beginning of T1, is given by (see Appendix D)

 t
v  L  L   L e 

• where  = CR. Substituting v- = L+ at t = T1 gives

  
 1    L  
 L
T1   ln     (13.31)
1  
 
 
 
• Similarly, during the discharge interval T2 the voltage v- at any time t, with t = 0 at
the beginning of T2, is given by
 t
v  L  L   L e 

• Substituting v- = L- at t = T2 gives

  
 1    L  
 L
T2   ln     (13.32)
1  
 
 
 
• Equations (13.31) and (13.32) can be combined to obtain the period T= T1 + T2.
Normally, L+ = -L-, resulting in symmetrical square waves of period T given by

1   
T  2 ln   (13.33)
1   
• Note that this square-wave generator can be made to have variable frequency by
switch­ing different capacitors C (usually in decades) and by continuously adjusting
R (to obtain continuous frequency control within each decade of frequency). Also,
the waveform across C can be made almost triangular by using a small value for
the parameter . However, triangu­lar waveforms of superior linearity can be easily
generated using the scheme discussed next.
• Before leaving this section, however, note that although the astable circuit has no
stable states, it has two quasi-stable states and remains in each for a time interval
determined by the time constant of the RC network and the thresholds of the
bistable multivibrator.
• FIGURE 13.24 (a) Connecting a bistable multi vibrator with inverting transfer characteristics in a
feed­back loop with an RC circuit results in a square-wave generator.
• FIGURE 13.24 (Continued) (b) The circuit obtained when the bistable multivibrator is implemented
with the circuit of Fig. 13.l9(a). (c) Waveforms at various nodes of the circuit in (b). This circuit is
called an astable multivibrator.
 13.5.2 Generation of Triangular
Waveforms
• The exponential waveforms generated in the astable circuit of Fig. 13.24 can be
changed to triangular by replacing the low-pass RC circuit with an integrator. (The
integrator is, after all, a low-pass circuit with a comer frequency at dc.) The
integrator causes linear charging and discharging of the capacitor, thus providing a
triangular waveform. The resulting circuit is shown in Fig. 13.25(a). Observe that
because the integrator is inverting, it is necessary to invert the characteristics of
the bistable circuit. Thus the bistable circuit required here is of the non-inverting
type and can be implemented using the circuit of Fig. 13.2.
• We now proceed to show how the feedback loop of Fig. 13.25(a) oscillates and
gener­ates a triangular waveform V1 at the output of the integrator and a square
waveform V2 at the output of the bistable circuit: Let the output of the bistable
circuit be at L+. A current equal to L+/ R will flow into the resistor R and through
capacitor C, causing the output of the inte­grator to linearly decrease with a slope of
-L+/CR, as shown in Fig. 13.25(c). This will continue until the integrator output
reaches the lower threshold VTL of the bistable circuit, at which point the bistable
circuit will switch states, its output becoming negative and equal to L-. At this
moment the current through R and C will reverse direction, and its value will
become equal to L-/R.
• It follows that the integrator output will start to increase linearly with a positive
slope equal to L-/CR. This will continue until the integrator output voltage
reaches the positive threshold of the bistable circuit, VTH. At this point the bistable
circuit switches, its output becomes positive (L+), the current into the integrator
reverses direction, and the output of the integrator starts to decrease linearly,
beginning a new cycle.
• FIGURE 13.25 A general scheme for generating triangular and square waveforms.
• From the discussion above it is relatively easy to derive an expression for the
period T of the square and triangular waveforms. During the interval T1 we have,
from Fig. 13.25(c),

VTH  VTL L

T1 CR

• from which we obtain

V  VTL 
T1  C R  TH  (13.34)
 L 
• Similarly, during T2 we have

VTH  VTL  L

T2 CR

• from which we obtain

V  VTL 
T2  C R  TH  (13.35)
  L 

• Thus to obtain symmetrical square waves we design the bistable circuit to

have L+ = - L-.
 13.6 Generation Of A Standardized Pulse - The
Monostable Multivibrator
• In some applications the need arises for a pulse of known
height and width generated in response to a trigger signal.
Because the width of the pulse is predictable, its trailing edge
can be used for timing purposes-that is, to initiate a particular
task at a specified time. Such a standardized pulse can be
generated by the third type of multi-vibrator, the monostable
multivibrator.
• The monostable multi-vibrator has one stable state in which it can remain
indefinitely. It also has a quasi-stable state to which it can be triggered and in
which it stays for a predeter­mined interval equal to the desired width of the output
pulse. When this interval expires, the monostable multi-vibrator returns to its
stable state and remains there, awaiting another trig­gering signal. The action of a
monostable multi-vibrator has given rise to its alternative name, the one shot.
• Figure 13.26(a) shows an op-amp monostable circuit. We observe that this circuit
is an augmented form of the astable circuit of Fig. 13.24(b).
• Specifically, a clamping diode D1 is added across the capacitor C1 and a trigger
circuit composed of capacitor C2, resistor R4, and diode D2 is connected to the non-
inverting input terminal of the op amp. The circuit oper­ates as follows: In the
stable state, which prevails in the absence of the triggering signal, the output of the
op amp is at L+ and diode D1 is conducting through R3 and thus clamping the
voltage VB to one diode drop above ground. We select R4 much larger than R1, so
that diode D2 will be conducting a very small current and the voltage Vc will be very
closely deter­mined by the voltage divider R1, R2. Thus Vc = L+ where  =
R1/(R1+R2). The stable state is maintained because L+ is greater than VD1.
• Now consider the application of a negative-going step at the trigger input and refer
to the signal waveforms shown in Fig. 13.26(b). The negative triggering edge will
be coupled to the cathode of diode D2 via capacitor C2, and thus D2 conducts
heavily and pulls node C down. If the trigger signal is of sufficient height to cause
Vc to go below vB, the op amp will see a net negative input voltage and its output
will switch to L-. This in turn will cause Vc to go negative to L-, keeping the op
amp in its newly acquired state. Note that D2 will then cut off, thus isolating the
circuit from any further changes at the trigger input terminal.
• FIGURE 13.26 (a) An op-amp monostable circuit. (b) Signal waveforms in the circuit of (a).
• The negative voltage at A causes D1 to cut off, and C1 begins to discharge
exponentially toward L- with a time constant C1R3. The monostable multi-vibrator
is now in its quasi-­stable state, which will prevail until the declining VB goes below
the voltage at node C, which is L-. At this instant the op-amp output switches
back to L+ and the voltage at node C goes back to L+. Capacitor C1 then charges
toward L+ until diode Dl turns on and the circuit returns to its stable state.
• From Fig. 13.26(b), we observe that a negative pulse is generated at the output
during the quasi-stable state.
• The duration T of the output pulse is determined from the exponential waveform of
VB.
 t
vB (t )  L  L  VD1 e C1R3

• by substituting vB(T) = L-,

 T
vB (T )  L  L  VD1 e C1R3

• For VD1 << L-, this equation can be approximated by

 1 
T  C1 R3 ln   (13.37)
1   
• Finally, note that the monostable circuit should not be
triggered again until capacitor C1 has been recharged to
VDl, otherwise the resulting output pulse will be shorter
than normal. This recharging time is known as the
recovery period. Circuit techniques exist for shortening
the recovery period.