Interfacing Methods and Circuits

1
Interfacing Methods and Circuits

• Chapter 11

2
Introduction

• A sensor/actuator can rarely operate on its own.
• Exceptions exist (bimetal sensors)
• Often a circuit of some sort is involved.
• can be as simple as adding a power source or a
  transformer
• can involve amplification, impedance matching,
  signal conditioning and other such functions.
• often, a digital output is required or desirable
  so that an A/D may be needed
• The same considerations apply to actuators

3
Introduction

• The considerations of interfacing should be part
  of the process of selecting a device for a
  particular application since this can simplify
  the process considerably.
• Example if a digital device exists it would be
  wasteful to select an equivalent analog device
  and add the required circuitry to convert its
  output to a digital format.
• The likely outcome is a more cumbersome,
  expensive system which may take more time to
  produce.
• Alternative sensing strategies and alternative
  sensors should always be considered before
  settling on a particular solution

4
Introduction

• Many types of sensors and actuators based on very
  different principles
• There are commonalities between them in terms of
  interfacing requirements
• Most sensors outputs are electric (voltage,
  current, resistance)
• These can be measured directly after proper
  signal conditioning and, perhaps, amplification.
• If the output is a capacitance or an inductance -
  require additional circuitry such as oscillators

5
Introduction

• There is a large range of signal levels in
  sensors.
• A thermocouples output is of the order of
  microvolts
• An LVDT may easily produce 5V AC.
• A piezoelectric actuator may require a few
  hundred volts to operate (very little current)
• A solenoid valve operates at perhaps 12-24V with
  currents that may exceed a few amperes.
• How does one measure these signals?

6
Introduction

• The circuitry required to drive and to interface
  them to, say a microprocessor are vastly
  different
• Require special attention on the part of the
  engineer.
• Must consider such issues as response (electrical
  and mechanical), spans and power dissipation as
  well as power quality and availability.
• Example Systems connected to the grid and
  cordless systems have different requirements and
  considerations in terms of operation and safety.

7
Purpose

• Discuss general issues associated with
  interfacing
• Outline general interfacing circuits the engineer
  is likely to be exposed to.
• No general discussion however can prepare one for
  all eventualities
• It should be recognized that there are both
  exceptions to the rules and extensions to the
  methods discussed here.

8
Purpose

• Example an A/D is a simple if not inexpensive
  method of digitizing a signal for the purpose
  of interfacing
• This approach however may not be necessary and
  too expensive in some cases.
• Suppose the hall element senses the teeth on a
  gear.
• The signal from the hall element is an ac voltage
  - only the peaks are necessary to sense the
  teeth.
• In this case a simple peak detector may be
  adequate.
• An A/D converted will not provide any additional
  benefit and is a much more complex and expensive
  solution.
• On the other hand, if a microprocessor is used
  and an A/D is available it may be acceptable to
  use it for this purpose

9
Content

• Operational amplifiers and power amplifiers
• A/D and D/A conversion circuits
• Bridge circuits
• Data transmission
• Excitation circuits
• Noise and interference

10
Amplifiers

• An amplifier is a device that amplifies a signal
  almost always a voltage
• The low voltage output of a sensor, say of a
  thermocouple, may be amplified to a level
  required by a controller or a display.
• Amplification may be quite large sometimes of
  the order of 106 or it may be quite small or even
  smaller than one, depending on the need of the
  sensor.

11
Amplifiers

• Amplifiers can also be used for impedance
  matching purposes even when no amplification is
  needed
• May be used for the sole purpose of signal
  conditioning, signal translation or for isolation
• Power amplifiers, which usually connect to
  actuators, serve similar purposes beyond
  providing the power necessary to drive the
  actuator.
• Amplifiers can be very simple a transistor with
  its associated biasing network or may involve
  many amplification stages of varying complexity.
• Amplifiers are sometimes incorporated in the
  sensor

12
Amplifiers

• We will use the operational amplifier as the
  basic building block for amplification.
• Operational amplifiers are basic devices and may
  be viewed as components.
• An engineer, especially when interfacing sensor
  is not likely to dwell into the design of
  electronic circuits below the level of
  operational amplifiers.
• Although there are instances where this may be
  done to great advantage, op-amps are almost
  always a better, less expensive and higher
  performance choice.
• Same idea for power amplifiers

13
Operational Amplifiers

• Operational amplifier is a fairly complex
  electronic circuit but
• It is based on the idea of the differential
  voltage amplifier shown in Figure 11.1.
• Based on simple transistors,
• The output is a function of the difference
  between the two inputs.
• Assuming the output to be zero when both inputs
  are at zero potential, the operation is as
  follows

14
Differential amplifier
15
Operational amplifier

• When the voltage on the base of Q1 increases, its
  bias increases while that on Q2 decreases because
  of the common emitter resistance.
• Q1 conducts more than Q2 and the output is
  positive with respect to ground.
• If the sequence is inverted, the opposite occurs.
  
• If, both inputs increase or decrease equally,
  there will be no change in output.

16
Operational amplifier

• An operational amplifier is much more complex
  than this but operates on the same principle.
• It contains additional circuitry (such as
  temperature and drift compensation, output
  amplifiers, etc.)
• These are of no interest to us other than the
  fact that they affect the specifications of the
  op-amp.
• There are also various modifications to op-amps
  that allow them to operate under certain
  conditions or to perform specific functions.

17
Operational amplifier

• Some are low noise devices
• Others can operate from a single polarity source.
  
• If the input transistors are replaced with FETs,
  the input impedance increases considerably
  requiring even lower input currents from sensors
• These are important but are variations of the
  basic circuit.
• We will consider it as a simple block shown in
  Figure 11.2 and discuss its general properties
  based on this diagram

18
The operational amplifier
19
Op-Amps - properties

• Differential voltage gain the amplification of
  the op-amp of the difference between the two
  inputs
• Also called the open loop gain
• in a good amplifier it should be as high as
  possible.
• Gains of 106 or higher are common.
• An ideal amplifier is said to have infinite gain.

20
Op-Amps - properties

• Common-mode voltage gain.
• By virtue of the differential nature of the
  amplifier, this gain should be zero.
• Practical amplifiers may have a small common mode
  gain because of the mismatch between the two
  channels but this should be small.
• Common mode voltage gain is indicated as Acm.
• The concept is shown in Figure 11.3.

21
Common mode signal an output
22
Op-amps - properties

• More common to specify the term Common Mode
  Rejection Ratio (CMRR)
• CMRR is the ratio between Ad and Acm

In an ideal amplifier this is infinite. A good
amplifier will have a CMRR that is very high
23
Op-amps - properties

• Bandwidth the range of frequencies that can be
  amplified.
• Usually the amplifier operates down to dc and has
  a flat response up to a maximum frequency at
  which output power is down by 3dB.
• An ideal amplifier will have an infinite
  bandwidth.
• The open gain bandwidth of a practical amplifier
  is fairly low
• A more important quantity is the bandwidth at the
  actual gain.

24
Op-amps - properties

• This may be seen in Figure 11.4
• The lower the gain, the higher the bandwidth.
• Data sheets therefore cite what is called the
  gain-bandwidth product.
• This indicates the frequency at which the gain
  drops to one and is also called the unity gain
  frequency.
• In Figure 11.4
• BW (open loop) is 2.5 kHz
• Unity Gain Frequency is 5 MHz

25
Bandwidth of op-amp
26
Op-amps - properties

• Slew Rate the rate of change of the output in
  response to a change in input, given in V/s.
• If a signal at the input changes faster than the
  slew rate, the output will lag behind it and a
  distorted signal will be obtained.
• This limits the usable frequency range of the
  amplifier.
• For example, an ideal square wave will have a
  rising and dropping slope at the output defined
  by the slew rate.

27
Op-amps - properties

• Input impedance the impedance seen by the sensor
  when connected to the op-amp.
• Typically this impedance is high (ideally
  infinite)
• It varies with frequency.
• Typical impedances for conventional amplifiers is
  at least 1 M? but it can be of the order of
  hundreds of M? for FET input amplifiers.
• This impedance defines the current needed to
  drive the amplifier and hence the load it
  represents to the sensor.

28
Op-amps - properties

• Output impedance the impedance seen by the load.
  
• Ideally this should be zero since then the output
  voltage of the amplifier does not vary with the
  load
• In practice it is finite and depends on gain.
• Usually, output impedance is given for open loop
  whereas at lower gains the impedance is lower.
• A good amplifier will have an output resistance
  lower than 1?.

29
Op-amps - properties

• Temperature and noise refer to variations of
  output with temperature and noise characteristics
  of the device respectively.
• These are provided by the data sheet for the
  op-amp and are usually very small.
• For low signals, noise can be important while
  temperature drift, if unacceptable must be
  compensated for through external circuits.

30
Op-amps - properties

• Power requirements. The classical op-amp is built
  so that its output can swing between Vcc
• Dual supply operation is common in op-amps
• The limits can be as low as 3V (or lower) and as
  high as 35V (sometimes higher).
• Many op-amps are designed for single supply
  operation of less than 3V and some can be used in
  single supply or dual supply modes.

31
Op-amps - properties

• Current through the amplifier is an important
  consideration, especially the quiescent current
  (no load)
• Gives a good indication of power needed to drive
  it.
• Particularly important in battery operated
  devices. The current under load will depend on
  the application but it is usually fairly small
  a few mA.
• In selection of a power supply for op-amps, care
  should be taken with the noise that the power
  supply can inject into the amplifier.
• The effect of the power supply on the amplifier
  is specified through the power supply rejection
  ratio (PSRR) of the specific amplifier.

32
Op-amps - data sheets

• The 741 op-amp is an older, general purpose
  amplifier.
• It is a fairly low performance device but is
  characteristic of the low-end amplifiers.
• Very common and quite suitable for many
  applications. LM741.PDF

33
Op-amps - data sheets

• The TLC27L2C is a dual, low power op-amp, suited
  for battery operated devices
• Part of a series of amplifiers using FETs as
  input transistors TLC27L2C

34
Inverting and noninverting amplifiers

• Performance of the amplifier depends on how it is
  used and, in particular on the gain of the
  amplifier.
• In practical circuits, the open loop gain is not
  useful and a specific gain must be established.
• For example, we might have a 50mV output
  (maximum) from a sensor and require this output
  to be amplified, say by 100 to obtain 5V
  (maximum) for connection to an A/D.
• This can be done with one of the two basic
  circuits shown in Figure 11.5, establish a means
  of negative feedback to reduce the gain

35
Inverting op-amp
36
Non-inverting op-amp
37
Inverting op-amp

• The output is inverted with respect to the input
  (180? out of phase).
• The feedback resistor, Rf, feeds back some of
  this output to the input, effectively reducing
  the gain.
• The gain of the amplifier is now given as

In the case shown here this is exactly 10
38
Inverting op-amp

• The input impedance of the amplifier is given as

Here it is equal to 1 k?. If a higher resistance
is needed, larger resistances might be needed
Or, perhaps, a different amplifier will be
needed (noninverting amplifier)
39
Inverting op-amp

• The output impedance of the amplifier is given as

AOL is the open loop gain as listed on the data
sheet Open loop gain is the open loop gain at
the frequency at which the device is operated
40
Inverting op-amp

• Example, for the LM741 amplifier, the open loop
  output impedance is 75W and the open loop gain at
  1 kHz is 1000. This gives an output impedance of

The bandwidth is also influenced by the feedback
41
Non-inverting amplifier

• The non-inverting amplifier gain is

For the circuit shown, this is 11 The gain is
slightly larger than for the noninverting
amplifier for the same values of R. The main
difference however is in input impedance.
42
Non-inverting amplifier

• Input impedance is

Rop is the input impedance of the op-amp as given
in the spec sheet Aol is the open loop gain of
the amplifier. Assuming an open loop impedance
of 1 M? (modest value) and an open loop gain of
106, we get an input impedance of 1011 ?.
(almost ideal)
43
Non-inverting amplifier

• The output impedance and bandwidth are the same
  as for the inverting amplifier.
• The main reason to use a noninverting amplifier
  is that its input impedance is very large making
  it almost ideal for many sensors.
• There are other properties that need to be
  considered for proper design such as output
  current and load resistance but these will be
  omitted here for the sake of brevity.

44
The voltage follower

• The feedback resistor in the noninverting
  amplifier is set to zero
• The circuit in Figure 11.6 is obtained.
• The gain is one.
• This circuit does not amplify.
• Why use it?

45
The voltage follower
46
Voltage follower

• The input impedance now is very large and equal
  to

The output impedance is very small and equal to
47
Voltage follower

• The value of the voltage follower is to serve in
  impedance matching.
• One can use this circuit to connect, say, a
  capacitive sensor or, an electret microphone.
• If amplification is necessary, the voltage
  follower may be followed by an inverting or
  noninverting amplifier

48
Instrumentation amplifier

• The instrumentation amplifier is a modified
  op-amp
• Its gain is finite and both inputs are available
  to signals.
• These amplifiers are available as single devices
• To understand how they operate, one should view
  them as being made of three op-amps (it is
  possible to make them with two op-amps or even
  with a single op-amp), as shown in Figure 11.7.

49
Instrumentation amplifier
50
Instrumentation amplifier

• The gain of an amplifier of this type is

In a commercial instrumentation amplifier all
resistances are internal and produce a gain
usually around 100. Ra is external and can be
set by the user to obtain the gain required.
51
Instrumentation amplifier

• The output of the instrumentation amplifier is

The main use of this amplifier is to obtain an
output proportional to difference between inputs.
Important in differential sensors, especially
when one sensor is used to sense the stimulus and
an identical sensor is used for reference (such
as when temperature compensation is needed)
52
Instrumentation amplifier

• Each of the inputs has the high impedance of the
  amplifier used
• The output impedance is low (inverting amp.)
• The main problem in a circuit of this type is
  that the CMRR depends on the matching of the
  resistances (R, R2 and R3) in each section of the
  circuit.
• These are internal and are adjusted during
  production to obtain the required CMRR.

53
Charge amplifier

• The so-called charge amplifier is shown in Figure
  11.8.
• Charge cannot be amplified but the output voltage
  can be made proportional to charge as follows
• The output of the inverting amplifier is

C0 is the capacitance connected across the
inverting input.
54
Charge amplifier

• Assuming that a change in charge occurs on the
  capacitor, equal to ?Q C0?V, the output voltage
  may be written as

In effect the charge generated at the input is
amplified
55
Charge amplifier

• If C is small, a small change in charge at the
  input can generate a large voltage swing in the
  output.
• The main method of connecting capacitive sensors
  such as pyroelectric sensors whose output is low
  (piezoelectric sensors, on the other hand produce
  a higher voltage).
• It is necessary for the input impedance to be
  very high and care must be taken in connections
  (such as the use of very good capacitors).
• Commercial charge amplifiers use FET transistors
  to ensure the necessary high input impedance.

56
Charge amplifier
57
Current amplifier

• Another example of the use of an amplifier to a
  specific end is the current amplifier

58
Current amplifier

• The input voltage is Viir.
• Just like the inverting amplifier, the output now
  is

Useful with very low impedance sensors. May be
used with thermocouples whose impedance can be
trivially low. They may be connected directly (r
then represents the resistance of the
thermocouple). The output is a direct function
of the current the thermocouple produces which
can be fairly large
59
The comparator

• An op-amp operated in open loop mode
• Because its gain is so high, a very small signal
  at the input will saturate the output.
• For practically any input, the output will be
  either Vcc or Vcc. Consider Figure 11.10.
• The negative input is set at a voltage V? and
  V0. Therefore the output is ?AolV??Vcc.
• Suppose we increase V. Output is (V?V?)Aol. As
  long as VltV-, the output remains Vcc. If VgtV?,
  the output changes to Vcc.

60
The comparator
61
The comparator

• The function of this device is to compare the two
  inputs and to indicate which one is higher.
• The comparator is useful beyond simple
  comparison.
• It will be used extensively in A/D and D/A
  conversion of signals and in many other aspects
  of sensing and actuation

62
Power amplifiers

• A power amplifier is a device or circuit whose
  power output is the input power multiplied by a
  power gain

That is, the amplifier is capable of boosting the
power level of a signal to match the needs of an
actuator.
63
Power amplifiers

• The obvious use of power amplifiers is in driving
  actuators, (speakers, voice coil actuators and
  solenoid actuators and motors).
• The power amplifier is really either a voltage
  amplifier or a current amplifier (also called
  transconductance amplifier).
• In a voltage amplifier, the input signal is a
  voltage.
• This voltage is amplifier and in the final stage
  a sufficiently high current provided so that the
  required power is met.

64
Power amplifiers

• In a current amplifier the opposite occurs.
• Power amplifiers are divided into linear and PWM
  (pulse width modulated) amplifiers.
• In a linear amplifier, the output (voltage) is a
  linear function of the input and can be anything
  between Vcc.
• In a PWM amplifier the output is either Vcc or
  zero and the power delivered is set by the time
  the output is on. The latter is controlled by the
  width of the pulse that controls the output.

65
Linear power amplifiers

• First step is to amplify the signal to the
  required output.
• Can be done using any amplifier
• We shall assume an op-amp was used for this
  purpose.
• Then this voltage is applied to an output stage
  
• It does not need to amplify but, rather, supplies
  the necessary current.
• A simple example is shown in Figure 11.11.

66
Linear power amplifier
67
Class A linear amplifier

• This is the so called Class A power amplifier.
• Set for a gain of 101 (noninverting amplifier).
• The output then drives the transistor whose
  output will swing, at most between 0 and V
• Will supply a current which is V/RL
• Class A designation indicates amplifiers for
  which the output stage is always conducting as in
  the case above. Also assumes output does not
  saturate.
• The BJT can be replaced with a MOSFET for higher
  currents.

68
Class A linear amplifier

• This type of amplifier is sometimes used to drive
  relatively small loads such as light indicators,
  small dc motors and some solenoid valves.
• In some cases the amplification is set high
  enough to saturate the amplifier in which case
  the amplifier operates as an on/off circuit
  rather than a class A amplifier
• Typically used to turn on/off relays, lights,
  motors, etc.

69
Class B amplifier

• A Class B or push-pull amplifier is shown in
  Figure 11.12.
• It is usually a better choice.
• It operates exactly as in the previous case
  except that under no input, the output is zero
  and there is no conduction in the transistors (or
  MOSFETs).
• When the input is positive, the upper transistor
  conducts supplying the load and when the input is
  negative, the lower transistor supplies the load.

70
Class-B (push-pull) power amplifier
71
Class B amplifier

• The voltage in the load can swing between Vcc
  and Vcc
• The current is again defined by the load.
• The output stage is made of a pair of power
  transistors, one PNP and one NPN (or of a P and
  an N type MOSFET).
• There are many variations of the basic
  amplifiers.
• For example, feedback may be added and it is
  common to protect the output stage from short
  circuits as well as from spikes due to inductive
  and capacitive loads.

72
Class B amplifier

• In terms of performance, the obvious are the
  power output and the type and level of input.
• For example, an amplifier may be specified as
  supplying 100W for a 1V input.
• Next is the distortion level.
• Distortions are specified as a percentage of
  output.
• The most common specification is the THD (total
  harmonic distortions) as of output.
• A good amplifier will have less than, say, 0.1
  THD.
• Other specifications are temperature rise and
  output impedance of the amplifier (must match
  load).

73
Class B amplifier

• Power amplifiers of various power level exist
  either as integrated circuits or as discrete
  components circuits.
• Usually the discrete circuits can supply higher
  powers.
• An example of an integrated amplifier is the
  TDA2040 which can supply 20W and is designed for
  use as an audio amplifier.
• Nevertheless it can drive other loads such as
  light bulbs, small motors, etc.

74
PWM amplifiers

• The PWM approach is shown schematically in Figure
  11.13.
• The power transistors are driven on and off so
  that the voltage on the load can only be zero or
  Vcc.
• The time the power is on is controlled by the
  timing circuit.
• This defines the average power at the load.

75
The PWM principle
76
PWM amplifiers

• The pulse width modulator is an oscillator which
  generates a square wave whose duty cycle can be
  controlled based on the required power.
• For example, in Figure 11.14, the timing circuit
  defines for how long the input signal is
  connected to the transistor, hence for how long
  it conducts.
• The power in the load is a function of this
  timing. This circuit is not particularly useful
  but others are.

77
PWM driving of a load
78
PWM driving

• Figure 11.15 shows an example often used to
  control speed and direction of small dc motors.
• It is called an H-bridge for obvious reasons.
• A pulse of constant amplitude but varying duty
  cycle connected to point A, will drive MOSFETs 1
  and 4, turning the motor into one direction.
• The duty cycle defines the average current in the
  motor and hence its speed.
• Connecting to point B, turns on MOSFETs 2 and 3
  reversing the process.

79
H-Bridge driven from a PWM source
80
H-bridge PWM driver

• Some precautions must be taken to ensure that
  only opposite transistors conduct
• This is one of the most common circuits used for
  bidirectional control of motors and other
  actuators.
• The controllers for these devices can be a small
  microprocessor
• Integrated PWM circuits and controllers are
  available commercially

81
A/D and D/A converters

• These are the means by which a signal can be
  converted from analog to digital or from digital
  to analog as necessary.
• The idea is obvious but implementation can be
  complex.
• There are certain types of D/A and A/D that are
  trivially simple.
• We will start with these and only then discuss
  some of the more complex schemes.
• In certain cases one of these simple methods is
  sufficient.

82
A/D and D/A converters

• Analog to digital and, to a lesser extent,
  digital to analog conversion are common in
  sensing systems since most sensors and actuators
  are analog devices and most controllers are
  digital.
• Most A/Ds required voltages much above the output
  of some sensors.
• Often the output from the sensor must be
  amplified first and only then converted.
• This leads to errors and noise and has resulted
  in the development of direct digitization methods
  based on oscillators (to be discussed below).

83
Threshold digitization

• In some cases, an analog signal represents simple
  data such as the presence of something.
• For example, in chapter 5 we discussed the
  detection of teeth on a gear using a hall
  element.
• The signal obtained is quite small and looks more
  or less sinusoidal with the peaks representing
  the presence of the teeth.
• In such a case it is sufficient to use a
  threshold amplifier which will then produce a
  digital output. An example is shown in Figure
  11.16a.

84
Threshold digitization
85
Threshold digitization

• The output from the the hall element varies from
  100mV to 150mV. This signal can be fed into a
  comparator as shown in Figure 11.17
• The negative input is set by the resistors to
  0.13V.
• Normally the output is zero until the voltage on
  the positive input rises above the threshold.
• When the input dips below 0.13V the output goes
  back to zero.
• The output in Figure 11.16b is obtained and now,
  each pulse represents a tooth on the gear.

86
Comparator threshold digitization
87
Threshold digitization

• Counting the teeth in a given time can give the
  speed of rotation of the gear or other data.
• A missing tooth, the corresponding pulse will be
  represented by a missing pulse
• This method is very effective when voltages at
  the input change across the comparison point
• At the comparison point itself, the output of the
  comparator is not properly defined and the output
  can change states back and forth creating pulses
  which are spurious.

88
Threshold digitization

• To avoid this a hysteresis is added to the
  comparator so that the transition from low to
  high occurs say, at V0 and the transition from
  high to low occurs at V0-?V.
• Hysteresis can be added to comparators through
  external components.
• Another approach is to use of a Schmitt trigger.
• The Schmitt trigger is essentially a digital
  comparator with a built in hysteresis as
  described above whose transition is around Vcc/2.

89
Threshold digitization

• Threshold digitization is a very simple method of
  digitization and is sufficient for many
  applications.
• It is commonly used for the purpose above but
  also in flow meters in which a rotating paddle
  operates a hall element or another magnetic
  sensor
• It is also useful for optical sensors which use
  the idea of interruption of the beam.
• It is not however suitable for measuring the
  level of a signal such as voltage from a
  thermocouple.

90
Direct voltage to frequency conversion

• In many sensors, the output is too small to use
  the method above or to be sent over normal lines
  for any distance.
• In such cases a voltage to frequency conversion
  can be performed at the location of the sensor
  and the digital signal then transferred over the
  line to the controller.
• The output now is not voltage but rather a
  frequency which is directly proportional to
  voltage (or current).

91
Direct voltage to frequency conversion

• These voltage-to-frequency converters or voltage
  controlled oscillators are relatively simple and
  accurate circuits and have been used for other
  purposes.
• Their main advantage over the threshold method
  above is that lower levels of signals may be
  involved and the problems with noisy transitions
  around the comparison voltage are eliminated.
• A circuit of this type is shown in Figure 11.18,
  as used with a light sensor.
• The circuit is an op-amp integrator.

92
Direct voltage to frequency conversion
93
Direct voltage to frequency conversion

• The voltage across the capacitor is the integral
  of the current in the noninverting leg of the
  amplifier.
• This current is proportional to the voltage
  across R2.
• As the voltage on the capacitor rises, a
  threshold circuit checks this voltage
• When the threshold has been reached, an
  electronic switch shorts the capacitor and
  discharges it.
• The switch then opens and allows the capacitor to
  recharge.
• The voltage on the capacitor is a triangular
  shape whose width (i.e. the integration time)
  depends on the voltage at the noninverting input.

94
Direct voltage to frequency conversion

• If no light is present on the sensor, it has a
  dark resistance and the voltage at the
  noninverting input will have a certain value.
• The output of the amplifier changes at a
  frequency f1.
• If now light falls on the sensor, its resistance
  goes down and the total resistance at the
  noninverting input falls.
• This reduces the input voltage and hence the
  integration time until the capacitor reaches the
  threshold increases.
• The result is that the amplifier changes state
  slower and the output is a lower frequency f2.
• Since small changes in frequency can be easily
  detected, this is a very sensitive method of
  digitization for small signal sensors.

95
Voltage to frequency conversion

• Other V/F converters require a much higher
  voltage and they are more suitable for A/D
  conversion after amplification of the lower
  signals or for sensors whose output is high to
  begin with. There are two basic methods.
• One type is essentially a free running oscillator
  whose frequency can be controlled by the input
  voltage.
• The second is a modification of Figure 11.18 and
  is called a charge-balance V/F converter.

96
Voltage to frequency conversion

• A simple V/F method is shown in Figure 11.19.
• It consists of a square wave oscillator (called a
  multivibrator) and a control circuit.
• The multivibrator operates by charging and
  discharging a capacitor.
• The on/off times of the waveform (hence
  frequency) are controlled by charging/discharging
  times of the capacitor.
• To control frequency voltage to be converted is
  amplified and fed as currents to the bases of the
  two transistors.
• The larger the base current, the larger the
  collector current and the faster the
  charge/discharge and hence the higher the
  frequency of the multivibrator.

97
Simple voltage to frequency conversion -
multivibrator
98
V/F conversion

• A different approach is shown in Figure 11.20.
• The amplifier acts as an integrator and the FET
  across the capacitor is the switch.
• The capacitor charges at a rate proportional to
  the current IV0/R which is proportional to the
  voltage to be converted.
• When the output has reached the threshold voltage
  of the schmitt trigger it changes state, turning
  on and this turns on the FET switch.
• Discharging of the capacitor occurs and the
  output resets to restart the process. Again, as
  before, only relatively large level voltages can
  be converted.

99
Simple voltage to frequency conversion -
integrator
100
Dual slope A/D converter

• The simpler (and slower) of the true A/D
  converters
• Based on the following principle a capacitor is
  charged from the voltage to be converted through
  a resistor, for a fixed, predetermined time T.
  The capacitor reaches a voltage VT which is

101
Dual slope A/D converter

• At time T, Vin is disconnected
• A negative reference voltage of known magnitude
  is connected to the capacitor through the same
  resistor.
• This discharges the capacitor down to zero in a
  time ?T

102
Dual slope A/D converter

• Since these are equal in magnitude we have

In addition, a fixed frequency clock is turned on
at the beginning of the discharge cycle and off
at the end of the discharge cycle. Since ?T and T
are known and the counter knows exactly how many
pulses have been counted, this count is the
digital representation of the input voltage. A
schematic diagram of a dual slope converter based
on these principles is shown in Figure 11.21.
103
Dual slope A/D conversion
104
Dual slope A/D converter

• The method is rather slow with approximately 1/2T
  conversions per second.
• It is also limited in accuracy by the timing
  measurements, accuracy of the analog devices and,
  of course, by noise.
• High frequency noise is reduced by the
  integration process and low frequency noise is
  proportional to T (the smaller T the less low
  frequency noise).

105
Dual slope A/D converter

• The dual slope A/D is the method of choice for
  many sensing applications in spite of its rather
  slow response because it is simple and readily
  built from standard components.
• For most sensors its performance and noise
  characteristics is quite sufficient
• Because of the integration involved, it tends to
  smooth variations in the signal during the
  integration.
• The method is also used in digital voltmeters and
  other digital instruments.

106
Successive approximation A/D

• This is the method of choice in A/D converter
  components and in many microprocessors.
• It is available in many off the shelf components
  with varying degrees of accuracy
• Depending on the number of bits of resolution it
  may resolve down to a few microvolt.
• The basic structure is shown in Figure 11.22. It
  consists of a precision comparator, a shift
  register a digital to analog converter and a
  precision reference voltage Vref.

107
Successive approximation A/D conversion
108
Successive approximation A/D

• The operation is as follows
• First, all registers are cleared, which forces
  the comparator to HIGH.
• This forces a 1 into the MSB of the register.
• The D/A generates an analog voltage Va which for
  MSB1 is half the full scale input.
• This is compared to Vin. If Vin is larger than
  Va, the output stays high and the clock shifts
  this into the next bit into the register.

109
Successive approximation A/D

• The register now shows 1100000000.
• If it is smaller than Vin, the output goes low
  and the register shows 010000000.
• Assuming that the input is still higher, the D/A
  generates a voltage Va(1/21/4)Vfs.
• If this is higher than the input, the register
  will show 011000000 but if it is lower, it will
  show 11100000 and so on, until, after n steps the
  final result will be obtained.
• The data is read from the shift register and
  represents the voltage digitally.
• This digital value can now be shifted out and
  used by the controller.

110
Successive approximation A/D

• A/D of this type exists with resolution of up to
  14 bits with 8 and 10 bits being quite common.
• An 8 bit A/D has a resolution of
  Vin/280.004Vin.
• For a 5V full scale, the resolution is 20 mV.
• This may not be sufficient for low level signals
  in which case a 10, 12 or 12 bit A/D may be used
  (a 14 bit A/D has a resolution of 0.3mV).
• There are also techniques of extending this
  resolution but it is almost always necessary to
  amplify signals from devices such as
  thermocouples if they must be digitized.

111
Successive approximation A/D

• The advantage of the successive approximation A/D
  is that the conversion is done in n steps (fixed)
  
• It is much faster than other methods.
• On the other hand the accuracy of the device
  depends heavily on the comparator and the D/A
  converter.
• Commercial devices are fairly expensive,
  especially if more than 10 bits are needed.
• This type of A/D has been incorporated directly
  into microprocessors and can sometimes be used
  for sensing as part of the overall circuitry.
• Some microprocessors have multiple A/D channels.

112
Digital to Analog Conversion

• Digital to analog conversion is less often used
  with sensors but is sometimes used with
  actuators.
• This occurs when a digital device, such as a
  microprocessor must provide an analog output.
• This should be avoided if possible by use of
  digital actuators (such as brushless dc motors
  and stepper motors) but there will be cases in
  which D/A will be necessary.
• It is often a part of A/D conversion

113
Digital to Analog Conversion

• There are different ways of accomplishing D/A
  conversion.
• The most common method used in simple converters
  is based on the ladder network shown in Figure
  11.23.
• It consists of a voltage follower. Its input
  impedance is high and the output of the follower
  equals the voltage at its noninverting input.
• The voltage is generated by the resistance
  network.

114
Ladder network D/A conversion
115
Ladder network D/A conversion

• The ladder network is chosen so that the
  combination of series and parallel resistances
  represent the digital input as a unique voltage
  which is then passed to the output.
• The switches are digitally controlled analog
  switches (MOSFETs).
• Depending on the digital input, various switches
  will connect resistors in series or in parallel.

116
Ladder network D/A conversion

• For example, suppose that the digital value 100
  is to be converted.
• The switches will be as in Figure 11.23.
• The voltage at the amplifiers input is exactly
  5V.
• The ladder can be extended as necessary for any
  number of bits.
• The accuracy and usefulness of a D/A depends on
  the quality and accuracy of the ladder network
  and the reference voltage used.

117
Bridge circuits

• Bridge circuits are some of the oldest circuits
  used in sensors as well as other applications.
• The bridge is known as the Wheatstone bridge
  (variations of the bridge exist with different
  names.)
• The basic Wheatstone bridge is shown in Figure
  11.24.
• It consists of 4 impedances ZiRijXi.

118
The impedance bridge
119
The impedance bridge

• The output voltage of the bridge is

The bridge is said to be balanced if
Under this condition, the output voltage is zero.
120
The impedance bridge

• If, for example, Z1 represents the impedance of a
  sensor, by proper choice of the other impedances
  the output can be set to zero at a given value of
  Z1.
• Any change in Z1 will change the value of Vo
  indicating the change in stimulus.
• Of course, one can do much more than that and
  bridges can be used for signal translation and
  for temperature compensation among other things.
• One important property of bridges is their
  sensitivity to change in stimuli

121
The impedance bridge

• The sensitivity of the output voltage to change
  in any of the impedances can be calculated as

Summing up gives the bridge sensitivity
122
The impedance bridge

• This relation reveals that if Z1Z2 and Z3Z4 the
  bridge is balanced
• If the change, is such that dZ1dZ2 and dZ3dZ4,
  the change in output is zero.
• This is the basic idea used in compensating a
  sensor for temperature variation and any other
  common mode effects.
• For examples, suppose that a pressure sensor has
  impedance Z1100 ? and a sensitivity to
  temperature dZ1 0.5 ?/?C.

123
The impedance bridge

• We use two identical sensors as Z1 and as Z2
• Sensor Z2 is not exposed to pressure (only
  exposed to the same temperature as Z1).
• Z3 and Z4 are equal and are made of the same
  material these are simple resistors.
• Under these conditions, there will be no output
  due to temperature changes
• The sensor is properly compensated for
  temperature variations.
• If however pressure changes, the output changes

124
The impedance bridge

• If all impedances in the bridge are fixed and
  only Z1 varies (this is the sensor), then dZ20,
  dZ30, dZ40 and the bridge sensitivity becomes

Or
125
The impedance bridge

• This bridge, especially with resistive branches
  is the common method of sensing with
• strain gauges, piezoresistive sensors,
• hall elements, thermistors
• force sensors and many others.
• Use of bridges allows a convenient reference
  voltage (nulling), temperature compensation and
  other sources of common mode noise.
• It is very simple and it can be easily connected
  to amplifiers for further processing

126
Temperature compensation of bridges

• Temperature compensation in sensors eliminates
  the errors due to temperature or any other common
  mode effect.
• It does not eliminate errors external to the
  sensors such as variations of Vi with
  temperature.
• These have to be compensated for in the
  construction of the bridge itself.
• There are many techniques by which this can be
  accomplished but this is beyond the scope of this
  course.

127
Bridge output

• The output from the bridge is likely to be
  relatively small.
• For example, suppose that the bridge is fed with
  a 5V source and a thermistor, Z4500? (at 0?C) is
  used to sense temperature.
• Assuming the bridge is balanced at 0?C, the other
  three resistances are also 500?.
• This gives an output voltage zero.
• Now, suppose that at 100?C the resistance of the
  thermistor goes down to 400W.

128
Bridge output

• The output voltage now is

Most sensors will produce a much smaller change
in impedance Some sort of amplification will be
necessary. The op-amp discussed above is ideal
for this purpose. There are many ways this can
be accomplished. Two methods are shown in Figure
11.25.
129
Amplified bridge
130
Active bridge
131
Amplified bridge

• In Figure 11.25a, the bridge is connected
  directly between the inverting and noninverting
  inputs.
• If we assume that the resistance of the
  resistance of the sensor changes as RxR0(1?),
  the voltage output of the bridge is

This circuit provides an amplification of (1n)
but requires that the voltage on the bridge be
floating
132
Active bridge

• Circuit in Figure 11.25b does not provide
  amplification but rather places the sensor in the
  feedback loop. This is called an active bridge
  and its output is

This circuit provides buffering (higher input
impedance, lower output impedance).
133
Data transmission

• Transmission of data from a sensor to the
  controller may take many forms.
• If the sensor is passive, it already has an
  output in a usable form such as voltage or
  current.
• It would seem that it is sufficient to simply
  measure this output directly to obtain a reading.
  
• In other cases, such as with capacitive or
  inductive sensors, indirect measuring is often
  used.
• The sensor is often likely to be in a remote
  location.

134
Data transmission

• Neither direct measurement of voltage and current
  or using the sensor as part of the circuit (in an
  oscillator) may be an option in such a case
• In such cases, it is often necessary to process
  the sensors output locally and to transmit the
  result to the controller.
• The controller then interprets the data and
  places it in a suitable form.

135
Data transmission

• The ideal method of transmission is digital.
• Often employed in smart sensors since they have
  the necessary processing power locally.
• In most cases a sensor of this type will have a
  local microprocessor supplied with power from the
  controller or have its own source of power
• The digital data may then be transmitted over
  regular lines or even through a wireless link.
• Since digital data is much less prone to
  corruption, the method is both obvious and very
  useful.

136
Data transmission

• Many sensors are analog and,
• Their output may eventually be converted into
  digital form but
• It is not always possible to incorporate the
  electronics locally.
• This may be because of cost or because of
  operating conditions such as elevated
  temperatures.

137
Data transmission

• Example, in a car there may be a half dozen
  sensors that control ignition, air intake and
  fuel, all of which are needed for control of the
  engine and are processed by a central processor.
• It is not practical to supply each sensor with
  power and electronics to digitize their data when
  the processor can do that for all of them.
• In other cases, such as, for example, the oxygen
  sensor, the sensor operates at elevated
  temperatures, beyond the temperature range of
  semiconductors making it impossible to
  incorporate electronics in them.

138
Data transmission

• In such cases the analog signal must be
  transferred to the controller.
• A number of methods have been developed for this
  purpose.
• Three of these methods, suitable for use with
  resistive sensors, or with passive sensors are
  discussed next

139
Four wire sensing

• In sensors that change their resistance, such as
  thermistors, and piezoresistive sensors, one must
  supply an external source and measure the voltage
  across the sensor.
• If done remotely, the current may vary with the
  resistance of the connecting wires and produce an
  erroneous reading.
• To avoid this the method in Figure 11.26 may be
  used.

140
Four wire sensing
141
Four wire sensing

• The sensor is supplied from a current source, i0.
  
• This current is constant since the internal
  impedance of a current source is very high.
• The voltage on the sensor is independent of the
  length of the wires and their impedance.
• A second pair of wires measures the voltage
  across the sensor
• Since a voltmeter has very high impedance there
  is no current (ideally) in this second pair of
  wires, producing accurate reading.
• This is a common method of data transmission when
  applicable.

142
Two wire sensing for passive sensors

• Passive sensors produce a voltage. It is
  sometimes possible to measure the voltage
  remotely (no current is involved in the
  measurement).
• Especially true for dc outputs such as in
  thermocouples.
• In sensors with high impedance it is much more
  risky to do so because of the noise the lines can
  introduce.
• In most cases a twisted pair line is used because
  it reduces the noised picked up by the line.

143
Two wire transmission for active sensors

• A common method of data transmission for sensors,
  and a method that has been standardized is the
  4-20 mA current loop.
• The output of the sensor is modified to modulate
  the current in the loop
• 4 mA corresponds to minimum stimulus
• 20 mA corresponds to maximum stimulus
• The configuration is shown in Figure 11.27.

144
4-20 mA current loop data transmission
145
4-20 mA current loop data transmission

• The sensors output must be modified to conform
  to this industry standard and this may require
  additional components.
• Many sensors are made to conform to this standard
  so that the user only has to connect them to the
  two-wire line.
• The power supply depends on the load resistance
  and the transmitters resistance but it is
  between 12 and 48V.

146
4-20 mA current loop data transmission

• Usually the sensors network allows for setting
  the range (minimum and maximum value of the
  stimulus) to the 4 mA and 20 mA range as shown.
• The current transmitted on the line is then
  independent of the length of the line and its
  resistance.
• The voltage measured across the load resistance
  is then processed at the controller to provide
  the necessary reading.

147
Other methods of transmission

• There are other methods of transmission that may
  be incorporated.
• 6-wire transmission is used with bridge circuits
  in which the 4 wire method above is supplemented
  by two additional wires which measure the voltage
  on the bridge itself.
• A new 1-wire protocol has become very popular for
  many devices including sensors.
• In this protocol both power to the device and
  data to/from it are passed on a single pair of
  wires,
• An effective and economical method for sensing.

148
Transmission to actuators

• There are only two ways the power can be
  transmitted to the actuator.
• One is to get the actuator close to the source
  that provides the power.
• This implies that lines must be very short.
• Possible in some cases (audio speakers, control
  motors in a printer, etc.).
• In some cases this is not practical and the
  controller and the actuator must be at
  considerable distance (robots on the factory
  floor, etc.).

149
Transmission to actuators

• In such cases one of the methods above may be
  used to transfer data but the power must then be
  generated locally at the actuator site.
• The controller now issues commands as to power
  levels, timings, etc. and these are then executed
  locally to deliver the power necessary.
• Much of this is done digitally through use of
  microprocessors on both ends.

150
Excitation methods and circuits

• Sensors and actuators must often be supplied with
  voltages or currents
• Either ac or dc.
• These are the excitation sources for the sensors
  and actuators.
• First and foremost is the power supply circuit.
• In many sensors the power is supplied by
  batteries
• Many others rely on line power through use of
  regulated or unregulated power supplies.

151
Excitation methods and circuits

• Other sensors require current sources (for
  example - Hall elements)
• Still others require ac sources (LVDTs)
• These circuits affect the output of the sensor
  and its performance (accuracy, sensitivity,
  noise, etc.)
• Are an integral part of the overall sensors
  performance.

152
Power supplies

• There are two types of power supplies
• Linear power supply
• Switching power supply.
• There are also so called dc to dc converters
  which are used to convert power from one level to
  another, sometimes as part of the circuit that
  uses the power.

153
Power supplies

• A linear power supply is shown in Fig. 11.28.
• Consists of a source, (line voltage) and a means
  of reducing this voltage to the required level (
  a transformer).
• The transformer is followed by a rectifier which
  produces dc voltage from the ac source.
• This voltage is filtered and then regulated to
  the final required dc voltage. A final filter is
  usually provided.
• This regulated power supply is very common in
  circuits especially where the power requirements
  are low.
• Some of the blocks may be eliminated depending on
  the application. If, for example the source is a
  battery the transformer and the rectifier are not
  needed and the filtering may be less important.

154
Linear regulated power supply
155
Linear power supply

• Consider the circuit in Figure 11.29.
• This is a regulated power supply capable of
  supplying 5V at up to 1A.
• Transformer reduces the input voltage to 16V rms.
  
• This is rectified through the bridge rectifier
  and produces 22V (16x1.4) across C1, C2.
• These two capacitors serve as filters the large
  capacitor reducing low frequency fluctuations on
  the line, the smaller capacitor is better suited
  for high frequency filtering.

156
Fixed voltage regulated power supply
157
Linear power supply

• The LM05 is a 5V regulator which essentially
  drops across itself 19V to keep the output
  constant.
• Does so for any input voltage down to about 8V.
• The capacitors at the output are again filters.
• The current is limited by the capacity of the
  regulator to dissipate power due to the current
  through it and the voltage across it.
• Other regulators are available that will
  dissipate more or less power.

158
Linear power supply

• These regulator exist at standard voltages,
  either positive or negative as well as adjustable
  variable voltage regulators.
• Discrete components regulators can be built for
  almost any voltage and current requirements.
• This circuit or similar circuits are the most
  common way of providing regulated dc power to
  most sensor and actuator circuits.

159
Linear power supply

• The advantage is that they are simple and
  inexpensive but they have serious drawbacks.
• The most obvious is that they are big and heavy,
  mostly because of the need for a transformer
  which must handle the output power.
• In addition, the power dissipated on the
  regulator is not only lost but it generates heat
  and this heat must be dissipated through heat
  exchangers.

160
Switching power supply

• An alternative method of providing dc power is
  through use of a switching power supply.
• Switching power supplies rely on two basic
  principle to eliminate the drawbacks of the
  linear power supply.
• The principle is shown in Figure 10.30.
• First, the transformer is eliminated and the line
  voltage is rectified.
• This high voltage dc is filtered as before.

161
Regulated switching power supply
162
Switching power supply

• The switching transistor is driven with a square
  wave
• It turns on for a time ton and off for a time
  toff
• When on, a current flows through the inductor
  charging the capacitor to a voltage which depends
  on ton
• When the switch is off, the current in L1 is
  discharged through the load supplying it with
  power for the off-time
• The voltage is stabilized by sampling the output
  and changing the duty cycle (ratio between ton
  and toff) to increase or decrease the output to
  its required value
• This change in duty cycle is done by use of a PWM
  (pulse width modulation) generator

163
Switching power supply

• In a practical power supply additional
  considerations must apply.
• First, it is necessary to separate or isolate the
  input (which is connected to the line) and
  output.
• In the linear PS this was accomplished by the
  transformer.
• Second, the switching, which must necessarily be
  done at relatively high frequencies, introduces
  noise into the system.
• This noise must be filtered for the PS to be
  usable

164
DC to DC converters

• DC to DC converters are a different type of
  switching power supply.
• They take the dc source and convert it