Motors&Drives.

book Page 205 Monday, January 10, 2005 1:54 PM

Chapter 5: Drive Control and Feedback Devices 205

Drive Control and

Feedback Devices

In this section, open- and closed-loop control will be reviewed. Most of the
information has been previously presented. However, the focus of this
chapter will be on the external devices that connect to the drive. These
external components allow the drive to interface to a system and provide
the necessary speed and torque control to maximize efficiency.

External devices (peripherals) are a vital part of any drive system—

whether operating as part of a remote operator station or part of an entire
production system. There is such a wide variety of peripherals that are
used in drive control that to include all possible combinations would be
beyond the scope of this book. However, the major categories of devices
and their use will be covered in this chapter.

Open-Loop Control
Open loop control was presented in Chapter 4, under the direct current
(DC) and alternating current (AC) drive sections. A brief review is given
here. Figure 5-1 indicates the basics of open-loop control.

Speed
Reference

AC or DC Motor
Drive

Figure 5-1. Open-loop control

The open-loop control is the simplest form of drive control. The drive
operates the motor, given the speed reference that is delivered from the
hand speed pot. If Figure 5-1 were of a DC drive, motor-speed regulation
Motors&[Link] Page 206 Monday, January 10, 2005 1:54 PM

206 Motors and Drives

would be 5–7% higher. If the figure were of an AC drive, motor-speed reg-

ulation would be 3–5% higher, possibly higher depending on motor
design.

Open-loop control is more typical in AC drives, where the inherent motor-

speed regulation is acceptable in the 3–5% range. DC drives are almost
never operated open loop because of the higher regulation characteristics.
Some type of feedback is used, such as armature voltage feedback (EMF)
or tach feedback. These methods will be discussed in greater detail later in
this chapter. Before these methods are discussed, a review of closed-loop
control is required. In addition, the types of devices involved will also be
presented.

Closed-Loop Control
As indicated by the name, this type of control method requires the use of
some peripheral device that closes the loop back to the drive. The device
that is commonly used is the tachometer, or tach generator, as it is some-
times called. Figure 5-2 indicates a simple closed-loop control scheme.

Speed
Reference
AC or DC Drive

Power &
Logic Control Motor

Summing
Circuit Speed Feedback
T

Figure 5-2. Closed-loop control

Closed-loop control is a bit more complex, but yields a great improvement

in motor-speed regulation. The key to the success of this scheme is the
tachometer (T) or feedback device. A speed reference is given to the AC or
DC drive. The actual speed, determined by the tach, is fed back to the
drive. Both the speed reference and feedback are combined into the sum-
ming circuit. The summing circuit produces an error signal. That error sig-
nal is now the speed reference that is sent to the power and logic control,
which translates the signal into faster or slower motor speed.

If this happened to be an AC drive configured in torque-control mode, the

feedback would be a current feedback instead of speed. If this was a posi-
tion control, such as a servo controller, the feedback would be a position
and use an encoder instead of a tach. The idea of this entire scheme is to
have zero error on the output of the summing circuit. If error occurs, it
Motors&[Link] Page 207 Monday, January 10, 2005 1:54 PM

Chapter 5: Drive Control and Feedback Devices — Closed-Loop Control 207

must be corrected. How fast it is corrected is a function of the logic control

in the drive.

The use of proportional integral derivative (PID) is critical to the success of

the entire system. The PID controller will be programmed to correct for the
error immediately or over a period of time. It will also be programmed as
to how much error should be corrected over a period of time. The dynam-
ics and stability of the system is dictated by the responsiveness or sloppi-
ness of the feedback loop and control system. As with any type of control
system, it is only as good as the weakest link. The remainder of this chap-
ter will be devoted to closed-loop components, their accuracy, and how
the responsiveness of the system will be affected.

Tachometers
A tachometer generator or tach is an electromechanical device that trans-
lates the rotational speed of a shaft to an electrical signal. An analog tach
generator is a small generator that produces an output voltage whose mag-
nitude is linearly proportional to shaft speed. An AC tach produces an AC
voltage and a DC tach produces a DC voltage. DC tachs produce a negative
voltage to indicate reverse rotation as shown in Figure 5-3.

Voltage (+)

Reverse (CCW) Forward (CW)

Speed

(-)

Figure 5-3. DC tach output voltage vs. speed

DC tachs are the most popular type of analog tach, since AC tachs do not
indicate direction and are generally less accurate than DC tachs. The out-
put of a DC tach is typically specified in volts per 1000 rpm. The V/1000
rpm voltage constant is the slope of the output voltage versus speed line.
Digital tachs, or pulse tachs as they are sometimes called, produce a series
of pulses. The frequency of those pulses is proportional to shaft speed. This
can be seen in Figure 5-4.

The output of a digital tach is typically specified in pulses per revolution

(PPR). Most pulse tachs are magnetic devices in which pulses are gener-
ated by gear teeth passing by a magnetic pickup.
Motors&[Link] Page 208 Monday, January 10, 2005 1:54 PM

208 Motors and Drives

High Speed - High Frequency

Low Speed - Low Frequency

Time
Figure 5-4. Pulse tach output

Encoders
An encoder is an electromechanical device that translates mechanical posi-
tion and/or velocity into electrical signals. A shaft encoder or rotary encoder
translates rotational or angular position and/or velocity into electrical sig-
nals.

An incremental encoder provides a signal that indicates position change or

incremental position rather than absolute position. In principle, an incre-
mental encoder is essentially the same type of device as a pulse tach, pre-
viously discussed. The term encoder usually implies a device with more
advanced features and performance. Terms such as incremental encoder,
encoder, pulse tach, and digital tachometer are basically used for the same
type of device. In industry, they are often used interchangeably.

Optical encoders, on the other hand, are devices that use pulses for rota-
tional signal generation. A light beam is transmitted through transparent
stripes or slots in a rotating disk. The photo detector picks up those light
beams and converts them into a series of pulses, sometimes called a pulse
train. Figure 5-5 shows this process.

Slotted Disc

Light Source

Photo-Detector

Figure 5-5. Optical encoder characteristics

Motors&[Link] Page 209 Monday, January 10, 2005 1:54 PM

Chapter 5: Drive Control and Feedback Devices — Closed-Loop Control 209

Optical encoders are the most widely used types of encoder. They are
capable of providing very high resolution.

Magnetoresistive encoders have a wheel or disk with numerous small mag-

netic poles imprinted on the disk surface in alternating north/south pairs.
Incremental position is indicated by a sensing element consisting of a com-
bination of resistors that change resistance on the basis of a magnetic field.
The sensing element provides input to electronic circuitry, which generate
pulses corresponding to each individual imprinted magnetic pole.
An absolute encoder produces a number of binary outputs that can be read
as a binary word, which indicate the absolute angular position of the shaft
relative to some unique index position. This type of encoder would be
used primarily in a system where position feedback would be required
(e.g., cut-to-length servo system).

This book is primarily concerned with motor speed or velocity in drive sys-
tems. Therefore, the focus will be on the incremental encoder or pulse
tach, rather than on absolute encoders devices.

Encoder Characteristic
The following paragraphs explain several encoder features that provide
functions and performance that go beyond simply indicating shaft speed.
An encoder with a single output cannot indicate direction of rotation.
Quadrature outputs are two outputs with a 90º phase displacement, as
shown in Figure 5-6.

Channel

Forward (CW) Rotation Reverse (CCW) Rotation

A Leads B B Leads A

Timel
Figure 5-6. Quadrature outputs of a pulse tach (encoder)

As shown in Figure 5-6, quadrature outputs can be used to determine

direction of rotation by detecting which of the two output signals is lead-
ing. The low-to-high transitions of the leading signal occur before the low-
to-high transitions of the lagging signal.

Incremental encoders can be used to indicate when the shaft is in one par-
ticular “index” position. This is done by using an index pulse or marker pulse
that occurs only once per revolution at the index position. Figure 5-7
Motors&[Link] Page 210 Monday, January 10, 2005 1:54 PM

210 Motors and Drives

shows this procedure. The width and phase relationship of the index pulse
may vary among encoder models.

Channel

One Revolution

Figure 5-7. Encoder marker pulses of index positions

Figure 5-8 indicates complementary or differential outputs. Complementary

outputs are pairs of outputs that are configured in such a way that one
output (A) is high whenever its complement (A) is low.

Channel

Figure 5-8. Complementary or differential outputs

A and A represent the signal voltages with respect to circuit common. The
voltage between the A output and the A output is the voltage difference,
or differential, between the A and A voltages. An encoder output signal
that does not include the complementary A and B signals is called a single-
ended output.

Differential signals have better noise immunity than single-ended signals.

Figure 5-9 indicates a differential line driver and receiver connected by a
twisted pair of shielded wires—a transmission line.

Line Line
Driver Receiver
A Shielded Cable

A
Figure 5-9. Differential line driver and receiver diagram
Motors&[Link] Page 211 Monday, January 10, 2005 1:54 PM

Chapter 5: Drive Control and Feedback Devices — Closed-Loop Control 211

The use of a differential line driver and receiver protects against common
mode noise between the transmitting and receiving locations. The use of a
shielded, twisted-pair transmission line protects the signal from electro-
magnetic fields that might interfere with signal transmission.

The resolution of an encoder is determined by the number of pulses per

revolution of the encoder shaft. The resolution determines the smallest
amount of shaft rotation that the encoder is able to indicate. If an encoder
generates 1024 pulses per revolution, it can indicate a movement of 1/
1024 of a revolution or 0.35 degrees of rotation. The effective resolution of
an encoder can be improved by using pulse multiplication. This procedure
is shown in Figure 5-10.

1X

2X

4X

Figure 5-10. Encoder pulse multiplication

The A and B signals shown in Figure 5-10 are the outputs of a quadrature
encoder. The 1X signal is a series of narrow pulses corresponding to the
low-to-high transitions of the A signal. These pulses are typically gener-
ated by the device that receives or uses the encoder signal.

The resolution of the 1X signal is equal to the encoder PPR. The 2X signal
contains pulses corresponding to both the low-to-high transitions and the
high-to-low transitions of the A signal. The resolution of the 2X signal is
twice the encoder PPR. The 4X signal contains pulses corresponding to the
low-to-high and high-to-low transitions of both the A and the B signals.
The effective resolution of the 4X signal is four times the encoder PPR.

To correctly apply encoders, the specifications of the encoder must be

determined. Then the applications specifics can be matched to the encoder
that would best meet the requirements. At this point, a review of typical
encoder specifications is in order.

Encoder Specifications
The most basic encoder specification is the resolution or number of pulses
per revolution of the encoder shaft. In general, using the highest possible
pulses per revolution allows the best possible drive performance. High PPR
allows speed to be measured more accurately and sampled more fre-
Motors&[Link] Page 212 Monday, January 10, 2005 1:54 PM

212 Motors and Drives

quently. The pulse frequency is determined by the encoder-operating

speed and the resolution according to the following formula:

Speed (rpm) × Resolution (PPR)

Pulse Frequency (Hz) =
60

The following example illustrates the pulse-frequency calculation for a

2048 PPR encoder operating at 1800 rpm. Note that the pulse frequency
unit of measurement, hertz, is equivalent to pulses per second. The
minute and revolution units cancel, leaving pulses-per-second or hertz as
the units of measurement.

revolutions pulses
1800 × 2048
minute revolution = 61440 pulses (Hz)
seconds second
60
minute

The encoder PPR must be selected so that the pulse frequency is high
enough to allow the required performance at the minimum operating
speed. At the maximum operating speed, the pulse frequency must not
exceed the maximum output frequency capability of the encoder. It also
must not exceed the maximum input frequency capability of the circuitry
that receives the signal.

Some drives can be adjusted to accept any PPR value within a specified
range. Other drives accept only specific PPR values, such as 1024 PPR or
2048 PPR. The accuracy of an encoder is the maximum percentage varia-
tion between the actual positions of the pulse edges or transitions and the
theoretical correct positions. The encoder accuracy is not the only factor
that determines the speed-measurement accuracy. Speed-measurement
accuracy is influenced by the encoder timing and counting accuracy. It is
also influenced by the mechanical backlash in the coupling to the
machine.

The measurement method may improve upon the encoder accuracy by

averaging a number of pulses in each measurement. When speed is aver-
aged over an extended period of time, the measurement accuracy will be
determined entirely by the timing and counting accuracy.

The basic specification for the output signals indicates whether the
encoder has a single output or quadrature output and whether or not a
zero marker is provided. If complementary output signals are provided,
the encoder specification may identify the IC-type number for the line
driver. The output voltage and current ratings should also be specified.
Single-ended outputs are often open collector, current sinking outputs.
Most encoders provide square-wave outputs (50% duty cycle), but some
models may be available with a fixed-pulse width.
Motors&[Link] Page 213 Monday, January 10, 2005 1:54 PM

Chapter 5: Drive Control and Feedback Devices — Closed-Loop Control 213

Output specifications must be checked for compatibility with the circuitry

that will receive the signal. Encoders require an external power source.
The most common requirements are 12 and 24 VDC. Encoders are also
available for operation at 5 VDC. However, these encoders are not recom-
mended for installations requiring a long cable distance between the
encoder and the drive.

Most encoders require a regulated input supply voltage and provide an

output at the same voltage level as the input voltage. If the encoder has its
own voltage regulator, its output signal level will be considerably lower
than the input supply voltage. If the drive is designed to provide a regu-
lated encoder supply voltage, it may not accept an encoder signal at a volt-
age level that is significantly below the supply voltage.

Before installation, the user should verify that the output voltage of the
encoder is compatible with the input voltage range of the drive’s encoder
input. In addition, all encoders have minimum and maximum operating
temperature ratings. If the encoder is subjected to temperatures outside
the specified limits, it may not operate correctly and can ultimately be
damaged. A possible source of encoder damage is heat. Heat is very likely
to be conducted to the encoder housing through the mounting structure
from motor or machine. An encoder should be selected with an operating
temperature rating that is suitable for the location and mounting method.
A variety of mechanical specifications define the type of encoder housing
available and the type of mounting arrangement possible. Encoders are
available with various degrees of protection from moisture and environ-
mental contaminants. The degree of protection is determined by the
encoder housing type and material, the shaft material, the type of shaft
seal, etc. Explosion-proof models are available for use in areas where fire
or explosion hazards may exist because of flammable gases or vapors. Var-
ious bearing ratings are available to accommodate a wide range of radial
and axial loads that might be subjected to the shaft of the encoder.

There will also be a maximum operating speed determined by the

mechanical construction of the encoder. For each PPR rating, there will be
a maximum speed corresponding to the encoder’s maximum output fre-
quency capability as described in a previous paragraph. The encoder man-
ufacturer’s installation and alignment procedures must always be carefully
followed to ensure reliable operation.

Once the encoder is properly installed, it must be wired to the specifica-

tions of both the encoder and drive manufacturer. Careful attention to
wiring is essential to ensure that a drive with encoder feedback performs
properly. The following recommendations cover the most common
requirements.

Each encoder output channel and the power supply should be connected
using an individually shielded twisted pair of wires. Encoder wiring and a
connection scheme are shown in Figure 5-11.
Motors&[Link] Page 214 Monday, January 10, 2005 1:54 PM

214 Motors and Drives

A
Encoder Drive

A or Com
B

B or Com
V+

V- or Com

Figure 5-11. Encoder wiring diagram

Three individual cables or a three-pair cable can be used. If a three-pair

cable is used, be sure that each pair of wires is individually shielded. The
shields are to be grounded as recommended by the drive instruction mate-
rial. Shielded cable is grounded at only one point to prevent ground loops
or circulating currents through the shield conductor. The best ground loca-
tion is usually at the drive. The encoder cable must be run in a steel con-
duit. Cables from more than one encoder can be run in the same conduit,
but the conduit cannot carry power wiring.

Encoder cable should be a low-capacitance type of cable, designed for

high-speed digital signals such as RS-422/485 serial communications sig-
nals.
Since the relationship between encoder shaft rotation and quadrature sig-
nal phasing is not standardized, this relationship can vary among encoder
models. The instruction material furnished with the encoder is the best
source of determining the encoder phasing. Phasing corresponds to the
selected motor direction. If the drive does not operate properly upon initial
startup, encoder phasing could be a suspect. It may be necessary to deter-
mine proper phasing by trial and error.

Note: Today’s microprocessor-based adjustable-speed drives are designed for use

with incremental encoders rather than analog tachometer generators. Although
some models may accept a feedback signal from a DC tach, the best performance is
obtained by using an encoder. Analog tachometer generators have better EMI
immunity compared with encoders. However, with the use of differential signals
and careful attention to wiring, EMI immunity of an encoder can be equal or better
than that of an analog tach.

If an analog tach is used, it is necessary to determine what range of signals

are compatible with the drive circuitry that it will be connected to. The cir-
cuit must be designed for either a DC or AC tach. The maximum input
voltage and tach-generating voltage at maximum speed are factors to con-
sider before purchasing a tach. Using a tach with the maximum compati-
Motors&[Link] Page 215 Monday, January 10, 2005 1:54 PM

Chapter 5: Drive Control and Feedback Devices — Closed-Loop Control 215

ble volts per 1000 rpm will provide the best performance over a wide
speed range.

Resolvers
A resolver is a “position” transducer, with characteristics that resemble a
small motor. This type of feedback device would be used mainly with
servo motor applications where precise feedback of rotor position is critical
to system accuracy. For example, in cut-to-length applications, linear posi-
tion of a sliding table or cutting arm would be a function of the position of
the rotor. Figure 5-12 shows a diagram of a resolver.

Stator
Output
(X)

Input Rotor

Stator
Output
(Y)

Figure 5-12. Resolver diagram

A resolver is very similar to an AC induction motor. A resolver contains a

single winding rotor that rotates inside fixed coils of wire, called stators. A
reference voltage is typically applied to the rotor winding. This rotating
winding has a magnetic field that induces a voltage in the stator windings,
which produce an analog output. This analog output is proportional to
shaft (rotor) rotation.

Basically, the resolver is a rotating transformer. The rotating primary

(rotor) induces a voltage in the fixed winding secondaries (stator wind-
ings). The analog output is what is fed back to the drive as an actual speed
signal, or what would be considered a position signal.
Motors&[Link] Page 216 Monday, January 10, 2005 1:54 PM

216 Motors and Drives

Because there are no electronics involved with resolvers, they are better
suited for dirty environments than encoders. Also, the resolver is a device
that is an absolute measuring instrument. It can retain its exact location
during a power outage. Typically the resolver can transmit information
over distances up to 1000 feet with little effect from electrical noise. The
resolution of some resolvers can be rated as high as 16,384 counts (14 bit
or 214).

Drive Control Methods: DC

As stated earlier, the commonly used methods of speed control are open
and closed loop. If speed regulation is not a factor, then a DC motor can be
operated in open-loop control. However, most applications require some
type of regulation to gain the most efficient use of the mechanics of the
system. Therefore, a means of sending the drive an actual speed signal is
essential to speed regulation. In DC drives there are basically two forms of
closed-loop control—tach feedback and armature voltage feedback (EMF
control). Though armature voltage feedback does not use an external
device, it is termed feedback and can be considered a form of closed-loop
control.

Armature Voltage Feedback (EMF Control, Speed Regulation)

Figure 5-13 shows an armature voltage feedback control scheme.

_ Armature Voltage Feedback (EMF)

Speed
+ Reference

Curr Pwr Out

Spd
Amp Arm
Amp
Summing
Circuit _ Current Feedback
+ IR Compensation Feedback

Figure 5-13. Armature voltage feedback (EMF control)

As shown in Figure 5-13, the drive requires a speed reference signal and a
feedback signal of opposite polarity. The feedback is used to balance the
control when the desired output speed is reached. All of the drive control
systems are within the dotted lines.

A speed reference is sent to the summing circuit. By sensing the armature

voltage at the drive output, the drive can sense the CEMF (Counter Elec-
tromotive Force) of the motor. This CEMF signal (negative polarity) is sent
Motors&[Link] Page 217 Monday, January 10, 2005 1:54 PM

Chapter 5: Drive Control and Feedback Devices — Drive Control Methods: DC 217

as feedback to the summing circuit. When the error is at zero, the drive
will stabilize at the desired speed.

Another summing circuit is located after the speed amplifier and before
the current amplifier. This summing circuit would use a shunt or other
device to sense the armature current. The negative current feedback is
sent to the summing circuit, with the resulting signal used to limit the
amount of current output. If the current level is within limits, then the
speed signal will be in control. But if the current exceeds the limits, it will
lower the speed control until the current is reduced to a safe level.

With armature voltage feedback, motor speed tends to droop between full-
load and no-load situations. To help compensate for this speed “droop,” a
feedback called IR compensation (an acronym for “current resistance”
compensation due to a voltage drop across the armature due to load –
E=IXR, Ohm’s Law) is included in the drive. This circuit senses the arma-
ture current and feeds a small additional signal back to the speed amplifier.

At the summing circuit, three signals exist: positive speed reference, nega-
tive armature voltage feedback (EMF), and positive IR compensation. The
IR compensation signal adds to the speed reference signal to compensate
for speed droop created by the load.

With three signals summing at the same point, there is a possibility for
instability. To set up IR compensation properly, speed and armature volt-
age feedback adjustments should be made with the IR compensation off.
While observing the motor during speed step changes, IR compensation is
gradually increased until oscillation occurs. Then IR compensation is
decreased until the oscillation (instability) stops. Speed regulation of 2–3%
is possible with this type of feedback control.

Armature Voltage Feedback (EMF Control, Torque Regulation)

The relationship between torque regulation and speed regulation in a
standard DC drive configuration illustrates the importance of torque
response. Since the armature current in a DC motor directly determines
torque, the DC controller is configured as a closed-loop current regulator,
using armature voltage feedback (EMF). The speed regulator then com-
mands the current regulator to produce whatever torque is required to
maintain the desired speed.

Torque-regulating drives are often used in load-sharing applications where

a speed-regulating drive controls the speed of the driven machine, while a
torque-regulated “helper” drive provides a controlled level of torque at
some other location on the machine. If the load does not restrict the speed
of a torque-regulated drive, the drive speed could exceed the safe operat-
ing limit. Therefore torque-regulating drives must have a speed-limiting
mechanism that prevents the speed from exceeding a safe limit if the
torque presented by the driven machine drops to zero.
Motors&[Link] Page 218 Monday, January 10, 2005 1:54 PM

218 Motors and Drives

With a DC drive, torque can be regulated directly by regulating armature

current. In any motor, torque is the result of the force between two mag-
netic fields. In a motor (DC), torque is easily and directly regulated by reg-
ulating the currents that control the flux in the two magnetic fields. The
field winding flux is the motor’s magnetizing flux, which is held constant
by providing a constant field current. The motor’s torque-producing flux is
the flux created by the armature current, which is controlled to regulate
torque. The torque produced at any speed is given by:

Torque = Κ×Φ×IA

where:

K = a constant

Φ = the magnetic flux produced by the stator field

IA = the armature current

Tachometer Feedback
When DC motor speed is of primary concern, it can be measured with a
transducer and regulated with a closed-loop regulator as shown if Figure
5-14.

Power

Motor
Speed
Setpoint
Regulator Drive
Tachometer

Feedback Signal

Figure 5-14. Closed-loop speed regulation—tach feedback

The transducer in Figure 5-14 is a tachometer generator. As previously

reviewed, a tach is a small generator that produces an output voltage that
is very accurately determined by its operating speed. There are also pulse
tachs, which provide a train of voltage pulses at an average frequency that
is exactly proportional to average speed.

The closed-loop speed regulator compensates for any changes in the char-
acteristics of the drive caused by changes in load or by outside influences
such as line voltage and ambient temperature. With a closed-loop speed
regulator, the most important characteristic of the drive is its ability to rap-
idly respond to changes in requirements for torque.

The transducer devices already presented would be involved in generating

the actual speed feedback signal. The accuracy of the system will be dic-
Motors&[Link] Page 219 Monday, January 10, 2005 1:54 PM

Chapter 5: Drive Control and Feedback Devices — Drive Control Methods: AC 219

tated by the regulation of the feedback device and the responsiveness of

the drive control.

Drive Control Methods: AC

The characteristics of open- and closed-loop control have already been dis-
cussed. In this section, the focus will be on AC drive control, with com-
monly used peripheral devices such as transducers, sensors, and other
control inputs. In addition, the performance (static and dynamic) and sta-
bility of the system will be explored.

Closed-loop regulation can be used with an adjustable-speed drive to reg-

ulate a variety of processes. Figure 5-15 shows the regulation of air pres-
sure in the duct of a ventilation system.

Air
Power Outlet
Fan
Pressure Motor Air
Setpoint Duct
Regulator Drive

Feedback Signal
Pressure
Transducer
Figure 5-15. Armature voltage feedback (EMF control)

As air-outlet dampers are opened and closed, the speed of the fan must be
increased or decreased to match the demand for air flow and maintain a
constant static pressure in the duct. The closed-loop control system uses a
pressure transducer in the duct to measure the static pressure. The air-
pressure feedback is sent to the drive, which adjusts the speed of the fan as
required. The regulator control loop not only adjusts for changes in air-
flow requirements but also compensates the characteristics of all of the
equipment such as the drive motor and fan that are “inside the loop.”

When an adjustable-speed drive is inside the loop in a closed-loop control

system, the speed-regulating accuracy of the drive is not the critical ele-
ment. The movement of air or water volume is not an “exacting science.”
An exact amount of flow (measured in cubic feet per minute [cfm]) is not
required in this particular application. The drive may have to pump air for
a few more seconds to meet the demand. In this type of a closed-loop con-
trol system, the drive needs only to provide the required torque and
respond to speed-correcting commands from the regulator.

However, if this were an application where drive speed directly determines

the accuracy of the process, then the drive would be a critical part of the
system. Such an application could be a coating line that processes paper
Motors&[Link] Page 220 Monday, January 10, 2005 1:54 PM

220 Motors and Drives

stock into 12-ounce coffee cups. Precision control is needed to ensure that
the cups are manufactured to the exact dimensions for 12-ounce capacity.

Feedback and Performance

The feedback devices of tachometers, encoders, and resolvers have already
been presented. This section will be devoted to the overall performance of
the AC drive system. The terminology will be presented and the perfor-
mance characteristics reviewed.

In earlier sections, only steady-state, or static, performance has been

reviewed. Static operation is where there are no changes in conditions.
Figure 5-16 shows a drive’s static and dynamic speed-regulating perfor-
mance.

Speed Static Speed Regulation

Steady State
Operating Point Steady State
Transition Operating Point
(Dynamic
Performance)
Static Static
Operation Operation

Torque Load
Change

Time
Figure 5-16. Static and dynamic performance of a system

The figure indicates static operating conditions before and after a load
change with a transition period of dynamic performance immediately after
the change. Static speed regulation is the change in steady-state speed that is
caused by a load change. Static performance measures the difference
between two operating points, without considering the performance dur-
ing the transition from one point to the other. At each point, operation is
measured only after the system has been operating at that point for some
length of time. Sufficient time is allowed so no further change will occur in
operation related to the transition from one point to another.

Dynamic performance describes the operation during the transition from one
operating point to another. The dynamic performance capability of a sys-
tem defines the system’s ability to respond to a load change or a reference
change.
Figure 5-17 is an enlarged view of Figure 5-16, showing the parameters
that quantify both the static response and the dynamic response to a step
change in load.
Motors&[Link] Page 221 Monday, January 10, 2005 1:54 PM

Chapter 5: Drive Control and Feedback Devices — Drive Control Methods: AC 221

Regulation
Drift

Speed
Transient Deviation
Steady State
Transient Response Time Regulation
Band

Time (seconds) Time (hours)

Figure 5-17. Step changes of load and the response

The transient deviation is the maximum deviation from set point immedi-
ately following the load change. The transient-response time is time required
for the output to return to the steady-state regulation band after going
through a period of damped oscillation. The steady-state regulation band is a
small “dead band” of output variation that is not recognized by the regula-
tor as a change. The regulation dead band is caused by regulator and trans-
ducer resolution limitations. The steady-state regulation is the change in
steady-state output resulting from a load change. Drift is the change in
steady-state output because of temperature changes and other long-term
influencing factors. Drift is usually specified for a 24-hour period.

It is difficult to use transient-response time to compare the performance of

two types of drives because total-system response time is determined by
load inertia. Figure 5-18 shows another way of quantifying the dynamic
change in output speed because of a step change in load torque.

Torque Load Change

100%
Low Inertia Load High Inertia Load
0%

Speed Resulting Transient Speed Change

40ms 80ms
100%

90%
Shaded Area = 0.3% Sec. Dynamic Speed Accuracy

Time
Figure 5-18. Dynamic speed accuracy

The dynamic speed accuracy is the area under the transient-response curve
measured in percent-seconds.

For the low-inertia load, the figure shows that the maximum transient
speed deviation is 15% and the response time is 40 ms. The dynamic speed
Motors&[Link] Page 222 Monday, January 10, 2005 1:54 PM

222 Motors and Drives

accuracy is the area of the shaded triangle or 15% x 40 ms/2 = 0.3%-sec-

onds. For the high-inertia load, the dynamic speed accuracy is 7.5% x 80
ms/2 = 0.3%-seconds. This example shows that the dynamic speed accu-
racy is about the same for a low-inertia load as it is for a high-inertia load.

Figure 5-19 shows the dynamic response of a drive resulting from a step
change in reference.

Reference - Step Change

100%
0%
Output - Response
Peak Overshoot
90%
Rise Time
Speed
Settling Time
10%
Time
Figure 5-19. Reference step change and response

The parameters that quantify the performance are rise time, peak over-
shoot, and settling time. The rise time is the time required for the output to
rise from 10 to 90% of its final value. The peak overshoot is the maximum
amount by which the output overshoots the final value. The settling time is
the rise time plus the time required for the output to reach a steady value
after going through a period of damped oscillation.

In many applications, it is important for a drive to accurately follow a

speed reference during acceleration and deceleration. In a web processing
machine for example, the speeds of the various machine sections must
match each other as a continuous web of material travels from one section
to another. When the master speed reference is increased or decreased, the
drives for each machine section must accurately follow the change. Figure
5-20 shows the dynamic deviation that occurs whenever the speed reference
changes.

Bandwidth, or small-signal bandwidth, is another parameter that is some-

times used to quantify a drive’s capability for accurately following a chang-
ing reference signal. Small-signal bandwidth is measured by applying a
small sinusoidal variation to the regulator reference and observing the
effect on the output. The bandwidth is the maximum frequency range of
input signal that the output can follow. Another name for bandwidth is
frequency response.

The bandwidth can be given in radians per second (ω) or in hertz (Hz). The
relationship between radians/sec and Hz is ω= 2πf. Bandwidth is inversely
Motors&[Link] Page 223 Monday, January 10, 2005 1:54 PM

Chapter 5: Drive Control and Feedback Devices — Drive Control Methods: AC 223

Dynamic
Deviation
Speed

Time
Figure 5-20. Dynamic deviation

proportional to the system time constant (t) or “Response Time” (ω= 1/t).
The response time is similar to the rise time shown in Figure 5-19. The
response time is the time required for the output to rise from 0 to 63% of
its final value.

The bandwidth of a drive defines the maximum capability of the control-

ler/motor combination with nothing connected to the motor shaft. The
bandwidth of a controller defines the maximum electrical output capabil-
ity of the controller without a motor connected to the output terminals.
The bandwidth of a system is the actual operating performance of the
drive and load when the drive is adjusted for optimum performance with
that specific load.

High-performance control systems often have multiple control loops as

shown in Figure 5-21.

Process Control Loop (for servo drives, position loop)

Bandwidth = ω

Speed Loop
Bandwidth = ω

Torque Loop
Bandwidth = ω

Process Speed Torque

Regulator Regulator Regulator Process

Figure 5-21. Bandwidth relationships in control loops

The outermost control loop regulates the process variable. An example of

this would be a position control loop in servo drive systems. A general-
Motors&[Link] Page 224 Monday, January 10, 2005 1:54 PM

224 Motors and Drives

purpose drive system might have a dancer position regulator that ulti-
mately controls the tension of a web or filament. The speed-regulator loop
is inside the process-regulator loop, and the torque-regulator loop is the
innermost loop.

To provide stable performance, each inner regulator loop must be 3–10

times faster than the next outer loop. That is, ω2 = 3–10 times ω1 and ω3 =
3–10 times ω2. If the process is subject to fast changes, ω1 is large. A high-
performance drive is required, and ω2 and ω3 must be large. Conversely, if
the driven machine has a high inertia or other characteristics that dictate
that its final output can change only slowly, then the drive does not need
to have wide speed- and torque-regulator bandwidths.

Stability of the System

The performance examples presented all have stable performance charac-
teristics. When the performance of a system is stable, the output is at a
steady value, except during periods of transition from one operating point
to another. During transitions, the output may oscillate, but the oscillation
is damped and rapidly decreases to a low value. In unstable systems, the
output may oscillate continuously for an extended period of time. Figure
5-22 illustrates responses to a step change in reference for a stable and
unstable system.

Critically
Damped

Slightly Overdamped
Underdamped Unstable

Figure 5-22. System responses—stable and unstable operation

Systems are usually adjusted for slightly under-damped or critically damped

operation. A system is critically damped when it responds with the fastest
rise time that is possible, without any overshoot.

System performance is determined by the interaction among all of the

components of the system. The components of a control system include
both the controlling system and the controlled system. An adjustable-
speed drive system includes the adjustable-speed controller, the motor, all
feedback and accessory devices, and the driven machine. The characteris-
tics of the driven machine, or load, are an essential factor in determining
the dynamic performance of an adjustable-speed drive system. Figure 5-23
Motors&[Link] Page 225 Monday, January 10, 2005 1:54 PM

Chapter 5: Drive Control and Feedback Devices — Drive Control Methods: AC 225

shows the load torque versus speed for a driven machine compared with
the torque capability of a drive.

Intermittent Torque
Capability of Drive

Torque Available For Accelerating Inertia Continuous Torque

Capability of Drive
100%

Load
Torque

Torque Available For Decelerating Inertia

0%
Speed

Figure 5-23. Accelerating and decelerating torque

The difference between the load torque and the intermittent torque capa-
bility of the drive is the torque available to accelerate or decelerate inertia.
The reflected load inertia, plus the motor inertia and the torque available
for acceleration or deceleration, determines the time to accelerate and
decelerate. The acceleration and deceleration time are the main compo-
nents of the drive’s response to a step change in speed reference.

Acceleration or deceleration time is given by:

(Load Inertia + Motor Inertia) × Speed Change

Time =
K × Torque

If the load inertia is expressed as WK2 (lb-ft2), speed change is given in

rpm, and torque is given in lb-ft, then the constant, K, is 308 for calculat-
ing time in seconds.
The drive’s torque response is a very important factor in determining the
drive’s dynamic performance. When the load torque increases suddenly,
or when the speed set point is suddenly changed, the drive is asked to
instantaneously change the level of torque that the motor is providing.
The drive’s torque response is the response time (typically milliseconds)
required for the drive to respond to a step increase of 0 to 100% torque
demand.

Accurately predicting the complete performance of a drive system and

driven machine requires detailed information. Drive and machine charac-
teristics must be known. To achieve optimum performance, the drive must
be adjusted (tuned) to the characteristics of the driven machine. One
aspect of tuning is to set the speed regulator gain adjustments for the best
regulation that can be achieved without getting too close to unstable oper-
ation. The system bandwidth must be tuned to a frequency that is below
any mechanical resonance frequencies of the driven machine.
Motors&[Link] Page 226 Monday, January 10, 2005 1:54 PM

226 Motors and Drives

For existing machine designs, drives can be selected based on comparing

the capabilities of available drives with the capabilities of drives that have
been successfully used in the past. For new machines, the keys to success
include experience with similar machines and close cooperation between
the machine design engineers and the drive manufacturer.

Sensors and Controls

A variety of sensors and controls are used to interface with the drive unit,
either AC or DC. In this section, attention will be given to the more com-
mon devices that control a drive or give information to the drive, in the
form of feedback.

Transducers
A transducer is a device that senses the condition, state, or value of an
item to be controlled and produces an output that reflects that condition,
state, or value. The following is a listing of typical transducers and sensors
used to feed information back to the drive. In some cases, the transducer
itself is the controlling element (determining a speed reference or set
point).

Temperature
Thermocouple and thermistor. This device changes resistance per the change
in temperature. Some thermocouples are mounted directly on machines
actually inside tanks to monitor temperature. This temperature is then fed
back to the drive to change the controlled value (increase or decrease the
flow of heat).

Flow
Electromagnetic flow meter. This device uses an electromagnet and fluid
motion to register an output. The moving fluid provides movement. An
EMF is detected by two electrodes embedded in the wall of the tube. The
electrodes are insulated from the liquid being measured.

Differential pressure flow meter. The fluid flow causes motion within the mea-
suring unit. A restriction in the orifice or pipeline causes an increase in
velocity and a decrease in pressure. This decrease is a rate of flow, and is
the controlling element.

Tension, Force, or Strain

Resistance gage. There are several forms of resistance gages. The ultimate
result is that the output is a change of resistance, per the unit being mea-
sured. A tension gage changes resistance per the tension applied to the sen-
sor. A strain gage changes resistance per the stretching of the wires inside
the sensor. This sensor type needs a power source to feedback a voltage
signal to the drive.
Motors&[Link] Page 227 Monday, January 10, 2005 1:54 PM

Chapter 5: Drive Control and Feedback Devices — Sensors and Controls 227

Displacement of Position
Potentiometers, proximity sensors, photo cells. These sensors change output
resistance per the controlling element. Potentiometers change resistance
per the location of the wiper arm or shaft. Linear pots are constructed in a
linear fashion with a slider arm, as opposed to a rotary arm as in a stan-
dard pot. Proximity sensors consist of two plates that change capacitance
per how close they are to the controlling element. Photo cells are available
as photovoltaic, which sends out a voltage per the amount of the control-
ling element. There are also photoconductive and photoresistive devices.
Photoresistive devices change resistance per the amount of light sensed by
a light-sensitive base plate.

Pressure
Diaphragm or bourdon tube. These are used to sense fluid pressures. The
movement of the internal device is proportional to the pressure changes.
The output is a variable resistance, which requires a voltage source to pro-
vide feedback to the drive. Three to fifteen PSI or static pressure transduc-
ers are commonly used in the HVAC industry to monitor the static
pressure in the heating or cooling ducts. These devices use the principle
listed above, but include an actual transmitter that would deliver a 4- to
20-mA output. That output would be accepted directly into the drive feed-
back circuit, with no additional power source needed. Pressure transducers
(4–20 mA) can effectively transmit signals several hundred feet back to the
drive with little loss of signal.

Level
Level transducers. Level transducers are available in a variety of controlling
element styles. A float device is nothing more than a variable resistor, with
the wiper arm connected to a flotation device that sits directly on top of
the fluid being monitored. Some photoelectric level transducers change
the voltage output per the density of the fluid being monitored. Capaci-
tive-type sensors change the amount of capacitance per the proximity of
the fluid to the sensing plates. A certain amount of conductivity would be
needed in the fluid being monitored.

Thickness
X-ray and ultrasonic. These devices change the output voltage or current per
the amount of material being sensed. An X-ray sensor has to monitor X-
ray-sensitive materials for the sensing element to be effective. Ultrasonic
sensors use an electronic oscillator to sense the thickness of the material. A
receiver–transmitter is used for this purpose, along with a bridge circuit
that does voltage comparisons.

Humidity
Hygrometer. This device uses a medium that changes dimensions according
to the amount of humidity in the atmosphere. A hair hygrometer uses
strands of human hair attached to a pivot arm. As the humidity changes,
the pivot arm moves because of changes in hair length. The pivot arm is
Motors&[Link] Page 228 Monday, January 10, 2005 1:54 PM

228 Motors and Drives

attached to a variable resistor, which can be connected to a power source

to generate a voltage feedback signal.

Density
Float devices. These devices change the amount of voltage output per the
density of the fluid being monitored. A rod (moveable core) is attached to
a float device that moves up or down, depending on material density. The
moveable core fits inside a linear differential transformer. The transformer
is energized with AC voltage, with the output changing value, depending
on how high or low the moveable core slides into the differential trans-
former. Some sort of rectification is needed before this type of signal could
be fed back to the drive.

There are a variety of other sensors and feedback devices not mentioned
above. The point to remember is that the drive needs an electrical feed-
back or reference signal to change speed or torque (current) output. All
drives accept a standard analog input value of 0–10 VDC or 4–20 mA DC.
These values could be a set point (reference) or a feedback. The drive can
respond to the analog input signals of the voltage or current values stated
above. The drive does not care what the controlling element is, as long as
the analog input is of the value and type it can respond to.

Chapter Review
The two types of drive control methods are open and closed loop. Open
loop is operating a motor directly from the drive unit. Closed loop includes
some type of feedback device that is connected directly to the motor shaft.
The speed regulation of the system can be improved by using a feedback
device.
Typical drive feedback devices include tachometers (tachs), encoders (or
pulse tachs), and resolvers. Tachometers are available in AC or DC ver-
sions. DC tachs are used in a variety of industrial applications, since they
are able to indicate direction and are generally more accurate than AC
tachs.

Encoders use a system of light beams directed toward a light sensor. The
light beams are created by shining a light source through a rotating, slotted
disk. The disk is connected to the encoder shaft, which is connected
directly to the motor shaft through a coupling.

Resolvers are position-type devices. These devices are typically used on

servo applications, where precise feedback of the rotor position is required.
Resolvers are similar to an AC induction motor, as well as acting like a
revolving transformer.

DC drive control methods include closed-loop control using a tach feed-

back device. It also includes armature voltage feedback or EMF control.
With EMF control, a sampling of the output voltage is used as feedback to
correct for output speed and current (torque).
Motors&[Link] Page 229 Monday, January 10, 2005 1:54 PM

Chapter 5: Drive Control and Feedback Devices — Check Your Knowledge 229

AC drive control methods also include tach or encoder feedback, as well as

operating the VFD by open-loop methods. A variety of transducers can be
used as feedback devices. Pressure, flow, level, temperature, and humidity
are just a few of the controlling elements sensed by transducers.

Dynamic and static speed regulation are part of the setup process of the
performance drive application. The accuracy of the speed or torque is
dependent on the feedback device, as well as the closed-loop control cir-
cuit within the drive. The amount of dampening will determine if the sys-
tem will be stable or produce oscillations.

Check Your Knowledge

1. Describe open-loop vs. closed-loop control.
2. What is a pulse tach and how does it work?
3. What is the difference between an AC and a DC tach?
4. How is the resolution of an encoder indicated?
5. Describe the EMF, current, and IR compensation feedback loops in a DC
drive system.
6. How is a pressure transducer used in a closed-loop AC drive system?
7. Describe the difference between static and dynamic performance of a
closed-loop system.
8. What is bandwidth?
9. How is stable operation obtained with an AC drive system?
10. What is the operating principle of temperature and humidity sensors?