Industrial Equipment

Appendix IV AC Motors & Variable Speed Drives

Industrial Equipment........................................................................................................... 1
Appendix IV AC Motors & Variable Speed DrivesRotor Slip in induction motors ....... 1
Rotor Slip in induction motors............................................................................................ 3
Motor slip is necessary for torque generation................................................................. 3
Slip depends on motor parameters. ................................................................................. 4
Methods to reduce slip. ................................................................................................... 5
Adjustable speed AC drive is often the best solution. .................................................... 5
Induction Motor Control Theory. ....................................................................................... 7
Stator design.................................................................................................................... 7
Rotor Design. .................................................................................................................. 7
Equivalent Circuit. .......................................................................................................... 8
Starting Characteristics. .............................................................................................. 8
Runnng Characteristics. ................................................................................................ 10
Design Classification. ................................................................................................... 11
Frame Classification. .................................................................................................... 11
Temperature Classification. .......................................................................................... 11
Power factor correction................................................................................................. 12
Single phase motors. ..................................................................................................... 13
Slip Ring Motors........................................................................................................... 13
Variable-frequency drives................................................................................................. 14
Technology development.............................................................................................. 14
Early VFDs ................................................................................................................... 14
Matching Motors and Controls ..................................................................................... 15
Nonlinear issues ............................................................................................................ 16
VFD Justification: A Mechanical Perspective.............................................................. 17
Recent developments .................................................................................................... 19
Specifying VFDs........................................................................................................... 20
Field installation............................................................................................................ 20

Figure 1 AC Motor Cutaway .............................................................................................. 3

Figure 2 Induction Motor Speed Curve .............................................................................. 4
Figure 3 AC Motor Torque & Speed .................................................................................. 5
Figure 4 Slip Compensation................................................................................................ 6
Figure 5 AC Motor Speed & Torque .................................................................................. 9
Figure 6 VFD & Pumps .................................................................................................... 18
Figure 7 VFD on Centrifugal Fans ................................................................................... 19

Rotor Slip in induction motors

Although slip is a common problem with standard motors, several options exist for
reducing its effects
The AC induction motor is often referred to as the workhorse of the industry because it
offers users simple, rugged construction, easy maintenance, and cost-effective pricing.
These factors have promoted standardization and development of a manufacturing
infrastructure that has led to a vast installed base of motors; more than 90% of all motors
used in worldwide industry are AC induction motors.
In spite of this popularity, the AC induction motor has two basic limitations. The standard
motor is neither a true constant-speed machine, nor is it inherently capable of providing
variable-speed operation. Both limitations require consideration, as the quality and
accuracy requirements of motor/drive applications continue to increase.
This article will explore the reason for slip and discuss ways to minimize it. In addition, it
will detail the best methods now available for controlling motor speed with power
electronics, including technology to minimize the negative effects of slip.

Motor slip is necessary for torque generation.

An AC induction motor consists of two basic assemblies: the stator and rotor. The stator
structure is composed of steel laminations shaped to form poles. Copper wire coils are
wound around these poles. These primary windings are connected to a voltage source
to produce a rotating magnetic field. Three-phase motors with windings spaced 120
electrical degrees apart are standard for
industrial, commercial, and residential
use.
The rotor is another assembly made of
laminations over a steel shaft core.
Radial slots around the laminations'
periphery house rotor bars, which are
cast-aluminium or copper conductors
shorted at the ends and positioned
parallel to the shaft. Arrangement of the
rotor bars looks like a squirrel cage,
hence the well-known term, squirrel
cage induction motor (Photo above).
The term induction motor comes from
Figure 1 AC Motor Cutaway
the alternating current (AC) that's
This cutaway view of a squirrel cage AC
induced into the rotor via the rotating
induction motor shows the stator and rotor
magnetic flux produced in the stator.
construction, the shaft with bearings, and the
Motor torque is developed from the cooling fan.
interaction of currents that flow in the
rotor bars and the stator's rotating magnetic field. In actual operation, rotor speed always
lags the magnetic field's speed, allowing the rotor bars to cut magnetic lines of force and
produce useful torque. This speed difference is called slip speed. Slip also increases
with load and is necessary for producing torque.

Slip depends on motor parameters.

According to the formal definition, the slip (S) of an induction motor can be found with
the following equation:

s = [(n2 - n)/ns] x 100%

where ns is synchronous speed
and n is actual speed.
For small values of motor slip,
the slip is proportional to the
rotor resistance, stator voltage
frequency, and load torque. It's
inversely proportional to the
second power of supply voltage.
The traditional way to control the
speed of a wound rotor induction
motor is to increase the slip by
adding resistance in the rotor
circuit. The slip of
lowhorsepower motors is higher
than those of high-horsepower
motors because of higher rotor
winding resistance in smaller Motor slip of selected aluminium and cast iron NEMA
motors with synchronous speed ranging from 3,600 RPM
motors.
to 900 RPM.

As seen in the Table above,

smaller motors and lower-speed motors typically have higher relative slip. However,
high-slip large motors and low-slip small motors are also available.
You can see that full-load slip varies
from less than 1% in high-hp motors
to more than 5% in fractional-hp
motors. These variations may cause
load-sharing problems when motors
of different sizes are mechanically
connected. At low load, the sharing is
normally not a problem, but at full
load, the motor with lower slip takes a
higher share of the load than the
motor with higher slip.
As shown in Fig. 1 at right, the rotor
speed decreases in proportion to the
load torque. This means that the rotor
slip increases in the same proportion.
Figure 2 Induction Motor Speed Curve

Relatively high rotor impedance is

required for good across-the-line, or This is a typical speed curve of an induction motor.
full voltage, starting performance. In The slip is the difference in rotor speed relative to that
of the synchronous speed. CD = AD BD = AB.
other words, high torque is required
against low current. Low rotor impedance is also necessary for low full-load speed slip
and high operating efficiency. The curves in Fig. 2 show how higher rotor impedance in

motor B reduces the starting current and increases the starting torque, but it causes a
higher slip than in standard motor A.

Methods to reduce slip.

Synchronous,
reluctance,
or
permanent-magnet (PM) motors can
solve the problem of slip because
there's no measurable slip in these
three types of motors. Synchronous
motors are used for very high- and
low-power applications, but to a
lesser extent in the mediumhorsepower range, where many
typical industrial applications fall.
Reluctance motors are also used, but
their output/weight ratio isn't very
good, so they're less competitive than
the squirrel cage induction motor. PM
motors, which are used with
Figure 3 AC Motor Torque & Speed
electronic adjustable speed drives
(ASDs) offer accurate speed control These curves depict torque/speed and current/speed
without slip, high efficiency with low for a standard motor (A) and a high-torque motor (B).
rotor losses, and the flexibility to
choose a very low base speed, eliminating the need for gear boxes. However, PM
motors are still limited to certain special applications, mainly because of high cost and
the lack of standardization.
Selecting an oversized AC induction motor is another way to reduce slip. Larger motors
typically have a lower slip value to begin with, and slip gets smaller with a partial, rather
than full, motor load. The disadvantage with oversizing the motor is that with a larger
motor comes higher energy consumption, which increases investment and operation
costs.

Adjustable speed AC drive is often the best solution.

The inherent limitations of the AC induction motor can be solved with ASDs. The most
common AC drives today are based on pulse-width modulation (PWM). The constant AC
line voltage with 60 cycles per second from the supply network is rectified, filtered, and
then converted to a variable voltage and variable frequency. When this output from the
frequency converter is connected to an AC motor, it's possible to adjust the motor speed.
When using an AC drive for adjusting the motor speed, there are many applications
where motor slip is no longer a problem. The speed of the motor isn't the primary control
parameter. Rather, it could be the liquid level, air pressure, gas temperature, or some
other controlling parameter.

High static speed accuracy and/or

dynamic speed accuracy are still
required in many drive applications,
such as printing machines, extruders,
paper
machines,
cranes,
and
elevators. There are also many
machines and conveyors where
speed control between sections
driven by separate motors has to
be
synchronized.
Instead
of
oversizing the motors to eliminate the
speed error caused by slip, it may be
better to use sectional drive line-ups
with separate inverters for each
individual motor. The inverters are
Figure 4 Slip Compensation
connected to a DC-voltage bus bar
supplied by a common rectifier. This The addition of slip compensation to the AC drive
is a very energy-efficient solution, helps reduce overall slip.
because the driving sections of the
machinery can use the braking energy from decelerating sections (regeneration).
Slip compensation can even be added to AC drives to further reduce the effect of motor
slip. A load torque signal is added to the speed controller to increase the output
frequency in proportion to the load. Slip compensation can't be 100% of the slip because
of rotor temperature variations that may cause overcompensation and unstable control.
But the compensation can achieve accuracies up to 80%, meaning slip can be reduced
from 2.4% to about 0.5%.

Induction Motor Control Theory.

Induction Motor Design
has a major effect on the
behaviour and
performance of an
induction motor. Very
often the details or class
of design of a motor are
not well understood or
promoted.

Stator design.
The stator is the outer body of the motor which houses the driven windings on an iron
core. In a single speed three phase motor design, the standard stator has three
windings, while a single phase motor typically has two windings. The stator core is made
up of a stack of round pre-punched laminations pressed into a frame which may be
made of aluminium or cast iron. The laminations are basically round with a round hole
inside through which the rotor is positioned. The inner surface of the stator is made up of
a number of deep slots or grooves right around the stator. It is into these slots that the
windings are positioned. The arrangement of the windings or coils within the stator
determines the number of poles that the motor has.
A standard bar magnet has two poles, generally known as North and South. Likewise,
an electromagnet also has a North and a South pole. As the induction motor Stator is
essentially like one or more electromagnets depending on the stator windings, it also has
poles in multiples of two. i.e. 2 pole, 4 pole, 6 pole etc.
The winding configuration, slot configuration and lamination steel all have an effect on
the performance of the motor. The voltage rating of the motor is determined by the
number of turns on the stator and the power rating of the motor is determined by the
losses which comprise copper loss and iron loss, and the ability of the motor to dissipate
the heat generated by these losses.
The stator design determines the rated speed of the motor and most of the full load, full
speed characteristics.

Rotor Design.
The Rotor comprises a cylinder made up of round laminations pressed onto the motor
shaft, and a number of short-circuited windings.The rotor windings are made up of rotor
bars passed through the rotor, from one end to the other, around the surface of the rotor.
The bars protrude beyond the rotor and are connected together by a shorting ring at
each end. The bars are usually made of aluminium or copper, but sometimes made of
brass. The position relative to the surface of the rotor, shape, cross sectional area and
material of the bars determine the rotor characteristics. Essentially, the rotor windings
exhibit inductance and resistance, and these characteristics can effectively be
dependant on the frequency of the current flowing in the rotor.

A bar with a large cross sectional area will exhibit a low resistance, while a bar of a small
cross sectional area will exhibit a high resistance. Likewise a copper bar will have a low
resistance compared to a brass bar of equal proportions.
Positioning the bar deeper into the rotor, increases the amount of iron around the bar,
and consequently increases the inductance exhibited by the rotor. The impedance of the
bar is made up of both resistance and inductance, and so two bars of equal dimensions
will exhibit a different A.C. impedance depending on their position relative to the surface
of the rotor. A thin bar which is inserted radialy into the rotor, with one edge near the
surface of the rotor and the other edge towards the shaft, will effectively change in
resistance as the frequency of the current changes. This is because the A.C. impedance
of the outer portion of the bar is lower than the inner impedance at high frequencies
lifting the effective impedance of the bar relative to the impedance of the bar at low
frequencies where the impedance of both edges of the bar will be lower and almost
equal.
The rotor design determines the starting characteristics.

Equivalent Circuit.
The induction motor can be treated essentially as a transformer for analysis. The
induction motor has stator leakage reactance, stator copper loss elements as series
components, and iron loss and magnetising inductance as shunt elements. The rotor
circuit likewise has rotor leakage reactance, rotor copper (aluminium) loss and shaft
power as series elements.
The transformer in the centre of the equivalent circuit can be eliminated by adjusting the
values of the rotor components in accordance with the effective turns ratio of the
transformer.
From the equivalent circuit and a basic knowledge of the operation of the induction
motor, it can be seen that the magnetising current component and the iron loss of the
motor are voltage dependant, and not load dependant. Additionally, the full voltage
starting current of a particular motor is voltage and speed dependant, but not load
dependant.
The magnetising current varies depending on the design of the motor. For small motors,
the magnetising current may be as high as 60%, but for large two pole motors, the
magnetising current is more typically 20 - 25%. At the design voltage, the iron is typically
near saturation, so the iron loss and magnetising current do not vary linearly with voltage
with small increases in voltage resulting in a high increase in magnetising current and
iron loss.

Starting Characteristics.
In order to perform useful work, the induction motor must be started from rest and both
the motor and load accelerated up to full speed. Typically, this is done by relying on the
high slip characteristics of the motor and enabling it to provide the acceleration torque.
Induction motors at rest, appear just like a short circuited transformer, and if connected
to the full supply voltage, draw a very high current known as the "Locked Rotor Current".
They also produce torque which is known as the "Locked Rotor Torque". The Locked
Rotor Torque (LRT) and the Locked Rotor Current (LRC) are a function of the terminal
voltage to the motor, and the motor design. As the motor accelerates, both the torque
and the current will tend to alter with rotor speed if the voltage is maintained constant.

The starting current of a motor, with a fixed voltage, will drop very slowly as the motor
accelerates and will only begin to fall significantly when the motor has reached at least
80% full speed. The actual curves for induction motors can vary considerably between
designs, but the general trend is for a high current until the motor has almost reached full
speed. The LRC of a motor can range from 500% Full Load Current (FLC) to as high as
1400% FLC. Typically, good motors fall in the range of 550% to 750% FLC.

Figure 5 AC Motor Speed & Torque

The starting torque of an induction motor starting with a fixed voltage, will drop a little to
the minimum torque known as the pull up torque as the motor accelerates, and then rise
to a maximum torque known as the breakdown or pull out torque at almost full speed
and then drop to zero at synchronous speed. The curve of start torque against rotor
speed is dependant on the terminal voltage and the motor/rotor design.
The LRT of an induction motor can vary from as low as 60% Full Load Torque (FLT) to
as high as 350% FLT. The pull-up torque can be as low as 40% FLT and the breakdown
torque can be as high as 350% FLT. Typical LRTs for medium to large motors are in the
order of 120% FLT to 280% FLT.
The power factor of the motor at start is typically 0.1 - 0.25, rising to a maximum as the
motor accelerates, and then falling again as the motor approaches full speed.
A motor which exhibits a high starting current, i.e. 850% will generally produce a low
starting torque, whereas a motor which exhibits a low starting current, will usually
produce a high starting torque. This is the reverse of what is generally expected.
The induction motor operates due to the torque developed by the interaction of the stator
field and the rotor field. Both of these fields are due to currents which have resistive or in
phase components and reactive or out of phase components. The torque developed is
dependant on the interaction of the in phase components and consequently is related to
the I2R of the rotor. A low rotor resistance will result in the current being controlled by the
inductive component of the circuit, yielding a high out of phase current and a low torque.

Figures for the locked rotor current and locked rotor torque are almost always quoted in
motor data, and certainly are readily available for induction motors. Some manufactures
have been known to include this information on the motor name plate. One additional
parameter which would be of tremendous use in data sheets for those who are
engineering motor starting applications, is the starting efficiency of the motor. By the
starting efficiency of the motor, I refer to the ability of the motor to convert amps into
newton meters. This is a concept not generally recognised within the trade, but one
which is extremely useful when comparing induction motors. The easiest means of
developing a meaningful figure of merit, is to take the locked rotor torque of the motor
(as a percentage of the full load torque) and divide it by the locked rotor current of the
motor (as a percentage of the full load current).
i.e
Starting efficiency =

Locked Rotor Torque

Locked Rotor Current

If the terminal voltage to the motor is reduced while it is starting, the current drawn by
the motor will be reduced proportionally. The torque developed by the motor is
proportional to the current squared, and so a reduction in starting voltage will result in a
reduction in starting current and a greater reduction in starting torque. If the start voltage
applied to a motor is halved, the start torque will be a quarter, likewise a start voltage of
one third will result in a start torque of one ninth.

Running Characteristics.
Once the motor is up to speed, it operates at low slip, at a speed determined by the
number of stator poles. The frequency of the current flowing in the rotor is very low.
Typically, the full load slip for a standard cage induction motor is less than 5%. The
actual full load slip of a particular motor is dependant on the motor design with typical full
load speeds of four pole induction motor varying between 1420 and 1480 RPM at 50 Hz.
The synchronous speed of a four pole machine at 50 Hz is 1500 RPM and at 60 Hz a
four pole machine has a synchronous speed of 1800 RPM.
The induction motor draws a magnetising current while it is operating. The magnetising
current is independent of the load on the machine, but is dependant on the design of the
stator and the stator voltage. The actual magnetising current of an induction motor can
vary from as low as 20% FLC for large two pole machines to as high as 60% for small
eight pole machines. The tendency is for large machines and high speed machines to
exhibit a low magnetising current, while low speed machines and small machines exhibit
a high magnetising current. A typical medium sized four pole machine has a magnetising
current of about 33% FLC.
A low magnetising current indicates a low iron loss, while a high magnetising current
indicates an increase in iron loss and a resultant reduction in operating efficiency.
The resistive component of the current drawn by the motor while operating, changes
with load, being primarily load current with a small current for losses. If the motor is
operated at minimum load, i.e. open shaft, the current drawn by the motor is primarily
magnetising current and is almost purely inductive. Being an inductive current, the power
factor is very low, typically as low as 0.1. As the shaft load on the motor is increased, the
resistive component of the current begins to rise. The average current will noticeably
begin to rise when the load current approaches the magnetising current in magnitude.

As the load current increases, the magnetising current remains the same and so the
power factor of the motor will improve. The full load power factor of an induction motor
can vary from 0.5 for a small low speed motor up to 0.9 for a large high speed machine.
The losses of an induction motor comprise: iron loss, copper loss, windage loss and
frictional loss. The iron loss, windage loss and frictional losses are all essentially load
independent, but the copper loss is proportional to the square of the stator current.
Typically the efficiency of an induction motor is highest at 3/4 load and varies from less
than 60% for small low speed motors to greater than 92% for large high speed motors.
Operating power factor and efficiencies are generally quoted on the motor data sheets.

Design Classification.
There are a number of design/performance classifications which are somewhat uniformly
accepted by different standards organisations. These design classifications apply
particularly to the rotor design and hence affect the starting characteristics of the motors.
The two major classifications of relevance here are design A, and design B.
Design A motors have a shallow bar rotor, and are characterised by a very high starting
current and a low starting torque. Typical values are 850% current and 120% torque.
Shallow bar motors usually have a low slip, i.e. 1480 RPM.
Design B motors have a deeper bar rotor and are characterised by medium start current
and medium starting torque. Typical design B values are 650% current and 180%
torque. The slip exhibited by design B motors is usually greater than the equivalent
design A motors. i.e. 1440 RPM.
Design F motors are often known as Fan motors having a high rotor resistance and high
slip characteristics. The high rotor resistance enables the fan motor to be used in a
variable speed application where the speed is reduced by reducing the voltage. Design
F motors are used primarily in fan control applications with the motor mounted in the air
flow. These are often rated as AOM or Air Over Motor machines.

Frame Classification.
Induction motors come in two major frame types, these being Totally Enclosed Forced
air Cooled (TEFC), and Drip proof.
The TEFC motor is totally enclosed in either an aluminium or cast iron frame with cooling
fins running longitudinally on the frame. A fan is fitted externally with a cover to blow air
along the fins and provide the cooling. These motors are often installed outside in the
elements with no additional protection and so are typically designed to IP55 or better.
Drip proof motors use internal cooling with the cooling air drawn through the windings.
They are normally vented at both ends with an internal fan. This can lead to more
efficient cooling, but requires that the environment is clean and dry to prevent insulation
degradation from dust, dirt and moisture. Drip proof motors are typically IP22 or IP23.

Temperature Classification.
There are two main temperature classifications applied to induction motors. These being
Class B and Class F.The temperature class refers to the maximum allowable
temperature rise of the motor windings at a specified maximum coolant temperature.

Class B motors are rated to operate with a maximum coolant temperature of 40 degrees
C and a maximum winding temperature rise of 80 degrees C. This leads to a maximum
winding temperature of 120 degrees C.
Class F motors are typically rated to operate with a maximum coolant temperature of 40
degrees C and a maximum temperature rise of 100 degrees C resulting in a potential
maximum winding temperature of 140 degrees C.
Operating at rated load, but reduced cooling temperatures gives an improved safety
margin and increased tolerance for operation under an overload condition. If the coolant
temperature is elevated above 40 degrees C then the motor must be derated to avoid
premature failure. Note: Some Class F motors are designed for a maximum coolant
temperature of 60 degrees C, and so there is no derating necessary up to this
temperature.
Operating a motor beyond its maximum, will not cause an immediate failure, rather a
decrease in the life expectancy of that motor. A common rule of thumb applied to
insulation degradation, is that for every ten degree C rise in temperature, the expected
life span is halved. Note: the power dissipated in the windings is the copper loss which is
proportional to the square of the current, so an increase of 10% in the current drawn, will
give an increase of 21% in the copper loss, and therefore an increase of 21% in the
temperature rise which is 16.8 degrees C for a Class B motor, and 21 degrees C for a
Class F motor. This approximates to the life being reduced to a quarter of that expected
if the coolant is at 40 degrees C. Likewise operating the motor in an environment of 50
degrees C at rated load will elevate the insulation temperature by 10 degrees C and
halve the life expectancy of the motor.

Power factor correction

Power factor correction is achieved by the addition of capacitors across the supply to
neutralise the inductive component of the current. The power factor correction may be
applied either as automatic bank correction at the main plant switchboard, or as static
correction installed and controlled at each starter in such a fashion that it is only in circuit
when the motor is on line.
Automatic bank correction consists of a number of banks of power factor correction
capacitors, each controlled by a contactor which in turn is controlled by a power factor
controller. The power factor controller monitors the supply coming into the switchboard
and adds sufficient capacitance to neutralise the inductive current. These controllers are
usually set to adjust the power factor to 0.9 - 0.95 lagging. (inductive)
Static correction is controlled by a contactor when the motor is started and when the
motor is stopped. In the case of a Direct On Line starter, the capacitors are often
controlled by the main DOL contactor which is also controlling the motor. With static
correction, it is important that the motor is under corrected rather than over corrected.
This is because the capacitance and the inductance of the motor form a resonant circuit.
While the motor is connected to the supply, there is no problem. Once the motor is
disconnected from the supply, it begins to decelerate. As it decelerates, it generates
voltage at the frequency at which it is rotating. If the capacitive reactance equals the
inductive reactance, i.e. unity power factor, we have resonance. If the motor is critically
corrected (pf = 1) or over corrected, then as the motor slows, the voltage it is generating
will pass through the resonant frequency set up between the motor and the capacitors. If
this happens, major problems can occur. There will be very high voltages developed
across the motor terminals and capacitors causing insulation damage, high resonant

currents can flow, and transient torque's generated can cause mechanical equipment
failure.
The correct method for sizing static correction capacitors, is to determine the
magnetising current of the motor being corrected, and connect sufficient capacitance to
give 80% current neutralisation. Charts and formula based on motor size alone can be
totally erroneous and should be avoided if possible. There are some power authorities
who specify a fixed amount of KVAR per kilowatt, independent of the size or speed. This
is a dangerous practice. power factor correction

Single phase motors.

In order for a motor to develop a rotating torque in one direction, it is important that the
magnetic field rotates in one direction only. In the case of the three phase motor, there is
no problem and the field follows the phase sequence. If voltage is applied to a single
winding, there are still multiples of two poles which alternate between North and South at
the supply frequency, but there is no set rotation for the vectors. This field can be
correctly considered to be two vectors rotating in opposite directions. To establish a
direction of rotation for the vector, a second phase must be added. The second phase is
applied to a second winding and is derived from the first phase by using the phase shift
of a capacitor in a capacitor start motor, or inductance and resistance in an induction
start motor. (sometimes known as a split phase motor.) Small motors use techniques
such as a shaded pole to set the direction of rotation of the motor.

Slip Ring Motors.

Slip ring motors or wound rotor motors are a variation on the standard cage induction
motors. The slip ring motor has a set of windings on the rotor which are not short
circuited, but are terminated to a set of slip rings for connection to external resistors and
contactors. The slip ring motor enables the starting characteristics of the motor to be
totally controlled and modified to suit the load. A particular high resistance can result in
the pull out torque occurring at almost zero speed providing a very high locked rotor
torque at a low locked rotor current. As the motor accelerates, the value of the
resistance can be reduced altering the start torque curve in a manner such that the
maximum torque is gradually moved towards synchronous speed. This results in a very
high starting torque from zero speed to full speed at a relatively low starting current. This
type of starting is ideal for very high inertia loads allowing the machine to get to full
speed in the minimum time with minimum current draw.
The down side of the slip ring motor is that the sliprings and brush assemblies need
regular maintenance which is a cost not applicable to the standard cage motor. If the
rotor windings are shorted and a start is attempted, i.e the motor is converted to a
standard induction motor, it will exhibit an extremely high locked rotor current, typically
as high as 1400% and a very low locked rotor torque, perhaps as low as 60%. In most
applications, this is not an option.
Another use of the slipring motor is as a means of speed control. By modifying the speed
torque curve, by altering the rotor resistors, the speed at which the motor will drive a
particular load can be altered. This has been used in winching type applications, but
does result in a lot of heat generated in the rotor resistors and consequential drop in
overall efficiency.

Variable-frequency drives
Electric motors have driven mechanical-system loads for many years. For most of that
time, constant-speed turning of the loads and the simplest and safest on-off controls
provided all the oversight a user needed. Today, though, energy-efficiency concerns and
precise control requirements are boosting the popularity of variable-frequency drives
(VFDs).

Technology development
The merit of the basic polyphase induction motor was always its simple configuration
and control: Energize the motor at its base speed on the line frequency of 60 hertz; set
its base speed by selecting the number of pole-pairs in design; and switch poles for
needed two-speed operation.
In the past, some building equipment, such as boiler fans and conveyors, required
adjustment of driven speed. Cumbersome solutions accommodated these adjustments.
Wound-rotor motors and large, variable rotor resistors would set speeds externally, or an
eddy-current drive inserted between motor and load allowed speed-slip to increase.
These methods tossed off a lot of energy in heat losses, but that was not the issue.
Rather, it was the need to vary speed that was important.
Direct-current (dc) motors have been available with silicon-controlled-rectifier drives for
variable-speed operation since the 1960s, but the higher maintenance needs of dc
motors deterred their use for building mechanical systems. Keeping brushes and
commutators fresh as they accumulate thousands of operating hours is simply not
practical.
The energy crisis of the mid-1970s, however, changed the variable-speed terrain,
probably forever. A lot of power was being wasted both in electrical drives and the
mechanical systems they served. Changing induction-motor speed by varying voltage
frequency to the motors kept these otherwise excellent drivers in business and
increased power efficiencies to an acceptable level. What was needed was a frequencyconversion method of similar reliability and reasonable cost. The efficiency of both the
underlying mechanical process and the motor-drive system serving the mechanical
equipment could be improved later.
But there is more to application efficiency than just dividing output watts by input watts.
Efficiency also includes evaluating any risks introduced to surrounding systems,
managing escalating parts counts, handling potential instabilities and counteracting
maintenance woes.

Early VFDs
As the 1980s opened, variable-frequency drives (VFDs) became available in ratings
appropriate to motor applications for building fans and pumps. Typically, this mechanical
equipment incorporates centrifugal designs, for which the following Affinity Laws relate
speed changes to flow volumes, static pressures and driving horsepower:

Flow volume moved varies directly with the driving speed ratio.

Static pressure developed varies with the square of the driving speed ratio.

Horsepower required varies with the cube of the driving speed ratio.

This means that, for centrifugal equipment, required driving power drops quickly as
motor speed slows. The counter side of this observation is that a rather slight increase in
speed--above rated maximum--leads quickly to a large overload on driving equipment.
Many early building-system VFDs came from other fields, such as conveying and
machine tools, and carried 150 percent overload ratings and speed ranges that were not
matched to centrifugal-load requirements. In time, variable-torque rated drives appeared
to meet the needs of this market, since equipment offerings vary directly with the
"speed" of the market and a little lag.
It was clear that the drive-makers were on to something that could produce real
efficiency improvements for air-handling and water-pumping systems. These
applications for VFDs, however, were rather expensive and often involved complicated
design considerations. There was much room for development.

Heating-water pumps (left) and variable-frequency drives (right) in

Pharmacia & Upjohns Kalamazoo, Mich., Building 300 are located close to each
other to ease maintenance and lessen the potential for electrical disturbances.
(Photo courtesy The Austin Company.)

Matching Motors and Controls

Motor-to-drive relationships long have been a concern for engineers specifing VFDs.
Previous motor applications did not require precise matches of motors to controllers
(such as magnetic starters). The National Electrical Manufacturers Association (NEMA)
ratings system for motors and starters--together with manufacturers' good designs-placed motors in the customary electrical environment, where standardization was well
developed.
There are two general VFD-design categories, which differ in how they pass power to
the motor after it is processed by the drive:

Voltage-source inverter (VSI) drives treat motors as parallel-connected loads,

controlling the overall performance envelope by adjusting drive output voltage.
Current-source inverter (CSI) drives, which are much more concerned about
motor impedance--inductive reactance in particular--since it drives current

through the motor as part of the performance envelope. CSI drives must be
matched carefully to their motors.
This motor-matching process always has been less important with VSI drives. Many
engineers have tied VFDs to motors of uncertain origin, particularly when retofitting
systems. But the best approach is to specify, purchase and install VFDs and associated
motors as a team--from one source, whenever possible. It is also important to keep track
of the operating motor's thermal budget, especially in rooms housing large groups of
drives.

Nonlinear issues
Harmonics proved to be another concern. Despite the efficiency gained from turning
mechanical loads at varying and lower speeds, the process of forming and delivering
variable-frequency power elicits some concern over losses. The drive power line-up
comprises basically three units: a rectification input section, changing 60 hertz
alternating-current (ac) supply to dc; a dc bus supported by capacitors; and an inverter
section, which recreates ac power of varying frequency and voltage, as directed by the
drive's controller.
Motors once operated, by design, from a sinusoidal voltage wave form. The wave forms
to which today's models are exposed are distinctly not sinusoidal. These waves include
a fundamental frequency of alternating voltage, to which the motor's base speed
responds, and a rather rich mixture of harmonics formed from the attempt of the six-step
switching process in the inverter to form a sine wave.
These harmonics produce two important effects. At higher frequencies, they lead to eddy
currents and skin-effect losses that produce no usable work, but do produce heat, which
must be dissipated. Second, harmonics cause motor currents that produce additive and
subtractive torques, which are neither useful nor efficient in their production.
While torque effects have seemed minor, motor-heating problems have remained a
concern. The first solution was to derate motor capacity and control ambient
environment. Low-temperature-rise U-frame motors and service-factor rated motors
were common steps to thermal survival. Integrated motor-cooling fans face effects
similar to those of the loads driven by the motors: reduced cooling flow at reduced
speeds. However, as drive inverters for variable-torque centrifugal applications
developed, so did the motors suited to these applications.
Harmonics also can travel up to the feeder power system supplying the drives. Inverters
did not allow input current to flow as a regular sine wave. Instead, current to the rectifier
came in bursts, and harmonics moved around the entire connected power-distribution
system. These harmonics were confined by isolating the power system for mechanical
loads from the rest of the facility. Identifying the "point of common-coupling" for this
somewhat separate system and enforcing a no-connection barrier around the
mechanical distribution performed well, especially on larger projects.
Placing inductance in the input circuit to a VFD also can control harmonic proliferation on
power systems and is used routinely. Isolation transformers are an even better
approach, but cost more and require more space.
Another problem was displacement power factor. Capacitors installed to correct for low,
lagging power factor on motor drives had a good operating and application history, both
in preserving system capacity and in satisfying electric-utility requirements. But

capacitors, shunt-connected to correct motor loads, could not work in the rich harmonic
environments around VFDs. Drives typically delivered good 90 percent-plus power-factor
values at high load, but dropped off as loads dropped off with reduced frequency and
speed.
Through the 1980s and into the 1990s, the population of personal-computer (PC) work
stations increased exponentially. Switch-mode power supplies common to these PCs
caused another level of nonlinear, nonsinusoidal power-system effects to appear
throughout building distribution and into utility distribution installations.
Together with a steadily rising VFD load, concerns over harmonic levels spurred the
development of standards and recommendations for control of these phenomena,
notably Standard 519 from the Institute of Electrical and Electronics Engineers. Ever
since, the industry has focused on measuring, developing, and meeting a structure of
limiting values for harmonics on power distribution systems. And, although it fosters
complaints and arguments, Standard 519 has served for many years as a specification
and applications keystone for VFDs.

VFD Justification: A Mechanical Perspective

VFDs offer real improvements to the operating efficiencies of many mechanical fan and
pumping systems. Mechanical systems are sized for performance at maximum design
conditions, but large parts of the total operating time are spent at conditions well away
from maximum. Slowing equipment drives often results in surprising opportunities for
operating savings, especially when a large number of hours accumulate.
In early variable-speed applications, high VFD expense had to be justified. Although
costs have decreased significantly, specifying variable-speed operation still requires
care to reduce electrical power distribution risks. Further, VFDs may never reach the
reliability level of magnetic motor starters. Bypass contactors on VFDs--an attempt to
deliver that level of reliability--are not appropriate for all applications (for example,
pressure-controlled supply-air). However, VFDs have become an accepted method for
meeting variable-flow needs and efficiency demands.

Figure 6 VFD & Pumps

Figure 7 VFD on Centrifugal Fans

Recent developments
Electronics also have contributed an important component to the further development
and proliferation of efficient VFDs. Insulated gate bipolar transistors (IGBT) allowed
distribution systems to switch large current levels at high kilohertz rates. Now, pulsewidth-modulated (PWM) inverter drives are becoming popular. Displacement power
factor at drive inputs can remain at mid-90 percent levels throughout the speedadjustment and power-load ranges, because the rectification stage is a diode bridge.

Distortion levels are reasonable, units are compact and operating efficiencies are at
high-90 percent levels.
PWM inverters control the kilohertz chopping rate of a wave--varying its envelope in
voltage and fundamental frequency--to control induction-motor operation. Pulse rate can
be set high enough to eliminate audible vibration. The range of fundamental frequencies
through which the drive accelerates and operates can be arranged to skip critical
frequencies where mechanical-equipment and structural-support resonances might be
excited to damaging levels.
VFD drive speed or frequency traditionally is positioned by external control systems,
typically the building temperature-control system. These settings have been passed to
the drive controller section by an analog 4-to-20 milliampere dc instrument loop and read
by an analog receiver at the VFD.
Building temperature-control systems also have adopted the ever-present PC as their
host controller, so widespread use of digital speed setting on a communications line and
protocol for VFDs appears. This simplifies speed transmission and should deliver
calibration-less speed-control loops.
In today's PWM inverters, most drive parameters and set-up requirements are controlled
by software. As a result, these units need little field calibration.
The power circuit through a VFD, and the attending electronic hardware, is highly similar
to other system applications of solid-state motor controllers (or "soft starters") and
uninterruptible power supply systems, as used for large computer systems and data
centers. Improvements in design for one system eventually will help all applications.

Specifying VFDs
As VFDs have become more popular, motor manufacturers have begun offering
products developed specifically for VFD applications. These units apply higher insulation
ratings--nominally 1,600 volts--to windings and provide adequate temperature-rise
headroom. However, some of these products can develop etching of their rotor bearings.
The cause: rapid PWM inverter output rise-times can excite the capacitance of bearing
lubricants and create a circuit to ground for stray currents.
Safety-disconnect switches between inverters and motors should include an auxiliary
control-circuit switch to disable drive output. Closing back into a drive will provoke
unnecessary faults. In addition, VFDs should be near motors wherever possible.
Also, specifiers should be cautious when applying VFD-powered motors in National
Electrical Code (NEC) Classified Hazardous areas. Thermal management of motors at
reduced speeds may press the hazard limits, particularly with increased surface
temperatures. Underwriters Laboratories Standards 674 and 1836 for Division 1 and
Division 2 locations should be consulted to confirm appropriate products for the
application. Minimum motor-speed interlocking and motor-winding temperature
monitoring should be specified for safer operation.

Field installation
Despite harmonic-distortion concerns, bringing VFDs online in the field is much more
straight-forward today than it has been in the past. Harmonic line analyzers can make
these installations even more trouble free. Documenting line conditions near and around

VFDs can aid future maintenance activities. Base-line data showing line conditions
without VFDs operating should be included in any records.
The oldest VFD installations are now 15 to 17 years. Most of these installations have
been on HVAC air-handling fans and pumps in the range of 10 to 450 hp, with the
largest number in 100- to 150-hp sizes. Most included input line reactors. Although these
units from the mid-1980s were very touchy to set up, problems have been minimal.
Further, the benefits of adjustable speed have now added up and matured through
several generations of development.