Woodbank does not monitor or record these emails
For many years the motor controller was a box which provided the motor speed control and enabled the motor to adapt to variations in the load. Designs were often lossy or they provided only crude increments in the parameters controlled.
Modern controllers may incorporate both power electronics and microprocessors enabling the control box to take on many more tasks and to carry them out with greater precision. These tasks include:
(speed, torque and efficiency of the machine or the position of its moving elements.)
Enabling self starting of the motor.
Protecting the motor and the controller itself from damage or abuse.
In an open loop control system the controlling parameters are fixed or set by an operator and the system finds its own equilibrium state.
In the case of a motor the desired operating equilibrium may be the motor speed or its angular position. The controlling parameters such as the supply voltage or the load on the motor may or may not be under the control of the operator.
If any of the parameters such as the load or the supply voltage are changed then the motor will find a new equilibrium state, in this case it will settle at a different speed. The actual equilibrium state can be changed by forcing a change in the parameters over which the operator has control.
Once the initial operating parameters have been set, an open loop system is not responsive to subsequent changes or disturbances in the system operating environment such as temperature and pressure, or to varying demands on the system such as power delivery or load conditions.
For continual monitoring and control over the operating state of a system without operator intervention, for more precision or faster response, automatic control systems are needed.
To meet these requirements closed loop systems are necessary. Also called feedback control systems, or negative feedback systems, they allow the user to set a desired operating state as a target or reference and the control system will automatically move the system to the desired operating point and maintain it at that point thereafter.
A sensor is used to monitor the actual operating state of the system and to feed back to the input of the controller an analogue or digital signal representing the output state. The actual and desired or reference states are continually compared and if the actual state is different from the reference state an error signal is generated which the controller uses to force a change in the controllable parameters to eliminate the error by driving the system back towards the desired operating point.
Loop GainThe error signal is usually very small so the controlling circuit or mechanism must contain a high gain error amplifier to provide the controlling signal with the power to affect the change.
The amplification provided in the loop is called the loop gain.
Loop DelayThe response is not always instantaneous as there is usually a delay between sensing the error, or aiming at a new position, and eliminating the error or moving to the new desired position. This delay is called the loop delay.
In mechanical systems the delay may be due to the inertia associated with the lower acceleration possible in getting a large mass to move when a force is applied.
In electrical circuits the delay may be associated with the inductive elements in the circuit which reduce the possible rate of current build up in the circuit when a voltage is applied.
Closed loop control systems must act very quickly to implement the error correction without delay, before the system has time to change to a different state. Otherwise the system will possibly become unstable.
When there is a time lag between sensing of the error and the completion of the corrective action and the loop gain is large enough the system the system may overshoot. If this happens the error will then be in the opposite direction and the control system will also reverse its direction of action in order to correct this new error. The result will be that the actual position will oscillate about the desired position. This instability is called hunting as the system hunts to find its aiming point.
In the worst case, the delayed error correcting response will arrive 180 degrees out of phase with the disturbance it is trying to eleiminate. When this happens the direction of the system response will not act so as to eliminate the error, instead it will reinforce the error. Thus the delay has changed the system response from negative feedback to positive feedback and the system will be critically unstable.
The diagrams below show the response of a control system to a small disturbance.
TheNyquistStability Criterionis used to predict whether or not a system is unstable from a knowledge of the loop gain and the loop delay as follows
If the loop gain is unity ot greater at the frequency of an input sinusoid where the time delay in the system is equal to half of a cycle period, the sytem will be unstable.
In practical terms, a system with high electrical or mechanical inertia will have a slow response (long delay). With a low magnitude, error correcting action (mechanical force or electrical voltage) the system will be slow in responding (speeding up) but because it is slow, it will also have a low momentum and will tend to settle at the desired operating point when the error correcting force is removed.
The delay in implementing the corrective action however depends on the loop gain.
If, in the same system, the error correcting force is high (amplified / higher loop gain), as in a fast acting system, the system will respond (get moving) more quickly (shorter delay) but it will have correspondingly higher momentum (higher speed of response). When the error correcting force is removed, like any high inertia system, the systems momentum will keep it moving and it will overshoot the target position. Applying the error signal in the opposite direction to bring the system back to its target will cause it to overshoot in the opposite direction.
Nyquist shows how much delay can be tolerated in a system with unity loop gain and defines the point at which the system becomes unstable
In the example of a DC electric motor, the desired operating state may be a particular speed. A tachometer is used to measure the actual speed and this is compared to the reference speed. If it is different, an error signal, whose magnitude and polarity correspond to the difference between the reference and the actual speeds, is fed to a voltage controller to change the motor speed so as to reduce the error signal. When the motor is operating at the desired speed the error signal will be zero and the motor will maintain that speed.
, named after three basic ways of manipulating the error information.
– Proportional error correction multiplies the error by a (negative) constant
, and adds it to the controlled quantity.
– Integral error correction incorporates past experience. It integrates the error over a period of time, and then multiplies it by a (negative) constant
and adds it to the controlled quantity. Equilibrium is based on the average error and avoids oscillation and overshoot providing a more stable system.
– Derivative error correction is based on the rate of change of the error and takes into account future expectations. It is used in so called Predictive Controllers. The first derivative of the error over time is calculated, and multiplied by another (negative) constant
, and also added to the controlled quantity. The derivative term provides a rapid response to a change in the system.
Combinations of all three methods of error processing are often used simultaneously in PID controllers to address different system performance priorities. Where noise may be a problem, the derivative term is not used.
PID controllers are also called 3 term controllers.
Motor controllers may be simple open loop systems or they may incorporate several nested closed loop systems operating simultaneously. For example closed loop controls may be used to synchronise the excitation of the stator poles with the angular position of the rotor or simply to control motor speed or the angular position of the rotor.
When an electrical machine is required to work as both a motor and a generator in both forward and reverse directions this is said to be four quadrant operation. A simple motor which only runs in one direction and is never driven as a generator is an example of a single quadrant application. A motor designed for automotive use which must run in forward and reverse directions and which must provide regenerative braking in both directions needs a four quadrant controller.
Control systems for four quadrant applications will obviously be more complex than single quadrant controls.
Controllers may have some or all of the following functions many of which have been implemented in integrated circuits.
One of the major attractions of brushed DC motors is the simplicity of the controls. The speed is proportional to the voltage and the torque is proportional to the current.
Speed control in brushed DC motors used to be accomplished by varying the supply voltage using lossy rheostats to drop the voltage. The speed of shunt wound DC motors can also be controlled byfield weakening. Nowadays electronic voltage control is employed. See below.
Simple open loop voltage control is sufficient when the motor has a fixed load, however open loop voltage control can not respond to changes in the load on the motor. If the load changes, the motor speed will also change. If the load is increased, the motor must deliver more torque to reach an equilibrium position and this needs more current. The motor consequently slows down, reducing the back EMF so that more current flows. To maintain the desired speed, a change in the voltage is needed to provide the necessary current required by the new load conditions. Automatic control of the speed can only be accomplished in a closed loop system. This uses a tachogenerator on the output shaft to feedback a measure of the actual speed. When this is compared with the desired speed, a speed error signal is generated which is used to change the input voltage to the motor to drive it towards the desired speed. Note – This is essentially avoltage controlsystem since the tachogenerator usually provides a DC voltage output which is compared with a reference input voltage.
Voltage control alone may be insufficient to cater for wide, fast changing load conditions on the motor since the voltage controller may call for currents in excess of the motors design limits. A separate current feedback loop may be required to provide automaticcurrent control. The current control loop must be nested within the voltage control loop. This allows the voltage control loop to deliver more current but it can not override the current control which ensures that the current remains within the limits set by the current control loop.
Brushless DC motors are powered by a pulsed DC supply to create a rotating field and the speed is synchronous with the frequency of the rotating field. Speed is controlled by varying the supply frequency. See alsoInvertersbelow.
The speed of AC motors generally depends on the frequency of the supply voltage and the number of magnetic poles per phase in the stator. Early speed controllers depended on switching in different numbers of poles and control was only available manually and in crude steps. Modern electronicinvertersmake continuously variable frequency supplies possible permitting closed loop speed control. For speed control in induction motors however the supply voltage must change in unison with the frequency. This requires a specialVolts/Hertz controller.
If the application requires direct control over the motor torque rather than the speed, in simple machines this can be accomplished by controlling the current, which is proportional to the torque, and omitting the speed control loop. For more precise control,vector controllersare used.
It is no longer necessary to use energy wasting rheostats to provide a variable voltage.
Modern controllers useswitching regulatorsor chopper circuits to provide a variable DC voltage from a fixed DC supply. The DC supply is switched on and off at high frequency (typically 10 kHz or more) using electronic switching devices such asMOSFETs,IGBTs orGTOs to provide a pulsed DC wave form. The average level of the output voltage can be controlled by varying the duty cycle of the chopper.
AC voltages can be similarly controlled using bi-directional pulses to represent the sinusoidal wave.
Various PWM schemes are possible. Only one is shown here. By varying the pulse width, the amplitude of the sine wave can be changed.
Variable voltages can also be generated by using fixed pulse widths but by varying instead the pulse amplitude (Pulse Amplitude Modulation – PAM) or the pulse repetition frequency (Pulse Frequency Modulation – PFM).
The DC output from choppers and PWM circuits is notoriously plagued by high harmonic content. Most DC motors however can tolerate a pulsed DC supply since the inductance of the motor itself and the mechanical inertia of the rotor help to smooth out the variations in the supply voltage. Since there is no current flowing when the switching device is off, the technique is relatively loss ggingmay occur if the chopper frequency is too low.
A voltage controller may be activated manually in an open loop system but for continuous voltage control, an inverter must be incorporated into a feedback loop in a closed loop system. The control system monitors the actual output voltage and provides a control signal, which may be an analogue or digital representation of the error signal, to the pulse width modulator to correct any deviations. When voltage control is used for speed control the error signal may be derived from a tachogenerator on the motor output shaft.
Electronic voltage control is also an essential part of many generator applications. In automotive systems the generator or alternator is driven at a variable speed which depends directly on the engine speed. It must give its full voltage output at the lowest speed but the voltage must be maintained as the engine speed rises. Alternators used in 12 Volt systems usually have built involtage regulation. In HEV applications a chopper regulator is used at the output of the generator to maintain the voltage at the DC link within strict limits to avoid damaging the battery. When the battery is fully charged, the batterys own management system disconnects it from the supply to prevent overcharging.
For low power applications a series orlinear regulatoris often used. It is less efficient than a switching regulator since the variations in voltage must be taken up, and the associated power dissipated, by the volt dropping series transistor but it provides a pure DC. Series regulators are not suitable for high power applications such as electric traction where efficiency is paramount.
With AC supplies,Thyristors (SCR)scan be used in series with the load to create a variable voltage by blocking the passage of current to the load for the initial part of the cycle and turning the current on by applying a signal to the gate of the SCR. A single SCR only affects one polarity of the waveform. To switch both the positive and negative going current requires two SCRs connected in parallel and in opposite polarity or a triac (bidirectional SCR). By varying the delay (the phase angle) before the current is turned on, the average current, and thus the average voltage seen by the load, can be varied as shown below.
This is the same principle as used in light dimmer switches.
Gate turn off thyristors (GTOs)can be used to switch off the current as well as switching it on allowing more control over the duration of the current through the device.
In many motor applications the motor current may lag the supply voltage due to the inductance in the circuit and it is often desirable to control the current directly, rather than the voltage, to obtain more precise or faster control of the current and hence the torque. In this case a shunt resistor or a current transformer is used to monitor the current. The difference between the actual and reference currents is used in a high gain feedback loop to provide the necessary current regulation.
Current control is particularly important for induction motors to protect the motor from excessive start up currents. A current feedback signal is used to change the firing angle of thyristors in the rectifier or inverter circuits to limit the current within its reference value.
This a generic term for circuits which may provide AC or DC outputs from either AC (mains frequency) or DC (battery) supply lines. They include power bridges for rectifying the AC supply and inverters for generating an AC waveform from a battery supply.
Buck and boost converters are DC-DC converters, the DC equivalent of AC transformers.
The buck converter is used to reduce the DC voltage. Thechopperabove is an example of a step down DC converter.
The boost converter is used to step up the DC voltage.
The circuit below can step up or step down the input voltage by varying the duty cycle of the transistor switch.
The transistor switch turns the supply voltage to the LC circuit on and off. When the transistor is on, the inductor is charged up and the diode cuts off the capacitor. When the transistor turns off, the inductor discharges, via the diode, through the capacitor charging it up. Note that the polarity of the output voltage is the reverse of the input voltage. With a low duty cycle when the transistor is off more than 50% of the time, the voltage which appears at the output is lower than the supply voltage and the circuit acts as a step down transformer. With a high duty cycle when the transistor is switched on more than it is off, the voltage builds up on the capacitor and the output voltage exceeds the supply voltage. Voltage regulation is thus provided by varying the duty cycle.
Inverters provide a controlled alternating current (AC) supply from a DC or AC source. There are two main classes of applications:
Inverters designed to deliver regulated AC mains power from sources which may have a variable input voltage (either AC or DC) or in the case of AC input power, a variable frequency input. Such applications may include emergency generating sets, uninterruptible power supplies (UPS) or distributed power generation from wind and other intermittent resources. All must deliver a fixed output voltage and frequency to the load since the applications expect it and may depend on it.
On the other hand, many applications require inverters to accept a fixed AC voltage and frequency from the mains and to provide a different or variable voltage and frequency for applications such as motor speed control. .
In both of these designs, a bridge rectifier is used to provide the intermediate DC power through a DC Link to a regular AC inverter.
The circuit below shows the principle of such an inverter designed for three phase applications.
Three phase variable frequency inverter
The three phase sinusoidal input is fed to a simple diode full wave bridge rectifier block delivering a fixed voltage to the inverter. The connection between the rectifier and the inverter is known as the DC Link. The inverter transistors are switched on in the sequence of their numbers as shown in the diagram with a time difference if T/6 and each transistor is kept on for a duration of T/4 where T is the time period for each complete cycle. The example above provides six possible current configurations and is known as a six step inverter.
The diodes connected across the switching transistors are known as freewheel or flywheel diodes. Their purpose is to provide a current bypass path around the transistor to protect it from the dissipation of the stored energy in the inductive load (the motor) when the transistor is switched off. The current through the diode freewheels until all the energy in the inductive load is dissipated.
The output line voltage wave form for each phase is shown below.
This inverter frequency reference may simply be a voltage applied to the input of a Voltage Controlled Oscillator (VCO) examples of which are commonly available as integrated circuit chips, or it may be derived from a microprocessor clock. Digital logic circuits are used to derive the timed trigger pulses to the inverter switches from the frequency reference source. In the case of generators delivering mains power, the frequency reference value will be fixed.
The amplitude of the output wave is determined by the level of the DC supply voltage to the inverter block but it can be varied by thyristor (SCR) control of the rectifier circuit to provide a variable voltage at the DC link.
Instead of transistor switches, the inverter may use MOSFETs, IGBTs or SCRs.
Free-wheeling diodes connected across the transistors protect them from reverse bias inductive surges due to motor field decay which results when the transistors turn off by providing free wheeling paths for the stored energy.
The waveforms for traction applications are often stepped waves rather than pure sinusoids since they are easier to generate and the motor itself smoothes out the wave.
Variable frequency inverters are used when variable speed control is required. The frequency of the wave is controlled by a variable frequency clock which initiates the pulses.
For speed control in AC machines the voltage and frequency must vary in unison. SeeAC motor speed control. In open loop systems the operating point is set by a speed reference and the equilibrium speed is determined by the load torque. A closed loop system allows a fixed speed to be set. This requires a tachogenerator to provide a feedback of the actual speed for comparison with the desired speed. If there is a difference, an error signal is generated to bring the actual speed into line with the reference speed by adjusting both the voltage and the frequency so as to eliminate the speed difference.
See alsobrushless DC motor speed controland examples ofgenerator speed controls.
Volts/Hertz controlis needed for speed control of induction motors. In an open loop system the control system converts the desired speed to a frequency reference input to a variable frequency, variable voltage inverter. At the same time it multiplies the frequency reference by the Volts/Hertz characteristic ratio of the motor to provide the corresponding voltage reference to the inverter. Changing the speed reference will then cause the voltage and frequency outputs from the inverter to change in unison.
In a closed loop system a speed feedback signal provided from a tachogenerator on the motor output shaft is used in the control loop to derive a speed error signal to drive a Volts/Hertz control function similar to the one outlined above.
As with large DC motors, speed control is normally accompanied by current control.
The cycloconverter converts AC supply frequency directly to a variable frequency AC without the intermediate DC link stage.
The system is complex and works by sampling the voltage of each phase of the AC supply and synthesising the desired output waveform by switching on to the load for the duration of the sampling period, the phase whose voltage is closest to the desired voltage at the instant of sampling. The output waveform is severely distorted and the capability of induction motors to cope with the very high harmonic content limits the maximum frequency for which the system can be used.
Cycloconverters are only suitable for very low frequencies, up to 30% of the input frequency. They are used for low speed high power drives to eliminate the need for a gearbox in heavy rolling and crushing mills and in traction applications for trains and ships.
All motors need a magnetising current and a torque producing current. In a brushed DC motor, these two currents are fed to two different windings. The magnetising current is fed to the stator or field winding and the torque producing current is fed to the rotor winding. This allows independent control of both the stator and the rotor fields. However in brushless motors such as permanent magnet motors or induction motors it is not possible to control the rotor field directly since there are no connections to it. Because the parameters to be controlled can not be measured, their values must be derived from parameters which can be measured and controlled. The only input over which control is possible is the input current supplied to the stator.
The actual stator current is the vector sum of two current vectors, the inductive (phase delayed) magnetising current vector producing the flux in the air gap and thein phase, torque producing, current. To change the torque we need to change thein phase, torque producing, current but because we want the air gap flux to remain constant at its optimum level, the magnetising current should also remain unchanged when the torque changes.
Vector Control or Field Oriented Control is a method of independently varying the magnitude and phase of the stator current vectors to adapt to the instantaneous speed and torque demands on the motor.
It enables parameters over which no direct control is possible to be changed by changing instead, parameters which can be measured and controlled.
For many applications vector control is not necessary, but for precision control, optimum efficiency and fast response, control over the rotor field is needed and alternative methods of indirect control have been developed. Because of the low cost of computing power, vector control is being used in more and more brushless motor applications.
Maximum current-to-torque power conversion, fast transient response, precise control of torque, speed and position.
Rotating flux to be maintained at 90 degrees to the rotor flux.
Available information (status of stator voltages and currents and rotor position and/or speed).
Two independent control loops to provide control of the magnetising and torque producing current vectors.
Mathematical transforms to analyse input signals from the stator and calculate any deviation from the desired conditions of the rotor.
Mathematical inverse transforms to convert the rotor error signal back into control signals to be applied to the stator to counteract the error.
A pulse width modulated (PWM) inverter providing power to the motor.
Stator input voltage waveforms of the correct amplitude, frequency and phase to effect the change.
Uses position sensors and complex mathematical transforms
Sensorless Uses even more complex mathematical transforms
(Both of the above methods use current sensors for current control of the stator windings)
Samples status and provides control signals at 20 kHz to provide continuous control.
Low speed control, efficiency improvement, smaller motors.
The good news is that a detailed knowledge of the process involved is not necessary since most of these tasks are implemented in integrated circuits and incorporated into the motor design. But read on to find out how the overall system is used.
Despite its many advantages, the venerable induction motor is relatively slow to respond to changes in load conditions or user commands for changes in speed. This is mainly because the rotor current can not instantaneously follow the applied voltage due to the delay caused by the inductance of the motors windings.
During the transition period the flux amplitude and its angle with respect to the rotor must be maintained so that the desired torque can be developed.
Torque also depends on the magnitude of the flux but this depends on the inductive component of the current and can not be changed instantaneously. In any case the flux density is set to its optimum point before saturation occurs.
Vector control is a way of changing the in phase current vector without changing the inductive magnetising current vector so that the machine response time is not subject to inductive delay.
The inductive phase lag noted above also causes an instantaneous loss of torque and reduced efficiency because the torque producing flux from the stator is not acting at 90 electrical degrees to the rotor field.
The torque on the rotor of any motor is at its maximum when the magnetic field due to the rotor is at right angles to the field due to the stator. SeeInteractive Fields
The vector control system provides instantaneous adjustments to the stator currents to control the position of the rotor with respect to the moving flux wave thus avoiding losses due to