Power electronic converters (part 1)

Intro

This section deals with the active components (e.g. diodes, thyristors, transistors, etc) and passive components (e.g. resistors, chokes, capacitors, etc) used in power electronic circuits and converters. Power electronics is that field of electronics which covers the conversion of electrical energy from one form to another for high power applications. It applies to circuits in the following power ranges:

• Power ratings up to the MVA range

• Frequency ratings up to about 100 kHz

Power electronics is a rapidly expanding field in electrical engineering and the scope of the technology covers a wide spectrum. Therefore, the emphasis will be on the components used in converters used for the speed control of electric motors. Components used for other applications such as power supplies, high frequency generators, etc won’t be covered in great detail.

Definitions

The following are the common terms used in the field of power electronics.

• Power electronic components, are those semiconductor devices, such as diodes, thyristors, transistors, etc that are used in the power circuit of a converter. In power electronics, they are used in the non-linear switching mode (on/off mode) and not as linear amplifiers.

• Power electronic converter or 'converter' for short, is an assembly of power electronic components that converts one or more of the characteristics of an electric power system. E.g., a converter can be used to change:

1. - AC to DC
2. - DC to AC
3. - Frequency
4. - Voltage level
5. - Current level
6. - Number of phases

The following graphic symbols are used to designate the different types of converter.

• Rectifier is that special type of converter that converts AC to DC
• Inverter is that special type of converter that converts DC to AC.
• AC converter is that special type of converter that converts AC, of one voltage and frequency, to AC of another voltage and frequency, which are often variable.
• An AC frequency converter is a special type of AC converter.

In a power electronic AC converter, it’s common to use an intermediary DC link with some form of smoothing.

• DC converter is one that converts DC of one voltage to DC of another voltage.

In a DC converter, it’s common to use an intermediary AC link, usually with galvanic isolation via a transformer.

• Electronic switch is one that electronically connects or disconnects an AC or DC circuit and can usually be switched ON and /or OFF. Conduction is usually permitted in one direction only.

The following components are those devices that are most commonly used as electronic switches in power electronic converters. Developments in semiconductor technology have made these power electronic components smaller, more reliable, more efficient (lower losses), cheaper and able to operate at much higher voltages, currents and frequencies. The idealized operating principles of these components can be described in terms of simple mathematical expressions.

• Power diodes

• Power thyristors

• Gate turn-off thyristors (GTO)

• MOS controlled thyristors (MCT)

• Power bipolar junction transistors (BJT)

• Field effect transistors (FET, MOSFET)

• Integrated gate bipolar transistors (IGBT)

• Resistors (provide resistance)

• Reactors or chokes (provide inductance)

• Capacitors (provide capacitance)

In power electronic circuits, semiconductor devices are usually operated in the bi-stable mode, which means that they are operated in either one of two stable conditions:

• Blocking mode: fully switched OFF

- Voltage across the component is high

- Current through the component is low (only leakage current)

• Conducting Mode: fully switched ON

- Voltage across the component is low

- Current through the component is high

Diodes and thyristors are inherently bi-stable but transistors are not. Transistors must be biased fully ON to behave like bi-stable devices.

Power diodes

A power diode is 2-terminal semiconductor device with a relatively large single P-N junction. It consists of a 2-layer silicon wafer attached to a substantial copper base. The base acts as a heat-sink, a support for the enclosure and also one of the electrical terminals of the diode. The other surface of the wafer is connected to the other electrical terminal. The enclosure seals the silicon wafer from the atmosphere and provides adequate insulation between the two terminals of the diode. The two terminals of a diode are called the anode (A) and the cathode (K). These names are derived from the days when Valves were commonly used.

SYMBOL:

IDEAL:

Forward conduction: Resistanceless Reverse blocking: Lossless Switch on/off time: Instantaneous

Many different mechanical designs are commonly used for diodes, some of which are shown below. Power diodes rated from a few amperes are usually stud mounted but it’s increasingly common (more economical) to have several diodes encapsulated into an insulated module. Examples are full wave rectifiers, 6-pulse diode bridges, etc.

++++ 1: Typical mechanical construction of diodes

++++ 2: Typical characteristic of a power diode.

The base of this type of diode module is usually not electrically active, so it can be mounted directly onto the heat-sink of a converter. Larger units for high current ratings are usually of the disc type, which provides a larger area of contact between the case and the heat-sink for better cooling.

When the anode is positive relative to the cathode, it’s said to be forward biased and the diode conducts current. When the anode is negative relative to the cathode the diode is said to be reverse biased and the flow of current is blocked. The typical characteristic of a power diode is shown below.

Unfortunately, power diodes have several limitations:

• In the conduction mode, when the diode is forward biased

- Real diodes are not resistanceless and there is a forward volt drop of between 0.5 to 1.0 volts during conduction

- As a result, there is a limit to how much current can continuously flow without overheating. This is the maximum rated current of the diode.

• In the blocking mode, when the diode is reverse biased

- there is a small leakage current

- there is a limit to how much voltage it can withstand before reverse breakdown and current can start to flow in the reverse direction. It’s sound common practice to select diodes with a reverse voltage limit of at least twice the value that will practically occur.

• The commutation time from the blocking mode to the conduction mode and vice versa takes a finite time.

A power diode must be rated for the electrical environment in which it’s to be used.

The following are the most important factors that must be considered when choosing a power diode for a converter application:

• Forward current rating. The current rating is based on a certain wave shape and should be taken as a guide only. The real selection should be based on the total power losses in the diode taking into account the actual wave shape, load cycle and cooling conditions.

• Forward voltage drop. This has an effect on current sharing between parallel circuits that include diodes.

• Forward surge current capability (rate of rise of current di/dt)

• Reverse voltage rating (sometimes referred to as PIV - peak inverse voltage)

• Reverse recovery current di/dt. This should be taken into account when considering the commutation transients in the diode circuit.

• I^2 t rating. This is a measure of the energy that a diode can handle in the case of a short circuit without permanent damage. It gives a guide to the correct choice of high speed fuses to protect the diode. Briefly, a protection fuse must be chosen with an I^2 t rating lower than the diode.

Depending on the application requirements, various types of diode are available:

• Schottky diodes

• These diodes are used where a low forward voltage drop, typically 0.4 volts, is needed for low output voltage circuits. These diodes have a limited blocking voltage capability of 50 to 100 volts.

• Fast recovery diodes

• These diodes are designed for use in circuits where fast recovery times are needed, for example in combination with controllable switches in high frequency circuits. Such diodes have a recovery time (tRR) of less than a few microsecs.

• Line frequency diodes

• The on-state voltage of these diodes is designed to be as low as possible to ensure that they switch on quickly in rectifier bridge applications. Unfortunately the recovery time (tRR) is fairly long, but this is acceptable for line-frequency rectifier applications. These diodes are available with blocking voltage ratings of several kV and current ratings of several hundred k_amps. In addition, they an be connected in series or parallel to satisfy high voltage or current requirements.

Power thyristors

Thyristors are often referred to as SCRs (silicon controlled rectifiers). This was the name originally given to the device when it was invented by General Electric (USA) in about 1957. This name has never been universally accepted and used. The name accepted by both the IEC and ANSI/IEEE is reverse blocking triode thyristor or simply thyristor. The name thyristor is a generic term that is applied to a family of semiconductor devices that have the regenerative switching characteristics. There are many devices in the thyristor family including the power thyristor, the gate turn-off thyristor (GTO), the field controlled thyristor (FCT), the triac, etc.

A thyristor consists of a 4-layer silicon wafer with 3 P-N junctions. It has two power terminals, called the anode (A) and cathode (K), and a third control terminal called the gate (G). High voltage, high power thyristors sometimes also have a 4th terminal, called an auxiliary cathode and used for connection to the triggering circuit. This prevents the main circuit from interfering with the gate circuit.

A thyristor is very similar to a power diode in both physical appearance and construction, except for the gate terminal required to trigger the thyristor into the conduction mode.

SYMBOL:

IDEAL:

Forward conduction: Resistanceless

Forward blocking: Lossless (no leakage current)

Reverse blocking: Lossless (no leakage current)

Switch on/off time: Instantaneous

As with power diodes, smaller units are usually of the stud type but it’s also increasingly common to have 2 or more thyristors assembled into a thyristor module. The base of this type of pack is not electrically active, so it can be mounted directly onto the heat-sink of a converter. Large thyristor units are usually of the disc type for better cooling.

++++ 3: Typical mechanical construction of thyristors

Most converters for the speed control of motors are air-cooled, the smaller units using natural convectional cooling over the heat-sink and the larger units using a fan for forced cooling.

A thyristor is a controllable device, which can be switched from a blocking state (high voltage, low current) to a conducting state (low voltage, high current) by a suitable gate pulse. Forward conduction is blocked until an external positive pulse is applied to the gate terminal. A thyristor cannot be turned off from the gate. During forward conduction, its behavior resembles that of a power diode and it also exhibits a forward voltage drop of between 1 to 3 volts. Like the diode, conduction is blocked in the reverse biased direction. A typical characteristic of the thyristor.

There are several ways in which a thyristor can be turned on or brought into forward conduction.

• Positive current gate pulse. This is the normal way that a thyristor is brought into conduction. The gate pulse must be of a suitable amplitude and duration, depending on the size of the thyristor.

• High forward voltage. An excessively high forward voltage between the anode and the cathode can cause enough leakage current to flow to trigger the turn on process.

• High rate of rise of forward voltage, dV/dt. A high dV/dt can produce enough leakage current to trigger the turn on process.

• Excessive temperature. The leakage current increases with temperature, so high temperature can aggravate the above two problems.

++++ 4: Typical characteristic of a thyristor

A thyristor must be suitable for the electrical environment in which it’s used. The following are some of the more important factors which must be considered when choosing a thyristor for a converter application:

• Same factors outlined above for diodes.

• The power losses in the thyristor comprise the conduction losses, switching losses (turn on and turn off), gate power losses, forward off state losses and reverse blocking losses. The data sheet usually provides curves for estimating power losses for various wave shapes.

• Peak forward voltage (PFV). This is the forward anode voltage that the device must withstand without switching on and without damage.

• Rate of rise of forward voltage dV/dt should not be too high, typically it should be less than about 200 Volt/µsec. A parallel RC snubber circuit is usually required to protect the thyristor.

• Rate of rise of anode current di/dt should not be too high, typically it should be less than about 100 amp/µsec. The current is initially concentrated around the gate and takes a finite time to spread over the conducting area.

If the rate of rise is too high, local overheating could damage the thyristor. Circuit inductance is usually required to limit the rate of rise of current.

• Holding current. The minimum forward current required for the thyristor to maintain forward conduction.

• Latching current. The minimum forward current that causes the thyristor to initially latch. This is usually higher than the holding current and is important because the gate pulse may be relatively short.

• Gate triggering requirements. A relatively small gate pulse will turn the thyristor on. Typically a value of 100 mA for 10 µsec is the threshold. In practice, a much higher value should be used for optimum thyristor operation.

Also, the turn on time is affected by the magnitude of the gate pulse.

The thyristor is turned off when it becomes reverse biased and /or the forward current falls below the holding current. This must be controlled externally in the power circuit.