The power diode is an uncontrollable switch capable of conducting very high currents and withstanding very high voltage levels - thousands amperes and thousands of volts. It operates in the same way as a small power diode, and has similar characteristics.

Two types of power diodes are in use in power electronics: bipolar diodes and Schottky Diodes. Power diodes are manufactured in different cases, suitable to dissipate high power losses. Some of them are shown in the figure.

A typical electrical diode circuit is shown in the figure. The current level in a circuit depends on both a supply voltage and a load. The diode only determines the direction of a current flow.
When a positive voltage is applied to the circuit, the diode turns on and acts as a closed switch. Next, a current flows through the load. If a negative voltage is applied, it tries to produce a negative current. The diode turns off and acts as an open switch. Device switching depends on supply voltage. Hence, the diode operates as an uncontrollable switch.

The bipolar power diode is based on a PN semiconductor junction, as shown in the figure. Hence, it is a minority carrier device. The simplest structure of a power diode includes an additional n^- layer. It increases the reverse voltage ratings.
The Schottky Diode is based on a metal semiconductor junction. The device is formed by placing a metal film on an n-type semiconductor. It has only one type of semiconductor and is a majority carrier device.
Schematic symbols of both devices are shown in the figure.

A real diode conducts very little current when it is reverse-biased. The maximum reverse current is I_RRM (Reverse Repetitive Maximum current). It determines the rated value U_RRM of the reverse voltage. For a bipolar diode, this voltage is typically 10÷12 hundreds of volts, but there are diodes with higher voltages.

The Schottky Diode has much lower rated reverse voltages than a bipolar diode. It is typically 60 V and cannot be higher than 100 V.
If the reverse voltage exceeds the rated value, U_RRM, the diode is damaged.

When a positive voltage is applied, the diode is turned on. A forward current, I_F, flows through the circuit. The forward voltage drop U_F is typically (1 ÷ 1,5) V. The knee (threshold) voltage U_FO and the forward resistance r_F are given in the datasheets. For safe operation, I_F should not exceed the rated average current, I_F(AV)M, given in the datasheets.

The forward-biased Schottky Diode has an important advantage over the bipolar diode. For the same current, U_F is smaller - typically 0,5 V. The I_F(AV)M current is smaller as well. Therefore, the Schottky power diode has the smaller power loss, (P_LOSS).

Switching characteristics show the behavior of a power diode during its transition from an off-state to an on-state and vice-versa. In an ideal circuit, an ideal diode switches on and off without delay.

A real diode needs some time to switch on. During this time, its resistance is high. Therefore, for a few microseconds, the voltage drop, U_Fmax, rises to 5÷10 V. When the diode is fully switched on, the voltage drop reduces to its normal value, U_F.

In a real circuit, a small stray inductance, L_s, exists which influences the switching-off process.

To switch off the diode, a negative voltage is applied. This is called reverse biasing. During the process, the current decreases, and for a while it even flows in a negative direction. It reaches a Reverse maximum level I_Rmax and returns to zero. The diode stops conducting in a reverse direction. The duration of the time that the diode is conducting in reverse is called the reverse recovery time - t_rr. The lower t_rr is, the higher the frequency application diode.
During the switching off phase, the abrupt current fall and presence of L_S produces a voltage spike U_Rmax across the diode. It could be very high and dangerous.

A reverse voltage spike which appears across the diode could be very high and could damage the diode. If this occurs, the improper reverse current flows through the circuit.
To prevent a diode from such high overvoltage, an additional RC group is included in parallel to the device. This group is called the snubber, and it reduces the voltage spike across the diode. A similar group must be connected to every power semiconductor device. The values of the resistor and the capacitor are usually recommended by the manufacturer of the device.

The time, t_rr, is a very important diode parameter. The illustration shows its influence on the electrical circuit. At low frequencies, the diode functions perfectly. Increasing the frequency reveals a small reverse recovery period - the real electronic circuit operation. At very high frequencies, the time, t_rr, approximately equals the duration of the negative half-cycle. The rectifying effect disappears. The diode is unsuitable for this high frequency. Schottky Diodes have a low t_rr. Depending on the t_rr value, the power bipolar diodes are divided into Rectifier Diodes (very high t_rr) and fast recovery diodes (low t_rr).

The thyristor is a power electronic device. It acts as a controllable switch, that is either open (off-state of the thyristor) or closed (on-state of the thyristor). It has three terminals: anode (A), cathode (K), and gate (G), as shown in the figure.

The anode and the cathode are the power terminals of the thyristor. They connect the device to the power circuit. With the third terminal (gate), the thyristor can be triggered into the on-state. Thyristors operate like controllable diodes and are also known as Silicon Controlled Rectifiers (SCR).

Thyristors are used for power conversion or power control. A typical electrical circuit for such applications consists of a Power Supply (voltage source), Load, Thyristors, and Gate Control Circuit. The voltage supplied could be AC or DC.

In it's off-state, the thyristor acts as an open switch and it doesn't conduct. In its on-state, the thyristor is a unidirectional device - a current flows from the anode to the cathode. Two conditions have to exist in order to switch the thyristor on: (1) the anode must be positively biased with respect to the cathode, and (2) a positive gate pulse must be produced by the control circuit. The thyristor has to continue conducting even after the gate current is removed.

A thyristor is a device with a four-layer, semiconductor structure: p1, n1, p2, and n2. It has three PN junctions: J1, J2, and J3. J3 is also known as the gate junction.

The gate and cathode are implemented in two different layouts depending on the thyristor's applications. A distributed gate structure is appropriate for high switching frequencies (more than 1kHz). It is also used in large diameter thyristors. The center gate structure is used in small diameter thyristors. It is used at low switching frequencies.

The thyristor can conduct very high currents - typically hundreds of amperes. Hence, the device's semiconductor structure has to dissipate high power levels - hundreds of watts. For this reason, thyristors are manufactured in many different packages, capable of dissipating these high power losses. Some of these are shown above.

The details of a thyristor's internal construction are also illustrated. Some thyristors have an additional lead for a cathode, as shown in the figure. This helps with the device's connection to the gate control circuit

When a positive voltage is applied to the anode with respect to the cathode of a junction, J2, it is reverse-biased. No current flows from the anode to the cathode. The device is in the forward blocking state (forward off-state).

If the anode voltage is negative with respect to the cathode, junctions J1 and J3 are reverse-biased. Junction J1 supports the reverse supply voltage, since junction J3 has a low breakdown voltage (3 - 4 V). No current flows through the device. It is in the reverse blocking state (reverse off-state). In both cases, the thyristor operates as an open switch.

The four-layer structure can be visualized as the interconnection of two three-layer devices. A two-transistor SCR model is illustrated in the figure. According to the model, if the base current of Q2 increases, the collector current of Q2 increases. Hence, the base current and the collector current of Q1 also increase. The two transistors drive each other harder and harder. They get locked into deep saturation in a few microseconds. This process is called theregenerative process.

The usual, and strongly recommended, way to trigger a thyristor to the on-state is by using a gate pulse. A small gate current from an external circuit initializes the base current of Q₂, and starts the regenerative process. In a few microseconds, the thyristor goes to its switched-on state and remains in this state even after the gate current is annulled.

A thyristor can be tested to see if it's in good working order with an ohmmeter. A normal, trouble-free device shows high resistance (several hundred kiloOhms) in both connections ("plus" from the ohmmeter to the anode, and "minus" to the cathode). A damaged thyristor is somewhat shorted-circuited in both directions.

A very peculiar fact is the low resistance that both ohmmeter connections have to a gate cathode junction in a properly operating device.

An idealized thyristor I-V characteristic has three important regions: reverse blocking, forward blocking, and forward on-state.

No current flows through the device in the first two regions. The voltage across the thyristor, U_T, equals the supply voltage, U_d.

When the thyristor is in its on-state, the current through the device, I_T, is determined from the external electrical circuit. The voltage drop is U_T = 0 V.

No power is dissipated by the ideal device in any of the three regions.

In reverse-biasing, the real thyristor has the same characteristics as a power diode. The maximum reverse current which the device can safely sustain is I_RRM.

When the real thyristor is in an off-state and forward-biased voltage is applied, a very small amount of current flows through the device. If we increase the voltage above a certain value, called the "breakover voltage," U_BO, the thyristor turns on, even without any gate current. This behavior is undesirable and should be avoided!

Let's suppose that a thyristor is in its off-state, forward biased, and there is a gate current present. In this case, the thyristor switches on at voltages lower than the rated voltage, U_DRM. This mode of operation is undesirable! A properly designed gate control circuit should not supply positive current when the thyristor is in its off-state.

The figure above shows how different gate current values influence the thyristor's capability to block forward voltages. These curves are called forward blocking characteristics.

To switch on a thyristor, two basic conditions should be met:

• Positive anode-cathode voltage, and
• Positive gate current (pulse or continuous current).

The forward voltage drop (on-state voltage) on the thyristor U_T is between 1 and 3V.

The current through the device must not exceed the rated values specified in the data sheets.

There are two important, but not so apparent, parameters: holding current, I_H, and leading current, I_L. They can be explained by focusing on the very beginning of the on-state.

When the thyristor is in its on-state, the current can be changed by load or supply voltage variations. The upper current limit, I_TRM, was already explained.

The lower current limit is called the holding current, I_H. I_H is the minimum current that can flow through the thyristor and still maintain the device in its on-state.

The leading current, I_L, is slightly larger than the holding current, I_H.

The current, I_L, shows the lowest required current through the thyristor during its switching on. This is the third requirement to turn on the device.

As an example, let's discuss the circuit above. It has relatively high inductance. Hence, after switching on the thyristor, the current slowly rises. If a short gate pulse is applied, the current (i_T) increases but stays below I_L. The thyristor switches off when the gate current nullifies.

If the gate pulse is wide enough, and the current (i_T) rises over I_L, then the turned on thyristor stays in its on-state. The electrical circuit operates properly.

The relationship between the gate-cathode voltage (U_G) and the gate current (I_G) is known as the gate characteristic. The gate characteristic is temperature dependant. For all devices of a given thyristor type, there exist a family of characteristics. This range of possible gate junction operating points has its upper and lower limits. The upper limits are the maximum gate current and gate power dissipation. The lower limits are the minimum gate current and the voltage under which the device cannot be turned on under the worst conditions.

The values U_gi, R_G, and I_G must be designed to ensure reliable switching on of the thyristor. The gate load line must be out of the forbidden zones shown.

The output stage of the gate control circuit is called the gate pulse amplifier (driver). The pulse amplifier has two purposes: (1) to produce a short (ten microseconds) pulse with optimal shape, and (2) to provide an electrical separation between the power thyristor and the control circuit. At the beginning of the pulse, the gate current (I_G) is higher in order to ensure the fast switching of the thyristor. Next, the current is reduced to minimize power dissipation, P_G.
Typical voltage waveforms at the most critical circuit points are shown in the figure.

Until now, we have assumed that the thyristor turns on immediately. The actual device needs a finite time in order to switch from a forward off-state to an on-state.

When the gate pulse is applied, the regenerative process starts, following a short turn-on delay-time, t_d. During the next time interval, t_r (rise-time), only a small region of the thyristor structure switches on. The thyristor voltage falls and the current rises.

In a few microseconds, the conducting zone spreads to the whole structure. The spreading time, t_spr, is the longest time in the switching on process.

At the beginning of the switching on process, only a small slice (closest to the gate terminal) conducts the current. If, at this moment, the thyristor is forced to conduct the rated current, an excessively high current density occurs in this small device slice. It can overheat the structure and can destroy the thyristor. The spreading of the conductivity must be ahead of the rate of the actual rise in thyristor current. If so, the current will be distributed uniformly in the whole semiconductor structure. The critical rate of rise of the anode current is provided in the datasheets as parameter di/dt. For example, 100 A/µs, 500 A/µs, 1000 A/µs.

During the switching off of a real thyristor, a negative current flows through it. This current decreases to zero after a time interval called the reverse recovery time (t_rr). If, at this moment, a positive voltage bias is applied, the device will turn on without a gate pulse. To avoid this undesirable effect, the negative bias should be held for a time known as the turn-off time (t_q). According to this parameter, the devices are divided in two groups: rectifier thyristors (t_q over 100 µs) and inverter thyristors (t_q = 8 ÷ 40 µs).

When the thyristor is switched-off, applying positive anode-cathode voltage causes an additional current to enter the gate junction. This current is proportional to the voltage rising rate and the space capacitance, C_J2. If this rate is too high, then the induced current is high enough to turn on the thyristor without a gate pulse. This usually provokes a short circuit in the power conversion system, or another kind of improper mode of operation. Therefore, the voltage rising rate should not exceed certain critical values, called the parameter du/dt, as provided in the datasheets.
The behavior of two different voltage rising rates in a thyristor is illustrated in the diagram.

In summary, there are two ways to switch-off a thyristor. The first is based on reducing the thyristor current ,i_T, under the holding current, I_H. This has few practical uses. The second, requires applying negative voltage across the device. The time interval during which the negative voltage is applied is called the schematic off-time, t_OFF. It should be longer than the thyristor off-time, t_q.

In an AC circuit, this is easy to realize. The line voltage consists of a positive and a negative half-cycle. During the positive cycle, a half wave gate pulse turns on the thyristor. During the negative half wave cycle, the thyristor turns off. This is called natural or line commutation.

When the thyristor is used in DC circuits, a method of forced commutation is used. A specially designed auxiliary circuit turns off the conducting thyristor. It is assumed that the capacitor is initially charged to the voltage, U_C0.

If the thyristor is in its on-state, a current flows through the circuit. Switching off the device begins when the switch, S, closes. The negative voltage from the capacitor is applied to the thyristor during the schematic off-time, t_OFF ≥ t_q. The thyristor switches off.

Many practical applications require the commutation of AC line voltage. Since a thyristor is a unidirectional device, it is possible to use a pair of thyristors connected in reverse parallel. A device with the same behavior and characteristics is called a triac. It is a bi-directional device which operates as an AC circuit breaker. It controls the power supplied to the load by switching the connection to the AC source off and on.

A typical triac application (as a circuit breaker) is shown in the illustration.

The triac semiconductor structure originates from the structure of two thyristors in reverse parallel. The triac terminals are called the Main Terminal 1 (MT1) and the Main Terminal 2 (MT2).

The two most important triac properties (illustrated in the figure) are:

• The triac is a bi-directional device.
• The triac can be switched on by either a positive or negative gate pulse.

The triac I-V characteristic is symmetrical as shown above.

Two distinctive regions can be identified. In its off-state, the device blocks both positive and negative voltage. No current flows through the electrical circuit when a triac is involved.

After a proper gate pulse is initiated, the device switches to its on-state and can conduct the current with a rated value in both directions.

For the triac, the maximal rated current is defined as its rms value, denoted in the datasheets as: I_TRMS.

The triac can be used in power controlling not only as an AC breaker (to switch on and off the load to the AC source), but also to realize a phase shift control (switching on the device at a controlled moment during the half-cycle of the supply voltage). Due to this phase control, the part of the sinusoidal voltage that is applied to the load, changes. Hence, it varies the rms value of the output voltage.

The rms value control described is very convenient for speed control in home appliances, for light dimming, and other applications.

The diac is a device similar to the triac but without a control gate. The diac can be switched on only by increasing the anode voltage. The switching voltage is, typically, several tens of volts.

Usually, the diac is used to form firing pulses suitable to turn on a thyristor or a triac. As shown in the figure, the capacitor is slowly charged via the potentiometer. As soon as the capacitor voltage reaches the diac switching voltage, a sharp pulse is produced. The moment this happens depends on the values of the capacitor and the potentiometer.

The GTO (Gate Turn-Off Thyristor) is one of the newest devices in the thyristor family. The GTO operates as a thyristor. The main difference between them is the manner of switching off the GTO. To turn off the device, it's enough just to apply a negative gate pulse. There is no need for a negative anode voltage, such as the one required by a thyristor. Therefore, it simplifies the power electronic circuits.

High currents in all power devices produce very high internal power losses, P_LOSS, which must be dissipated from the device. Transferred into heat, P_LOSS increases the semiconductor's temperature and may overheat the device.

Specially designed heat sinks that conduct the heat from a power device to an ambient one are used for this purpose. They are made of aluminum and have various shapes, as shown above. Two cooling possibilities are available: air cooling - either natural or forced (by fans), and water cooling.

The rate of heat transfer between two bodies with a temperature difference of ΔΘ depends on the thermal resistance, R_Θ. The resistance, R_Θ is defined as R_Θ = ΔΘ / P_LOSS.

To determine the total resistance of a cooling assembly: R_Θja = R_Θjc+ R_Θchs + R_Θhsa and R_Θja =(Θ_j - Θ_a) / P_LOSS .If a user knows the power dissipation, P_LOSS, Θ_j and Θ_a, it is possible to calculate a needed R_Θja. For a given device, R_Θjc is a data sheet parameter. This permits easy calculation of R_Θhsa and allows the choice of a proper heat sink.

Connect the power supply to its appropriate polarity. Activate the gate control circuit. Observe the waveforms on the scope screen. Change the capacitor value by clicking on it. Observe the differences in the shape of the voltages across the thyristor on the oscilloscope display. Explain the influence of the capacitor value on the schematic t_off time.

Observe the different voltage shapes in the circuit by clicking on the red blinking symbol. Each symbol indicates the point where the oscilloscope will be connected. Explain how the t_off parameters will change if several green and red lamps are connected in parallel to the shown lamps.