Power Electronics

Power electronics integrates the use of three areas of specialization in engineering, namely, Electronics, Power, and Control.  It deals with the use of electronic for the conversion and control of electronic for electronic power in various industrial, commercial, residential and aerospace applications.  The evolution in power electronic is the synthesis of multiple technological disciplines. Today, a specialist in this area is supposed to have expertise in power semi-conductor devices, converter circuits, electrical machines, analog/ digital electronic, control theory, computer-aided design, microcomputers, and the newly emerging VISI electronic.

Applications of power electronics Systems

• DC-AC regulated power supplies
• Electro-chemical processes
• Heating and Lighting Control
• Electronic Welding
• Power Line Var.  and Harmonic Compensation
• High voltage DC system
• Variable speed constant frequency system
• Solid State Circuit Breaker
• Induction Heating

Types of Converter systems (power conditioning system)

• Rectifier (Uncontrolled & Controlled)
• AC voltage controller
• DC-DC converters
• Inverters (Uncontrolled & Controlled)
• Cycloconverters

Generalized power converter

Input filter -
filters off any harmonic noise generated in the power  converter, such that the feedback switching harmonic jammed into the input line are minimized.

Power circuits -
interconnects all the power switching devices together to form a bulk power module.

Output filter -
smoothes out any switching harmonic in the output so as to obtain the desired output waveform.

Switching Control -
provides gate signals to all the power devices according to the switching strategy used and the external control signals.

Electrical Power control
Achieved by :    (1)    a switch
                        (2)    an adjustable stepless  impedance
When the power flow is regulated by the controlled variation of series resistance, there occurs a waste of heat and loss of system efficiency. For this reason, the control of high power circuits is invariably approached via the use of semi-conductor power devices as switches.

Limitations of practical semi-conductor switches

• Finite on-resistance
• Finite leakage current in “ off ” state  
• Finite turn-on and turn-off times.
• Switching losses.

Classifications of semi-conductor switches
Device conducts automatically when forward bias is applied, e.g. DIODE.
Device begins to conduct in the forward direction upon command of a control signal and continue to conduct until the next current zero crossing , e.g. THYRISTOR.
Forward conduction can be initiated and interrupted by control signals applied to the control terminal , e.g. TRANSISTOR, GTO, MOSFET, IGBT.

DIODES
The  basic concepts will not be discussed since they should be learnt in previous years.  However, several criteria will be discussed.
A.    Reverse voltage breakdown
Increasing applied reverse bias eventually to a junction reverse voltage breakdown and the diode current is controlled by the external circuit. Breakdown may be due to one of the following three phenomena :
1.  Punch-through voltage
The reverse voltage extends the depletion layer to at least one of the ohmic contacts and the device presents a short-circuit to that voltage in excess of the punch-through voltage in excess of punch-through voltage.  Punch-through occurs with devices which employ a low-concentration region, as is usual with high voltage devices.

2.  Avalanche Breakdown
Avalanche breakdown is the most common mode of breakdown and occurs when the electric field in the depletion layer at junction exceeds a certain level which is dependent on the doping level of the lighter doped region. Minority carriers associated with the leakage current are accelerated to kinetic energies high enough for them to ionize silicon atoms on collision , thereby creating a new hole- electron pair . These are accelerated in opposite directions, because of the high electric field strength , colliding and ionising again and again - hence lead to carrier multiplication or avalanche.    

3.    Zener breakdown
Zener breakdown occurs with heavily doped junction regions. It occurs when the depletion layer is too narrow for avalanche yet electric field grows very large and electrons tunnel directly from the valence band to the conduction band. These modes of reverse voltage breakdown are not necessarily destructive provided that the current is uniformly distributed . If the current density in a particular area is too high, a local hot spot may occur, leading thermal destruction .

B.    Turn-off characteristic
Once a diode is conducting, the junction cannot instantaneously revert to the blocking mode because during forward conduction there is an excess of minority carriers in each diode region (i.e. holes in n-region and electrons in p-region) and these must be removed at turn-off.

The recovery charge has two components, one due to internal excess charge recombination and the other due to the reverse diode current. These charges must first be swept out before the diode can regain its reverse blocking capability. The time for reverse current to flow is called the reverse recovery time, trr = t4 + t5.  The time t5 is defined by projecting Irr though 0.25Irr.  Vrr is the reverse over-shoot voltage produced by the high dirr /dt effect and the external circuit inductance L. Afterwards, dirr /dt  reduces to zero , the circuit supports zero volts and the diode blocks the reverse  voltage .

Schottky Barrier Diode
Apart from the conventional type p-n function diode. There exists a low-voltage , high-speed type , called the Schottky barrier diode (or Schottky Diode).  It has very low forward voltage drop but with a very high reverse leakage current relative to application imposing a reverse bias of less than 400V. Another key feature is the high switching speed.

POWER TRANSISTORS

• Bipolar Junction transistor (BJT)
• Metal-Oxide Semi-conductor Field Effect transistor (MOSFET)

A.  BJT
Types :PNP and NPN
In switching application, the common emitter configuration is generally used.

Operating Regions

• Cut-Off†
• Active (linear)
• Saturation †

†    The operating modes that the transistors will be used in power electronics.
 

Switching Characteristics

A.  Turn-on time
It consists of a delay time ( td ) and a rise time (tr ). The delay time is mainly due to the charging of the BE junction capacitance and can be significantly reduced by increasing the applied rate and magnitude of the forward base current ( Ibf ). The rise time (tr ) depends on the time constant determined by capacitances of the transistors.

B.  Turn-off time
It is the sum of storage time ( ts ) and the fall time ( tf ). The storage time is the time required to remove the saturating charge from the base.  The fall time ( tf  ) depends on the time constant determined by the capacitance of the reverse-biased BE junction.

Power Darlington  Transistor
Since a power bipolar transistors consume considerable input control power, an intermediate power driver must be cascaded with transistor before a unit may be considered for operation from a signal - power level source.

The complementary connection is particularly useful as it uses as low level PNP driver to convert a high power NPN bipolar into a high power PNP unit.  It is found that NPN units can be constructed on an integrated basis, sometimes with three devices in cascade, with protecting diodes and thermal stabilizing resistors also included.

POWER MOSFET

Types :  

• Depletion type (normally on )
• Enhancement type (normally off)

Operating Region

• cut-off region†
• constant current region (saturation)
• constant resistance region (triode)†

†    The operating modes that the transistors will be used in power electronics.

Since MOSFETs do not have inherent minority-carrier delay as in bipolar devices.  The turn-on and turn-off times depend on the ability of the gate drive circuit to charge and discharge a tiny input capacitor Ciss = Cgs+Cgd . Typical switching times are 150 to 200ns.

Synchronous Rectifier
For the n-type MOSFETs, a negative drain voltage is applied to the device with their gates connected to the source and the p-n junction becomes forward-biased.  When the drain voltage exceeds a knee voltage, VN of approximately 0.7V at room temperature, the p-n junction begins to conduct.  If a positive gate voltage is applied to the device which create a channel, an alternate path for current flow between drain and source is created.  Consequently, the power MOSFET will exhibit a very low voltage drop, significantly lower than a p-n junction diode.

THYRISTORS
Thyristor is a family name of bipolar devices, which comprise 4 semi-conductor layers, including

1. Silicon-Controlled Rectifier (SCR)
2. Triac (Bi-directional thyristor)
3. Reverse-conducting thyristor (RCT)
4. Gate -turn-off thyristor (GTO)

1.  SCR
When the anode voltage VAK is positive, the junctions J1 and J3 are forward-biased, but junction J2 is reverse-biased and the leakage current flows from A to K.  The SCR is then said to be in forward blocking or off-state condition and the leakage current is known as off-state current ID.
If VAK is increased to sufficiently large value i.e.> VBO ( forward breakdown voltage ) and avalanche breakdown occurs in the reverse - biased junction J2.  Since the other junction J1 and J3 are already forward-biased, there will be free movement of carrier across all three junctions resulting in large forward anode current.  The device will then be in a conducting state or on state.  The on-state voltage drop across the 4 layers is small.
         In the on-state, IAK is limited by the external impedance or resistance.  However, to maintain the required amount of carrier flow across the junction , the anode current IAK must be more than a value known as Latching Current ( IL ), otherwise , the device will revert to the blocking condition as the VAK is reduced.  Once a SCR is conducting, it behaves like a conducting diode and there is no control over the device.  The device will continue to conduct because there is no depletion layer on the junction J2 due to free movements of carrier. If the forward anode current is reduced below a level known as the Holding Current (IH ), depletion region would develop around junction J2 due to the reduced no. of carriers and the SCR would be in the blocking state.

         An increase of reverse voltage can cause SCR failure by punch-through of the reverse-biased junction J1.  In high voltage SCR this is prevented by using a thick N2 layer.

Thyristor turn-on
1.    Gate triggering
2.    Forward breakdown
3.    Irradiation methods (light-activated SCR)
4.    dv/dt triggering
5.    Temperature elevation
 

Thyristor turn-off
1.    Insertion of series impedance
2.    Reversal of the anode voltage
 The main objective is to make the anode current is lower than the holding current. Typical turn-off times lie in the range of 10-300s.

Ratings of Thyristors
1.    di/dt Limitation
In many applications, a thyristor is subjected to a very steep rise of current at and after switching ‘on’.  The ‘vertical’ regenerative triggering action occurs exceedingly quickly but the ‘horizontal’ migration of triggering charge in the base region take place very slowly.
The result is that only the region near the gate contact actually triggers.  Therefore, unless limited by external means, the full-load current passes through a small fraction of cathode surface near the gate.  The thermal capacity of silicon is very small and it is a poor thermal conductor with the result that a hot spot is created.  The result is damaged to the crystalline structure of die or cracks in it due to mechanical stress as silicon is brittle.  The performance can be improved by using centre gate, dual gate, and gate-cathode amplifying dual gate.

2.    dv/dt Limitation
Even if a small amplitude fast transient of forward voltage is applied to the device, it may trigger as displacement current will flow through the depletion layer of J2.

Anode voltage ratings
1.    Crest working reverse voltage        ( VRWM )
2.    Repetitive peak reverse voltage              ( VRRM )
3.    Non-repetitive peak reverse voltage         ( VRSM )
4.    Crest working off-state voltage                ( VDSM )
5.    Repetitive perk off-state voltage              (VDRM )
6.    Non-repetitive peak off-state voltage    (VDSM )
 

Current Ratings
1.    Continuous on-state current
2.    Mean-on state current
3.    Repetitive peak on-state current
4.    Non-repetitive surge on-state current
5.    Surge current capability for fusing
6.    Repetitive peak reverse current
7.    Rate of rise of forward current

Gate ratings
1.    Peak forward gate voltage
2.    Peak reverse gate voltage
3.    Peak forward rate current
4.    Peak reverse rate current
5.    Peak gate power
6.    Mean gate power

Power Losses of Switching Devices
1.    Switching Losses
2.    Off-state leakage power loss
3.    Conduction Power Loss
4.    Device input Device power Loss

1.    Switching Losses

An approximation of straight line switching is assumed.  For resistive load,


For an inductive load ,



2.    Off-state leakage power loss

              : Duty cycle of the switch

3.    Conduction Power Loss



4.    Drive Input Device Power Loss
Usually very small compared to other losses. For bipolar transistors, .
Thus,
Total Power Loss  =  Psw(on) + Psw(off) + Poff + Pon + Pg

Thermal considerations and Heat sinks
Power Losses in semi-conductor devices are dissipated in the form of heat.  This heat must be transferred in the device to maintain the operating junction temperature within the specific range. It should be noted that the reliability and life expectancy of any power semi- conductor are directly related to the maximum device junction temperature experienced.

Pd     - average power loss in the device  (W)
Rjc     - thermal resistance from junction to case (oC/W)
Rcs     - thermal resistance from case to sink (oC/W)
Rsa     - thermal resistance from sink to ambient (oC/W)
Rca     - thermal resistance from case top ambient (oC/W)
 Total thermal resistance ,
In general , ,

Rjc and Rcs are normally specified by the power device manufacturers.  If the device power loss Pd is known, the required thermal resistance of the heat sink can be calculated for a specified ambient temperature Ta.  The next step is to choose a heat sink and its size which would meet the thermal resistance requirement.  
                                   
Methods of reducing switching times
1.    Speed-up capacitor
2.    Speed-up diode
3.    Overdrive the gate signal and protect by collector clamping diode.


Driving of switching devices
A.    Power Transistors

• provide adequate base current to saturate  the output power transistor.
• provide electrical isolation between the power circuit and the switching command signals.
• being high switching speed, low power consumption and simple circuitry.

Remark :
When operated in the saturation region, the conduction loss of a power transistor is minimized.  However, the base drive current should not be too high to heavily saturate the transistor since the turn-off is lengthened due to large quantity of base charge.

B.    Power MOSFET

• Do not suffer secondary breakdown
• Drive by a current source followed by a voltage source.
• Protect the gate from damaging overvoltage, zener diode protection is necessary.
• Suitable for low power, high frequency applications.

C.    Thyristors

Three basic types :

• DC firing signals
• Pulse signals
• AC phase signals

Commutation is achieved by a capacitor
Self commutation by parallel resonant turn-off
Controlled resonant turn-off
Parallel capacitance turn-off

Protection

1.     Snubber circuits

• a.     to shift the device switching loss to the snubber circuit
• b.     to avoid second breakdown
• c.    to control    dv/dt

2.    Series connection

• Steady-state voltage sharing is forced by using a resistor.
• Transient response is controlled by a low non-inductive resistor and capacitor in series are placed in parallel.

3.    Parallel connection
When the load current is greater than the thyristor current rating, operation is parallel.  Current division may vary due to different thyristors unless matched thyristors are used. To attempt to get equal current sharing, external balancing reactors can be used .