Q2Industrial Electronics

(a) Triggering Methods for Thyristor

A thyristor (SCR) can be triggered (turned on) from its forward-blocking state into conduction by several distinct methods:

• Forward voltage (breakover) triggering: if the forward anode-to-cathode voltage is increased beyond the device's rated forward breakover voltage VBO, avalanche multiplication at junction J2 initiates the regenerative turn-on process without any gate signal — this method is generally avoided in normal operation since it does not allow controlled timing of turn-on and can subject the device to excessive stress.
• Thermal triggering: an increase in junction temperature reduces the width of the depletion region and increases the leakage current, which can raise the effective current gains (α1+α2) toward unity, causing the device to trigger due to heat alone — this is generally an undesirable, uncontrolled triggering mode that circuit designers seek to avoid through proper thermal management.
• Light triggering (LASCR): in a light-activated SCR, photons striking the exposed junction near the gate region generate electron-hole pairs, providing the equivalent of a gate current and triggering conduction — used in applications requiring electrical isolation between the control (triggering) circuit and the power circuit, such as in HVDC valve triggering and series-connected thyristor strings.
• dv/dt triggering: a rapidly rising anode-to-cathode voltage couples displacement current through the junction J2 depletion-layer capacitance, which can be large enough to trigger the device even without an intentional gate signal — this is generally an unwanted triggering mode (a device limitation) that is mitigated using RC snubber circuits to limit the rate of voltage rise across the thyristor.
• Gate triggering: the standard, controlled method used in essentially all practical power-electronic applications, in which a positive current pulse is deliberately injected into the gate terminal (with the anode already forward-biased), directly raising the NPN section's current gain and initiating the regenerative turn-on at a precisely controlled instant — gate triggering itself is further subdivided into DC gate triggering (a continuous DC signal), pulse gate triggering (a single or repetitive short pulse, most common in practice as it minimizes gate power dissipation), and AC gate triggering (a signal derived from the same AC supply, often via a phase-control circuit using an RC network or a UJT relaxation oscillator, used for phase-angle control in rectifier and AC voltage controller circuits).

(b) Comparison of IGBT, FET, and MOSFET

(Since a power MOSFET is itself a specific type of FET, the following comparison treats MOSFET and general FET together, contrasted against the IGBT, which combines MOSFET-like gate control with bipolar-like conduction characteristics.)

(i) Operation mode: a power MOSFET (and FET generally) is a unipolar, majority-carrier device that conducts current through a voltage-controlled conducting channel formed between drain and source when a sufficient gate-source voltage is applied, with no minority-carrier storage/recombination delay involved in its switching action. An IGBT is a hybrid device, combining a MOSFET-like gate input structure with an additional bipolar PNP output stage, so that its conduction involves both majority-carrier (MOSFET channel) and minority-carrier (bipolar) current components — this bipolar component provides conductivity modulation of the device's drift region, substantially lowering the on-state voltage drop for a given current rating compared to an equivalent-voltage-rated MOSFET, at the cost of a somewhat slower turn-off (due to the 'current tail' caused by stored minority-carrier charge that must recombine before turn-off completes).

(ii) Input/output impedance: both the MOSFET/FET and the IGBT are voltage-controlled devices with an insulated (MOS) gate structure, giving both an extremely high input impedance (essentially only the gate's capacitance draws any dynamic charging current, with negligible steady-state DC gate current in either case) — this is a major advantage shared by both devices over current-controlled devices such as the bipolar junction transistor or the thyristor, since it greatly simplifies the drive circuit and minimizes control-circuit power consumption. The output impedance of a MOSFET in its fully-on (triode) region is characterized by a well-defined on-resistance RDS(on), which is a resistive (ohmic) characteristic, whereas the IGBT's output characteristic in conduction more closely resembles that of a bipolar transistor, with a relatively constant, load-current-dependent on-state voltage drop (VCE(sat)) rather than a purely resistive characteristic, particularly significant for higher current ratings where the IGBT's bipolar conduction advantage becomes more pronounced.

(iii) Amplification: in a linear amplifier context (their typical use in industrial electronics is as a switch, but each also has a characteristic transfer/amplification behavior), the MOSFET/FET's drain current is related to gate-source voltage through its transconductance gm, giving a certain voltage-controlled current amplification, while the IGBT similarly exhibits a transconductance characteristic relating collector current to gate-emitter voltage, but with generally higher transconductance for a given die size due to the additional bipolar current gain contributed by its PNP output section, allowing IGBTs to control larger currents with the same small MOS-gate-driven input signal, at generally higher power-handling capability per unit chip area than an equivalently-rated power MOSFET.

(iv) Switching property: power MOSFETs are inherently faster-switching devices than IGBTs, since their unipolar (majority-carrier only) conduction mechanism involves no minority-carrier storage delay, allowing MOSFETs to switch at very high frequencies (hundreds of kHz to several MHz in some applications) with low switching losses, making them the preferred choice for high-frequency switch-mode power supplies and similar applications. IGBTs, due to the bipolar minority-carrier storage effect described above, exhibit a slower turn-off (with a characteristic current tail as stored charge recombines), limiting their practical switching frequency to a lower range (typically up to a few tens of kHz for most IGBT ratings, though modern trench-gate and field-stop IGBT designs have progressively pushed this frequency capability higher) — however, IGBTs are generally preferred over MOSFETs for higher-voltage, higher-current industrial motor drive and inverter applications, since their bipolar conduction-modulation advantage gives substantially lower conduction losses at these higher power levels, a trade-off between switching speed and conduction efficiency that is central to selecting the appropriate power semiconductor device for a given industrial electronics application.

Summary Comparison

• Gate/input signal: MOSFET and IGBT — voltage-controlled, high input impedance, negligible steady-state gate current; both far simpler to drive than current-controlled devices such as BJTs or thyristors.
• On-state conduction: MOSFET — purely resistive RDS(on), voltage drop rises linearly with current; IGBT — near-constant VCE(sat), better suited to high-current, high-voltage ratings.
• Switching speed: MOSFET — very fast, negligible tail current; IGBT — slower turn-off due to bipolar current tail.
• Typical application range: MOSFET — low-to-medium voltage (up to a few hundred volts), high-frequency SMPS, choppers; IGBT — medium-to-high voltage (hundreds of volts to a few kV), motor drives, UPS, and industrial inverters.
• Typical voltage/current rating ceiling: MOSFET generally practical up to a few hundred volts before on-resistance rises steeply with rated voltage; IGBT extends efficiently into the multi-kilovolt, multi-hundred-ampere range needed for industrial drives and traction converters.

Practical Gate Drive Circuit Considerations

Although MOSFETs and IGBTs both present a high DC input impedance, their gate-source (or gate-emitter) capacitance must still be charged and discharged rapidly at each switching transition, and the transient gate-drive current required to achieve a fast transition (needed to minimize switching loss, since power dissipated during the linear transition region of turn-on/turn-off is proportional to transition time) can be substantial even though the average gate current is negligible — practical drive circuits therefore use a dedicated low-impedance gate driver stage (often an isolated driver IC with a totem-pole output) capable of sourcing and sinking several amperes of peak transient current into the gate capacitance. A negative (or at least zero) gate-drive voltage during the off-state is commonly used in industrial IGBT drive circuits to improve noise immunity against spurious turn-on caused by dv/dt-induced displacement current through the device's Miller (gate-collector) capacitance, and an appropriate turn-on gate resistor is chosen to control the rate of rise of drain/collector current, balancing switching loss against electromagnetic interference and voltage overshoot caused by parasitic circuit inductance.

IGBT Latch-Up Concern

Because the IGBT's internal structure inherently contains a parasitic thyristor (a four-layer PNPN path formed by its bipolar output section together with an adjoining parasitic NPN transistor in the device structure), operating the IGBT beyond its rated collector current or under excessive dv/dt stress can cause this parasitic thyristor path to latch into an uncontrolled, gate-independent conduction state, exactly as an ordinary SCR would latch on — once latched, the device can no longer be turned off by the gate signal and will continue conducting until the load current is externally interrupted, generally destroying the device through excessive power dissipation. Modern IGBT designs mitigate this latch-up risk through careful doping profile and cell geometry design (shorting the parasitic NPN's base-emitter junction internally to raise the latch-up current threshold well above the device's rated operating current), but industrial IGBT applications must nonetheless respect the manufacturer's safe operating area (SOA) limits on peak current, dv/dt, and junction temperature to avoid inadvertently triggering this latch-up failure mode.