Search This Blog

Silicon Controlled Rectifier

A silicon controlled rectifier or semiconductor-controlled rectifier is a four-layer solid-state current-controlling device. The principle of four-layer p-n-p-n switching was developed by Moll, Tanenbaum, Goldey and Holonyak of Bell Laboratories in 1956. The practical demonstration of silicon controlled switching and detailed theoretical behavior of a device in agreement with the experimental results was presented by Dr Ian M. Mackintosh of Bell Laboratories in January 1958.  The name "silicon controlled rectifier" is General Electric's trade name for a type of thyristor. The SCR was developed by a team of power engineers led by Gordon Hall and commercialized by Frank W. "Bill" Gutzwiller in 1957.
Some sources define silicon controlled rectifiers and thyristors as synonymous, other sources define silicon controlled rectifiers as a proper subset of the set of thyristors, those being devices with at least four layers of alternating n- and p-type material.  According to Bill Gutzwiller, the terms "SCR" and "controlled rectifier" were earlier, and "thyristor" was applied later, as usage of the device spread internationally.

SCRs are unidirectional devices (i.e. can conduct current only in one direction) as opposed to TRIACs, which are bidirectional (i.e. current can flow through them in either direction). SCRs can be triggered normally only by currents going into the gate as opposed to TRIACs, which can be triggered normally by either a positive or a negative current applied to its gate electrode.


The silicon control rectifier (SCR) consists of four layers of semiconductors, which form NPNP or PNPN structures, having three P-N junctions labeled J1, J2 and J3, and three terminals. The anode terminal of an SCR is connected to the p-type material of a PNPN structure, and the cathode terminal is connected to the n-type layer, while the gate of the SCR is connected to the p-type material nearest to the cathode.
An SCR consists of four layers of alternating p- and n-type semiconductor materials. Silicon is used as the intrinsic semiconductor, to which the proper dopants are added. The junctions are either diffused or alloyed (alloy is a mixed semiconductor or a mixed metal). The planar construction is used for low-power SCRs (and all the junctions are diffused). The mesa-type construction is used for high-power SCRs. In this case, junction J2 is obtained by the diffusion method, and then the outer two layers are alloyed to it, since the PNPN pellet is required to handle large currents. It is properly braced with tungsten or molybdenum plates to provide greater mechanical strength. One of these plates is hard-soldered to a copper stud, which is threaded for attachment of a heat sink. The doping of PNPN depends on the application of SCR, since its characteristics are similar to those of the thyristor. Today, the term "thyristor" applies to the larger family of multilayer devices that exhibit bistable state-change behaviour, that is, switching either on or off.
The operation of an SCR and other thyristors can be understood in terms of a pair of tightly coupled bipolar junction transistors, arranged to cause the self-latching action:

Modes of operation

There are three modes of operation for an SCR depending upon the biasing given to it:
  1. Forward blocking mode (off state)

  2. Forward conduction mode (on state)

  3. Reverse blocking mode (off state)

Forward blocking mode

In this mode of operation, the anode (+) is given a positive voltage while the cathode (−) is given a negative voltage, keeping the gate at zero (0) potential i.e. disconnected. In this case junction J1 and J3 are forward-biased, while J2 is reverse-biased, due to which only a small leakage current exists from the anode to the cathode until the applied voltage reaches its breakover value, at which J2 undergoes avalanche breakdown, and at this breakover voltage it starts conducting, but below breakover voltage it offers very high resistance to the current and is said to be in the off state.

Forward conduction mode

An SCR can be brought from blocking mode to conduction mode in two ways: Either by increasing the voltage between anode and cathode beyond the breakover voltage, or by applying a positive pulse at the gate. Once the SCR starts conducting, no more gate voltage is required to maintain it in the ON state.
There are two ways to turn it off:
  1. Reduce the current through it below a minimum value called the holding current, or

  2. With the gate turned off, short-circuit the anode and cathode momentarily with a push-button switch or transistor across the junction.

Reverse blocking mode

When a negative voltage is applied to the anode and a positive voltage to the cathode, the SCR is in reverse blocking mode, making J1 and J3 reverse biased and J2 forward biased. The device behaves as two reverse-biassed diodes connected in series. A small leakage current flows. This is the reverse blocking mode. If the reverse voltage is increased, then at critical breakdown level, called the reverse breakdown voltage (VBR), an avalanche occurs at J1 and J3 and the reverse current increases rapidly. SCRs are available with reverse blocking capability, which adds to the forward voltage drop because of the need to have a long, low-doped P1 region. (If one cannot determine which region is P1, a labeled diagram of layers and junctions can help.) Usually, the reverse blocking voltage rating and forward blocking voltage rating are the same. The typical application for a reverse blocking SCR is in current-source inverters.
An SCR incapable of blocking reverse voltage is known as an asymmetrical SCR, abbreviated ASCR. It typically has a reverse breakdown rating in the tens of volts. ASCRs are used where either a reverse conducting diode is applied in parallel (for example, in voltage-source inverters) or where reverse voltage would never occur (for example, in switching power supplies or DC traction choppers).
Asymmetrical SCRs can be fabricated with a reverse conducting diode in the same package. These are known as RCTs, for reverse conducting thyristors.

Thyristor turn-on methods

  1. forward-voltage triggering

  2. gate triggering

  3. dv/dt triggering

  4. temperature triggering

  5. light triggering
Forward-voltage triggering occurs when the anode–cathode forward voltage is increased with the gate circuit opened. This is known as avalanche breakdown, during which junction J2 will break down. At sufficient voltages, the thyristor changes to its on state with low voltage drop and large forward current. In this case, J1 and J3 are already forward-biased.


SCRs are mainly used in devices where the control of high power, possibly coupled with high voltage, is demanded. Their operation makes them suitable for use in medium- to high-voltage AC power control applications, such as lamp dimming, power regulators and motor control.
SCRs and similar devices are used for rectification of high-power AC in high-voltage dc power transmission. They are also used in the control of welding machines, mainly MTAW (metal tungsten arc welding) and GTAW (gas tungsten arc welding) processes similar. It is used as switch in various devices.

No comments

Popular Posts