Bipolar transistor

A bipolar transistor is an electronic system containing emi-conductor of the family of the transistors. Its principle of operation is based on two junctions PN, one on line and the other in reverse. The polarization of opposite junction PN by a weak electric current (sometimes called transistor effect) will make it possible to order a current much more important. It is the principle of the amplification of current.
The discovery of the bipolar transistor made it possible to effectively replace the electron tubes in the years 1950 and thus to improve the miniaturization and the fiabilisation of the electronic assemblies.

Transistor point-contact

Retort first bipolar transistor invented by two researchers of the Bell Laboratories and tested successfully on December 16th, 1947.
John Bardeen and Walters Brattain under the direction of William Shockley had set up an work group on the semiconductors since 1945.
A first prototype developed by Shockley did not function correctly and it is with the assistance of the physicists Bardeen and Brattain that it succeeds in detecting and correcting the various problems involved in the electric fields in the semiconductors. December 16th, 1947, Bardeen and Brattain reflect in place a small device made up of germanium and two contacts out of gold which made it possible to amplify the signal in input of a factor 100. December 23rd, they presented it to the remainder of the laboratory. John Pierce, an engineer in electricity, gave the name of transistor to this new component which was officially presented at the time of a press conference to New York on June 30th, 1948.

Transistor with junctions PN

Shortly after the discovery of Bardeen and Bartain, Shockley tried another approach based on junction P-N, a discovery of Russell Ohl going back to 1940.
Work of Shockley opened the way for the realization of the bipolar transistors composed of a sandwich NPN or PNP. However, their manufacture posed real issues because the semiconductors were insufficiently homogeneous. A chemist of the Bell Laboratory, Gordon Teal, developed in 1950 a process of purification of germanium. Morgan Sparks, Teal and other researchers could manufacture junctions PN then a sandwich NPN.

Improvement of the manufactoring processes

The two following years were devoted in the search of new processing and manufactoring processes of germanium. Silicon was more difficult to work than germanium because of its higher melting point but it offered a better stability in front of the thermal changes. Nevertheless, it is not before 1954 that the first silicon transistor could be carried out. In 1952, the first apparatuses with transistors were marketed. The Bell Laboratories imposed their know-how during all the decade with in particular the tuning of masking by oxide by Carl Frosch. This technique offered new prospects for the mass production for the silicon transistors. The photolithography on the silicon plates, a process developed by Jules Andrus and Walter Jump in 1955, strongly contributed on arrival of more precise and effective tech news of machining. Still today, the photolithography constitutes a crucial stage in the realization of the transistors.

Types and Symbols

Bipolar symbols of transistors
Caption: B: Bases - C: Collector - E: transmitter
The catalogs of transistors comprise a high number of models. One can classify the bipolar transistors according to various criteria:
the type: NPN or PNP. These two types are complementary, that is to say the direction of the currents and tensions for the PNP is the complement of those of the NPN. Transistors NPN having in general characteristics better than the PNP, they are used. The continuation of the article will thus discuss only the circuits using of transistors NPN
power: the transistors for the amplification of small signals dissipate only a few tens or hundreds of milliwatts. The transistors average power support a few Watts; the transistors of power, used for example in the audio amplifiers of power or the stabilized power supplies can support, on the condition of being placed on an adequate heatsink, more 100W
frequency band: transistors for low frequencies (function correctly until a few MHz), averages (until a few tens of MHz), high (until some GHZ), even higher (maximum frequencies of oscillation of several hundreds of GHZ).
The figure opposite watch the symbol and indicates the name of the 3 electrodes of the transistors. One can thus distinguish 3 interesting potential differences: Vbe, Vbc and Vce; and 3 currents: basic current Ib, of transmitter IE and collector Ic. However, these 6 variables are not independent. Indeed, one can write:
Vce = Vcb + Vbe and IE = Ic + Ib
Certain manufacturers propose many networks of characteristics, but this tendency is in the process of disappearance. Moreover, it should be known that the typical parameters of the transistors change with the temperature, and strongly vary from one transistor to another, even for the same model.

Principle of operation

Physical principle of transistor NPN

We will take the case of a type NPN for which the tensions Vbe and Vce and the current entering at the base are positive.
In this type of transistor, the transmitter, connected to the first zone NR, is polarized with a tension lower than that of the base, connected to the P. zone the diode transmitter/base is thus on line polarized, and of the current (injection of electrons) circulates of the transmitter towards the base.
In normal functioning, the junction base-collector is polarized in reverse, which means that the potential of the collector is quite higher than that of the base. The electrons, which for the majority diffused to the zone of field of this junction, are collected by the collecting contact.

Simple model of a transistor under linear operation

Ideally all the current from the transmitter is found in the collector. This current is an exponential function of the tension base-transmitter. A very small variation of the tension induces a great variation of the current (the transconductance of the bipolar transistor is much higher than that of the field-effect transistors).
The current of the holes circulating of the base towards the transmitter added to the current of recombination of the electrons neutralized by a hole in the base corresponds to the basic current Ib, coarsely proportional to the collector current Ic. This proportionality gives the illusion that the basic current controls the collector current. For a model of transistor given, the mechanisms of recombinations are technologically difficult to control and gain Ic/Ib can only be certified higher than a certain value (for example 100 or 1000). The electronic assemblies must take account of this uncertainty (see low).
When the tension collector-bases is sufficiently positive, it quasi totality of the electrons is collected, and the collector current does not depend on this tension; it is the linear zone. In the contrary case, the electrons station in the base, recombine, and the gain falls; it is the zone of saturation.

Principles of design

at first sight, the bipolar transistor seems to be a symmetrical device, but in practice dimensions and the doping of the three parts are very different and do not allow to reverse transmitting and collecting. The principle of the bipolar transistor rests indeed on its geometry, on the difference in doping between its various areas, even on the presence of a heterojunction.
The current of the holes of the base towards the transmitter must be negligible compared to the current of electrons come from the transmitter. That can be obtained
with a very high doping of the transmitter compared to the doping of the base. A heterojunction can also block the current of holes completely, and authorize a high doping of the base.
The recombinations of the electrons (minority) in the base rich in holes must remain weak (less than 1% for a gain of 100).
That imposes that the base is very fine.
The surface of collector is often larger than the surface of the transmitter, to ensure that the path of collection remains short (perpendicular to the junctions)

Model of Ebers-Moll

Model of Ebers-Moll of a transistor under linear operation
The model of Ebers-Moll results from the superposition of the modes Forward and Reverse.
It consists in modelling the transistor by a power source placed between the collector and the transmitter.
This power source comprises two components, ordered respectively by the junction BE and junction BC.
The behavior of the two junctions is simulated by diodes.

Electric characteristics

Idealized characteristics of a bipolar transistor
Ic ⁄ Vbe characteristic of a bipolar transistor
The figure opposite watch pace of characteristic Ic ⁄ Vce. Two principal zones are distinguished:
zone of saturation, for Vce tensions; 1V; in this zone, Ic depend at the same time on Vce and Ib;
linear zone: the collector current is nearly independent of Vce, it depends only on Ib.
When the transistor works in the linear zone, it can be regarded as an amplifier of current: the output current, Ic is proportional to the current of input, Ib. the Ic ⁄ Ib report ⁄ ratio, called gain while running of the transistor, is one of the fundamental characteristics of this one; it is generally noted by the Greek letter I². I² of the illustrated transistor is worth 100. It is important to hold account owing to the fact that, for a given transistor, I² increases with the temperature. In addition, the standard I² of transistors in the same way present a great dispersion. That obliges the manufacturers to indicate classes of gain. If one takes for example a very widespread transistor like the BC107, the gain while running varies from 110 to 460. The manufacturer then tests the transistors after manufacture and adds a letter after the number, to indicate the class of gain has, B, C
The Ic ⁄ Vbe figure shows that, for a transistor working in the zone of saturation, the Vbe tension varies very little. In lower part of Vbe = 0,65V, the transistor does not lead. When this value is exceeded, called tension of threshold, the collector current increases exponentially. One shows thus that the collector current Ic is equal to, oà ¹ Is corresponds to the current of saturation of the junction transmitter bases and VEA the tension of Early.
Ic = Is * [1 + (Vce ⁄ VEA)] * exp (Vbe ⁄ Vth)
In practice, Vbe generally lies between 0,65V (for Ic of some my) and 1V (for the transistors of power traversed by one Ic important, EP. 1A).
In addition to the gain while running, one uses some other electric characteristics to qualify the operation of a transistor:
its frequency of transition FT, characteristic its speed of operation (produced accessible gain-tape); the able the transistor is to reach a high transconductance for a low capacity, the more the frequency of transition is large; grà¢ce with the technological advancements, one thus nowadays reaches FT of several tens of gigahertz. The bipolar transistors are in that higher than the field-effect transistors.
its tension of Early VEA, all the more large as the transistor behaves like an ideal source of current; resistance transmitter-collector corresponds to the ratio between the tension of Early and the collector current.
its transconductance (gain tension-current or slope of the active component), directly related to the collector current (at first approximation, it is worth gm = Ic ⁄ Vth où one with the thermal tension Vth = kT ⁄ q). Biensur, each transistor being designed to function correctly in a certain range of current, it is useless to amplify the current beyond a certain limit to increase the gain.

General principles of implementation

As the parameters of a transistor (and particularly I²) vary with the temperature and from one transistor to another, it is not possible to calculate the properties of the circuits (gain in tension) with greater accuracy. The 4 basic principles given below make it possible to simplify calculations.
The currents collector and transmitter of a transistor can be regarded as equal, except in the event of thorough saturation.
So that a current Ic circulates in the transistor, it is necessary to provide him a basic current equal (for an operation in the linear zone) or superior (for an operation in the zone of saturation) to Ic ⁄ I ².
When the transistor is conducting, the tension Vbe base-transmitter lies between 0,6 and 1V.
The tension collector-transmitter has little influence on the collector current as long as one works in the linear zone of the characteristics.
The following law is useful for the more elaborate assemblies.
For two transistors identical to same temperature, the same Vbe tension defines same running Ic.

Amplifying assemblies

Generally, one can distinguish two great types of operation of the transistors:
operation in the linear zone of the characteristics; it is used when it is a question of amplifying signals coming from a source or of another (microphone, antenna);
operation in commutation: the transistor commutates between two states, the blocked state and the state saturated (weak Vce, it is the item A). The fast circuits avoid this state has, which corresponds to an excess of carriers in the base, because these carriers are long to evacuate, which lengthens the switching time of the state saturated towards the blocked state.
In the paragraphs which follow, we will discuss operation of the transistor like amplifier. Operation in commutation is discussed in end of the article.


Polarization makes it possible to place the point of rest of the transistor (state of the transistor when any signal is not applied to him) at the desired place of its characteristic. The position of this point of rest will fix the tensions and currents noted quiescent and as well as the class of the amplifier (has, B, AB or C).
Ibo, Ico, Vce0   Vbe0
Because of the capacities of connection and decoupling, the relation current ⁄ output voltage of the transistor assemblies is often different between the modes statics and dynamics. In order to study the behavior of the assembly at the time of the static mode and the dynamic mode, one calculates the lines of load in these two cases. The point of polarization of the assembly is at the intersection of these two characteristics.

Straight line of static head

Simple diagram of polarization of a bipolar transistor

The way simplest to polarize an assembly of the transmitting type common is represented on the diagram opposite. The transmitter is with the mass, the base is connected to the supply voltage Vcc via R1, the collector is connected has Vcc via R2. For reasons of simplifications, the assembly is not charged. The relations between resistances R1 and R2 and the various tensions are the following ones:
R1 = Vcc - Vbe0 ⁄ Ibo R2 = Vcc - Vce0 ⁄ Ico
the formula can be rewritten like this
R2 = Vcc - Vce0 ⁄ βIbo
This simple diagram suffers however from a great defect: calculated resistances strongly depend on the gain while running I² of the transistor. However, this gain while running changes from one transistor with other (and that even if the transistors have the same references) and strongly varies according to the temperature. With such an assembly, the point of polarization of the transistor is not control. More complex assemblies are thus preferred to him but whose point of polarization depends less on the gain while running I² of the transistor.

Diagram practices polarization.

To avoid this problem, one has recourse to the complete diagram indicated below. Resistances R1 and R2 form a tension divider which fixes either the current bases but the tension between base and the zero. The relation between the currents and tension can be written as follows:
Ico = β( Veq - Vbeo) ⁄ Req + (β + 1) R4
Req = R1 R2 ⁄ R1 + R2
Veq = Vcc * R2 ⁄ (R1 + R2)
If Req is small in front of (I² + 1) R4, the relation current/tension can be written
Ico = (Veq - Vbe0) ⁄ R4
The current of polarization is then independent of the gain while running I² of the transistor and is stable according to the temperature. This approximation also amounts choosing R1 and R2 so that the current which them cross-piece is large in front of Ib0. Thus, the tension applied to the base of the transistor depends little on the basic current Ib0.
The line of static head is a straight line plotted in the figure which gives Ic according to Vce. It passes by item Vcc on the x axis, and the point Vcc ⁄ (R3+R4) on the axis of the Y. For a supply voltage, a R3 load and a resistance of R4 transmitter given, this line of load indicates the point of operation.

Dynamic characteristics

diagram of a common transmitting assembly.

The complete diagram of a common transmitter amplifier is represented on the figure opposite. Compared with the diagram used during the calculation of the point of polarization, the diagram used comprises in more the capacitors of connection C1 and C2, the capacity of C3 decoupling as well as a Rl load.
The capacitors of connection prevent the tension and D.C. current to be propagated in all the assembly and to find themselves in input and output or to modify the polarization of the other assemblies present in the final circuit. The capacities of decoupling make it possible to remove certain components (here R4) assembly in a certain frequency band.
The value of the capacitors of coupling C1 and C2 is selected so that those have a sufficiently low impedance in all the range of the frequencies of the signals to amplify:
compared to the resistance of input of the stage for the C1 capacitor;
compared to the resistance of load for the C2 capacitor;
The selected value of C3 so that its impedance weak is compared with that of R4 in the desired frequency band.
The capacitors C1, C2 and C3 had not been represented until now, because they have an infinite impedance with the continuous one. The Rl load was not, it also, presents because the C2 capacitor prevented the D.C. current dà" with polarization crossing it and thus from influencing the static characteristics of the assembly.

Diagram are equivalent small signals of a common transmitting assembly in low frequencies

In order to calculate the characteristics of the assembly in dynamic mode, one has recourse to a model small signals of the transistor. This model makes it possible to describe the behavior of the transistor around its point of polarization. The model used here is simplest possible. It models the transistor gràçe has a Rbe resistance and a power source whose intensity is proportional to the basic current. If one wishes a finer modeling of the transistor, it is necessary to use a more complex model (Ebers-Moll for example). Rbe resistance models the slope of the straight line Vbe (Ib) at the point of polarization and is calculated as follows:
Rbe = Vt ⁄ Ibe0 = kT ⁄ qIbe0
with: Vt the thermal tension, K the Boltzmann constant, Q the elementary charge, and T the temperature of the transistor in kelvins.a Vt ambient temperature is worth 25mv
With this model, one obtains easily :
Vs = (R3 * R1 ⁄ R3 + R1 ) * IC
Ib = Ve ⁄ Rbe
Ic = βIb
If one notes G the gain in tension of the stage and S, his transconductance. one obtains :
G = Vs ⁄ Ve = βR3R1 ⁄ Rbe * (R3 + R1)
S = Ic ⁄ Ve = β ⁄ Rbe
The transconductance can be defined as follows: it is the variation of the collector current due to a variation of the tension base-transmitter; it is expressed in A ⁄ V. It is primarily determined by the D.C. current of transmitter IE (fixed by the circuit of polarization).

Power dissipated in the transistor

For an amplifying assembly in class has, the power dissipated in the transistor is worth:
P = Vce.Ic + Vbe.Ib
Where Vce and Vbe is the continuous potential differences between the collector and the transmitter, the base and the transmitter, and Ic, Ib are respectively the collector currents and basic. This power does not vary when a signal is applied to the input of the amplifier. As the gain while running (beta) of the transistor is generally very high (a few tens to a few hundreds), the second term is generally negligible.
Why calculate the power dissipated in the transistor, To evaluate the temperature of the junction it of the transistor, which cannot exceed approximately 150°C for a normal functioning of the amplifier.
The temperature of junction will be calculated using the thermal Law of Ohm.
In our example, the power dissipated in the transistor is worth 4.2.10-3 + 0,65.20.10-6 = 4.2mW. The temperature of the junction, if the ambient temperature is of 25°C and thermal resistance junction-environment of 500°C ⁄ W, is worth 25 + 500.4,2.10-3 is 27,11°C.

The transistor in commutation

Assembly of a transistor for operation in commutation
One calls operation in repeating spring, an operating process of the transistor where this last either is blocked, or traversed by a sufficiently important current so that it is saturated (that D. Vce reduced to less 1V). In the figure opposite, when the Int switch is opened, Ib is null, therefore Ic are null and Vc = Ucc (not B on the characteristics of the transistor). On the other hand, when one closes Int, a current (Ucc - Vbe) ⁄ RB circulates in the base. The transistor thus will try to absorb a collector current Ic equal to I². Ib. However, generally, load RL is selected so that Ic are limited to a value lower than I². Ib, typically 10.Ib. The transistor is then saturated (not has on the characteristics).

Power dissipated in the transistor

The power dissipated in the transistor can be calculated by the formula:
P = (Vce.Ic + Vbe.Ib) .RC

Bipolar isolated grid transistor

Usual symbol of the IGBT
The bipolar isolated grid transistor IGBT is a semiconductor device of the family of the transistors which is used like electronic switch, mainly in the assemblies of the electronics of power.
This component, which combines the advantages of previous technologies that is to say, the simplicity of the control field effect transistor compared to the bipolar transistor, while maintaining low conduction losses of the latter led to many advances applications in power electronics, both in regard to the reliability of the economic aspect.
Transistors IGBT made it possible to consider developments until then nonviable in particular in variable speed like in the applications of the electric machines and the convertors of power which accompany us each day and everywhere, without we of it being particularly conscious:
electric household appliances
house automation


The first attempt relating to this component is its realisation in discrete components, with a field-effect transistor of low power ordering a bipolar transistor of power (BipMos assembly). The goal is to simplify the control circuits inherent in the applications of the transistors of power in commutation, strong complexes in the years 1970-1980.
Technology IGBT was patented in the United States on 14 December 1982 by Hans W. Beck and Carl F. Wheatley, Jr., under the name of Power MOSFET with year Anode Area (N° Patent: 4,364,073)
It is a recent technology, which succeeds the thyristors, the Darlington transistors and thyristors GTO.
The first generation of transistors IGBT presented big problems of locking, which were corrected in the second generation appeared with the beginning of the year 1990. The end of the century knew three new generations of transistors IGBT, which increased the performances for currents and tensions important (IGBT with structures trench, CSTBT.
The characteristics of the IGBT make that in the years 2000 it largely was essential in all the fields of the electronics of power vis-a-vis the other types of components for the ranges of tension 600 V on 3300 V and that it bores in the tensions higher vis-a-vis the GTO, like in the tensions lower vis-a-vis the MOSFET, although it is slower.


Diagram are equivalent of the IGBT
The IGBT is a hybrid transistor, gathering a field-effect transistor of type MOSFET in input and a bipolar transistor at output. It is thus ordered by the grid voltage (between the grid and the transmitter) which is applied to him, but its characteristics of conduction (between collector and transmitter) are those of bipolar. The equivalent diagram of transistor IGBT opposite watch a third transistor, which represents in fact a property parasitises responsible for the latching.
This structure gives him the energy low costs of ordering of a MOSFET, with the weaker losses of conduction (on surface of chip given) of bipolar. Moreover, the IGBT can manage a tension much higher than that managed by the MOSFET.


The conductance is defined by the resistance of the transistor when this one is busy: it is called in the case of also Ron a FET or VCEsat for bipolar. It is an important characteristic because it determines the heating of the component according to the Ice current: the weaker VCEsat is, the more the acceptable current can be strong. In the case of the IGBT, the conductance is minimised by the use of a bipolar transistor at output, and by the optimisation of the saturation of this one. For that, it is possible to decrease Ron of the MOSFET of input, and to increase the gain of the bipolar transistor. However a too important gain will involve a high risk of latching
Last technologies SPT (Software-Punch-Through), known as SPT+, make it possible to decrease the direct voltage drop further VCEsat of the rdre from 25 to 30%.


The weakness of the IGBT (compared with the MOSFET) results primarily in its switching speed, in particular at the time of the passage of the state passing in a blocked state: the holes present in the zone of epitaxy N (Drift transistor zone) must recombine or be evacuated when the grid voltage passes in lower part of the threshold of commutation. Technology Pt has a buffer zone (buffer) near the zone of drift transistor to accelerate the absorption of the holes. Transistors IGBT-PT will be thus faster, but will have more raised VCEsat tension.
The maximum frequencies of commutation can be notably increased by the use of circuits of assistance to commutation passive dissipative), but especially active (nondissipative), of type ZVS (Zero Switch Voltage, commutation to the zero of tension), ZCS (Zero Current Switch, commutation to the zero of current) or others. These circuits, while ensuring of commutations douces&, allow a drastic reduction in the losses of commutation, while facilitating largely the setting in conformity of the equipment relating to electromagnetic compatibility. Nevertheless, because of their complexity and of their cost, they are used still little in the strong powers.


The IGBT presents four layers N-P-N-P which can under certain conditions of becoming busy with the manner of a thyristor, because of presence of the parasitic transistor between transmitter and bases principal bipolar transistor: it is the locking effect. Under these conditions the transistor will remain busy, with destructive effects, until the supply is cut off. The manufacturers managed to decrease this main issue of transistor IGBT, and this in various manners: reduction of the transconductance of the bipolar transistor of output, use of new technologies of engraving like the IGBT Trench. These evolutions, as well as the improvement of the processes of ordering of grid, make that the phenomenon of locking is currently well controlled and does not pose any more problems with the development of the industrial use of the IGBT.

Surface of security

The surface of security or zone of reliable operation or SOA (which are the English initials for Safe Operating Area) indicate the authorised zones of operation of the transistor in the voltage plan. In these zones, the transistor can work without suffering damage during the time when at the same time an important current crosses the semiconductor and an important tension is present at its terminals, that is to say apart from saturated operation (conducting and weak voltage drop). In all the cases these zones of operation can be only transitory, because the powers dissipated in instantaneous values are several orders of magnitude above the nominal acceptable power of the component. Three critical phases are distinguished:
the short-circuit. It is about the zone known as of SCSOA (for Short Circuit SOA) or surface of security of short-circuit. When the load ordered by the transistor is in short-circuit, the required current is in theory infinite. In practise, the current ICE in the transistor is limited by tension VGE on the grid and the value of the transconductance, like by the external circuit. The risk for the IGBT is then locking. According to the family used this risk is minimised with the detriment of the transconductivity or VCEsat. Certains IGBT has an internal circuit of limitation of the current of short-circuit to some multiples of the rated current.
commutation ON-OFF with an inductive load. It is about the zone known as of RBSOA (for Reverse Bias SOA) or opposite surface of security. At the time of this phase of transition one passes from a state where a current stable (and important) ICE is established in the load and the transistor in a state where the transistor is blocked. Tension VCE believes then of a few volts in the increased supply voltage of the FCEM of the inductive load. This FCEM must be limited, for example, by a diode known as of free wheel on its terminals. During this phase the current is constant - because when the load is inductive, it tends to be opposed to the current fluctuation - this until the end of the recombination of the carriers and with the blocking of the junction by increase in the barrier of potential. It follows a risk of breakdown of the component by formation of located hot points and thermal runaway; this phenomenon is known for the bipolar transistors of power under the name of the second breakdown (Second Down Station-waggon in English). The IGBT are however much more robust than the bipolar ones for the behaviour in opposite surface of security. Circuits of assistance to commutation with blocking, by deriving the current from the inductive load (in an auxiliary capacitor for example) for the period of blocking authorise a commutation with quasi-null losses for silicon and avoid the risk of the second breakdown.
the use of the transistor in linear mode. The study of this phase is of interest more limited, because it is not the usual operating process of the IGBT. An special attention on this operating process is however necessary at the time of the setting in work of the circuits protective of the component against short-circuit.


The transconductance of a IGBT is the relationship between the output current and the tension of input. This report ⁄ ratio depends on many parameters, in particular the size of the transistor, the output temperature or current. Contrary to the bipolar transistors, the MOSFET and the IGBT do not have a gain of transconductance which falls with the output current.


The following table shows the typical performances of some products of the market of the transistors.
It releases the general tendency:
VCEsat increases and the frequency of use decreases when the behaviour in tension increases;
the MOSFET and the GTO become competing at the ends of the range.
Compared average characteristics
  MOSFET 600V IGBT 600V IGBT 1700V IGBT 3300V IGBT 6500V GTO 6000V
VCEsat with 125°C 2,2 V 1,8 V 2,5 V 3,5 V 5,3 V 3 V
typical frequency 15-100 Khz 6-40 Khz 3-10 Khz 1-5 Khz 0,8-2 Khz 0,3-1 Khz
Products of certain manufacturers can deviate significantly from the values mentioned because concerning different optimisations (improving one of the parameters to the detriment of the other) or using very recent technologies.


Cross-section of basic cell (double diffusion, Pt)
The structure of a IGBT is based on that of vertical MOSFET diffused doubling: the thickness of the support is used to separate the drain from the source. The typical thicknesses of the wafers are about 70 with 100µm. A zone known as of epitaxy, doped N, allows the appearance of a channel when electrons are injected by the grid (VG>0, state passing).
The technique of double diffusion is used to create doped wells P ⁄ P+ near the source. The presence of a doped area P+ decreases the risk of latchup, while increasing the tension of threshold of commutation.
The principal difference between a vertical MOSFET and a IGBT is the existence of a layer of P+ substrate (strongly doped) side collecting drain. This layer injects holes in layer N, which causes to decrease the voltage drop to the state passing and to transform it into bipolar transistor.
In a blocked state, it is the layer N which supports the tension. This maximum tension will be all the more important as the layer NR of epitaxy will be doped little and/or thick.
To the state being on, the power will be limited by the width of the channel. The vertical structures allow the parallelization of several basic cells, in order to amplify the acceptable current and to decrease resistance to the state passing RDSon.

Various structures of IGBT

Section of a IGBT Not Through Punch Section of a IGBT Through Punch Section of a IGBT Through Punch in trench
IGBT NPT with plane grid is the structure simplest to realise. It uses thinner chips, without additional N+ layer. The transconductance will be less low, it is thus more robust in situation of short-circuit.
IGBT Pt (English initials: Punch Through) with plane grid uses thick chips comprising a layer N+ buffer. It has in theory a weaker voltage drop in a busy state.
This buffer layer between the zone of epitaxy NR and the zone of P+ injection of the collector makes it possible to obtain a distribution of the trapezoidal electric field.
One also finds transistors called DS-IGBT (for Depletion Stop IGBT), or FS-IGBT (for Field Stop IGBT), which show the same characteristics as the PT-IGBT, with a buffer layer less doped. That makes it possible to use the simpler manufacturing techniques of a NPT-IGBT.
The preceding structures known as with grid planes have the advantage of being easy to realise. Nevertheless a technology known as of grid in trench is also used: the zone of epitaxy is cut out under the grid so as to decrease the phenomena of latching and to thus allow more important densities of current. This geometry is as more compact and generally more powerful as the geometry with plane grid.


Module IGBT 3300 V 1200 has
Interior of a module IGBT 600 V 400 has
The IGBT are manufactured with techniques similar to that of the integrated circuits (like the MOSFET, but contrary to the GTO and the thyristors. This has as a consequence which the size of the chip is limited to approximately 1cm2, whereas one can make monolithic diodes of 150mm of diameter (176cm2).
The large IGBT are thus multichip modules, made up of many chips in parallel, generally brazed on an Al-Sic or copper sole through which one ensures their cooling.
The majority integrate also an antiparallel diode (or of wheel-free), itself multichip. This diode is in fact a very great part of module IGBT, because its characteristics (in particular of recovery) must be compatible with the IGBT itself, required crucial. Besides this represents one of the first applications for the silicon carbide semiconductors.
One finds mainly IGBT channel NR. the complementary structure channel P possible, but is limited to the small powers, because as for the bipolar transistors and the MOSFET, the characteristics obtained are less good (higher losses for example).
These components are practically available in all the current cases, since small plastic case (TO-220) for currents of a few amps to a few tens of amps and tensions collector-transmitter of 600 with 1500volts, to the modules of strong power of a few hundreds of amps and a few kilovolts.

Ranges and uses

These components are available in a range of tension going of 600 (and less) to 6500 volts, and of the currents up to 2400 amps per module. The most current values of tension are:
600V: value adapted to connexion on an alternative network 230V;
1200V: value adapted to connexion on an alternative network 400V;
1700V: value adapted to connexion on an alternative network 660V;
3300V: value used in railway traction 1500V continuous;
6500V: value used in railway traction 3000V continuous;


The usual applications of the IGBT are the inverters, rectifiers, choppers for the feedings with cutting and variable speed, but also for the FACTS.

Junction Field Effect Transistor

Schematic diagrams of a JFET with channel NR
A transistor of the type JFET is a field-effect transistor whose grid is directly in contact with the channel. One distinguishes the JFET with a channel of the type NR, and those with a channel of the P. type.

Types and Symbols

Transistor JFET channel P.
Just as one distinguishes two types of bipolar transistors, the NPN and the PNP, one distinguish also two types of transistors FET to junction (JFET): the JFET channel NR and the JFET P. channel the JFET channel NR is used the most (like besides bipolar transistor NPN), its symbol is indicated below in the diagrams of setting in ? uvre. The symbol of the JFET channel P is identical, with share the arrow which changes direction.
As for the bipolar transistors, there are the choice between a great number of models, according to the power put in work and the waveband to be amplified.
The main difference between the bipolar transistors and the JFET, it is that the bipolar transistors are ordered while running (it is necessary to provide to the transistor a certain basic current so that it leads), whereas the JFET is ordered in tension (the current drain depends on the Vgs tension).

Principle of operation

Diagrams of a JFET with channel NR during a normal functioning (VDS 0, VGS 0) with two different values of polarisation.
The transistor is formed by a layer of semiconductor little doped (channel) placed between two layers of semiconductor of doping opposite and connected between them to form the electrode which one names the grid. The ends of the channel form two other named electrodes the drain and the collector) and the source and the transmitter). For a JFET with channel of the type NR, the grid is of P. type In normal functioning the tension between the drain and the source is positive (VDS 0) and that between the grid and the source (junction PN) is negative (VGS 0). The increase in this opposite tension makes grow the zones of déplétion (nonconducting) around the channel until the pinching of this one. The conduction of the channel is thus modulated by tension VGS.

Electric characteristics

The two figures opposite represent:
the static characteristics Id according to Vds and Vgs; these characteristics are rather similar to the characteristics of the bipolar transistors, put aside the fact that Ib is replaced by Vgs;
the Id characteristic according to Vgs, for values of Vds higher than a few volts.
The Id/Vgs characteristic is described by the equation:
ID = IDSS [1 - VGS ⁄ VP
where Idss (Id of short-circuit, shorts circuit in English) is the current one of saturation, obtained when Vgs is null, and Vp (tension of pinching, pinch off) is the tension Vgs which cancels Id. For the illustrated transistor, Idss = 10mA and Vp = -4V.
The primary application of the JFET, it is like amplifier small signals. For this application, one polarises the JFET in the middle of the linear zone (notice that the names of the zones are reversed compared to the bipolar transistors).

Calculation of an amplifier small signals


The diagram opposite watch how the point of operation uninterrupted of a JFET is fixed. Divider R1-R2 fixes the potential of the grid; let us notice that here, contrary to the case of the bipolar transistors, it is not necessary to take account of the current roasts, because this one is quasi-non-existent. One can thus write:
VGM = UCC * R2 ⁄ R1 + R2, where VGM is the potential difference between the grid and the mass.
The potential of the source is worth Id.R4, it is higher than the potential of the grid since the Vgs tension must be negative. The potential of the drain is worth VDD Id.R3.

Calculation of the gain

Let us apply a small sinusoidal tension ve to the grid, through a capacitor so that the source of alternating signals (generator of signals, microphone, antenna, another amplifying stage) does not modify the polarisation of the stage.
Are successively id =, vs = id. R3 and finally G = vs ⁄ ve = S.R3
where ve is the alternating voltage of input (we will use the tiny ones to indicate the tensions and alternative courses), id the AC current of drain, vs the output voltage on the level of the drain, G the gain in tension of the stage, and S the transconductance. This one can be defined as follows: it is the variation of the current drain due to a variation of the tension grid-source; it is expressed in A/V. It is primarily determined by the D.C. current of drain. Indeed, by deriving the equation which gives Id, one finds:
S = So. (1 - Vgs/Vp) with So = - 2.Idss/Vp.
Note: let us not forget that Vgs and Vp are negative.
In our example, So = 8mA/V, S = 3mA/V and G = 6. This gain is much smaller than that which one can obtain with a bipolar transistor amplifier.
As for the resistance of input of the stage, it is equal to R1//R2 since the resistance of input of the JFET is extremely high (considering the grid current is quasi-no one). When one wishes a resistance of raised input, one generally omits R1; in this case, the resistance of input of the stage is simply equal to R2.

Grid metal-oxide field-effect transistor

Photographs representing two MOSFET
Type P Type P
Type NR Type NR
enrichment impoverishment
Caption: D: Drain - S: Source - G: Roast
A grid field-effect transistor isolated more usually named MOSFET (English Metal acronym Oxide Semiconductor Field Effect Transistor - which results in field-effect transistor with structure metal-oxide-semiconductor), is a type of field-effect transistor. Like all the transistors, the MOSFET modulates the current which crosses it using a signal applied to its named central electrode grid. It finds its applications in the numerical integrated circuits, in particular with technology COMPLEMENTARY METAL OXIDE SEMICONDUCTOR, like in the electronics of power.
These transistors are divided into two categories:
MOSFET with enrichment. They are used because of their nonconduction in the absence of polarization, of their strong capacity of integration like for their easier manufacture.
MOSFET with impoverishment. Those are characterized by a conducting channel in the absence of polarization of grid (VGS = 0).
The transistor is characterized by the load of its majority carriers which determines if it is of type P or NR. the symbols of the MOSFET make it possible to differentiate its type and its category. The letters on the three electrodes correspond to the drain, the source and the grid.


The MOSFET was conceived in a theoretical way in 1920 by Julius Edgar Lilienfeld which patented it as being component being used to control the current. However, necessary technology with its construction was not available before 1950. Indeed, the characteristics of the MOSFET require techniques of manufacture nonavailable at that time. The advent of the integrated circuits allowed its realization. Thus, M.M Atalla and Dawon Khang of the Bell Laboratories built the first MOSFET into 1960 which will make its appearance in the integrated circuits in 1963. A little later the development of technology COMPLEMENTARY METAL OXIDE SEMICONDUCTOR ensured the commercial and technological future MOSFET in integrated electronics.

Principle of operation

Transistor MOSFET calls upon only one type of charge carrier (it is thus component unipolar). The basic rule rests on the effect of the electric field applied to the structure metal-oxide-semiconductor that is to say the electrode of grid, insulator (dioxide of silicon) and the semiconductor layer (called substrate); generally in micro-electronics the metal layer is replaced by polycrystalline silicon.
When the potential difference between the grid and the substrate is null it does not occur anything. Progressively of the increase in this potential difference the free loads in the semiconductor are pushed back junction semiconductor ⁄ oxide, first of all creating a zone known as of déplétion, then when the potential difference is sufficiently large it appears a zone of inversion. This zone of inversion is thus a zone oà ¹ the type of charge carriers is opposed to that of the remainder of the substrate, thus creating a channel of conduction.

Operation of the MOSFET with channel NR

Cross-section of a MOSFET with channel NR
The following example takes into account the case of a channel NR, which is most frequent; the channel P has an identical operation by reversing polarizations.
The transistor generally consists of a substrate of the type P, slightly doped, in which one diffuses by epitaxy two N+ zones which will become the source and the drain. Silicon above the channel is oxidized dioxide of silicon - SiO2) then metallized to produce the grid, which constitutes a capacity between the grid and the substrate.
In general, the source and the substrate are connected to the mass. The drain is carried to a higher potential of those of the source and substrate, which creates an electric field between the source, the substrate and the drain.
At rest, two cases are possible
That is to say the capacity roasts ⁄ substrate is floating with vacuum: there are almost no carriers to lead possible running, the two junctions source-substrate and substrate-drain is polarized in reverse; it is the case of a MOSFET with enrichment.
That is to say the capacity roasts ⁄ substrate is in inversion, which means that electrons of the substrate are attracted in the vicinity of oxide. Those constitute a minority surge of carriers which will be available to lead the current between the source and the drain. The transistor is conducting, the MOSFET is known as with impoverishment.
In both cases, the current source-drain is modulated by the grid voltage. For the transistor with enrichment, it is necessary to apply a positive tension to the grid to bring the capacity grid-substrate in inversion: the transistor led starting from a certain threshold. For the transistor with impoverishment (déplétion), the channel leads when the grid is with the mass, it thus should be led to a negative tension to put an end to conduction.
When the led transistor, an increase in polarization between the drain and the source amplifies the current (non-linéairement). starting from a tension of drain higher than the grid voltage minus the tension of threshold, the electrostatic field enters the substrate and the grid is reversed locally in the vicinity of the drain. The electrons disappear at this place, the current saturates. Any increase in the tension of drain beyond the tension of saturation leads to a disappearance even more important electrons, and to a weak increase (even null) in the current.
àtension source-drain constant, the current of saturation varies like the square of the tension grid-substrate.

Operating processes

Operating processes of a MOSFET with channel NR
Characteristics Voltage of a MOSFET with channel NR

Tension of threshold

The tension of threshold is defined as being tension VGS between the grid and the source for which the zone of inversion appears, that is to say the creation of the channel of conduction between the drain and the source. This tension notes VTH, TH being the abbreviation of English threshold (threshold). When the tension grid-source VGS is lower than the tension of threshold VTH, it is said that the transistor is blocked, it does not lead. In the contrary case, one says that it is busy, it leads the current between the drain and the source.

Linear zone

IDS = β (VGS - VTH - ½VDS) * VDS
β = W ⁄ L * µ Cox
W : width of the channel
L : length of the channel
µ : mobility of the charge carriers (mobility of the electrons in the case of a MOSFET with channel NR)
Cox : oxide capacity of grid

Not pinching


Saturated zone

IDS = IDSSAT * (L ⁄ L - λ)
λ = λ0ln [1 + (VDS - VDSSAT ⁄ VDSSAT)]
λ0 = √ ∈si ⁄ ∈ox * xjTox
si: permittivity of silicon
ox: permittivity of oxide of grid
xj: depth of junction
Tox: thickness of oxide of grid


A very useful analogy to easily include/understand the operation of a FET, without using concepts of electrostatics, is to compare it with a tap water. The grid is the command similar to the screw pitch of the tap which controls the water flow (running). After a quarter of turn, it may be that only a weak filament of water runs. Then, the current increases quickly with a weak rotation. Lastly, in spite of turns in the vacuum, the current does not increase any more, it saturates. Lastly, if one wants to increase the flow of the tap, it is necessary to increase the bore (potential difference grid-substrate).

A quantum transistor of only seven atoms

A transistor of the size of a small molecule was already carried out but it is the first time that one manufactures also small, a constitutional body of 7 of the same atoms species is called element or chemical element. Composition: - a core of nucleons: protons and neutrons practically concentrating all the mass of the atom.
You point out, in 1989, researchers of IBM managed to handle xenon atoms individually, writing the name of their company with 35 of them. Twenty years later, of the researchers of the UNSW Center for Quantum Technology Computer (CQCT) and of the University of Wisconsin-Madison used the same technique, namely a microscope with tunnel effect, to handle silicon and phosphorus atoms.
With 7 phosphorus atoms, they built a quantum box in a silicon crystal. It is it which behaves like a transistor, of a size of 4 nanometers.
A stage on a long way
It is the first time that one thus builds an electronic system with a microscope with tunnel effect and it could be a question well of a big step in the control of the nanomonde. There still, the hope is to be able to produce computers even smaller, faster and less greedy in energy. According to the researchers of the CQCT, that opens also the gate with the realization of a quantum computer using silicon.
One is not yet there. To produce such a quantum computer is a thing, to build able to compete with the traditional models is another. It is known indeed that it is necessary to cross the frightening obstacle of the décohérence and even if nature, with photosynthesis, gives signs that it is possible, it is not known as that a great power of calculation can be reached.

The first molecular transistor

Scientists of the University of Yale and institute of science and technology (the word technology has two meanings in fact:) of Gwangju in South Korea (Southern is a name:), succeeded in creating the first transistor starting from a simple molecule. The team proved that a benzene molecule connected to gold contacts could behave like a transistor with silicon.
By adjusting the tension, the researchers could raise or lower the states of energy (In the commonsense reasoning energy indicates all that makes it possible to carry out a work, faiquer heat,) molecule; they showed that it could be used like a traditional transistor at the molecular level.
It is as a ball which rolls to the top and over a hill, where the ball represents the electric current and the height (the height has several significances according to the field approached.) of the hill represents the various states of energy of the molecule, "indicated Mark Reed, professor in engineering (engineering indicates the active whole of the functions of the design and the studies to the responsibility for) and applied sciences with Yale.
We could adjust the height of the hill, allowing the current to pass when its value was low, and to stop it when it was strong. The result is functionally identical to the traditional transistors, but with unemolécule of some atoms.
This work is based on searchs for Reed in the Nineties, which had shown that individual molecules could be "imprisoned" between electrical contacts. Although this new transistor is a scientific projection some, Mark Reed admits that practical applications such as "the molecular computers" smaller and more rapids, will not be born (the day or the day is the interval which separates to raise it to lay down Sun, it is the period between two) - if they see it - before a few decades. We are not on the point (C-W communication) to create the next generation of integrated circuits,".
For as much, this is the ultimate goal as regards electronic miniaturization, recently, a australo-Finnish team mentioned the development of a transistor with single atom. These last years, the engineers moved away from silicon for materials of a matter of natural or artificial origin that the man works to make objects.) exotic of them such as graphene to reduce the size of the transistors, 10 atoms constituting the record. The molecular transistor still reduced these limits, with probably a nanometer length.

NOMFET: an organic transistor opens the way with new generations of computers neuro-inspired

For the first time, researchers of CNRS (the National center of the scientific research, more known under its initials CNRS, is the greatest organization of) and of the ECA developed (C-W communication) a transistor mimant at him only the principal functionalities of a synapse. This organic transistor (the organic chemistry is a branch of chemistry concerning the description and the study of a big class of molecules) realized containing pentacene and of gold nanoparticules, named NOMFET (Nanoparticle-Organic Memory transistor), opens the way with new generations of computers neuro-inspired, able to answer in a way similar to the nervous system (the nervous system is a formed networked system of the bodies of the directions, nerves, brain, spinal-cord,). The study is published on January 22nd, 2010 in the review Advanced Functional Materials.
Model of fluctuation of the period in transitory mode of the NOMFET following various values of the train of impulse in input (0.05 Hz - 3 Hz).
In the development of new strategies for the data processing, an approach consists with mimer the operation of the biological systems, such as the networks of neurons, to carry out electronic circuits with the new capacities. In the nervous system, the synapse is the junction between two neurons. It allows the transmission of the electric messages of one neuron the other and the adaptation of the message according to the nature of the signal
General terms
A signal is a simplified message and generally coded.
There exists in form) entering (plasticity). For example, if the synapse receives impulses very close to entering signals, it will transmit a more intense potential of action. Conversely, if the impulses are distant, this last will be weaker.
It is this plasticity which the researchers made a success of with mimer with the transistor NOMFET.Le transistor, base unit of an electronic circuit, can be used as simple switch (a switch (derivative of rupture) is a device or body, physics or virtual, making it possible to stop or) it can then transmit or not a signal or offer many functionalities (amplification, modulation, coding).
The innovation of the NOMFET lies in the original combination of an organic transistor and gold nanoparticules. These nanoparticules encapsulated, fixed in the channel of the transistor and covered with pentacene has a ratchet effect (the ratchet effect is a physicochemical phenomenon affecting the performances of the electric fencers.) allowing them mimer operation of a synapse during the transmission of the potentials of action between two neurons. This property confers thus on the electronics component (an electronics component is an element intended to be assembled with others in order to realize one or more)
capacity to evolve/move according to the system in which it is placed. The performance is to be compared with the seven transistors COMPLEMENTARY METAL OXIDE SEMICONDUCTOR (has minimum) necessary hitherto for mimer this plasticity. The devices carried out were optimized until nanometric sizes in order to be able to integrate them into large scales. The computers neuro-inspired thus produced are capable of functions comparable with those of our brain.
Contrary to the computers out of silicon used in abundance in the computers for intensive calculation, the computers neuro-inspired can solve problems much more complex like the visual recognition.

Flexible Transistors based on graphene

Solution graphene monolayer.
Thus, researchers from the CEA, CNRS, Université Lille 1, of Northwestern University have developed a new original method of manufacturing flexibility and mobility transistors electronics capable of operating Very High Frequency (GHz) and manipulated using graphene in solution, compatible with printing techniques.
Such electronic components should allow the development of efficient electronic circuits, integrated into everyday objects.
Graphene, single plane of carbon atoms in hexagonal structure, has exceptional properties. In particular, the high mobility of electrons in this material shall promote the very high frequency electronic components made of graphene. Moreover, its mechanical properties make it a flexible material. These two advantages could be exploited in the manufacture of electronic circuits and components for various sectors: development of flexible displays, transistors and electronic components and manufactured high performance at low cost. Currently, several synthetic routes of graphene exist. One of them allows to produce in the form of a solution of particles of a few hundred nanometers in diameter, in water stabilized by surfactants.
For this conductive ink, the synthetic route used allows to select only those leaflets monolayers that provide remarkable electronic properties and not a mixture of monolayer and multilayer graphene. Another unique feature: the production of components can be performed on a wide variety of media such as glass, paper, or an organic substrate. Researchers from the CEA, CNRS, Université Lille 1 and Northwestern University have, for the first time, developed an original method for manufacturing flexible transistors from graphene solubilized on polyimide substrates.
They then studied thoroughly high-frequency performance. In the method developed, the graphene sheets in solution are deposited onto the substrate under the effect of an alternating electric field applied between electrodes made previously. This technique of dielectrophoresis (DEP) to direct the deposition of graphene and obtain a locally high density of deposited layers. This density is crucial for excellent high frequency performance. Thus, the charge mobility in transistors is made of about 100 cm2/Vs, which is much higher than the performance obtained with molecules or semiconducting polymers. These transistors are therefore reaching very high frequencies in the range of 8 GHz, never far obtained in organic electronics
These results show that graphene prepared in the form of a conductive ink material is particularly competitive for the realization of electronic functions in a flexible range of high frequencies (GHz) inaccessible to conventional organic semiconductors used. This new generation of transistors opens significant opportunities in many fields of applications such as flexible displays (foldable or rollable), integrated electronics in textiles and everyday objects such as RFID tags capable of processing and transmitting Information.
Graphene is a two-dimensional crystal (monoplane) whose stack carbon is graphite. It was isolated in 2004 by Andre Geim, of the Physics Department, University of Manchester, who has been for this discovery the Nobel Prize for Physics in 2010 with Konstantin Novoselov. It can be produced in two ways: by mechanical extraction of graphite (graphene exfoliated) whose technique was developed in 2004, or by heating a crystal of silicon carbide, which allows the release of silicon atoms (graphene epitaxy).
Graphene was first identified in 2004 by the team of Andre Geim at Manchester University in England. This discovery earned him, he and his colleague Konstantin Novoselov, the Nobel Prize in Physics in October 2010. If the structure of graphene is a "textbook case" in the calculation of electronic band structure, scientists have long believed that such a structure could not really exist.

Graphical representation of graphene.
Graphene is naturally in the graphite crystals, where it comes in the form of a stack of sheets. Several techniques designed to make it usable emerged in recent years.
exfoliated graphene
The idea is to extract a very thin layer of graphite crystal using a tape, then repeating the process a dozen times on the samples and products so that they be as thin as possible. They are then deposited on a sheet of silicon dioxide as an optical identification will select the samples consist of a single layer.
epitaxial graphene
This is to produce graphene from silicon carbide. A sample of the latter is heated under vacuum at 1300 ° C so that the silicon atoms of the outer layers in evaporate. After a very specific time, the remaining carbon atoms are reorganized into thin layers of graphene.
graphene produced by CVD
Graphene is produced by catalytic decomposition at high temperature of a carbon-containing gas (methane, ethylene ...) on a metal, in general, copper, nickel or iridium, . The optimal reaction temperature depends on the type of gas and metal. There are two main types of reaction:
On metals such as copper, the decomposition of carbon-containing gas produces carbon atoms which remain on the surface because of their very low solubility in the metal, and interact to form a surface layer graphene .
On metals nickel type, it is the wide variation in solubility of carbon in the metal as a function of temperature which, once the carbon product has diffused into the metal at high temperature, to find expelled surface thereof when the temperature decreases. This technique generally produces few graphene layers.
The principle involves the oxidation of graphite in an acidic medium and then using hydrazine as reducing solvent to purify graphene.
Graphene is conductive.
The electronic band structure that can be called graphene semiconductor material of zero-gap.
One of the most spectacular property of graphene own electrons at the Fermi level whose apparent mass is zero, and is the only physical system and dividing it into massless fermions, this which is of great interest for physics fondamentale.L one of the most striking effects is the appearance in a magnetic field of a quantum Hall effect at room temperature.
The theoretical electron mobility is 200 000 cm2.V-1.s-1 which makes that this material is particularly attractive for electronics and high-frequency terahertz.
Graphene is a two-dimensional crystal the electrons move on graphene at speeds of 1000 km · s-1, 30 times the speed of electrons in silicon . Thanks again to its properties of two-dimensional crystal and a newfound ability to self-rapid cooling, not a graphene transistor heats up very little.


It is in 1964 what appears on the market a device ensuring the setting in conduction and the blocking of two alternations of an alternating voltage by one only electrode (the trigger). This component with three electrodes was called TRIAC.


According to whether a1 anode or a2 anode is positive compared to the other, the triac will start in the first or the third quadrant.
The release of the triacs can be carried out in the four following modes
When have it feeds Triac into alternate there are 4 possibilities of release:
Mode 1 and 2: the alternating voltage changes the polarity of the Anodes A1, A2 and the signal of release is always positive. (little recommended system).
Mode 1 and 3: the alternating voltage on A1, A2 and the signal of release is identical to the current principal (economic release)
Mode 4 and 2: the alternating voltage on A1, A2 and the signal of release is opposed to the principal current (without interest, disadvised).
Mode 4 and 3: the alternating voltage on A1, A2 and the signal of negative release by report ⁄ ratio A1 (powerful industrial release)
Here some references of triacs
Reference Characteristics Case
BT136-1000 4A/1000V TO220
BT137-1000 8A/1000V TO220
BT138-1000 12A/1000V TO220
BT139-1000 16A/1000V TO220
BT136F1000 4A/1000V Insulated
BT137F1000 8A/1000V Insulated
BT138F1000 12A/1000V Insulated
BT139F1000 16A/1000V Insulated
BTA06-400B 6A/400V TO220
BTA06-1000T 6A/1000V Igt 5 my TO220
BTA08-400B 8A/400V TO220
BTA12-400B 12A/400V TO220
BTA16-400B 16A/400V TO220
BTA25-1000B 25A/1000V RD91
BTA40-1000B 40A/1000V RD91
BTA64 6A/400V insulated  
BTA84 8A/400V insulated  
BTA87 8A/700V insulated  
BTA100 10A/400V insulated  
BTA107 10A/700V insulated  
BTA124 12A/400V insulated  
BTA127 12A/700V insulated  
BTA164 16A/400V insulated  
BTA167 16A/700V insulated  
BTA254 20A/400V insulated  
BTA267 20A/700V insulated  
BTA404 40A/400V insulated  
BTA414 40A/400V insulated  
BTA417 40A/700V insulated  
BTB44 4A/400V  
BTB87 8A/700V  
BTB127 12A/700V  
BTB156 15A/1000V  
BTB164 16A/400V   
BTB166 16A/1000V   
BTB167 16A/700V   
MAC15A8 15A/1000V TO220
TIC206M 4A/1000V Igt 10mA TO220
TIC216M 6A/1000V Igt 10mA TO220
TIC225D 8A/400V Igt 5mA TO220
TIC225M 8A/1000V Igt 10mA TO220
TIC226D 8A/400V TO220
TIC226M 8A/1000V TO220
TIC236M 12A/1000V TO220
TIC246D 15A/400V TO220
TIC246M 15A/1000V TO220
TIC263M 25A/1000V TO3P
TLC336T replaced by Z0405MF  
TO509D 5A/400V Igt 10mA  
TO510D 5A/400V Igt 25mA  
TO812D 8A/1000V Igt 50mA  
T2512M 25A/1000V Igt 50mA  
T4012M 40A/1000V Igt 50mA  
Z0107MA 0.8A/1000V TO92
Z0405MF 4A/1000V TO220
Z0409D 4A/400V Igt 25mA TO220
Z0410D 6A/400V Igt 25mA TO220




It is a conducting semi element which includes ⁄ understands two structures of thyristor NPNP in opposite direction. One notices on the diagram two thyristors N4P1N1P2 and P1N1P2N2.
Cross simplified of a triac



The thyristor is component which becomes completely conducting, in D.C. current, following an electric impulse on its electrode called "trigger" or "G". Not only this conduction is honest and brutal but it is permanent even after suspension of this current of trigger.
The thyristor is an one-way element contrary to the triac which is bidirectional, this component comparable to a diode is thus ordered the current passes in only one direction of the anode towards cathode. The third electrode the trigger makes it possible to order release
The thyristor is rectifying because of its conduction in only one direction, and it is connected with a diode. It is used as switch, while applying a signal to its electrode of control G (Trigger) it passes from the state blocked in a busy state and can thus replace a contactor. The thyristor behaves as an amplifier of power because the current of command about the milliampere makes it possible to order the principal current several Amps.
Into alternate the thyristor can be used as regulator, for example in an assembly of the gradator type of light.
Some values of thyristors:
Reference Characteristics Igt (my) Case
2N682 25A/50V   TO-48
2N683 25A/100V   TO-48
2N685 25A/200V   TO-48
2N689 25A/400V   TO-48
2N690 25A/600V   TO-48
2N883 0.5A/400V   TO-48
2N5207 35A/1200V   TO-48
2N6405 16A/800V   TO-220
2P4M 2A/400V   TO-220
50RIA60  50A/600V   TO-65
BRX49 replaced by PO102DA    
BRY39 2.5A/70V 2 gâch.   TO-18
BRY55-60 0.8/60V   TO-92
BRY55-200 1A/200V   TO-92
BT151-500 12A/500V   TO-220
BT151-650 12A/650V   TO-220
BT169D 0.5A/400V   TO-92
P0102DA 0.8A/400V   TO-92
TIC106D 6A/400V   TO-220
TIC106M 5A/600V   TO-220
TIC116M 8A/600V   TO-220
TIC126M 12A/600V   TO-220
TYN1012 12A/1000V   TO-220
P0102AB 0.8A/100V    
S4014MH 40A/600V    
X0402BE 4A/200V    
After having examined this diagram I think that you will include ⁄ understand the operation of the thyristor. A brief running of trigger by K1 leaves the conducting thyristor. Only a cut by K2 will leave it insulator. Let us see other interesting characteristics: Maximum tension A ⁄ K can reach values raised, of 100 with 1200V according to the models. It is thus a contactor high voltage. The current of trigger minimal "Ig" to start conduction A ⁄ K is about 10mA, sometimes 1mA for the significant models. This current enters by "G" and leaves by "K" towards the mass. Its duration does not have any importance.3) the response time is very short (a few nano seconds). The intensity of Iak conduction is also high, from 0.3 with 35A according to the modèles5) the thyristor cannot return in a blocked state (insulator) only if the busy intensity Iak falls to the lower part from a minimal value. This threshold says "current of stop" is about 2% of the maximum intensity of the model.
The thyristor is a semiconductor made up of four layers of silicon as shown in the figure Ci above, the electrodes has as an Anode, G for Trigger and K for Cathode are laid out around.


The diac is a bidirectional diode: it can be blocked or busy in the two directions, according to the direction of the AC current. Its crucial role is to be used for release of a triac.
The diac does not lead the current (except for a negligible leakage current) as long as its nominal voltage is not reached. This tension is, according to the model, towards 32 or 40 V. When this tension is reached, it occur a phenomenon of conduction in avalanche and the tension of threshold of the component falls in the neighbourhoods of 5 V (typical value). The current which crosses the diac is then sufficient to start a triac.


The diac blocks the tensions in the two directions, until its nominal voltage V (BO) is reached. The output voltage Vo falls then with a quite less value.
Use of the triac as a gradator
For a use of the triac as a gradator, one often resorts to a network RC, associated if necessary with a diac, bidirectional diode allowing to obtain a dephasing even more important. One varies the intensity in the load via the variable resistor.
Schematic diagram of a triac used as a gradator, associated here with a diac. The potentiometer of tuning makes it possible to vary the luminous intensity of the bulb (100 W maximum).
Two important points: in the case of an inductive load (engine), it is necessary to add a protective circuit of the triac, by connecting in parallel a resistance and a capacitor, and a circuit of suppression, comprising a coil accompanied or not by capacitors. Moreover, as soon as the power exceeds 100 W, the triac must be equipped with a squanderer (radiator).
Lastly, it never should be forgotten that the triac is directly connected to the sector and that it is consequently advisable to take in this respect all the useful precautions, to start with a perfect insulation of the assembly.

Use of the triac out of switch

For a use of the triac out of switch, it is preferable to call upon a component specialised, the optoone, still called photo-coupler (reference MOC 3041, for example), which is designed for this application and which has moreover two appreciable advantages: an insulation of 7500 V and a current of command of about 10 my only. The implementation of this device is simpler, since it is enough to apply a positive level low tension to the LED the optoone, which orders from its turn the triac. The opto-triac fact thus figure of interface, to some extent, between the control circuit and the ordered circuit.
Order triac by opto-triac for a use out of switch.

execution time customer :
runtime server : 0.133 seconds