KR830001097B1 - Control circuit for gate diode switch - Google Patents

Abstract

No content.

Description

Control circuit for gate diode switch

1 is a cross-sectional view of a gate diode switch.

2 shows a control circuit and a switch according to the invention.

3 shows a control circuit and a switch element of an embodiment according to the present invention.

4 shows an interactive switch that can be controlled as the control circuit of FIG.

5 shows a control circuit and a switch of another embodiment of the present invention.

6 shows a control circuit and a switch of another embodiment of the present invention.

The present invention relates to a control circuit for controlling a gate diode switch element. D. this. In Houston's paper (IEEE Transactions on Election Devies, Vol. 23, No. 8, 1976), a double carrier injection (pinched off) to provide a blocking state and also to provide a continuity state ( A semiconductor high pressure switch is described that includes a region capable of exhibiting high electrical conduction by injection of electrons and holes. Such a device, which may be preferred as a gate diode switch (G.D.S), is promising as a semiconductor device that can be replaced by an electromechanical switch because it can control a high voltage. As will be described in detail later, these individual component elements can be directly integrated into the integrated circuit fabrication technology and can also be in bidirectional switching configurations.

It is desirable to use semiconductor integrated circuit technology to fabricate control circuits for GDS. The control circuit for applying the blocking voltage to the gate (or grid) must be able to apply a voltage higher than the voltage at the anode and cathode to the gate, and must be able to supply at least a magnitude of current flowing through the switch element. Therefore, the problem of using semiconductor integrated circuit technology is somewhat difficult. Since the gate diode switch GDS of the type mentioned above is a relatively new device, there is little data on the control circuit for controlling this device.

The semiconductor control circuit for operating the GDS is preferably manufactured on the same semiconductor substrate together with the switch element to be controlled.

The circuit is characterized in that the cathode comprises a second gate diode switch (GDS2) connected to the gate of GDS 1 and a voltage controlled branch circuit coupled to GDS 2 for controlling electrical conduction between the anode and the cathode. This is one solution for controlling the state of the first gate diode switch GDS 1 according to the above.

The state of GDS 2 is effectively controlled by a voltage controlled branch circuit that selectively regulates the gate-to-node voltage. A relatively low voltage pulse triggers the voltage control circuit. The voltage control circuit has the ability to supply a high voltage and a small current. Therefore, any steady-state current flowing through GDS 2 should be at least a reasonable value for the voltage control circuit to switch GDS 2 from conduction to disconnection. Since GDS 2 is in a blocked state, the gate potential of GDS 1 is at a level not higher than that of the anode and cathode, so that GDS 1 is in a conductive state and electrical conduction occurs between the anodes. In order to switch the GDS 1 to the blocking state, it becomes the potential of the gate, and an amount of current flowing between the anode and the cathode flows to the gate. The anode of GDS 2 is linked to a precursor that is set higher than the potential at the anode of GDS 1.

When GDS 2 is in a conductive state, the potential at the gate of GDS 1 (as well as the cathode of GDS 2) is higher than the potential at the anode of GDS 1, and GDS 2 supplies a sufficient positive current so that GDS 1 is in a blocked state. maintain. Several other embodiments will be mentioned later.

The present invention will be described in detail with reference to the drawings.

1 shows a preferred form of a GDS structure 10 comprising a support member 12, which is insulated from the support member 12 by an insulating layer 14 and p−. It has a single crystal semiconductor body 16 having a conductivity type and a major surface 11.

An anode region 18 having a p + conductivity type is included in the body 16, which region 18 is formed to extend to the surface 11. Gate region 20 and cathode region 24 having an n + conductivity type are also included in body 16. The region 22 having a p conductivity type and having a portion extending to the surface 11 surrounds the cathode 24 and serves to suppress punch-through, and also between the region and the region 20. The surface 11 acts to prevent the reversal of the body portion 16 near. The gate region 20 is located between the anode region 18 and the region 22 and is separated from the two regions by the body portion 16. The resistivity of the regions 18, 22 is lower than the resistivity of the bodies 18, 20, 24. The resistivity of the region 16 is about the middle of the resistivity values of the cathode region 22 and the body 24.

The electrodes 28, 30 and 32 are connected with low resistance to the surface portions of the regions 18, 20 and 24 respectively. The insulating layer 26 is covered by the main surface 11 to insulate the electrodes 28, 30, 32 from all regions except those that are electrically connected. The electrode 36 is connected to the support member 12 with low resistance through a heavily doped region 34 having the same electrical conductivity as the support member 12.

Advantageously, the support member 12 and the body 16 are each formed of silicon, and the support member 12 may have any conductivity type of n or p conductivity type. Each of the electrodes 28, 30, 32 can advantageously be connected to the semiconductor region with a low resistance. The electrode 32 is also covered by the area 22. This covering of the electrode in the semiconductor region is known as field plating, which facilitates high voltage operation by increasing the voltage causing breakdown.

A plurality of separate bodies 16 may be formed on one support member 12 to constitute a plurality of switch elements. The structure 10 may be formed between the anode region 18 and the cathode region 24 when in a conductive state. It acts as a switch with the characteristic that it becomes a low resistance passage and is in a high resistance state between the two regions 18 and 24 when it is in the blocking state. The potential applied to the gate region 20 determines the state of the switch. If the potential of the gate region 20 is lower than the declination of the anode region 18 and the cathode region 18, electrical conduction occurs between the anode region 18 and the cathode region 24. In the conduction state, holes are injected into the body 16 from the anode region 24 and electrons are injected into the body 16 from the cathode region 24. The number of holes and electrons injected into the body is a number sufficient to form a plasma (state in which the number of holes and electrons in the body is the same) which makes the body 16 a conductive state. As such, the resistance of the body 16 is effectively lowered so that when the structure 10 is operated in a conductive state, the resistance between the anode region 18 and the cathode region 24 is relatively low. This behavioral form is specified as a double carrier (hole, electron) injection.

FIG. 2 shows a control circuit (shown within a 210 large square dotted line) connected to a gate diode switch GDS 1 of the type shown in FIG. 1 having an anode, a cathode and a gate terminal. GDS 1 is shown as a circuit symbol representing a gate diode switch.

Control circuit 210 includes anode, cathode and gate terminals, first and second current limiters CL1 and CL 2, transistor Q1, diode D1, D2, D3, resistors R1, R2, R3, capacitor C1 and FIG. Gate diode switch GDS 2 in the form shown in FIG. It is coupled to both the anode of D1 and D3 and the first terminal of CL1, terminal 212. The collector of Q1 is connected to the cathode and terminal 211 of D3 and the cathode of D1 is connected to the gate and terminal 220 of GDS 2. The base of Q1 is connected to the input terminal 216 through the diode D2. The emitter of Q1 is coupled to one terminal of R1, that is, terminal 217, and the second terminal of R1 is connected to terminal 218 and power supply VSS. The second terminal of CL1 is connected to the power supply + V1 or terminal 214. CL2 is connected to the cathode of GDS 2 and the gate and terminal 222 of GDS 1. C0 2 is connected as a second terminal to power source -V3 and terminal 228. The third current limiter CL 3 is coupled to terminal 220 as a first terminal and to the power supply -V4 or terminal 226 as the second terminal. CL3 and -V4 are both incidental. The voltage -V4 may be equal to VSS or -V3.

The anode of GDS 2 is connected to one terminal of R3, that is, terminal 221. The second terminal of R3 is connected to the first terminal and the terminal 223 of R2 and the first terminal of C1. The second terminal of R2 is connected to the power supply + V2 or terminal 224. The second terminal of C1, however, is connected to the ruler. + V1 is chosen to be more positive than + V2. The combination of D1, D2, D3, Q, CL1, R1, CL3 (shown in dashed rectangle A) is provided as a voltage controlled branch circuit, and the potential of the terminal 220 (gate terminal of GDS2) is controlled to control the state of GDS2. Set it. C1 and R3 are incidental. Without C1 and R3, the terminal 221 and the terminal 223 may be directly connected.

C1 serves as a limited source of charge used to shut off GDS1. Except for C1, it is necessary to flow a large steady-state current through GDS2 to supply sufficient current to the gate of GDS1 in order to switch the GDS1 in the conduction state to the blocking state.

The default behavior is as follows:

When +220 volts and -220 volts voltage are applied to the anode and cathode terminals of GDS1, respectively, the electrical conduction occurs between the anode and the cathode if the potential of the anile gate (terminal 222) is slightly lower than +220 volts. Electrical conduction is interrupted by raising the potential of the gate (terminal 222) above +220 volts and supplying a positive current flowing through the gate of the GDS1 (terminal 222). Current limiter CL1 limits current to + V1 = + 280 volts, VSS = 0 volts, + V = + 250 volts, -V3 = -250 volts, -V4 = -250 volts and 50, 5 and 5 micro amps, respectively , CL2 and CL3, the circuit 210 serves to flow a current to the terminal 222 necessary for controlling the potential of the terminal 222 and the state of the GDS1. The type of current limiter is shown in the Source book of Electronic, Circuits, John Markus, Mcgraw Hill Book CD, 1968, page 171.

Consider one case where electrical conduction is possible through GDS1. When an input signal of 0 to 0.4 volt potential level is applied to terminal 216, at this level Q1 remains blocked and approximately at terminal 212. A potential of + V1 (approximately +280 volts) is shown. Without CL3, D1 remains conductive in the forward direction until the potential of terminal 220 is higher than the potential of terminal 212 and is cut off when the potential of terminal 220 is higher than the potential of terminal 212. CL1 and CL3 are set such that the voltage appearing at terminal 220 when the Q1 is in the cutoff state is at a level higher than the voltage of + V2. In this case, a voltage of approximately +280 volts appears at terminal 220. The +280 volts voltage of the terminal 220 biases the GDS2 into the cutoff state and causes the potential + V2 to be not applied to the terminal 222. The voltage drop occurs at terminal 222 due to negative potential -V3 (-250 volts) until the gate-to-node junction of GDS1 is forward biased. The potential of the terminal 222 is not higher than the anode potential of GDS1, but stabilizes as about the anode potential of GDS1. Thus, GDS1 is biased into a conductive state and electrical conduction occurs between the anode and the cathode. The slight current flowing from the anode of GDS1 to the gate is limited by CL2.

If GDS2 is in a conductive state before a 0 to 0.4 volt input voltage is applied to terminal 216, then a positive current flows from + V1 through D1 to the gate of GDS2. CL1 is set to allow a large current to flow through CL1 rather than through CL2 to flow enough positive current to the gate of GDS2 to block the conduction state between the anode and cathode. The electrical current through the GDS2 is only 5 bike amps, so only a relatively small positive current flows into the gate of the GDS2 to cut off the conduction state. Therefore, there is no need to use a high current device to flow the current required to turn off the GDS2.

If the potential of terminal 216 rises to a 2-5 volt level that causes GDS1 to shut off, this input voltage level biases Q1 to conduction and causes Q1 to operate in a saturation region. The potential of terminal 212 drops to about +1.6 volts. (When the input voltage of the terminal 216 is 2 volts, the voltage of the collector reserve intervener VCE is 0.3 volts and the voltage drop by D3 is 0.7 volts in saturation with respect to Q1). The potential at terminal 212 is then a function of the input voltage level and the forward voltage drop by VCE and D3 of Q1. If CL3 is not present, the potential of terminal 220 is at + V2 or more negative because of the loss through D1. Since the junction diode comprising the anode and gate of GDS2 is forward biased and raises the potential of terminal 220, the potential of terminal 220 does not drop below one diode voltage drop at the anode potential of GDS2. . In the case where CL3 is present, the potential of the terminal 220 drops sharply to the potential of the anode potential of GDS2 minus one diode voltage drop. In both cases, GDS2 is switched to a conductive state. In this case, the potential of the terminal 222 becomes the potential of + V2 minus the voltage drop of R3 and R2 and the forward voltage drop of the anode-cathode of GDS2.

The voltage drop across R2, R3, and GDS2 is selected so that the potential of terminal 222 is higher than the anode potential of GDS1 as an amount sufficient to bring GDS1 into the cutoff state. In addition, sufficient positive current flows through the gate of GDS1 to put the GDS1 into a blocking state. Once GDS1 is blocked, the current flowing to the gate is interrupted. The geometry and impurity concentration of GDS1 accurately determines how high a positive potential must be supplied to the gate to the anode and cathode to enable GDS1 to shut off. .

The minority carriers (electrons), which are emitted from the GDS1's cathode and collected as gates, constitute a current equal to the positive current flowing through R2, R3, GDS2 to the gate from + V ₂ . This current flow can be practical and as a result it is necessary to have a high voltage and high current device such as GDS2 to shut off GDS1. The use of high voltage and high current transistors in such control circuits is much more costly.

R2 and R3 limit the current flowing through GDS2 to the gate of GDS1 from + V2, and R3 also limits the current flow from C1.

GDS1, which is used as a telephone switching element, is normally only 48 volts between the anode and cathode when in shutdown, but ± 220 volts is present between the anode and the cathode due to ringing, testing, public telephone control, induced commercial AC voltage, etc. The control circuit 10 is designed with such a high voltage element so as to be operable even in the case.

When Q1 operates in the saturation region, the coupling between the base collectors is forward biased. D3 is provided to prevent current flow from the input terminal 216 through the collector base junction of Q1.

The circuit of FIG. 2 comprising CL3, R2, R3, C ₁ can be fabricated on one integrated circuit chip with GDS1 and GDS2 in the form shown in FIG. The fabricated control circuit can block the 500 volts voltage across the anode and cathode of GDS1, and can also block the 100 milliampere current flowing through it.

If R3 is directly connected to + V2 and C1 and R2 are absent, the values of R1 and R3 are suitable for 100 and 3000 ohms, respectively.

When C1 and R2 are used, C1 and R2 reduce the time required to change the switch state of GDS1 from conduction to disconnection. Preferred values of C1 are 0.1 micro ferrets when R1 = 1000 ohms, R2 = 2 × 10 ⁵ ohms, and R3 = 3000 ohms.

3 shows a control circuit 310 connected to a gate diode switch GDS31 having an anode and a cathode gate terminal. The control circuit 310 is similar to the control circuit 210 of FIG. 2 except that diodes D1 and D3 are removed and use a current reflector circuit comprising Q2 and Q3. Q2 and Q3 are switching elements whose base is specified as the control terminal and the collector and the data are specified as the first and second output terminals, respectively.

The actors of Q2 and Q3 are connected together to terminal 314 and power supply + V31. The bases of Q2 and Q3 are connected together to the collector of Q2 and the first terminal and terminal 330 of CL31. The collector of Q3 is connected to the gate of GDS31 and the first terminal of CL33 and the circuit terminal 320. In essence all other components and connections are similar to those in the circuit of FIG.

The combination of D32, Q31, R31, Q2, CL31, and CL33 (indicated by the rectangular dotted line B) is specified as a voltage control branch circuit, and serves to set the potential of the terminal 320 to control the state of the GOS32.

As a suitable high level voltage (usually + 2-5 volts) applied to terminal 316, Q31 is biased, causing electrical conduction from the power supply to power supply VSSO via Q2, CL31, Q31, R31 + V31. Q2 and Q3 are virtually identical transistors, and the same current flows through these two transistors. When Q31 is in the operating state, the potential of the terminal 320 is obtained by subtracting the VCE (collector-emitter voltage) of Q3 from the potential of + V31. When a low level input signal (0 to 0.4 volts) is applied to the terminal 316, Q31 is cut off and no electric conduction through Q31 and Q2 occurs. The potential of the terminal 320 becomes approximately -V34 until the junction between the gates of the anodes of the GDS32 is biased in the forward direction, and the potential is somewhat lower than the potential of + V32. + V32 is set to a higher potential than + V32, and the potential of -V34 is set negatively higher than the potential of + V32. The operation of GDS32 to control the state of GDS31 is substantially the same as that mentioned for the operation of GDS2 in FIG. Corresponding to the power supply of FIG. 2, it becomes easy to control the state of GDS31 in which ± 220 volts is applied between the cathode and the anode when the same potential is used for the power supply of FIG.

Changing the potential of terminal 320 operates GDS32 in a similar manner corresponding to GDS2 in FIG. As such, the state of the GDS31 is controlled in the same manner as the state of the corresponding GDS1 in FIG. 2 as an input signal of opposite polarity. Complementary transistors Q31 and Q2 or Q3 can be formed on the same integrated circuit chip with GOD32.

4 shows a bidirectional switch comprising gate diode switch elements GDS3 and GDS4 in such a way that the anode of GDS3 is connected to the cathode of GDS4, the cathode of GDS3 is connected to the anode of GDS4 and the two gates are connected together. It is. One advantage of the gate diode switch of FIG. 1 is that the two switch elements can be bidirectionally connected to withstand high voltages without causing an avalanche breakdown. The gates of GOS3 and GDS4 can be connected to the terminal 222 of the control circuit 210 of FIG. 2 and the terminal 322 of FIG. 3 to control this device in the same manner as described above. That is, the states of GDS3 and GDS4 can be controlled in the same manner as the states of GDS1 in FIG. 2 and GDS31 in FIG.

Various modifications may be made without departing from the spirit of the invention. For example, various other control circuits may be substituted for the gates of GDS2 and GDS32 in FIGS. 2 and 3 to provide the voltage level and current capacity needed to control the state. The npn transistor can also be replaced by a pnp transistor when the polarity of the power supply is changed. It is also possible to remove R1 and R31, and the emitters of Q1 and Q31 can be connected directly to VSS and VSSD respectively. In this case, it is preferable to insert a current limiting means which is usually a resistor in series with the input terminals 216 and 316, respectively.

5 shows another embodiment of the present invention as a control circuit 510 connected to the gate terminal 528 of the gate diode switch GDS 51. The control circuit 510 controls the state of the GDS51 to include transistors Q51 and Q52, diodes D51 and D52, gate diode switches R51 R52, current limiters CL51 and CL52, and resistors R15 and R52.

Components within dashed rectangle 5A control the potential between the anode cathodes of the GDS52. R52 is ancillary and can be eliminated.

When voltages of +220 volts and -220 volts are applied to the anode and cathode of the GDS51, respectively, electrical conduction occurs between the anode and the cathode when the gate potential of the GDS51 (terminal 528) is slightly lower than +220 volts. Electrical conduction is interrupted by increasing the potential of the gate (terminal 528) above +220 volts and by flowing current through the gate of the GDS51 (terminal 528). + V51 = 250 volts, VSS = 0 volts, V52 = -250 volts, and current limiters CL51 and CL52 limit current to 50 and 5 microamps, respectively, while circuit 510 controls the state of the GDS51. Supply the required potential and current.

In the case of generating electrical conduction through the GDS51, an input signal of 0 to 0,4 volts is applied to the input terminal 516. As the low level voltage, Q51 is in a cutoff state, and the potential of the terminal 581 is approximately + V51. In this state, Q52 is cut off, which in effect results in turning off the circuit between +51 and terminal 526 (the anode of GDS52). As such, the GDS52 is shut off because no current flows between Ed and the cathode. Terminal 528 is isolated from + V51 when GDS52 is in a blocked state and remains at negative potential of -V52 (-250 volts) until the gate-to-node junction potential of GDS51 is forward biased.

When the potential of the terminal 528 rises to a somewhat lower potential near the anode potential of the GDS51, the GDS51 becomes conductive and a current flows between the anode and the cathode.

The current flowing from the anode of GDS51 to the gate is limited as CL52.

As is apparent if a pulse of 3 to 5 volts is applied to terminal 516, the GDS51 is in a shut down state. Q51 operates in the saturation region while conducting. This allows the junction between the emitter bases of D51 and Q52 to be forward biased. This allows Q52 to conduct, enabling electrical conduction from + V51 to -V52 via the emitter-collector of Q52, the anode cathode of GDS52 and CL52. With Q52 conducting, the collector emitter intervoltage (VCE) of Q52 is set to a magnitude lower than the forward voltage drop caused by D52. As such, since the potential of the anode (terminal 526) is higher than that of the gate, the GDS52 remains in a conductive state.

The GDS52 is in a conductive state, and the potential of the terminal 528 is about + V51. This potential level is higher than the potential level at the anode of the GDS51 so that the GDS51 is in a blocking state. The geometry and impurity concentration of the GDS51 accurately determines how much higher potential must be applied to the gate than the anode in order for the GDS51 to be blocked.

In order to switch the GDS51 to the shut-off state, not only should the potential level be applied to the gate of the GDS51, but also a current comparable to the magnitude of the current flowing between the anode and the cathode of the GDS51 must be applied to the gate of the GDS51. Most of the current flowing to the gate of GDS51 is from + V51 through the gate-cathode of diodes D52 and GDS52. The remaining current flows from + V51 through the collector-emitter of Q52 and the anode-cathode of GDS52. This current can be real, and as a result it is necessary to use a high voltage and high current device such as GDS52 to switch the GDS51 to a shut-off state.

The current gain of Q52 serves to limit the flow of current from GDS52 to the gate of GDS51. This will prevent the GDS51 and GDS52 from burning out. In many telephone swarming applications, the GDS51 operates with only 48 volts between the anode and cathode when in a disconnected state, but can be operated even if +220 volts is applied between the anode and cathode due to ringing, induction of utility voltage, etc. 510 is designed to perform the high pressure blocking effect.

6 shows a control circuit 610 connected to the gate magnetic terminal of the gate diode switch GDS61. The control circuit 610 is similar to the control circuit 510 of FIG. 5 except that npn transistors Q63 and Q64 and diodes D63 and D64 are added.

Q63 and Q64 have a Darlington configuration in which collectors are connected together and connected to terminal 620, and emitters of Q63 are connected to base and terminal 634 of Q64. The collector of Q62 is connected to the base and terminal 632 of Q63. The emitter of Q62 is connected to the anode and terminal 620 of GDS62. The anode of D62 is connected to terminal 620 and the casings of D64 are connected to terminal 624, while D62, D63 and D64 are connected together in series between terminal 620 and terminal 624. Q61, CL61, D61, Q62, Q63, Q64, D62, D63, D64, R61 and R62 are provided as control branch circuits (shown in dashed rectangle 6A) for controlling the anode potential of GDS62 with respect to the cathode. R62 is ancillary and can be eliminated.

Current semiconductor technology has some difficulties in obtaining a pnp transistor with a high current gain. The combination of Q62, Q63, and Q64 acts as an equivalent to a pnp transistor with virtually high current gain. As such, Q62, Q63, and Q64 perform the same function as Q52 of FIG. D63 and D64 are needed to compensate for the increase in emitter bay voltage drop by Q63 and Q64. Q62, Q63, and Q64 are in a conductive state, and the voltage at the get (terminal 624) of GDS62 is slightly lower than the voltage at the anode of GDS62 (terminal 626). In this case, the GDS62 is in a conductive state.

The circuit of FIG. 6 including R62 blocks a 500 volt voltage between the anode and cathode of the GDS61 and blocks the 100 milliampere current flowing through the GDS61.

The embodiments mentioned herein have been described as general principles of the invention. Various modifications may be made without departing from the spirit of the invention. For example, other switching devices, such as MOS transistors, can be replaced by bipolar transistors while giving the appropriate voltage magnitude and polarity.

Claims (1)

A first region 18 having a relatively low resistivity and a first conductivity type, a second region 24 having a second conductivity type which is the opposite conductivity type of the first conductivity type, and a gate region 20 having the second conductivity type. The semiconductor body 16 includes the first and second regions connected to the output terminal of the switching element, and the first and second and gate regions are separated from each other by a part of the body 16, and the gate region. A depletion layer region as a first voltage applied to the semiconductor body is formed to prevent current flow between the first and second regions, and a relatively low resistance between the first and second regions as a second voltage applied to the gate region. In the control circuit for controlling the first switching device GDS1 in which the current path is formed, the output terminal of the second switching device GDS2 having the same type as the first switching device GDS1 has a gate of the first switching device GDS1. Connected to the first and A control circuit, characterized in that the voltage control branch circuit (A) is connected to the second switching element (GDS2) to control the conduction between the second regions.

Images