PCT Patent Application for Improved Transient Blocking Unit

FIELD OF THE INVENTION

This invention relates to use of a transient blocking unit (TBU) to protect an electrical load from over-voltage and/or over-current conditions.

BACKGROUND

Many circuits, networks, electrical devices and data handling systems are operated in configurations and environments where external factors can impair their performance, cause failure or even result in permanent damage. Among the most common of these factors are over-voltage and over-current. Protection against these factors is important and has been addressed in the prior art in various ways.

Fuses that employ thermal or magnetic elements are one common protection measure. In other cases, protection circuits are available. Some examples are described in U.S. Pat. Nos . 5,130,262; 5,625,519; 6,157,529; 6,828,842 and 6,898,060. Protection circuits are further specialized depending on conditions and application. For example, in the case of protecting batteries or rechargeable elements from overcharging and over-discharging one can refer to circuit solutions described in U.S. Pat. Nos. 5,789,900; 6,313,610; 6,331,763; 6,518,731; 6,914,416; 6,948,078; 6,958,591 and U.S. Published Application 2001/00210192. Still other protection circuits, e.g., ones associated with power converters for IC circuits and devices that need to control device parameters and electric

parameters simultaneously also use these elements. Examples can be found in U.S. Pat. Nos. 5,929,665; 6,768,623; 6,855,988; 6,861,828.

When providing protection for very sensitive circuits, such as those encountered in telecommunications the performance parameters of the fuses and protection circuits are frequently insufficient. A prior art solution embodied by transient blocking units (TBUs) that satisfy a number of the constraints is considered in international publications PCT/AU94/00358; PCT/AU04/00117; PCT/AU03/00175; PCT/AU03/00848 as well as in

U.S. Pat. Nos. 4,533,970; 5,742,463 and related literature cited in these references.

In a TBU, two or more transistors are arranged such that they normally provide a low series resistance. However, when an over-voltage or over-current transient is applied to the TBU, the transistors switch to a high impedance current blocking state, thereby protecting a load connected in series to the TBU. Variations and/or refinements of the basic TBU concept are considered in US Pat. Nos. 3,916,220, 5,319,515, 5,625,519, 5,696,659, 5,729,418, 6,002,566, 6,118,641, 6,714,393, 6,865,063, and 6,970,337.

A conventional TBU provides combined current limiting and current disconnect performance, as shown on Fig. 1. A TBU having zero applied voltage is in a low impedance state, where the current through the TBU rises rapidly as the voltage across the TBU increases. The current through the TBU is limited to be no greater than a trigger current I _t , so once this current level is reached, the TBU current does not change as the TBU voltage further increases. When the TBU voltage exceeds a disconnect voltage Vd, the TBU switches to a high impedance state, effectively isolating downstream electrical devices and circuits from the transient. Accordingly, conventional TBUs are designed

to withstand a maximum power dissipation P _max that is at least ItVd.

This constraint on TBU power dissipation can cause problems in practice. For example, powered span telecommunication systems typically have operating voltages of 50 to 110 VDC (the voltage can be as high as 180 VDC) , in combination with currents much less than 200 mA. Protecting such a system with a 200 mA TBU would be desirable, but difficulties can occur when power is applied to the span (e.g., at start up) or when a TBU is inserted following a "break then make" protocol. To accommodate the line-charging transient by limiting the current to 200 mA without disconnecting, a conventional TBU would require a power handling capacity of at least 20-40 W (since V _d would need to be on the order of 110 to 180 V) . Providing such high power handling capacity is costly, and it is also highly inefficient, since TBU power dissipation in normal line operation is far less than 20-40 W in this example.

One approach for alleviating this problem is to provide a low power TBU (e.g., having V _d on the order of 5 V for a 1 W, 200 mA TBU) , and to protect this TBU from normal transients associated with powering up the span. However, protecting the TBU from normal span transients undesirably adds complexity to the system. Accordingly, it would be an advance in the art to provide a TBU that more efficiently accommodates normal span transients without going into a full disconnect mode.

SUMMARY

As indicated above, a conventional TBU limits the current to a trigger current It, and disconnects when the voltage exceeds the disconnect voltage Vd. In contrast, TBUs according to the present invention have a disconnect condition that is related to the TBU temperature (e.g., a TBU die temperature).

In a first embodiment of the invention, a maximum TBU power P _max is derived from a maximum TBU temperature T _raax , such that if the TBU power dissipation does not exceed P _raax , then the TBU temperature does not exceed T _max , and that thermal damage to the TBU will not occur for TBU temperatures less than T _max . Thus the temperature can be held to values less than Tmax by requiring the power dissipation to be less than P _raax . In operation, the impedance of the TBU increases in response to increasing applied voltage such that the TBU power does not exceed P _raax . For example, Fig. 2 shows typical behavior for such a TBU, where the TBU partially disconnects (i.e., the TBU impedance increases, but not to its maximum level) as voltage increases in such a way as to approximately follow a curve of constant TBU power dissipation. In a second embodiment of the invention, a temperature sensor responsive to the TBU temperature is included in the TBU. In operation of this TBU, the TBU impedance increases in response to increasing applied voltage such that the sensed TBU temperature does not exceed a maximum TBU temperature T _max . For example, Fig. 3 shows typical behavior for such a TBU, where the TBU partially disconnects as voltage increases in such a way as to approximately follow a curve of constant TBU temperature.

An advantage of this second embodiment is that direct temperature monitoring automatically accounts for possible TBU heat sink variability (either from device to device, or over time) . In contrast, P _raax for TBUs of the first embodiment will depend on the level of heat sinking provided to the TBU (e.g., improving the heat sinking of a TBU will increase P _max for a fixed T _raax ) . Thus, in the first embodiment, P _max is determined by T _max and by the TBU heat sinking performance. For example, if a simple thermal resistance model is applicable, then P _max is on the order of R _t h (T _max -T ₀ ) , where R _t h is the thermal resistance provided to the TBU by the heat sink and To is room temperature.

BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1 shows a relation between I _t , V _d and P _raax for a TBU.

Fig. 2 shows partial disconnection of a TBU according to a first embodiment of the invention.

Fig. 3 shows partial disconnection of a TBU according to a second embodiment of the invention.

Fig. 4 is a schematic diagram of a conventional unipolar TBU. Fig. 5 is a schematic diagram of a first example of the invention.

Fig. 6 is a schematic diagram of a second example of the invention.

Fig. 7 is a schematic diagram of a third example of the invention.

Fig. 8 is a schematic diagram of a fourth example of the invention.

Fig. 9 is a schematic diagram of a fifth example of the invention . Fig. 10 is a schematic diagram of a sixth example of the invention .

Fig. 11 is a schematic diagram of a seventh example of the invention.

DETAILED DESCRIPTION

Conventional TBU operation is best appreciated by beginning with the unipolar example of Fig. 4. The circuit of Fig. 4 has a depletion mode n-channel NMOS transistor 402 (Ql) and a depletion mode p-channel JFET 404 (Q2) . The source of Ql is

FLT-103/PCT 5

connected to the source of Q2, the gate of Ql is connected to the drain of Q2, and the drain of Ql is connected to the gate of Q2. The TBU input is the drain of Ql and the TBU output is the drain of Q2. As I _TBU flows through Ql and Q2, corresponding source-drain voltage drops Vl and V2 are generated. The gate to source voltage for Q2 is Vl and the gate to source voltage for Ql is V2. As the gate to source voltages for Ql and Q2 increase, Vl and V2 also tend to increase (since Ql and Q2 are depletion mode devices) , and this self-reinforcing feedback drives the TBU to a high impedance state when V _TB u exceeds the disconnect voltage V _d , thereby disconnecting the TBU. Once disconnected, a small leakage current (which is typically negligible) continues to flow through the TBU.

Fig. 5 shows a first example of the invention. Additional depletion mode p-channel JFETs 502 (Q2) , 504 (Q3) , and 506 (Q4) are connected in parallel with JFET 404 (Ql) . In this circuit, each of the p-channel JFETs Q1-Q4 has a different pinch-off voltage (V _p ) and a different series resistance R _0n . More specifically, V _pl < V _p2 < V _p3 < V _p4 and R _on i < R _O n2 < R _O n3 < Ron4 • Furthermore, V _pn and the on-resistance of NMOS FET 402 are less than the corresponding parameters of Ql .

Approximately, the operation of the circuit of Fig. 5 is as follows. For V _TB U < V _p i, transistors Q1-Q4 are all conducting, and a first trigger current I _t i = V _pn / (R _on i I I R _O n2 I I Ron3 I I R _O n4) • For Vp ₁ < V _TBU < V _p2 , transistor Ql is switched off, and the current decreases to a second trigger current I _t2 = V _pn / (R _on2 I I R _O n3 I I Ron4) • Similarly, for V _p2 < V _T BU < V _P 3, transistors Ql and Q2 are both switched off, and the current further decreases to a third trigger current I _t 3 = V _pn / (R _on3 | | R _on4 ) . For V _p3 < V _TB u < V _p4 , transistors Q1-Q3 are switched off, and the current further decreases to a fourth trigger current I _t4 = V _pn /R _on 4. Finally, for VTBU > V _p4 , transistors Q1-Q4 are all switched off, and the TBU is in full disconnect mode, where only a leakage current flows. By appropriately selecting the pinchoff voltages V _p i-V _p4 (e.g., such FLT-103/PCT 6

that VpiIti ⁼ Vp2l _t 2 = V _p3 It3 = Vp/jlfj = Pmax) , an approximation to a curve of constant TBU power dissipation can be provided, e.g. as shown on Fig. 2. Although four stages are employed in this example, any number of stages can be employed in practicing the invention.

In most cases, it is preferred for the TBU to be implemented as a single integrated circuit. Such implementation of the circuit of Fig. 5 requires fabrication techniques that can provide p-channel devices having different pinch-off voltages on the same die. One approach is to vary the gate width of the p-channel devices, in order to vary the effective depth of the n+ gate region of the p-channel JFETs. Increasing gate width decreases pinch-off voltage and decreasing gate width increases pinch-off voltage, other parameters being equal. Another approach is to use different n+ gate diffusions to provide the various JFET pinch-off voltages.

Fig. 6 shows a second example of the invention. The circuit of Fig. 6 is like the circuit of Fig. 5, except that Zener or avalanche diodes 602, 604, 606, and 608 are placed in series with the gates of transistors Q1-Q4. In this example, transistors Q1-Q4 can have identical pinch-off voltages, and diodes 602-608 can be employed to change each transistors effective pinch-off voltage to a distinct value, thereby allowing the circuit of Fig. 6 to operate as described above in connection with Fig. 5. This embodiment allows the use of a simple process flow providing nominally identical p-channel JFET transistors. In the circuit of Fig. 6, it may be useful to connect gate to source in each JFET with a resistor (not shown) , typically having a resistance greater than about 100 kω, in order to provide diode bias current and prevent charge trapping on the gate. Providing a diode bias current can be helpful for controlling the effective pinch-off voltage more reliably.

Fig. 7 shows a third example of the invention. The circuit of Fig. 7 is like the circuit of Fig. 6, except that resistor 706 and transistors 702 and 704 are added. The circuit inside the dotted line on Fig. 7 is the circuit of Fig. 6 and can be regarded as a TBU "core". In some cases, if the TBU core is fully disconnected, the voltage across the TBU core can increase to a level where damage to the TBU core can occur. The circuit of Fig. 7 addresses this issue, since transistors 702 and 704 can switch off in response to excessive TBU core voltage, thereby providing additional voltage handling capability.

In this example, transistor 702 is a p-channel depletion mode JFET having a high (effective) pinch-off voltage. Preferably, the pinch-off voltage of transistor 702 is selected to be the voltage at which minimal leakage current is desired. The high pinch-off voltage of transistor 702 can be provided by direct fabrication of a high V _p transistor, or by addition of a series diode to a low V _p transistor as described in connection with Fig. 6.

The circuit of Fig. 7 can be regarded as a TBU core within a TBU. In operation, if the TBU core is fully disconnected, but the voltage across the TBU core is not sufficient to switch off transistors 702 and 704, these transistors conduct. Current flow through transistor 702 provides a relatively large leakage current flow, even though the TBU core is fully disconnected. When the TBU core voltage increases to a second disconnect level, transistors 702 and 704 switch off, thereby driving the overall TBU into full disconnect and reducing leakage current flow through the TBU to a minimum. Since transistor 704 can be a high voltage transistor, the overall voltage required to turn off the circuit can be significantly above the voltage handling capability of the TBU core. In this manner, TBU circuits can be made to leak small amounts of current up to very high voltages (e.g., > 200 V) without exceeding the TBU power handling capability. This approach can be regarded as providing a final FLT-103/PCT 8

"step" on the I-V curve of Fig. 2 having low current and high voltage. The circuit of Fig. 7 can react very quickly to fast transients (because of the TBU core) , while also providing stable high voltage operation and progressive leakage reduction at high voltages .

Fig. 8 shows a fourth example of the invention. The circuit of Fig. 8 is like the circuit of Fig. 7, except that the gate diodes on JFET string 750 are omitted (which requires in this configuration that each of the JFETs be manufactured with distinct pinch-off voltages as described above) , as is the connection between the gates of JFETs 750 and the source of high voltage FET 704. Like the circuit of Fig. 7, this circuit also provides a high voltage TBU having a partial disconnect capability. At low voltage, the channel of FET 704 is conducting, so the source and drain of FET 704 are effectively at about the same voltage. Therefore, at low voltage the circuit operates as described in connection with Fig. 6, with TBU disconnection controlled by the combination of transistor 402 and JFET string 750. However, for sufficiently high voltages across the circuit, transistor 704 is biased off, thereby providing additional TBU voltage handling capability.

If resistor 706 is not present in the circuit of Fig. 8 (i.e. if it is replaced with a diode, or with a wire to provide a direct connection) , the maximum voltage the TBU can block is the gate-drain breakdown voltage of JFET string 750. If resistor 706 is present, then the maximum voltage the TBU can block is the smaller of Vl and V2. Here Vl is the breakdown voltage of high voltage FET 704, and V2 = I _av R, where R is the resistance of resistor 706 and I _av is the maximum gate to drain avalanche current of JFET string 750. Since Vl and V2 are typically both significantly larger than the gate-drain breakdown voltage of JFET string 750, TBU voltage handling capability is improved by resistor 706. Resistor 706 can also

be a current source. Further details of the high voltage approach of Fig. 8 are described in PCT application WO 069753.

The circuits of the preceding examples operate by defining several trigger currents I _t j and disconnect voltages V _d j, selected to ensure that a TBU power dissipation limit P _raax is not exceeded (i.e., ItjVdj < P _m ax for each j) . In turn, the power dissipation limit is set such that a TBU temperature maximum T _max is not exceeded, where T _max is selected to be low enough to prevent thermal damage of the TBU in operation. As indicated above, alternative embodiments of the invention employ a temperature sensor to directly control TBU disconnection such that T _raax is not exceeded.

Fig. 9 shows a fifth example of the invention. In this example, temperature is directly monitored. The effective pinch-off voltage of transistor 404 is increased by placing diode 804 in series with its gate. Transistor 802 is placed in parallel with diode 804, and is a normally off device. The circuit of Fig. 9 operates by comparing the voltage across diode string 806 (which is relatively temperature independent) with the voltage across Zener diode 810 (which is highly temperature dependent) . With appropriate trimming (i.e., by selecting the values of resistors 808 and 812), the circuit can be made to provide a gate drive to transistor 802 that allows increasing amounts of current to flow parallel to diode 804 as the TBU temperature increases, thereby progressively decreasing the effective pinch-off voltage of transistor 404. In turn, this progressively reduces the TBU trigger current, thereby partially disconnecting the TBU in such a manner as to keep the temperature below T _max . This approach can be regarded as providing a continuous partial TBU disconnection, as opposed to the stepwise disconnection described above. When an absolute disconnection voltage level is reached, the circuit goes into a full disconnect.

In order to obtain such progressive disconnection, diode 804 and shorting transistor 802 are connected to the gate of the TBU transistor having the higher pinch-off voltage. In this example, TBU transistor 404 is thus selected to have a higher pinch-off voltage than transistor 402. If the situation is reversed (i.e., if transistor 404 has a lower pinch-off voltage than transistor 402), the TBU will not act unless the maximum temperature is reached, at which point it will go into a full disconnect (i.e., no partial TBU disconnection occurs in this case) . In this case the circuitry effectively provides an electronic PTC (positive temperature coefficient) device for controlling TBU switching, and the resulting TBUs are applicable to high current/high power applications .

The diodes in diode string 806 preferably have a low temperature coefficient in order to provide a stable temperature reference. A preferred approach for providing these diodes is to employ Zener diodes having a breakdown voltage of about 5V, which inherently have low temperature sensitivity.

Fig. 10 shows a sixth example and a preferred embodiment of the invention. This example is a bidirectional TBU based on the circuit of Fig. 9. A second NMOS FET 902 is added to the circuit, as are commutating diodes 904, 906, 908, and 910. For positive current (i.e., flowing from left to right on Fig. 10), transistors 402 and 404 act as a TBU as described above, since diodes 904 and 910 conduct, and diodes 906 and 908 do not conduct. For negative current (i.e., flowing from right to left on Fig. 10) , transistors 902 and 404 act as a TBU as described above, since diodes 906 and 908 conduct, and diodes 904 and 910 do not conduct. Similarly, bidirectional TBUs corresponding to the unidirectional TBUs of Figs. 5-8 are also examples of the invention. Either or both of diodes 904 and 906 can be replaced by functionally equivalent resistors and/or current sources to provide bidirectional TBU operation.

Fig. 11 shows a seventh example of the invention. In this example, a positive temperature coefficient (PTC) device 1102 is disposed in series between the TBU transistors 402 and 404. As the TBU heats up, the resistance of the PTC device increases, thereby decreasing the trip current of the TBU. Preferably, the PTC device resistance increases dramatically at a predetermined threshold temperature (typical PTC threshold temperatures are from about 100 ⁰ C to about 140 ⁰ C) . Suitable PTC devices for practicing the invention are available commercially. The preceding description is by way of example as opposed to limitation. The invention can also be practiced by making various modifications to these examples. TBUs according to the invention can include any type or polarity of transistor. The pinch-off voltage in the above examples can be regarded more generally as a switching voltage, where input voltages above the switching voltage cause the device to turn off. More generally, the invention is also applicable to other voltage controlled switching elements suitable for making a TBU, such as voltage controlled relays and microelectromechanical (MEMS) switches. The invention is applicable to any kind of uni-directional TBU or bi-directional TBU. Current limiters can be used in place of any or all of the resistors in TBU circuits according to the invention .

The preceding examples consider cases where partial TBU disconnection is performed in discrete steps to approximate an I-V curve of constant power dissipation and where partial TBU disconnection is performed in a continuous manner to prevent a temperature limit from being exceeded. Principles of the invention should also be applicable to discrete partial TBU disconnection to prevent a temperature limit from being exceeded (e.g., to provide a response as shown on Fig. 3). Similarly, the above principles should also apply to continuous partial TBU disconnection to approximate an I-V curve of constant power dissipation .