The present invention relates to an integrated circuit arrangement for protecting a semiconductor device during reverse voltage operation.
In many applications semiconductor devices have to be equipped with a reverse voltage protection. In unprotected devices the supply voltage can be shorted by the substrate diode of the semiconductor chip. Considering a MOSFET as an example, a substrate diode can be formed by the pn-junction between a p-doped substrate and an n-doped drain zone located adjacent to the substrate in an epitaxial layer which has been deposited onto the substrate. In order to inhibit such short circuits, the substrate can be isolated during reverse voltage operation. As a consequence the potential of the substrate is floating and undefined, i.e. the potential of the substrate depends on the state of other components integrated in the same substrate. A floating substrate entails the risk of a latch-up of parasitic thyristor structures inherent in many integrated circuits.
There is a general need for an a integrated circuit arrangement which is capable of withstanding a certain reverse voltage without the risk of a high current flow and a latch-up due to parasitic semiconductor structures.
One embodiment of the invention relates to an integrated circuit arrangement comprising a semiconductor body having a substrate and at least one substrate terminal. The semiconductor substrate can be supplied with any external potential via the at least one substrate terminal. The integrated circuit arrangement further comprises at least one semiconductor component integrated in the semiconductor body. The semiconductor component may be a semiconductor switch and may be connected between a first and a second supply terminal providing a first and a second supply potential respectively. In order to provide for an adequate reverse voltage protection the integrated circuit arrangement additionally comprises switching means adapted for connecting the at least one substrate terminal to one of said first and second supply terminal dependent on which supply terminal provides the lower supply potential. The potential of the substrate therefore is at least almost equal to the lower supply potential.
The invention can be better understood with reference to the following drawings and description. The components in the figures are not necessarily to scale, instead emphasis being placed upon illustrating the principles of the invention. Moreover, in the figures, like reference numerals designate corresponding parts. In the drawings:
a shows a MOS-transistor as an exemplary semiconductor component integrated in a semiconductor body, wherein parasitic thyristor-structures are illustrated by means of circuit diagrams symbols.
b shows an equivalent circuit diagram for the circuit arrangement shown in
a shows—as an exemplary semiconductor component—a MOS-transistor which is integrated in a semiconductor body. Parasitic diode and thyristor structures which are formed between active areas of the semiconductor component and between active areas and a semiconductor substrate are illustrated by means of circuit diagram symbols. The example shows an n-channel D-MOS transistor (double diffused vertical MOS transistor) disposed above an p-doped substrate. Of course the invention is also applicable to p-channel MOSFETs in n-doped substrate material and to pure bipolar technologies. Dependent on the manufacturing technology n-doped wafers can be used, too.
The embodiment depicted in
A pn-junction between the body zone 22 and the drain zone 21 (or the drain contact zone 21a) forms a so-called body diode DBD. The pn-junction between the drain zone 21 and the substrate 1 forms a substrate diode DS. The body diode DBD and the substrate diode together can also represented by a pnp-transistor QP, wherein the body diode DBD represents the emitter-base diode and the substrate diode DS represents the collector-base diode of the pnp bipolar transistor QP. That is, the p-doped body zone 22, the n-doped drain zone 21 and the p-doped substrate 1 form a (vertical) pnp-transistor QP. The n-doped drain zone 21, the p-doped substrate contact zone 11 and the epitaxial layer 2 “outside” the substrate contact zone 11 forms a (lateral) npn-transistor QN. The substrate is connected to the collector of the pnp-transistor QP and to the base of the npn-transistor QN. Resisters RS indicate the non ideal conductivity of the substrate.
The electrical equivalent circuit of the semiconductor component described above is depicted in
During normal operation the first supply potential VDD is higher than the second supply potential VSS and the substrate is connected to the second supply potential VSS in order to inhibit the latch-up of the parasitic thyristor TPAR formed by the bipolar transistors QN an QP (cf.
In order to prevent high substrate current in case of a reverse voltage the substrate may be isolated from the second supply potential VSS. As a consequence the potential of the substrate 1 is floating and determined by the state of adjacent semiconductor components integrated in the same semiconductor body. Having a floating substrate 1 the parasitic thyristor TPAR connecting the drain of the high side switch M1 and the body of the low side switch M2 can not be neglected as already mentioned above.
There is a need for an “intelligent” circuit for, on the one hand side, preventing a latch up of the parasitic thyristor TPAR and, on the other hand side, preventing a short circuit via the substrate diodes DS in reverse voltage operation as well as in normal operation.
Switching of the substrate potential is effected by means of a transistor half-bridge comprising a first semiconductor switch M1A and a second semiconductor switch M0, wherein the first semiconductor switch M1A is connected between a first substrate terminal B1 and the first supply terminal (VDD) and wherein the second semiconductor switch M0 is connected between the first substrate terminal B1 and the second supply terminal (VSS). That is, the common node of the semiconductor switches M1A and M0 is connected to one substrate terminal (e.g. substrate terminal B1) and the switching state of the half-bridge determines the potential of the substrate 1. A first and a second comparator K1, K0 control the switching states of the first and the second semiconductor switch M1A, M0 of the half-bridge such, that the first semiconductor switch M1A in an on-state and the second semiconductor switch M0 is in an off-state if the first supply potential VDD is lower than the second supply potential VSS (i.e. reverse voltage operation), and vice versa. An output terminal of the first comparator K1 is therefore connected to a control terminal of the first semiconductor switch M1A, and an output of the second comparator K0 is connected to a control terminal of the second semiconductor switch M0. Both comparators K1, K0 have an inverting and a non-inverting input, wherein the inverting input of the first comparator K1 and the non-inverting input of the second comparator K0 are connected to the first supply terminal (VDD), and the non-inverting input of the first comparator K1 and the inverting input of the second comparator K0 are connected to the second supply terminal VSS.
Optionally a further semiconductor switch M1B is connected between the first supply terminal (VDD) and a second substrate terminal B2, wherein a control terminal of the further semiconductor switch M1B is connected to the control terminal of the first semiconductor switch M1A. This further semiconductor switch M1B allows to contact different substrate terminals (B1, B2, etc.) in order to provide a uniform electric substrate potential (VDD) to the substrate 1 in case of a reverse voltage operation. During normal operation it can be useful to connect only one substrate terminal B1 to the second supply potential VSS and to connect further substrate terminals (B2, etc.) to the second supply potential VSS via a parallel resistor R1 (parallel to the substrate resistance RS). This is especially expedient, if a “hard” connection to VSS is unwanted to allow special guard rings to operate properly inhibiting adverse effects of a reverse current.
If—during normal operation—a reverse current flows from the source to the drain terminal, of transistor M2, for example, then transistor QN can be activated and will inject minority carriers into the substrate leading to possible malfunction of other parts of the circuit integrated in the same substrate. Several techniques may be applied to reduce the activation of npn transistor QN. Most of them are based on guard ring arrangements that are shorting by a means of a switch the base-emitter junction of npn transistor QN during reverse current operation. To be effective, the substrate resistance close to the npn transistor QN has to be high, allowing to the substrate potential to be pulled to the lowest possible potential, which is the voltage drop VSS−VBD across body diode DBD during reverse current operation. Therefore some substrate terminals (e.g. B2) are not directly switched to the second supply potential VSS but via a resistor R1. That is, the function of resistor R1 is to decouple two or more substrate locations on the same chip. While substrate terminal B1 is connected via the low-resistance semiconductor switch M0 to VSS, the further substrate terminal B2 is connected via a higher resistance, given by the parallel connection of R1 with the substrate resistance.
One simple embodiment of the invention is depicted in