CMOS-inverter

The invention relates to a CMOS inverter comprised of a series connection of a normally off p-channel MOS-FET and a normally off n-channel MOS-FET whose common drain connection represents an output of the inverter.

There are instances of CMOS circuits in which different operating voltages are necessary. For example, in electrically erasable and programmable memories (E.sup.2 PROMS) CMOS-inverters whose operating voltage lies in a so-called high-voltage range of around 20 V (for example, clearly above a so-called low voltage operating voltage of around 5 V), bust be activated by low voltage inverters. The p-channel MOS-FET of a high voltage inverter, however, on account of the maximum output voltage of the low voltage inverter being too low, can thus not be blocked so that a static forward current flows.

An object underlying the invention is to provide a CMOS-inverter of the above-captioned type which, even in the case of an input H-level in the low voltage range, gurrantees a reliable blocking of the p-channel MOS-FET.

This object is achieved in that there is connected in series with the p-channel MOS-FET an additional normally on n-channel MOS-FET whose control input is connected with the inverter output.

The FIGURE shows a CMOS-high voltage inverter according to the present invention.

The FIGURE shows a CMOS-high voltage inverter which is comprised of a series connection of a normally off n-channel MOS-FET 1 and a normally off p-channel MOS-FET 2. The control inputs of these MOS-FETs jointly form the inverter input which is connected to a control voltage U.sub.E. The common drain connection of the two MOS-FETs forms the inverter output with an output signal U.sub.A. Connected in series with the p-channel MOS-FET 2 is a normally on n-channel MOS-FET 3 whose control input is connected with the inverter output. The drain connection of the additional n-channel MOS-FET 3 is connected with an operating voltage U.sub.pp which lies in a high voltage range of e.g. 20 V.

The input voltage U.sub.E is derived from a pre-connected low voltage CMOS-inverter 4 which, in the illustrated example, is comprised of a series connection of two normally off MOS-FETs and is operated with a low voltage operating voltatge U.sub.cc of about 5 V. The input signal U.sub.E can assume one of the two logic levels L or H respectively, whereby the H-level also lies in the low voltage range. The magnitude of the input voltage U.sub.E, with which a value of the output signal U.sub.A of 0 volts is obtained, is determined by the difference of the starting voltage U.sub.Tn of the normally on n-channel MOS-FET 3 and the amount of the starting voltage U.sub.Tp of the p-channel MOS-FET 2.

For a statically low-current state of the above-described high voltage CMOS-inverter with an input voltage U.sub.E which corresponds approximately to the low voltage operating voltage U.sub.CC, in the following advantageous dimensioning example is provided. The low voltage operating voltage lies at 5 V, the amount of the starting voltage U.sub.Tn of the normally on n-channel MOS-FET 3 typically lies at 3 V, and the amount of the starting voltage U.sub.Tp of the p-channel MOS-FET 2 typically lies between 0.5 V and 1 V. In order to switch the inverter with low dynamic forward current, the series connection of the p-channel MOS-Fet 2 and the normally on n-channel MOS-FET 3 must be dimensioned so as to be weakly conductive in relation to the n-channel MOS-FET 1. This is the case if, in the case of the normally on n-channel MOS-FET 3 and/or in the case of the normally off p-channel MOS-FET 2, the ratio of channel width and channel length is smaller than in the case of the normally off n-channel MOS-FET 1.

While various changes and modifications might be proposed by those skilled in the art, it will be understood that I wish to include within the claims of the patent warranted hereon all such changes and modifications as reasonably come within my contribution to the art.