Dynamic Logic Gate

1. Field of the Invention

This invention relates to an integrated circuit logic gate compatible with a four-phase, major-minor clocking arrangement and a six-phase metal oxide semiconductor (MOS) system.

2. Prior Art

Integrated circuit field effect transistor (FET) logic circuits, which are strobed by a four-phase, major-minor clocking scheme, can be arranged to form six basic logic gates. Reference may be made to U.S. Ser. No. 659,057, filed Feb. 18, 1976, for a brief description of these aforementioned six basic logic gates.

A first logic gate, commonly designated as a type 2 logic gate (which precharges when multi-phase clock signal .phi..sub.1 comes true and discharges or evaluates when multiphase clock signal .phi..sub.2 comes true) and a second logic gate, commonly designated as a type 4 logic gate (which precharges during .phi..sub.3 and evaluates during .phi..sub.4) are referred to as major gates by those skilled in the art, inasmuch as each gate evaluates during major clock phases .phi..sub.2 and .phi..sub.4, respectively. For a more detailed description of the major logic gate types 2 and 4, reference may be made to U.S. Pat. No. 3,601,627 issued Aug. 24, 1971. However, none of the six basic logic gates is implemented so that both of the major gate types, 2 and 4, can be directly fanned-in without the addition of space consuming interface gates, such as inverters and the like. This is because no single logic gate is known which is adapted to accept information signals when the output of both type 2 and 4 logic gates are good (i.e. the output terminals are isolated from the output terminals thereof).

One example of a MOSFET dynamic logic system including six clock phase inputs is U.S. Pat. No. 3,747,064 issued July 17, 1973.

An improved compact, dynamic logic gate is disclosed which is compatible with a six-phase MOS system and a four-phase, major-minor clocking scheme. In a preferred embodiment, the logic gate precharges during a first recurring major clock time interval, evaluates or discharges during a first in-between time interval (i.e. the clock phase after the completion of the first major time interval and before the occurrence of a second recurring major clock time interval), and isolates the logic gate output node from the input node during both the second major clock time interval and a second in-between clock interval (i.e. the clock phase between the second and first recurring major time intervals). As a result, at least four conventional logic gates, including each of the major logic gate types, designated 2 and 4, can fan directly into the logic gate of the instant invention without the requirement of complex interfacing gates, such as inverters and the like. Moreover, the improved logic gate is ratioless and consumes no d.c. power.

FIG. 1 is a diagram representative of the fan-in and fan-out characteristics of the six conventional logic gates of a four-phase major-minor clocking scheme.

FIG. 2 illustrates the waveforms of a four-phase, major-minor clocking scheme utilized to drive the instant logic gate.

FIG. 3 shows a circuit for implementing the improved logic gate of the instant invention.

FIG. 4 is a diagram representative of the fan-in and fan-out characteristics of the six gates of FIG. 1 and the improved logic gate of the instant invention.

FIG. 1 illustrates the fan-in and fan-out characteristics of the six conventional logic gates of a four phase, major-minor clocking scheme having in-between phases to separate the major clock intervals. The number of each gate indicates in which phase of the present four-phase clocking scheme the particular gate discharges or evaluates. The arrows indicate the possible fan-out of each gate. Gate types denoted as 2 and 4 are referred to as major logic gates. As will be known to those skilled in the art, and as indicated by FIG. 1, logic gates types 2 and 4 evaluate during the .phi..sub.2 and .phi..sub.4 major clock phases, respectively. Moreover, the output of both the type 2 and 4 major logic gates are good (i.e. isolated from the input) during each of the inbetween clock phases, to be discussed in more detail hereinafter.

FIG. 2 is illustrative of a well-known four-phase, major-minor clocking scheme which is utilized to strobe the improved logic gate of the instant invention. The waveform of each multi-phase clock signal, .phi..sub.1, .phi..sub.1+2, .phi..sub.3, and .phi..sub.3+4 has relatively HI (i.e. true) and LOW (i.e. false) signal levels, corresponding, for example, to -V volts and ground, respectively. .phi..sub.1+2 and .phi..sub.3+4 are designated as major clock phases. The intervals occurring between the major clock phases are designated as in-between clock phases.

FIG. 3 illustrates an integrated circuit for implementing, in a first preferred embodiment, the improved, in-between phase logic gate 10 of the instant invention. The circuit includes metal oxide semiconductor field effect transistors (MOSFETs) Q.sub.1, Q.sub.2, Q.sub.3 and Q.sub.4. One conduction path electrode and the control (i.e. gate) electrode of FET Q.sub.1 are connected together and to a suitable multi-phase clock signal generator (not shown) to receive a supply of multi-phase clock signals, .phi..sub.3+4 (the waveform of which is illustrated in FIG. 2). The second conduction path electrode of FET Q.sub.1 is connected to a first conduction path electrode of FET Q.sub.2. The gate electrode of FET Q.sub.2 is connected to the clock signal generator to receive a supply of multiphase clock signals .phi..sub.1+2 (the waveform of which is also illustrated in FIG. 2). The second conduction path electrode of FET Q.sub.2 is connected to the clock signal generator to receive the clock signal designated .phi..sub.3+4. Thus, the conduction paths of FETs Q.sub.1 and Q.sub.2 are electrically connected together in series.

One conduction path electrode and the gate electrode of FET Q.sub.3 are connected together and to the multi-phase clock signal generator to receive the clock signal .phi..sub.3+4. The second conduction path electrode of FET Q.sub.3 is connected to a first conduction path electrode of isolation FET Q.sub.4. The gate electrode of isolation FET Q.sub.4 is connected to a common electrical junction 1 with one plate of a capacitor C, between the conduction paths of FETs Q.sub.1 and Q.sub.2. The second plate of capacitor C is connected to a source of reference potential, such as ground. The output of the instant logic gate, designated node 2, is taken between the conduction path of FETs Q.sub.3 and Q.sub.4. The second conduction path electrode of isolation FET Q.sub.4 is connected to a suitable logic network 4. By way of example, logic network 4 conveniently includes a FET Q.sub.5. One conduction path electrode of FET Q.sub.5 is connected to the second conduction path electrode of FET Q.sub.4. The second conduction path electrode of FET Q.sub. 5 is connected to the multiphase clock generator to receive the multi-phase clock signal, designated .phi..sub.3+4. Thus, the conduction paths of FETs Q.sub.3 - Q.sub.5 are electrically connected together in series. The input of the instant logic gate, designated node 3, is connected to the gate electrode of FET Q.sub.5.

The operation of the instant logic gate 10 can be described by concurrently referring to FIGS. 2 and 3. Logic gate 10 precharges during the intervals of time designated t.sub.3 and t.sub.4. As shown in FIG. 2, during the precharge time intervals, a relatively HI level clock signal, .phi..sub.3+4, (-V) is applied to each of the gate electrodes of FETs Q.sub.1 and Q.sub.3. Hence, each of these devices is rendered conductive. The relatively HI level clock signal is applied to one plate of capacitor C and to the gate electrode of FET Q.sub.4 via the conduction path of FET Q.sub.1, whereupon capacitor C is precharged and FET Q.sub.4 is rendered conductive. The I level clock signal is also applied to output node 2 via the conduction path of FET Q.sub.3, and output node 2 is precharged. Output node 2 and capacitor C are each precharged to a voltage level -V minus a threshold level drop in voltage, V.sub.t, caused by the respective impedances of FETs Q.sub.1 and Q.sub.3 . A relatively LOW level clock signal, .phi..sub.1+2, (ground) is applied to the gate electrode of FET Q.sub.2, whereupon FET Q.sub.2 is rendered non-conductive during the precharge intervals of time.

The instant circuit evaluates during the SA in-between clock time interval. The SA in-between phase occurs between the t.sub.4 and t.sub.1 major time intervals when each of the multi-phase clock signal levels is relatively LOW (ground). During the SA in-between time interval, relatively LOW level clock signals are applied to the gate electrodes of each of FETs Q.sub.1, Q.sub.2, and Q.sub.3, whereupon each of these aforementioned devices is rendered non-conductive. FET Q.sub.4 remains conductive during the SA in-between time interval, inasmuch as the gate electrode thereof is connected to capacitor C, and capacitor C is precharged during the preceding t.sub.3 and t.sub.4 precharge time intervals. As a result, gate 10 evaluates as a function of the level of the information signal applied to input node 3 of logic network 4. In a first example, if the information signal applied to input node 3 (and, thus, to the gate electrode of FET Q.sub.5) is relatively LOW, FET Q.sub.5 is rendered non-conductive. Hence, output node 2, which is precharged during the t.sub.3 and t.sub.4 clock time intervals, remains charged at a relatively HI level (V - V.sub.t). In a second example, if the information signal applied to input node 3 is relatively HI, FET Q.sub.5 is rendered conductive during the SA in-between time interval. Thus, output node 2 is clamped to ground through the series connected conduction paths of FETs Q.sub.4 and Q.sub.5. It can be seen that, in the present embodiment, logic network 4 causes the instant logic gate 10 to function as an inverter with respect to the logic level of the signals at input node 3 and output node 2.

During the next time intervals, designated t.sub.1 and t.sub.2, a relatively HI level clock signal, .phi..sub.1+2, is applied to the gate electrode of FET Q.sub.2, whereupon FET Q.sub.2 is rendered conductive. Also during the t.sub.1 and t.sub.2 time intervals, a relatively LOW level clock signal, .phi..sub.3+4, is applied to each of the gate electrodes of FETs Q.sub.1 and Q.sub.3, whereupon FETs Q.sub.1 and Q.sub.3 are rendered non-conductive. During the t.sub.1 and t.sub.2 time intervals, capacitor C is discharged (i.e. clamped) to ground through common electrical junction 1 and the conduction path of FET Q.sub.2. As a result, the gate electrode of isolation FET Q.sub.4, is also clamped to ground, and FET Q.sub.4 is rendered non-conductive. Thus, during the t.sub.1 and t.sub.2 time intervals, input node 3 is isolated from output node 2 (by FET Q.sub.4), and the level of the information signal applied to input node 3 of logic network 4 is inconsequential to the operation of gate 10.

During the following clock time interval, which occurs between the t.sub.2 and t.sub.3 major time intervals and is designated in-between phase SB, a relatively LOW level clock signal (i.e. ground) is applied to each of the gate electrodes of FETs Q.sub.1, Q.sub.2, and Q.sub.3. As a result, FETs Q.sub.1, Q.sub.2 and Q.sub.3 are rendered non-conductive. Isolation FET Q.sub.4 remains non-conductive, inasmuch as capacitor C is discharged during the preceding t.sub.1 and t.sub.2 time intervals. Hence, during the SB time interval, output node 2 remains isolated from input node 3, and the level of the information signal applied to input node 3 of logic network 4 remains inconsequential to the operation of gate 10.

In succeeding clock cycles, the improved logic gate 10 precharges, evaluates, and provides isolation, as described above. As will be understood by those skilled in the art, and in a second preferred embodiment, if the .phi..sub.3+4 and .phi..sub.1+2 clock signals are interchanged (i.e. clock signal .phi..sub.3+4 is applied to the gate electrode of FET Q.sub.2 and clock signal .phi..sub.1+2 is applied to the gate and first conduction path electrodes of FETs Q.sub.1 and Q.sub.3 and to the second conduction path electrodes of FETs Q.sub.2 and Q.sub.5) another improved in-between phase logic gate is produced. However, this second in-between phase logic gate precharges during time intervals t.sub.1 and t.sub.2, evaluates during in-between time interval SB, and isolates the output node from the input node during time intervals t.sub.3, t.sub.4, and SA.

Therefore, a unique in-between phase, dynamic logic gate is disclosed which precharges during a first major clock time interval (t.sub.3 and t.sub.4 or t.sub.1 and t.sub.2), evaluates only during the in-between clock interval (SA or SB) before the next (i.e. second) major clock time interval, and isolates the output node from the input node during both the second major clock time interval (t.sub.1 and t.sub.2 or t.sub.3 and t.sub.4) and the second in-between clock interval (SB or SA). Moreover, the improved logic gate of the instant invention is ratioless and uses no direct current power. The instant logic gate is also compatible with a six-phase MOS system, but utilizes a minimum number of components.

What is more, the improved logic gate can accept either of the major logic gates (i.e. types 2 and 4) as direct inputs without the requirement of complex interface gates (e.g. inverters and the like). As disclosed above, the improved logic gate 10, in the first and second preferred embodiments, evaluates only during the SA or SB in-between clock time intervals, respectively. Since the outputs of both of the major logic gate types 2 and 4 are good (i.e. isolated) during each of the SA and SB in-between phase time intervals, logic gates 2 and 4 may fan directly into improved logic gate 10, as represented diagrammatically in FIG. 4.

It will be apparent that while a preferred embodiment of the invention as shown and described, various modifications and changes may be made without departing from the true spirit and scope of the invention. For example, FETs Q.sub.1 - Q.sub.5 of FIG. 3 are p-channel transistor devices. However, it is to be understood that these or any other suitable multi-terminal semiconductor device may be employed. Thus, the HI and LOW signal levels of the multi-phase clock signals illustrated in FIG. 2 are chosen to be compatible with the type of multi-terminal semiconductor devices utilized. Moreover, the logic network 4 of logic gate 10 is chosen for exemplary purposes only. It is to be understood that logic network 4 may be comprised of any suitable number of multi-terminal semiconductor devices or the like and the corresponding interconnections therebetween.