Testing method for dynamic logic keeper device

Details Testing method for dynamic logic keeper device Testing method for dynamic logic keeper device

1998-04-27

2001-07-31

Decady, Albert (Department: 2133)

Error detection/correction and fault detection/recovery

Pulse or data error handling

Digital logic testing

Reexamination Certificate

active

06269461

ABSTRACT:

BACKGROUND OF THE INVENTION
1. Technical Field of the Invention
This invention pertains to dynamic logic circuits. In particular, this invention is directed to testing dynamic logic circuits that include a half-latch, also known as a keeper device. A novel apparatus and method is described for testing whether keeper structures in dynamic logic are defective.
2. Background Art
A keeper device maintains a dynamic logic gate's precharge state and adds significant robustness to the circuit. In comparison to static logic circuits, dynamic logic circuits require a pre-charge pulse (RESET signal) that is followed by an evaluation phase wherein input signal data causes the gate to evaluate (the input signal data) by outputting a high or low output signal. In static logic, typically there are PFET pull-up devices and NFET pull-downs that are both connected to receive input signal data sent to the static logic gate.
A useful feature of a dynamic logic gate is the use of a half-latch and inverter (shown in
FIG. 1
) to maintain a precharge state of the gate.
FIG. 1
illustrates an example generic logic gate having a RESET input
11
, a half-latch
10
, inverter
13
, an output, and an evaluate function determined by the structure of logic tree
12
. The logic tree may be connected to the primary node A at only one point or at several points, depending on the logic function to be performed. A logic gate such as in this example can be defective in many ways. The objective of testing is to identify those defective gates. One of the defect modes affects the keeper device
10
and makes it inoperative. For the gate shown in
FIG. 1
, the PFET keeper
10
can have an open source, an open drain, or the gate stuck at a “1”. When such a defect exists the gate will typically function “properly” as a result of the inherent capacitance in the node A at the top of the logic tree
12
. It will not have the robustness required to ensure proper operation at all conditions, however. Currently there is no good way to test dynamic logic gates and ensure that the keeper devices are indeed intact.
Dynamic logic has both a precharge and an evaluate time during its normal operation. During precharge, a gate is reset so that its output(s) and primary internal node(s) are precharged to a given state, e.g., a high state (binary “1”). Then when data (In
1
. . . InX) arrives, the gate evaluates and, if appropriate (e.g. the logic tree couples the primary node to ground), the internal node discharges and the output transitions to the opposite state. Dynamic logic gates are frequently designed with an inverter on the output and a half latch to hold the primary internal node at its precharge state. This is comparable to a kind of memory function because the primary node is held in a high or low (“1” or “0”) precharge state.
In the operational sequence for a dynamic gate with an NFET dominated logic tree, the reset signal, pulsed low, pulls the primary node high, which causes the output to go low (through the inverter). The output, coupled to the half-latch PFET, keeps the half-latch on and the primary node is pulled and maintained high. The PFET half-latch thus maintains the gate in its reset state until data arrives. Even if “reset” goes high, the half-latch maintains the primary node high in a noisy environment.
SUMMARY OF THE INVENTION
A problem exists verifying that a dynamic logic gate is fault free, with respect to its keeper device. Three solutions are proposed. First, a method is presented for testing dynamic logic gates without including the novel apparatus of the present invention. Second is a method for testing logic gates which implements the novel apparatus. Third, the novel apparatus itself includes a device for very slowly bleeding charge from a primary node of dynamic logic gate to determine whether the bleeding causes an erroneous gate output, hence, uncovering a defective keeper device whose function it is to maintain the charge state of the primary node.
Other features and advantages of this invention will become apparent from the following detailed description of the presently preferred embodiment of the invention, taken in conjunction with the accompanying drawings.


REFERENCES:
patent: 5122738 (1992-06-01), Simpson et al.
patent: 5399918 (1995-03-01), Taylor et al.
patent: 5440243 (1995-08-01), Lyon
patent: 5541537 (1996-07-01), Kim et al.
patent: 5557620 (1996-09-01), Miller, Jr. et al.
patent: 5592143 (1997-01-01), Adams et al.
patent: 5671151 (1997-09-01), Williams
patent: 5831451 (1998-11-01), Bosshart
patent: 5831990 (1998-11-01), Queen et al.
patent: 5910735 (1999-06-01), Allen
patent: 5949266 (1999-09-01), Hinds et al.
patent: 6002292 (1999-12-01), Allen et al.
patent: 6014041 (2000-01-01), Somasekhar et al.

Affiliated with

Also associated with

Rating

Testing method for dynamic logic keeper device does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Testing method for dynamic logic keeper device, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Testing method for dynamic logic keeper device will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2528912