摘要:
Boundary scan testing methods that detect manufacturing defects in AC-coupled differential pairs may be based on a selected signal parameter: phase or frequency. The data is encoded by the signal parameter. In one embodiment, the signal parameter is compared to reference parameter data provided by the transmitter. In a second embodiment, the reference parameter data is sent from an external source. All the components on board are synchronized with this external source. In a third embodiment, the reference parameter data is embedded in each AC signal between the transmitter and receiver. Two lines of one differential link are used to send different patterns.
摘要:
Hierarchically-controlled automatic test pattern generation (ATPG) is provided. One embodiment comprises a method for automatically generating test patterns for testing a device under test. Briefly described, one such method comprises the steps of: receiving a hierarchical model of a device under test, the hierarchical model comprising at least one low-level design component and at least one high-level design component which contains the low-level design component; selecting a fault to be detected in the device under test; and performing an automatic test pattern generation (ATPG) algorithm on the design components based on the hierarchy of the hierarchical model.
摘要:
A gate transition counter. A ring oscillator provides a plurality outputs, each delayed from the adjacent output by a gate delay. The outputs of the ring oscillator are captured by an array of latches upon receipt of a halt signal. The last latch drives a ripple counter. The preferred implementation uses five inverters in the ring oscillator so that each complete cycle of the ring oscillator represents ten gate delays. A ripple counter counts the number of gate delays by ten. The latch outputs and the ripple counter outputs can be converted to a binary representation of the number of gate delays to provide a count with the smallest time increment that can be produced by the circuit.
摘要:
Systems and methods for testing integrated circuits are provided. One such method comprises: providing a target fault list corresponding to an integrated circuit, the target fault list including at least a first fault and a second fault; measuring a relationship between the first fault and the second fault, the relationship corresponding to which of the first fault and the second fault is more readily detected by automatic test pattern generation; ordering the first fault and the second fault within the target fault list in a manner corresponding to the relationship; and performing automatic test pattern generation based upon an order of the faults of the target fault list. Systems and other methods also are provided.
摘要:
An integrated circuit is provided with Fully Automated Scan Testing (FAST)-lite flip-flop. The integrated circuit has data, scan in, master-hold, clock, scan-into-master, and master to-scan-out inputs. A first transistor circuit is connected to the data, master-hold, and clock inputs and has a first transistor circuit output. A second transistor circuit is connected to the can in and scan-into-master inputs and has a second transistor circuit output. A first flip-flop is connected to the first transistor circuit and second transistor circuit outputs and has a first flip-flop output. A third transistor circuit is connected to the second transistor circuit output and the master-to-scan-out input and has a third transistor circuit output. A second flip-flop latch is connected to the third transistor circuit output has a second flip-flop output. The FAST-lite flip-flop uses the normal functionality of the first flip-flop and second flip-flop to operate either in a normal mode or a test mode for scan testing. The FAST-lite flip-flop is so designated because it uses one less flip-flop than the prior art FAST flip-flop.
摘要:
A method for operating a data processing system to provide an estimate of the leakage current expected from an integrated circuit having a known test vector applied thereto. The method generates a graph having a first node connected to a terminal maintained at a first power supply potential and a second node connected to a terminal maintained at a second power supply potential, the first and second power supply potentials having a potential difference of VDD. Each edge represents a source-drain connected transistor connected between two nodes as a switch having a resistance that depends on whether the transistor is “on” or “off” when the test signals are present. Each node in the graph is assigned a value of 0 or 1, together with the strength of that value. The assignment process also verifies that the circuit is in a valid static state. A leakage current estimate is generated for each path connecting the first and second nodes, and estimates are combined to arrive at the estimate for the integrated circuit. In one embodiment, the leakage current is calculated using an equation having eight components. Four of these represent the leakage current arising from N-type transistors in the circuit and the other four represent the leakage current arising from P-type transistors in the circuit.
摘要:
A method and apparatus of carrying out a computer assisted analysis function on a hierarchical circuit model. The method is carried out by inputting the hierarchical circuit model, specifying at least one circuit block within the hierarchy as a target of the function on the target block, and simplifying the hierarchical circuit model by deleting circuit blocks not affecting the analysis function, to produce a simplified hierarchical circuit model. A computer assisted analysis function can then be carried out on the simplified hierarchical circuit model. The model is simplified by carrying out a block by block analysis of the hierarchical circuit model to determine if a circuit block is a circuit block desired in the computer assisted analysis and if so keeping the circuit block as part of the simplified hierarchical circuit model; and by carrying out a block by block analysis of the hierarchical circuit model to determine if a circuit block comprises a parent circuit block containing a child circuit block desired in the computer assisted analysis and if so keeping the parent circuit block as part of the simplified hierarchical circuit model.
摘要:
The present invention is generally directed to an improved automatic test pattern generator for generating test patterns that are used by an integrated circuit testing device. In accordance with one aspect of the invention, a method is provided for generating a set of test vectors for testing an integrated circuit, each test vector of the set of test vectors containing a plurality of bits defining test inputs for the integrated circuit. The method includes the steps of defining a list of faults for the integrated circuit, and generating at least one test vector that defines values for those inputs necessary to detect at least one target fault selected from the list of faults, the values comprising only a portion of the bits of the at least one test vector, wherein a remainder of the bits in the at least one test vector are unspecified bit positions. The method further includes the step of setting the values of a plurality of the unspecified bit positions using a non-random filling methodology.
摘要:
Improvement in the transmission of Boundary Scan Test mode data may be achieved through the assignment of boundary scan test mode traffic to selected bit patterns that facilitate clock recovery and frame alignment in the serial channel. The encoding of boundary scan test traffic as such may be achieved through either multiplexed transmission to the serializer/deserializer (SERDES) alongside a regular channel encoder or incorporated into the channel encoder.
摘要:
A method for more efficient handling of data used in testing electronic integrated circuits and boards consisting of specifying data in a flat tabular format. Signal data are contained in columns and cycle data are contained in rows. Any given signal in a cycle will typically be a "0" or "1". If a signal does not change from one cycle to the next, the character "0" or "1" is replaced in the latter cycle by the dot "." character. As a result of the "." character replacements, the data is much more highly compressible by standard compression software than before, thus requiring less storage space for the data.