摘要:
A method and circuit for a self timed DRAM. The circuit includes interlock circuits coupled to an extension of the DRAM. The extension does not store “real” data but mimics the operations of the DRAM. The interlock circuits, in conjunction with the extension monitor and control read and write timings of the DRAM and self adjust these timings via feedback. To properly track DRAM cell timings, the interlock circuits and extension use the same cell design and load conditions as the DRAM. The method includes: activating a wordline and reference wordline, interlocking the sense amplifiers, column select and write back functions of the primary DRAM array by monitoring the identical reference cells and the state of the bitline in the extension DRAM array.
摘要:
A method of memory BIST (Built-In Self Test) and memory repair that stores a redundancy calculation on-chip, as opposed to scanning this data off-chip for later use. This method no longer requires level-sensitive scan design (LSSD) scanning of memory redundancy data off-chip to the tester, and therefore does not require re-contacting of the chip for electrical fuse blow.
摘要:
A system for testing a DRAM includes DRAM blocks, the system further includes a processor based built-in self test system for generating a test data pattern, for each DRAM block, performing a write of the test data pattern into the DRAM block, performing a pause for a predetermined period of time, and performing a read of a resulting data pattern from the DRAM block. For each DRAM block, the performing the write of the test pattern into the DRAM block is performed before the performing the pause for the predetermined period of time, and the performing the read of the resulting data pattern from the DRAM block is performed after the performing the pause for the predetermined period of time, and at least a portion of the pause for the predetermined period of time of two or more the DRAM blocks overlap in time.
摘要:
A method and system for testing an embedded DRAM that includes DRAM blocks. The method including: generating a test data pattern in a processor based BIST system, for each DRAM block, performing a write of the test data pattern into the DRAM block, performing a pause for a predetermined period of time, and performing a read of a resulting data pattern from the DRAM block; where for each DRAM block, the write of the test data pattern into the DRAM block is performed before the pause, and the read of the resulting data pattern from each DRAM block is performed after the pause; where at least a portion of the pause of two or more of the DRAM blocks overlap in time; and for each DRAM block comparing the test data pattern to the resulting data pattern.
摘要:
Disclosed is a hybrid built-in self test (BIST) architecture for embedded memory arrays that segments BIST functionality into remote lower-speed executable instructions and local higher-speed executable instructions. A standalone BIST logic controller operates at a lower frequency and communicates with a plurality of embedded memory arrays using a BIST instruction set. A block of higher-speed test logic is incorporated into each embedded memory array under test and locally processes BIST instructions received from the standalone BIST logic controller at a higher frequency. The higher-speed test logic includes a multiplier for increasing the frequency of the BIST instructions from the lower frequency to the higher frequency. The standalone BIST logic controller enables a plurality of higher-speed test logic structures in a plurality of embedded memory arrays.
摘要:
Wafer test and burn-in is accomplished with state machine or programmable test engines located on the wafer being tested. Each test engine requires less than 10 connections and each test engine can be connected to a plurality of chips, such as a row or a column of chips on the wafer. Thus, the number of pads of the wafer that must be connected for test is substantially reduced while a large degree of parallel testing is still provided. The test engines also permit on-wafer allocation of redundancy in parallel so that failing chips can be repaired after burn-in is complete. In addition, the programmable test engines can have their code altered so test programs can be modified to account for new information after the wafer has been fabricated. The test engines are used during burn-in to provide high frequency write signals to DRAM arrays that provide a higher effective voltage to the arrays, lowering the time required for burn-in. Connections to the wafer and between test engines and chips are provided along a membrane attached to the wafer. Membrane connectors can be formed or opened after the membrane is connected to the wafer so shorted chips can be disconnected. Preferably the membrane remains on the wafer after test, burn-in and dicing to provide a chip scale package. Thus, the very high cost of TCE matched materials, such as glass ceramic contactors, for wafer burn-in is avoided while providing benefit beyond test and burn-in for packaging.
摘要:
A power up detection circuit is provided which includes a power supply terminal, an output terminal, an impedance device coupling the output terminal to the power supply terminal and a latch including a first inverter having a first device connected between the output terminal and a point of reference potential and a second device connected between the output terminal and the power supply terminal, the devices are designed so that subthreshold current passing through the first device is greater than the effective subthreshold current passing through the impedance device and the second device, and a second inverter including third and fourth devices which are designed so that a smaller subthreshold current passes through the third device than the subthreshold current passing through the fourth device. The power up circuit may further include a capacitor connected between the power supply terminal and gate electrodes of the first and second devices.
摘要:
A built-in, i.e., on-chip, self-test system for a VLSI logic or memory module. A deterministic data pattern generator is provided on the VLSI chip, and operates to test a chip module and provide a fail/no-fail result, along with data identifying where the fail occurred. This location data is captured and made available for subsequent utilization. The built-in test circuitry is programmable, and is provided with a looping capability to provide enhanced burn-in testing, for example.
摘要:
An off-chip driver circuit is provided which includes a pull-up device disposed between an output terminal and a first voltage dropping diode which is connected to a first voltage supply source and a first voltage limiting circuit connected to the common point between the pull-up device and the voltage dropping diode. The off-chip driver circuit further includes an input inverter circuit having an output connected to the control element of the pull-up device. The inverter circuit has a P-channel field effect transistor and an N-channel field effect transistor serially connected with a second voltage dropping diode which is connected to the first voltage supply source and a second voltage limiting circuit connected to the common point between the second voltage dropping diode and the P-channel field effect transistor of the input inverter. First and second switches are also provided to short out the first and second voltage dropping diodes, respectively, when all circuits connected to the output terminal use a common voltage supply. A pull-down device serially connected to a pass device is provided between the output terminal and a point of reference potential. A buffer circuit having an output connected to the pull-down device is coupled to a second voltage supply source having a voltage significantly lower than the voltage of the first voltage supply source.
摘要:
A system and method for providing DRAM device-level repair via address remappings external to the device. A system includes a memory controller having an interface to one or more memory devices via a memory module. The memory devices include addressable redundant and non-redundant memory blocks. The memory controller also includes a mechanism for utilizing one or more redundant memory blocks in place of one or more failing non-redundant memory blocks via an address remapping external to the memory device. The remapping occurs while the system is on-line.