[go: up one dir, main page]

HK1019263A - Floating bitline test mode with digitally controllable bitline equalizers - Google Patents

Floating bitline test mode with digitally controllable bitline equalizers Download PDF

Info

Publication number
HK1019263A
HK1019263A HK99104243.9A HK99104243A HK1019263A HK 1019263 A HK1019263 A HK 1019263A HK 99104243 A HK99104243 A HK 99104243A HK 1019263 A HK1019263 A HK 1019263A
Authority
HK
Hong Kong
Prior art keywords
bitline
floating
test mode
timing cycle
leaky
Prior art date
Application number
HK99104243.9A
Other languages
Chinese (zh)
Inventor
Kirihata Toshiaki
Wong Hing
Krsnki Bozidar
Original Assignee
Siemens Aktiengesellschaft
International Business Machines Corporation
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Siemens Aktiengesellschaft, International Business Machines Corporation filed Critical Siemens Aktiengesellschaft
Publication of HK1019263A publication Critical patent/HK1019263A/en

Links

Description

Field of the Invention
The present invention relates to the testing of Dynamic Random Access Memories. More specifically, it relates to a digitally controllable floating bitline test mode for DRAM's that can be performed during a wafer test.
Background to the invention
A DRAM includes a memory cell array for outputting holding data to a bitline when selected, and sense amplifiers for amplifying outputs. Memory cell defects and memory array defects have many sources and as a result, many signatures. While single, isolated cell failures may be scattered throughout the array, very often, multiple cells in the same vicinity fail. One characterization of these failures may be a bit (or column) line failure (i.e., failing cells with the same bit address). The sources of these failures vary, and as such, extensive on chip testing of memory arrays is required. The testing of these VLSI memory chips is a crucial process in their development and production.
One of the functional failures of the DRAM has been attributed to the leaky bitline (BL), which is primarily caused by short-circuiting between the BL and the wordline, WL. The leaky BL causes the BL voltage to drift during a signal development phase. This occurs when the BL is in a floating state due to the disabled BL equalizer and the sense amplifier not being set. The floating state in the signal development phase is generally set at 10ns. A major leaky BL can be easily detected during an ICC5 test because the stand-by current (CMOS), is over specification (typically 100µA). In stand-by phase, the BL equalizer is enabled, and the current is supplied to the BLs to hold the voltage.
A medium leaky BL, cannot be detected in the ICC5 test, but can be detected in a function test, since the leaky BL reduces the sensing voltage, which is typically 100mV. The leaky BL causes more than 100mV voltage drifts which results in a sensing failure. The detectable leakage current in the function test can be calculated by: BL Capacitance(100fF) x BL Drift Voltage (100mV) / time (5ns) which is 2µA.
A slightly leaky BL, which has less than 2µA, can still pass the function test. However, the slightly leaky BL often causes a failure in a module test or burn-in test, thereby reducing the module and burn-in yield. Even if it still passes the module test, it may be defective in the field, and result in reliability problems.
If the BL floating state could be controlled by the test mode, a leaky BL can be more accuratly detected by extending the BL floating state. This floating bitline test mode makes it possible to detect a leaky bitline in a wafer test. The detected defective BL can be repaired with column redundancy, thereby increasing the module yield, and improving reliability.
FIG. 1 shows a standard approach to generate a bitline floating state, in which a bitline sensing start is delayed by a predetermined time when a test mode is enabled. This method is inflexible, and has increased complexity because of the difficulty in determining a reasonable predetermined delay time during design.
Another floating bitline test mode shown in FIG. 2 includes the external control of internal timings. This type of test mode allows a bitline sensing start to be directly controlled by a VLSI tester. Although this increases the flexibility of the test mode, it requires the changing of other internal timings such as, column address set-up and hold times, data strobe, and cycle time according to the sense amplifier (SA) delay.
SUMMARY OF THE INVENTION
It is therefore an object of the present invention to provide a floating bitline test mode that has increased flexibility.
It is another object of the invention to provide a floating bitline test mode that does not require the changing of other control timings during the test.
It is a further object of the invention to generate a floating bitline for enabling a test mode that utilizes dummy timing cycles between the initiation of the test mode, and the starting of the normal read operation.
These and other objects of the invention are achieved by providing a floating bitline test mode for DRAM's with digitally controllable bitline equalizers. The test mode is enabled in the standard way by a Write CAS Before RAS (WCBR) cycle with address. The bitline floating time is digitally controllable by changing the number of dummy cycles between the time in which a test mode is enabled, and the time at which the read operation starts. By choosing a certain number of dummy cycles, a leaky bitline will be noticeably defective, while allowing normal bitlines to be non-defective. The number of dummy cycles selected is controllable by an operator through a VLSI tester. Once a leaky bitline is detected using this method, it is possible to repair the defective bitlines with column redundancy, which overcomes problems in yield reduction in module burn-in tests.
BRIEF DESCRIPTION OF THE DRAWINGS
A more complete appreciation of this invention, and many of the attendant advantages thereof, will be readily apparent as the same becomes better understood by reference to the following detailed description when considered in conjunction with the accompanying drawings, in which like reference symbols indicate the same or similar components, wherein:
  • FIG. 1 is a graphic representation of a first floating bitline test mode according to the prior art;
  • FIG. 2 is a graphic representation of a second floating bitline test mode according to the prior art;
  • FIG. 3 is a graphic representation of a floating bitline test mode with digitally controllable equalizers according to the present invention;
  • FIG. 4 is a schematic diagram of an embodiment of the equalizer control circuit according to the invention; and
  • FIG. 5 is a graphic representation of simulated waveforms of the normal operation mode and the floating bitline test mode according to the invention.
DETAILED DESCRIPTION OF THE INVENTION
The invention generally relates to semiconductor memories and particular to testing semiconductor memories during manufacturing. To facilitate discussion, the invention is described in the context of Dynamic Random Access Memories (DRAM). However, the invention is broader and is applicable to all types of semiconductor memories. Such memories, for example, include synchronous (DRAMs) and merged DRAM-logic (embedded) circuits.
Referring to FIGS 3 and 4, the equalizer control circuit 100 according to the invention is shown. For purposes of simplification, it is assumed that the WCBR only activates a negative pulsed signal TEST ¯ . However, in actual implementation, WCBR would need to allow a refresh operation. The NAND flip flops 102 and 104 and inverters 106, 108, and 110 are set to make EQ go high in a chip power up phase.
In normal read mode, equalizers in the corresponding sub-array are first reset when the row address strobe ( RAS ¯ ) is enabled. Word Line (WL) is then activated, and after a signal development time, sense amplifiers are set. When RAS ¯ is disabled, word line (WL) is first reset, and after WL has been disabled, sense amplifiers (SA) are reset and equalizers (EQ) are enabled again.
When a test mode is enabled by detecting a WCBR cycle, pulsed signal TEST ¯ periodically goes down (state 4). As shown in FIG. 3, this negative pulsed TEST ¯ disables all equalizers (EQ) in all of the sub-arrays. When RAS ¯ is enabled (i.e., goes high), the read operation starts, as in a normal mode. The EQ signal which had been disabled remains low, while WL and SA work as if in a normal mode. This allows the determination of a leaky bitline BL/BL. The timing for this cycle is determined by the digitally controllable dummy cycles as depicted between timing states 4 and 5. As shown, a leaky bitline can be sufficiently leaked during the dummy timing cycles, resulting in a sensing failure.
When RAS ¯ is disabled (i.e., brought low) at the end of the dummy cycle, word line WL is first reset which subsequently disables SA. The Equalizers are automatically reset by detecting the time at which WL is disabled, as in a normal read operation.
The following state table shows the relation between equalizer signal (EQ) with respect to TEST, WL and RAS.
1 2 3 4 5 6 7 8
1 0 0 1 1 1 0 0 1 1
0 1 1 1 1 1 1 0 1 1
1 1 1 0 1 1 1 1 1 1
WL 0 0 0 0 0 0 1 1 0 0
SA 0 0 0 0 0 0 0 1 1 0
EQ 1 1 1 0 0 0 0 0 0 1
As shown, when signal TEST ¯ is pulsed at timing state 4, EQ goes low to generate the floating bitline mode. EQ remains low until the next RAS is activated and disabled as in a normal random access mode. More specifically, the EQ remains low until the WL in the next random access mode is disabled. A number of dummy cycles 40 - 41 are inserted until the next random access mode is started (state 5). This allows the time, at which BL is floated, to be digitally and flexibly controlled. When WL goes low at the timing state 7, SA goes low immediately thereafter (timing state 8), and EQ is reset in response to the disabled state of WL and SA.
In an alternative embodiment (not shown), only selected sub-array equalizers can be reset with address. In this mode, the signal EQ in the activated sub-array can be reset, and the read operation continued for the other sub-arrays.
FIG. 5 shows simulated waveforms for the normal read and floating bitline test modes. In normal read mode, when RAS is disabled, the EQ signal is also disabled, and the word line WL signal is subsequently enabled. When WL is disabled, EQ resets and returns to a high state. When operating in the floating bitline test mode, according to the invention, the negative pulsed TEST signal causes EQ to be disabled (i.e., go low). A number of dummy cycles are inserted until the next RAS activation. When the next RAS is activated, EQ remains low while RAS and WL operates as if operating in a normal mode, and when WL is disabled EQ resets and reutrns to a high state, thereby terminating the floating bitline mode.
Through the application of digitally controlled dummy cycles, the floating bitlines are easily controlled without requiring a change in any other timings. A VLSI tester can control the number of dummy cycles to be implemented for each floating bitline test mode.
It should be understood that the present invention is not limited to the particular embodiment disclosed herein as the best mode contemplated for carrying out the present invention, but rather that the present invention is not limited to the specific embodiments described in this specification except as defined in the appended claims.

Claims (11)

  1. A method for detecting leaky bitlines in semiconductor memories comprising the steps of:
    generating at least one dummy timing cycle;
    disabling a bitline equalizer to create a floating bitline test mode; and
    detecting a defective bitline by reading from the semiconductor memory as in a normal mode.
  2. The method according to claim 1, wherein said step of generating at least one dummy timing cycle is performed by a VLSI tester.
  3. The method according to claim 1, wherein said step of disabling the bitline equalizer comprises the step of generating a negative pulsed test signal after a WCBR has been completed.
  4. The method according to claim 3, wherein said at least one dummy timing cycle is initiated upon the disabling of the bitline equalizer.
  5. The method according to claim 2, wherein said at least one dummy timing cycle is digitally controllable by the VLSI tester.
  6. The method according to claim 2, wherein the number of dummy timing cycles is digitally controllable by the VLSI tester.
  7. The method according to claim 1, further comprising the step of repairing a detected leaky bitline with a column redundancy.
  8. The method according to claim 1, wherein the semiconductor memory is a Dynamic Random Access Memory (DRAM).
  9. A method of detecting a leaky bitline in DRAM comprising the steps of:
    disabling a bitline equalizer to create a floating bitline test mode;
    initiating at least one dummy timing cycle at the beginning of the test mode using a VLSI tester, said at least one dummy timing cycle being digitally controllable by said VLSI tester; and
    determining the presence of a defective bitline during the at least one dummy timing cycle.
  10. The method according to claim 9, further comprising the step of repairing a defective bitline with a column redundancy.
  11. The method according to claim 9, wherein said step of disabling the bitline equalizer comprises the step of generating a negative pulsed test signal by detecting the completion of a WCBR operation.
HK99104243.9A 1997-09-25 1999-09-29 Floating bitline test mode with digitally controllable bitline equalizers HK1019263A (en)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
US937528 1997-09-25

Publications (1)

Publication Number Publication Date
HK1019263A true HK1019263A (en) 2000-01-28

Family

ID=

Similar Documents

Publication Publication Date Title
KR101492667B1 (en) Method and apparatus for testing a memory device
KR960002010B1 (en) Semiconductor memory
US5638331A (en) Burn-in test circuit and method in semiconductor memory device
USRE38956E1 (en) Data compression circuit and method for testing memory devices
US20040240285A1 (en) Method and apparatus for data compression in memory devices
JPH07262798A (en) Semiconductor memory stress circuit and stress voltage supply method
US6510533B1 (en) Method for detecting or repairing intercell defects in more than one array of a memory device
US6185137B1 (en) Semiconductor memory device with decreased current consumption
US20030058717A1 (en) Word-line deficiency detection method for semiconductor memory device
US7859938B2 (en) Semiconductor memory device and test method thereof
EP0907185B1 (en) Floating bitline test mode with digitally controllable bitline equalizers
JPH0821607B2 (en) Dynamic storage device and burn-in method thereof
KR100568253B1 (en) Semiconductor memory device and write control method thereof
US5926423A (en) Wafer burn-in circuit for a semiconductor memory device
US7162663B2 (en) Test system and method for testing memory circuits
US7170797B2 (en) Test data topology write to memory using latched sense amplifier data and row address scrambling
JPH0628893A (en) Semiconductor memory device
HK1019263A (en) Floating bitline test mode with digitally controllable bitline equalizers
JP2001307499A (en) Semiconductor memory
CN100421184C (en) Memory device and method for burn-in test
US20040204891A1 (en) Semiconductor memory device having a test mode for testing an operation state
KR20080057385A (en) Test method of memory cell failure in semiconductor memory device
KR100195194B1 (en) Burn-in stress circuit for semiconductor memory device
KR100618703B1 (en) Early Detection and Improvement of Bad Patterns in Semiconductor Memory
KR100426445B1 (en) Peripheral Circuit Built-in Test Method for Semiconductor Memory Devices