Font Size: a A A

Stability Of The SRAM Cells For Deep Sub-micron Technologies

Posted on:2012-11-03Degree:MasterType:Thesis
Country:ChinaCandidate:L ZhaoFull Text:PDF
GTID:2218330368992191Subject:Microelectronics and Solid State Electronics
Abstract/Summary:PDF Full Text Request
SRAM cell stability has become a major concern in deep sub-micron technologies, due to the progressive increase in intra-die variability and VDD scaling. So we need a simple and efficient cell stability judging method to improve the circuit design time. Besides, the increase in integration has been achieved by miniaturizing the device structures. This miniaturization for increased packing density, however, has had unexpected effects, and the result has frequently been a loss of reliability for semiconductor memory devices, such as alpha-particle induced soft-errors.Firstly, this paper introduces the traditional method for judging the stability of SRAM cells, which is Static-Noise Margin (SNM). Then, we utilize a new N-curve metric to analyze the SRAM cell read stability and write-ability. We demonstrate that the N-curve contains information both on the voltage and current, thus allowing a complete functional analysis of the SRAM cell with only one N-curve. Additionally, it's easier to get N-curve than SNM by circuit simulation without any more mathematical manipulation of the measured data. So, the N-curve is applied to the traditional 6T SRAM cell to investigate how VDD , cell ratio (r) and pull-up ratio (q) impact the SRAM cell read stability and write-ability. The conclusions are: Higher VDD improves read stability but degrades write-ability, higher r improves read ability and lower q improves write-ability.To study the problem of alpha-particle injection induced Soft errors in SRAM cells, this paper firstly makes use of a inverter model to study the impact of alpha-particle injection on SRAM cell, and analyzes how the PMOS equivalent resistor and node capacitor parameters affect the output voltage of the inverter. Then the inverter's output is used as the input source of the SRAM cell circuit simulation, so we can clarify how an alpha-particle injection influences the stability of a bistable circuit in a SRAM cell. The alpha-particle injection is modeled by a single exponential current source. We find out that the cell soft error tolerance is higher with larger Critical Charge or node capacitor and smaller PMOS equivalent resistor.Finally, the N-curve is used to investigate the impact of alpha-particle injection on SRAM cell stability, and the result is that alpha-particle injection degrades read stability, but improves write-ability.
Keywords/Search Tags:SRAM cell stability, N-curve, alpha-particle, Soft Error
PDF Full Text Request
Related items