{"title":"将宽CMOS晶体管分解成最小尺寸栅极的好处","authors":"Hans Kristian Otnes Berge, S. Aunet","doi":"10.1109/NORCHP.2009.5397795","DOIUrl":null,"url":null,"abstract":"In this paper we show how decomposition of a wide CMOS transistor into a multi-finger FET with gates of minimum size can be beneficial for the reduction of delay and power-delay products in logic gates. This design possibility, which we call a minimum-split transistor (MST), seems to be largely overlooked in the literature. In a 90 nm CMOS process we compare the design to wide transistors. By exploiting the narrow-width effect, reduced parasitic capacitances from a shorter active channel and increased gate-drain spacing, we achieve up to 75–85% higher operation speed at a similar or reduced power consumption. The worst-case timing delay is reduced by 35–40% along with the nominal values. The design technique is considered valuable, in particular for critical time paths. The paper takes the perspective of subthreshold logic design at 200 mV, but the technique is also useful above threshold. A statistical experiment also investigates how Vth variation in MSTs changes with the number of parallell gates.","PeriodicalId":308859,"journal":{"name":"2009 NORCHIP","volume":"36 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2009-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"17","resultStr":"{\"title\":\"Benefits of decomposing wide CMOS transistors into minimum-size gates\",\"authors\":\"Hans Kristian Otnes Berge, S. Aunet\",\"doi\":\"10.1109/NORCHP.2009.5397795\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In this paper we show how decomposition of a wide CMOS transistor into a multi-finger FET with gates of minimum size can be beneficial for the reduction of delay and power-delay products in logic gates. This design possibility, which we call a minimum-split transistor (MST), seems to be largely overlooked in the literature. In a 90 nm CMOS process we compare the design to wide transistors. By exploiting the narrow-width effect, reduced parasitic capacitances from a shorter active channel and increased gate-drain spacing, we achieve up to 75–85% higher operation speed at a similar or reduced power consumption. The worst-case timing delay is reduced by 35–40% along with the nominal values. The design technique is considered valuable, in particular for critical time paths. The paper takes the perspective of subthreshold logic design at 200 mV, but the technique is also useful above threshold. A statistical experiment also investigates how Vth variation in MSTs changes with the number of parallell gates.\",\"PeriodicalId\":308859,\"journal\":{\"name\":\"2009 NORCHIP\",\"volume\":\"36 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2009-11-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"17\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2009 NORCHIP\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/NORCHP.2009.5397795\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2009 NORCHIP","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/NORCHP.2009.5397795","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Benefits of decomposing wide CMOS transistors into minimum-size gates
In this paper we show how decomposition of a wide CMOS transistor into a multi-finger FET with gates of minimum size can be beneficial for the reduction of delay and power-delay products in logic gates. This design possibility, which we call a minimum-split transistor (MST), seems to be largely overlooked in the literature. In a 90 nm CMOS process we compare the design to wide transistors. By exploiting the narrow-width effect, reduced parasitic capacitances from a shorter active channel and increased gate-drain spacing, we achieve up to 75–85% higher operation speed at a similar or reduced power consumption. The worst-case timing delay is reduced by 35–40% along with the nominal values. The design technique is considered valuable, in particular for critical time paths. The paper takes the perspective of subthreshold logic design at 200 mV, but the technique is also useful above threshold. A statistical experiment also investigates how Vth variation in MSTs changes with the number of parallell gates.