Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.
Of course, it does matter. Because the above rule is one of many rules that you should satisfy to have a good matching. Common centroid is in addition to the above rule and it is for eliminating first-order temprature and process variation across the wafer.
Of course, it does matter. Because the above rule is one of many rules that you should satisfy to have a good matching. Common centroid is in addition to the above rule and it is for eliminating first-order temprature and process variation across the wafer.
If I want to arrange my differential pair transistor as above, I have to discontinue the active for each block. Because if i don't do that, the current will flow in opposite direction each other. Am I Right?
I would prefer adding a dummy polt in between the "improved" version to improve profiling between the diffusion. also able to complete the diffusion ends with sharp changes from diffusion. I would also add dummy magnet-like POLY to improve POLY end mis-matches.
hi ryusgnal - please read post on "Matching Layout Techniques" about dummy transistor...but if hung_wai_ming can explain further the magnet-like poly the better.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
