Lect.20.comments

not ground 2 it clearly changes vt and gm terms make

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: ues (and how to simplify) •  You need to simplify! •  Neglect ro except for current sources (i.e. it will not affect your signal path modeling) •  Body effect maXers: 1) bulk node of NMOS goes to most negaZve supply!! (not ground); 2) it clearly changes VT and gm terms (make first cut approximaZons, then let SPICE help you refine them, but don’t have VSB terms!!) •  Min. width is 2µm for Current Sources. Other devices can be 1µm. That means that raZo- metric scaling will be your main choice in geSng voltage gain (you don’t need a lot ;); but you can increase L in the loads (see ωT discussion above) Comments about gm and Design •  Recall: gm=2*IDrain/VOV •  IDrain is what we have to worry about for Power! •  VOV is also a design variable, both in the sense of a circuit variable (knob) we can vary and as a layout variable (knob) that we can scale. But remember… •  As you vary W/L you do change your biasing (and that’s why you...
View Full Document

Ask a homework question - tutors are online