d******a
发帖数: 57 | 1 Cutting the point at V1 and applying a test incremental voltage at the gate
of M8, Loop gain should be gm8/gm9/(1+gm8Rs)
For stability, loop gain should be less than 1, So I figure that in practice
, when Rs is an off chip resistor, one needs to choose a large enough one...
Why in Martin&Johns book, they say M8 should be larger than M9 in order to
keep the loop gain <1? I thought keeping M8 larger is only for a reasonable
gm9
as gm9 is [1-sqrt(m)]/Rs where m=(W/L)8/(W/L)9
Please help | j*****q
发帖数: 33 | 2 gm9 is 2*[1-sqrt(m)]/Rs?
loop gain和gm是不是两个衡量的标准？如果提供直流的gm稳定，和loop gain有很大关
系吗？
你的loop gain是怎么算出来的啊？
gate
practice
..
reasonable
【在 d******a 的大作中提到】
: Cutting the point at V1 and applying a test incremental voltage at the gate
: of M8, Loop gain should be gm8/gm9/(1+gm8Rs)
: For stability, loop gain should be less than 1, So I figure that in practice
: , when Rs is an off chip resistor, one needs to choose a large enough one...
: Why in Martin&Johns book, they say M8 should be larger than M9 in order to
: keep the loop gain <1? I thought keeping M8 larger is only for a reasonable
: gm9
: as gm9 is [1-sqrt(m)]/Rs where m=(W/L)8/(W/L)9
: Please help
|
|
