※ 引述《windyddd (ddd)》之铭言： : ※ 引述《coa481 (手机快复活-_-||)》之铭言： : : 请教各位前辈: : : 在模拟OPA时，单就该OPA的特性做模拟时，如: : : 开回路增益、Phase Margin等，所得到的值 : : 与接成闭回路(假设直接由输出接回输入)所得到值，哪一个才是对的阿?? : : 我试的结果，当我用开回路模拟的话，Gain很小，很多MOS都在Liner，甚至有off的 : : 而接成闭回路後，不但Gain上升，PM也稳定约70度 : : 正常而言，应该是失败吧~可是又会被闭回路所看到的结果给迷惑了... : : 特别是有一位业界的朋友，说他们都直接模拟闭回路，这样我更迷惑了... : : 另外，以一个简单的two-stage OPA来讲，假设输入对用PMOS : : 那这2颗的body，可以直接接它们的source吗? : : (正常不是都要接到最正的电压?) : : 请帮我解惑~谢谢。 : I recommend you read Jacob Baker's book on Operational Amp. I : He clearly depicts every detail on testing open loop gain, : freq. response, compensation..so on so forth. : Good Luck Why do we need to know the phase margin? The system is unstable because it is a closed-loop system. That's why I said that it doesn't make sense to talk about the phase margin when the system is an open-loop system. We are suppposed to measure the phase margin of the closed-loop system using the bode plot of the loop gain rather than the open-loop gain or the closed-loop gain. When the feedback factor, beta, is equal to unity, it's a so-called worst case. Under this condition, the loop gain is equivalent to the open-loop gain. That's why somebody measures the phase margin of the closed-loop system based on the open-loop gain. They want to know the phase margin for the worst case. --

※ 发信站: 批踢踢实业坊(ptt.cc) ◆ From: 61.216.173.75