How to make gm/id plot in Cadence Virtuoso ADE (English pronunciation)

In Cadence IC6.1.8, you can use the "calculator function" to plot gm/id vs id/W and gm/id vs gm*ro without much effort.
This video shows the complete simulation flow. Hope it helps you.
NMOS in gm/id plot:
• How to make gm/id plot...

[@ashwiniganesh662]

Thank you for your video. Looking forward to see more of your tutorial in english as well.

[@strotta]

Great tutorial. Thank you so much!

[@samvenker3137]

Also note gmoverid is a BSIM param and NOT equivalent to gm over id if plotted separately

[@007viviadam]

Why is that?

[@Gruftgrabbler]

Thanks for doing this video. I don't particularly understand why a ID/W Simulation is of interest. When Designing the circuit one never types directly a value for W instead always W/L.
Is there an explanation available to some of you why this type of simulation was done?
Thank you

[@AhmedAli-gj8os]

Vov or GM/ID or ID/W they all represent the bias point, for example in RF AND mm-wave designs they prefer to use ID/W

[@Pathfinder-vz3be]

Hello, thank you for sharing your video.
is there any reason to choose W=10um initially?
just selection for lambda(Vds coefficient) area?

[@pablomarco5118]

Thank you very much

[@jaredpowell1476]

Thank you!!

[@skgupta563]

Hello
Why are you sweeping L , we have to fix L for a technology, right?
And I don't get how to calculated W from this experiment.

[@pravinmane4408]

Can you please upload other videos in English? Thank you for video

[@xxggww8740]

How to use gm/id methodology to design circuit with FinFet process?
In FinFet, W(width) is a function of number of fins. Just get gm/id vs id/(number of fins)? or gm/id vs id/(number of fingers)? or gm/id vs id/multiplier?

[@ericyeh3787]

In finfet, since the current is no longer linear with width or Nfin, it is difficult to use id/W or id/Nfin to calculate the normalized current density.
I currently use the id/multiplier, but the disadvantage is that I must fix a Nfin to every finfets. I am also studying related topics. If I have new ideas, I will reply to you.

[@xxggww8740]

@@ericyeh3787 How do you consider the length of finfet? Do you also sweep length? Do you use stack of mos transistor to realize larger length? If using stack of mos transitors, how to decide the stack number(say, stack=4, 5, 6...), how to make gm/id plot? A video tutorial will be helpful. Thanks!

[@9935403017]

how to find expression for gm/W in calculator?

[@wendorisy2557]

你好老师 想问下能用这个方法将ft和gm/id的曲线图画出来吗？ 我这里尝试了下一直显示eval err ac仿真和sin波都有加 不知道问题出在了哪里

[@samvenker3137]

Additionally you are assuming OS params are available in every PDK, but this is not the case. It would be better to create the expressions with getInfo("{instance}" result? "DcOpInfo") as it is more general

[@ericyeh3787]

Thanks for sharing, good to know this information.

[@taofu7486]

一旦gm/id确定后,管子的Vgs也要随着gm/id改变吧。那在design的时候一般是先确定gm/id再确定Vgs的吗

[@ericyeh3787]

使用gm/id最大的好處是你不再需要執著於絕對的vgs 是多少？而是只要給定準確的電流就能保證gm的數值。

[@СоломияНовицкая]

Specify the goal and need money

[@CC-ec1il]

老师我来催更了 (⊙v⊙)

[@ericyeh3787]

我有錄製一段新的設計是關於constant bias的，但我剛剛注意到你有提問fold cascode的設計，這我下一集來談一下

[@CC-ec1il]

@@ericyeh3787 嘿嘿 Thanks~

[@CC-ec1il]

​@@ericyeh3787 请问老师有把constant bias的视频上传吗？ 也想学习一下。

[@ericyeh3787]

constant gm bias：
ruclips.net/video/vngAxZ4CFi8/видео.html
wide swing current bias：
ruclips.net/video/_E-8AtA2ho0/видео.html

[@CC-ec1il]

@@ericyeh3787 感谢老师~