The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> RF Design >> a question about ft
https://designers-guide.org/forum/YaBB.pl?num=1146242099

Message started by frankaurora on Apr 28^th, 2006, 9:34am

Title: a question about ft
Post by frankaurora on Apr 28^th, 2006, 9:34am

to a given process such as 0.18u cmos,if L is a definite value,i think ft is constant,but some book say ft depend on bais condition,am i wrong?

thanks for replying :)

Title: Re: a question about ft
Post by milkdragon on Apr 28^th, 2006, 12:44pm

wt=gm/Cgs
where gm is porortional to your sqrt(Id)

So you will see ft change with bias

Title: Re: a question about ft
Post by Visjnoe on Apr 29^th, 2006, 4:59am

Hello,

ft ~ tech_const × (Vgs-Vt)/L²

So, ft only depends on the overdrive voltage of your device (bias) and its length. High frequency performance thus requires minimal length and large Vgs-Vt.

Kind Regards,

Peter

Title: Re: a question about ft
Post by frankaurora on Apr 29^th, 2006, 5:38am

but refer to Tomas Lee book,it says in short-channel device wt≈const*Esat/L(saturation),so wt is independent of bias?

Title: Re: a question about ft
Post by Visjnoe on May 1^st, 2006, 2:14am

Hi Frank,

the statement in Lee's book is correct, but it describes ft when the transistor is operating in velocity saturation,
so the dependency on Vgs-Vt disappears. This effect becomes important for deep-submicron technologies and is less apparent for older technologies (in which the formula I posted previously is valid).

Kind Regards

Peter

Title: Re: a question about ft
Post by frankaurora on May 6^th, 2006, 4:20am

:D
i got it,thanks a lot!