The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Modeling >> Semiconductor Devices >> Rds vs process
https://designers-guide.org/forum/YaBB.pl?num=1375742938

Message started by RobG on Aug 5^th, 2013, 3:48pm

Title: Rds vs process
Post by RobG on Aug 5^th, 2013, 3:48pm

Hello -
I understand that Rds (output impedance) of a MOSFET in saturation goes down with shorter channel length, but does going to a finer process necessarily decrease the impedance if the length doesn't change? For example, would a L=180 nm device in a 40 nm process have significantly higher (or lower) Rds than a L = 180 nm device in a 28 nm process?

Best regards,
Rob

Title: Re: Rds vs process
Post by raja.cedt on Aug 6^th, 2013, 8:29am

Dear Rob,
Yes, impedance drops from one generation to another scaled technology even for the same length, but i am not sure weather it is significant drop or not. I am not good at theory but all secondary effects like DIBL and velocity saturation will worsen with scaling.

Thanks,
raj.

Title: Re: Rds vs process
Post by RobG on Aug 6^th, 2013, 8:44am

Thanks - I talked with a process guy (admittedly not a fine line guy) and he thought the thinner oxide would give the gate more control over the channel which would counteract DIBL and channel length modulation.

Also, if it wasn't obvious, I am assuming same operating point (Id and Vgs-Vt) for the two transistors.

Title: Re: Rds vs process
Post by RobG on Aug 6^th, 2013, 8:49am

One more thought - why would DIBL change going from 40nm to 28nm if the respective devices had the same length (e.g. 160nm)? Does the doping change? If anything, I think the doping would be increased since it is even more important to reduce DIBL at 28nm.