J
Jagadeesh M
Guest
Folks,
We have a situation where we have a huge POWER transistors. The
width(Width-per-finger) of the transistors is huge, by default RCX models
each finger as a single MOS device. It will collapse all the
Metal-to-contact resistors and it will connect the the center point of src/drain
metal1 to the transistor. This method of extraction is fine for a normal
application. But for huge power transistors this will be inaccurate since you
are modeling the MOS with big W(let's assume W=200um) as a single device. The
best way to model it is to break a single finger of transistor into multiple
devices along its width so that the single MOS is modelled as a distributed
device. This will be a accurate way to model this. How can acheive this using
RCX ? Let me know if you need mode details.
Rgds,
Jagi.
We have a situation where we have a huge POWER transistors. The
width(Width-per-finger) of the transistors is huge, by default RCX models
each finger as a single MOS device. It will collapse all the
Metal-to-contact resistors and it will connect the the center point of src/drain
metal1 to the transistor. This method of extraction is fine for a normal
application. But for huge power transistors this will be inaccurate since you
are modeling the MOS with big W(let's assume W=200um) as a single device. The
best way to model it is to break a single finger of transistor into multiple
devices along its width so that the single MOS is modelled as a distributed
device. This will be a accurate way to model this. How can acheive this using
RCX ? Let me know if you need mode details.
Rgds,
Jagi.