Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

A problem of cascade of inverter driver!

Status
Not open for further replies.

gdhp

Advanced Member level 4
Joined
Jan 7, 2005
Messages
116
Helped
2
Reputation
4
Reaction score
1
Trophy points
1,298
Activity points
931
cascaded inverters as drivers

Hi all
i am designing a driver using cascade of inverters to drive a large capcitance

(20pf)! But the delay is too large beyond our requirements. The delay should be

less than 4.5ns. the power is 1.5v and i use the p35(n35) MOSFET.

So any other method to do the same thing?

i use the UMC0.25u. i want to know if the p25(n25) MOSFET can work at 3.3v

power.

Thank you all!!
 

increasing w/l is no a good method. because increase last inverter , also increase the total delay!
 

My experience is that beyond 3 levels, the cascade doesnt work as the delay introduced by itself offsets the delay due to output cap. One method could be to progressively size the inverters in the cascade so as to reach the desired drive...
 

increase the current!
at the same time, use the min L
 

is there any other method to do the same things?
 

this ppt will help u .this is by rabaey he gave an explanation for sizing a chain of inverters for optimizing performance.if sizing a chain of inverters didn't work use super buffers
 

Why cascade structure? It's always slower than comparable normal structure because of more nodes to charge/discharge.

If you really want to use cascade structure, you can design the top pmos and bottom nmos in weak inversion mode at quiescent state. One of them will be brought into strong inversion mode which allows large current source or sink for load capacitor.
 

I want to know why use cascode inverter ?
 

because you r designing a driver, if the input impedence is fixed and the load is 20pf. probably your only choice is increase the bias current. it is because the problem is introduced by the low gain or low driving capabilty of each stage.

it is very hard now, because you may be limited by the capability of the technology.

if the signal bandwidth is not large, you may be able to use inductive peaking technique to increase the gain, hence use less stages.

Added after 3 minutes:

there may be some technique or circui will increase the gain a little bit.

try using negtive resistance to increase the gain of each stage.

it might help.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top