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.

Register Log in

IS this structure can work as common centriod and interdigitized

Status
Not open for further replies.
MahmoudHassan

MahmoudHassan

Full Member level 6
Joined
Oct 4, 2010
Messages
349
Helped
44
Reputation
90
Reaction score
40
Trophy points
1,328
Activity points
3,913
Hello

I have two transistors of 6 multipliers each and another transistor (named X) that should be matched with them with multiplier 2
I am trying to make them common centroid and interdigitized tottally
the center of the Y axis is through the X transistor is this OK

Capture.JPG
 

I think it's fine.

But don't use mirror symmetry, use repeat symmetry: all sources at the left, all drains at the right side - or the other way round.
 
Many thanks erikl
But why don't use mirror symmetry and use repeat symmetry, what is the problem of that?
 


Erilk,

Do you know the when Maloberti created that document?
 

erikl said:
The asymmetry between source & drain generation due to the necessary 7° ion implant to avoid channeling.

See p.9 of Maloberti's Analog CMOS Layout tutorial: View attachment 145121
Click to expand...

I think this is why we should mirror them in common centroid rather than using a repeated pattern
I mean I will not interchange Source and Drain connections but I will mirror so that the common centroid is more symmetric
 

How do you propose laying out the 2x current mirror? You would have to have one facing left and one facing right which will make your interconnect a little more complicated. Stepping them L-R will be the easiest and arguably the best layout. Alternatively, you may consider rotating your transistor 90 degrees in one row and sharing the gates. Some believe that a continuous poly gate will result in better matching.
 
2X is not a current mirror it is a bias voltage to the base of two transistors and it is being biased using a current from the biasing current
I liked the idea of rotating 90 degrees the first time I hear about it
 

MahmoudHassan said:
I liked the idea of rotating 90 degrees
Click to expand...

This is better anyway. I just thought you need this crazy aspect ratio - sometimes you need such an aspect ratio to save silicon area in a larger cell layout.

Don't forget L & R dummies in this rotated case.
 
Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top