The Designer's Guide Community
Forum
Welcome, Guest. Please Login or Register. Please follow the Forum guidelines.
Aug 17th, 2024, 2:21am
Pages: 1
Send Topic Print
substrate connection of cascoding device (Read 3371 times)
Julian18
Community Member
***
Offline



Posts: 55

substrate connection of cascoding device
Mar 31st, 2008, 6:35am
 
Hi, there
    Seems this is too trivial, but ,I just want to know if you have triple-well process and free to connect the substrate to source, what would you do to connect the substrate of cascoding device, to the source or to the GND? what is the pros and cons if I just connect the substrate to the source.


Thanks.
Back to top
 
 
View Profile   IP Logged
vivkr
Community Fellow
*****
Offline



Posts: 780

Re: substrate connection of cascoding device
Reply #1 - Mar 31st, 2008, 7:07am
 
Hi Julian,

Excellent question. But you are talking of bulk node and not substrate. There is only 1 substrate on your chip.

Connecting bulk-source together

PROS:
1. More signal swing available.
2. Impact ionization reduced. This might be critical if you are using a process with L <1 um and trying to achieve very high output imepdance through the cascode, particularly if
you are gain boosting the cascodes. The additional current due to impact ionization will flow out of the bulk terminal, acting as an additional impedance in parallel to your cascode output
impedance. Shorting source-bulk eliminates the problem. Note that impact ionization is almost never modelled in most cases. A simple test is to sweep VDS for a large VGS value and look at the bulk current. If you see zero current or only a very small fixed current regardless of the drain potential, then you know you are up for it.

CONS:
1. Slightly reduced gain. In a cascode, the body effect increases output impedance (one of the few +ve benefits of the body effect).
2. More parasitics at the source node, specifically those related to the well in which the cascode device now sits. This can be important.

In any case, do not forget to properly bias the outer well with a clean supply.

Regards
Vivek
Back to top
 
 
View Profile   IP Logged
solidstate
Junior Member
**
Offline



Posts: 16
Munich, Germany
Re: substrate connection of cascoding device
Reply #2 - Apr 1st, 2008, 1:01pm
 
If you have more than one cascode (e.g. in a differential architecture), you might also consider that matching decreases if you connect bulk and source. Well-to-well spacing is much larger than device-to-device spacing, so more area is required and the wells will be at slightly different potential.
Back to top
 
 
View Profile   IP Logged
gsheng
New Member
*
Offline



Posts: 4

Re: substrate connection of cascoding device
Reply #3 - Apr 2nd, 2008, 12:14am
 
PSRR is better if bulk is connected to source.
Back to top
 
 
View Profile   IP Logged
loose-electron
Senior Fellow
******
Offline

Best Design Tool =
Capable Designers

Posts: 1638
San Diego California
Re: substrate connection of cascoding device
Reply #4 - Apr 17th, 2008, 3:42pm
 
Vivek had some very good points - with one other issue -- all of these seperate wells will grow the size quickly in the layout stages due to wide seperation of wells needed. That becomes both an area penalty and often a BW penalty due to the extra metal parasitics.
Back to top
 
 

Jerry Twomey
www.effectiveelectrons.com
Read My Electronic Design Column Here
Contract IC-PCB-System Design - Analog, Mixed Signal, RF & Medical
View Profile WWW   IP Logged
Pages: 1
Send Topic Print
Copyright 2002-2024 Designer’s Guide Consulting, Inc. Designer’s Guide® is a registered trademark of Designer’s Guide Consulting, Inc. All rights reserved. Send comments or questions to editor@designers-guide.org. Consider submitting a paper or model.