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.

Why N-well Process but not P-well Prcoess????

Status
Not open for further replies.

scottieman

Full Member level 2
Joined
Aug 8, 2005
Messages
136
Helped
12
Reputation
24
Reaction score
1
Trophy points
1,298
Location
On Top of Silicon
Activity points
2,629
Hi all,

Suddenly have a question in my mind on CMOS technology. To have both NMOS and PMOS, we need a p-substrate for NMOS and a n-substrate for PMOS on the same silicon. That is why we do have a N-well in p-sub wafer.

However, why don't we do the opposite. Use n-sub with P-well to achieve the goal of CMOS.

Anyone can tell me the reasons behind it?

Thanks alot
Scottie
 

pratyusha

Junior Member level 1
Joined
Jan 12, 2006
Messages
16
Helped
1
Reputation
2
Reaction score
1
Trophy points
1,283
Activity points
1,452
one of the reason may be , in design we use multiple positive voltages and common ground. so entire psub is connected to gnd and n-well can be isolated and connected to multiple supplies . but if we take nsub how can we connect it to multiple supplies. then we have to use twin well process which is costly process.
so this may be one of the reason. i am glad to know the main reason .
 

Fom

Advanced Member level 2
Joined
Mar 10, 2004
Messages
633
Helped
84
Reputation
168
Reaction score
31
Trophy points
1,308
Location
Taiwan
Activity points
4,453
Some devices are manufactured with N-sub also.
For example some LCD controllers that charge pump negative high voltage.
 

pbs681

Full Member level 3
Joined
Aug 19, 2004
Messages
173
Helped
15
Reputation
30
Reaction score
3
Trophy points
1,298
Activity points
1,278
scottieman said:
Hi all,

Suddenly have a question in my mind on CMOS technology. To have both NMOS and PMOS, we need a p-substrate for NMOS and a n-substrate for PMOS on the same silicon. That is why we do have a N-well in p-sub wafer.

However, why don't we do the opposite. Use n-sub with P-well to achieve the goal of CMOS.

Anyone can tell me the reasons behind it?

Thanks alot
Scottie

We try to use nmos in our design if we have option between nmos and pmos. Why? one of the reason is that, the mobilty of the carrier for nmos is higher than pmos. So, with smalll w/l of nmos we can get what pmos can give. That means small real estate for nmos in IC which mean lower cost for product. There are few more reason in term of performance in which nmos is better. In fact, before we have CMOS tech, circuitry was made up by nmos. So, I don't see any good reason to use n-sub instead of p-sub.
 

scottieman

Full Member level 2
Joined
Aug 8, 2005
Messages
136
Helped
12
Reputation
24
Reaction score
1
Trophy points
1,298
Location
On Top of Silicon
Activity points
2,629
pbs681 said:
scottieman said:
Hi all,

Suddenly have a question in my mind on CMOS technology. To have both NMOS and PMOS, we need a p-substrate for NMOS and a n-substrate for PMOS on the same silicon. That is why we do have a N-well in p-sub wafer.

However, why don't we do the opposite. Use n-sub with P-well to achieve the goal of CMOS.

Anyone can tell me the reasons behind it?

Thanks alot
Scottie

We try to use nmos in our design if we have option between nmos and pmos. Why? one of the reason is that, the mobilty of the carrier for nmos is higher than pmos. So, with smalll w/l of nmos we can get what pmos can give. That means small real estate for nmos in IC which mean lower cost for product. There are few more reason in term of performance in which nmos is better. In fact, before we have CMOS tech, circuitry was made up by nmos. So, I don't see any good reason to use n-sub instead of p-sub.

Thanks.

Still, as I known from the web searching, alot of old technology were with n-sub rather than p-sub. That is what I want to ask, why people try to change from n-sub to nowadays p-sub. What I believe is there must be a resone or some reasons for this substrate change.

Scottie
 

derekqiao

Member level 2
Joined
Nov 11, 2004
Messages
42
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,286
Activity points
335
I agree with pratyusha, but the multiple operation voltage is the key reason.
 

jcpu

Full Member level 4
Joined
Dec 17, 2005
Messages
215
Helped
18
Reputation
36
Reaction score
6
Trophy points
1,298
Activity points
2,968
answer #1, when silicon crytal is pulled by one seed out of the furnance.
It's p- type, so it's cheapest to use p- wafer.

answer #2, Foundry implant boh Nwell and Pwell. Some need only draw Nwell, and they use reverse tone photo mask to implant Pwell.

answer #3, lots of people use n+ wafer for high speed logic, but that is more costly. And process would be slightly different.

In a hurry, bye!
 
Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top