Misr International University Faculty of Engineering Electronics and Communications Department

MICROWIND LAYOUT EDITOR TOOL

TUTORIAL 2: FULL CUSTOM PHYSICAL IMPLEMENTATION OF THE INVERTER CIRCUIT USING LAMBDA DESIGN RULES
DIGITAL ELECTRONICS II COURSE

Prepared By Eng. Waleed El - Halwagy

CMOS Construction
As the CMOS is composed of NMOS and PMOS we should at first present their properties and construction.

NMOS Versus PMOS Properties and Construction

GND
Field Oxide p+ Tap n+

NMOS
Poly Silicon SiO2 Gate Oxide

VDD
Field Oxide n+ n+ Tap p+

PMOS
Poly Silicon SiO2 Gate Oxide

p+

P- Substrate 1. 2. 3. 4. 5. 6. The Substrate : P Type. The Drain and Source : n+ diffusion. The Tap Area : p+ . The p + Tap is connected to ground. The current flows from D to S. The Drain voltage > The Source voltage 1. 2. 3. 4. 5. 6.

N- Substrate The Substrate : N Type. The Drain and Source : p+ diffusion. The Select Area : n+ . The n + select is connected to VDD. The current flows from S to D. The Source voltage > The Drain voltage

CMOS Inverter Construction

Elevation View
VDD
Field Oxide
n+ Tap p+
Poly Silicon SiO2 Gate Oxide

OUT
Poly Silicon SiO2 Gate Oxide

GND
n+ n+ p+ Tap

p+

N- Well

P- Substrate

CMOS Inverter Construction

Top View Constructed Via Microwind 3.1

CMOS Inverter Construction

3D View Extracted Via Microwind 3.1

Fabrication Process of the CMOS Inverter

Mask 1: N-well Mask

N- Well

P- Substrate

Mask 2: Active Mask

N- Well

P- Substrate

Mask 3: Poly Silicon Mask

N- Well

P- Substrate

Mask 4: N+ Select Mask

n+ Tap

n+

n+

N- Well

P- Substrate

Mask 5: P+ select Mask

n+ Tap

p+

p+

n+

n+

p+ Tap

N- Well

P- Substrate

Mask 6: Contact Mask

n+ Tap

p+

p+

n+

n+

p+ Tap

N- Well

P- Substrate

Mask 7: Metal 1 Mask

n+ Tap

p+

p+

n+

n+

p+ Tap

N- Well

P- Substrate

The Design Rules and Layout

It provides a set of guidelines for constructing the various masks needed in the patterning process.

Scalable Design Rules : all dimensions are given as a function of

Micron Rules : all design rules are expressed in absolute dimensions

The Design Rules

It provides a set of guidelines for constructing the various masks needed in the patterning process

The Design Rules acts as the interface or even the contract between the circuit designer and the process engineer. Circuit designers generally want tighter, smaller designs, which lead to higher performance and higher circuit density. The process engineer on the other hand, wants a reproducible and high yield process. Consequently, design rules are a compromise that attempts to satisfy both sides. Design rules consists of:
Minimum width requirements. Minimum spacing requirements. Minimum Surface requirements. Requirements between objects on the same or different layers.

The Design Rules

Scalable Design Rules : all dimensions are given as a function of

Even for the same minimum dimension, design rules tend to differ from company to company, and from process to process. This makes porting an existing design between different processes a time consuming task. To address this issue we can use the scalable design rules, which defines all the design rules as a function of a single parameter . Scaling of the minimum dimension is accomplished by simply changing the value of . This results in a linear scaling of all dimensions. For a given process, is set to a specific value and all design dimensions are consequently translated into absolute numbers. Minimum Feature size = 2

The Design Rules

Disadvantages of the Scalable design rules approach

Linear scaling is only possible over a limited range of dimensions (for example between 0.25 m and 0.18 m). When scaling over large ranges, the relations between different layers tend to vary in a non-linear way that can not be converted by the linear scaling rules. Scalable design rules are conservative, they represent a cross section over different technologies, and they must represent the worst case rules for the whole set. This results in over dimensioned and less dense designs. For these and other reasons, scalable design rules normally are avoided by industry. (while not entirely accurate, the lambda rules are still useful to estimate the impact of a technology scale on the design area).

The Design Rules

Micron Rules: all design rules are expressed in absolute dimensions

As circuit density is a prime goal in industrial designs, most semiconductor companies tend to use the micron rules, which express all design rules in absolute dimensions and thus can exploit the features of a given process to a maximum degree. Scaling and porting designs between technologies under these rules is more demanding and has to be performed either manually or by using advanced CAD tools.

What is a Layout ?

A layout consists of a combination of polygons, each of which is attached to a certain layer. The functionality of the circuit is determined by the choice of the layers, as well as the interplay between objects on different layers. A transistor A MOS transistor is formed by the cross section of the diffusion layer and the poly silicon layer.

The Objective Is to design the minimum sized inverter in the 0.18 m technology and develop its seven masks.

DESIGN AN 0.18 m technology CMOS INVERTER ( 2 = 0.18 m )

The technology is specified by its minimum line width (minimum feature size) which is usually taken as the channel length of the transistor and it is denoted by 2

Design using minimum sized NMOS and take the W PMOS = 3 WNMOS and take the channel length of both transistors as 2

First : Select the Design Technology from the Microwind

To get the Design Rules of the chosen technology

First : The minimum sized NMOS Transistor Design

GND
Field Oxide
p+ Tap n+
Poly Silicon

SiO2 Gate Oxide

n+

P- Substrate

First: Design the minimum sized NMOS transistor.

What do we need to construct an NMOS transistor ?

P

substrate Poly silicon for the gate. n+ diffusion regions for the drain and source. P+ tap region that is connected to ground. Contacts to connect the active area with the metal layer.

Remark:
The

Microwind assumes the Silicon ignot used in fabrication is doped with Boron, that is its a P-type silicon ignot .

What are the design rules we need to know to be able to construct the NMOS ?
The Minimum Poly Width = 2 The Minimum extra Poly surrounding the n diffusion = 3 The Minimum Poly Area = 16 2

The Minimum extra n diffusion surrounding the poly = 4

The Minimum n diffusion Width = 4 The Minimum n diffusion Area = 16 2

The Minimum extra n diffusion surrounding the contact = 2 The Minimum spacing between the contact and the Poly = 3
The Minimum Contact Width = 2

What are the design rules we need to know to be able to construct the NMOS ?

The Minimum p diffusion Width = 4 The Minimum spacing between contacts = 4 The Minimum p diffusion Area = 16 2

How to Check that the design Rules of the layout are satisfied ?

Design Rule Check ( DRC )

If any of these design rules are not satisfied, the Microwind will signify it.

If any of these design rules are not satisfied, the Microwind will signify it.

Now what is the minimum size and area of the NMOS Transistor in the 0.18 m technology ?

L : is the length of the poly silicon. W : is the width of the n+ diffusion region. From the design rules:
Minimum

poly silicon length = 2 . Minimum n+ diffusion width = 4 . Taking into consideration that the minimum extra poly surrounding the n+ diffusion is 3 . Taking into consideration that the minimum extra n+ diffusion surrounding the poly is 4 .

Now what is the minimum size of the NMOS Transistor in the 0.18 m technology ?
L=2 W=4 Area = 10 x 10
The Minimum extra Poly surrounding the n diffusion = 3

The Minimum extra n diffusion surrounding the poly = 4

The Minimum Poly Width L = 2

The Minimum extra n diffusion surrounding the poly = 4

The Minimum n diffusion Width W= 4

The Minimum extra Poly surrounding the n diffusion = 3

Now lets add the contacts.

The contacts Design Rules:

Minimum contact width = 2 Minimum extra diffusion surrounding the contact = 2. Minimum spacing between the poly and contact = 3.

W = 4 2

LLDD = =44
2

As we can see that the last two design rules are not satisfied when LD = 4 and W = 4 , so we have to expand them as follows :

L D = 3 (poly contact) + 2 (contact width)+ 2 (contact diff.) = 7 W = 2 (contact diff) + 2 (contact width)+ 2 (contact diff.) = 6

Now lets add the contacts.

L = 2 W=6 Area = 16 x 12

Extra Poly 3
Contact diff= 2 W = 6

Contact width= 2
Contact diff= 2

Contact diff 2

Contact Poly width contact 2 3

Poly Width L =2

Poly contact 3

Contact width 2

Contact diff 2

Extra Poly 3

Optimizing the Aspect ratio and area of the NMOS Transistor

From the above discussion we conclude that:

Minimum L = minimum poly width = 2. Minimum W = 2 ( minimum contact - diff) + minimum contact width = 6 .

The Question now is: Can we optimize this aspect ratio ?

We can not decrease L because we can not implement a dimension that is less than 2. But from the design rules, the minimum diffusion width is 4. We were forced to implement it as 6 due to the contact design rules constraints.

By a small trick we could let W = 4 without altering the contact design rules constraints. This can be accomplished by reducing the diffusion width to 4 at the poly silicon surface and widening it in the region surrounding the contact.

Optimizing the Aspect ratio and area of the NMOS Transistor

7 L =2 Extra Poly 3 7

diffusion width at the poly W = 4

Diffusion width at the contact 6

L = 2 W=4 Area = 16 x 10

Extra Poly 3

Adding the P-Tap region

Taking into consideration the contact design rules restrictions Area of p tap = 6 x 6
L = 2 W=4 Area = 22 x 10 Extra Poly 3

2
Contact 2 diff

Diffusion width at the poly W = 4

Contact Contact 4

Contact 2 width Contact 2 diff

Extra Poly 3
Contact Contact diff width 2 2 Poly contact 3 Poly Width L =2 Poly Contact Contact Contact Contact Contact contact width diff diff width diff 2 2 3 2 2 2

Second : The PMOS Transistor Design LPMOS = 2 and W PMOS = 3 WNMOS

VDD
Field Oxide
n+ Tap p+
Poly Silicon SiO2 Gate Oxide

p+

N- Substrate

Now Design The PMOS with LPMOS = 2 and W PMOS = 3 WNMOS

What do we need to construct a PMOS transistor ?

N well Poly silicon for the gate. p+ diffusion regions for the drain and source. n+ tap region that is connected to ground. Contacts to connect the active area with the metal layer.

There are some additional design rules that we will need to know concerning the N-well before we proceed in our design of the PMOS. The design rules concerning with the minimum width and area of the diffusion as well as the relation between the poly and the contacts with the diffusion are the same for both n-type and p-type.

Design Rules Concerning the N-well

The minimum extra n-well surrounding the p diffusion = 2
The minimum extra n-well surrounding the p diffusion = 6 The Minimum n-well Area = 144 2

The Minimum n-well Width = 10

The Number of Contacts in the PMOS

As a rule: the more the number of contacts, the better the performance. This is because the resistance through which the current flows will decrease. The following equation gives a relation between the diffusion width and the number of contacts.
W = 2 ( contact diff.) + N ( contact width ) + ( N 1 ) ( contact contact )

W 18

Example : W = 18 18 = 2 ( 2 )+ N ( 2 ) + ( N 1 ) ( 4 ) ------- N = 3 contacts Example : W = 17 17 = 2 ( 2 )+ N ( 2 ) + ( N 1 ) ( 4 ) -------- N = 2.83, that is N = 2 contacts

2 2 2 2 2 2

L 2

3 2 2

2 2
2

4 2 2

W 12

24

6 3
30

L = 2 W = 12 Area = 30 x 24

The CMOS is Constructed by connecting the NMOS and PMOS transistors together VDD
Field Oxide n+ p+ Tap
Poly Silicon SiO2 Gate Oxide

OUT
Poly Silicon SiO2 Gate Oxide

GND
n+ n+
p+ Tap

p+

N- Well

P- Substrate

The CMOS is constructed by interconnecting both the NMOS and PMOS

There are some additional design rule that must be taken in consideration when interconnecting the NMOS and PMOS to construct the CMOS Inverter.

The minimum spacing between the n-well of the PMOS and the n+ diffusion (drain) of the NMOS = 6 .

The metal used in the connections

Minimum metal width = 6 Minimum metal surface = 16 Minimum Spacing between metal layers = 4 Minimum extra metal surrounding the contact = 2

We need metal to

NMOS: Connect the p+ tap with the source to ground PMOS: Connect the n+ select with the source to VDD Connect the NMOS and PMOS drains to the output. Connect the poly to the input.

NMOS : L = 2 W = 4 PMOS : L = 2 W = 12 CMOS Area = 58 x 24

The minimum metal spacing = 4

The minimum extra metal surrounding the contact = 2

The minimum spacing between the n-well and n-diffusion = 6

The Seven Masks Extraction

After we have finished the CMOS layout design and computed its dimensions, we are ready to extract the seven inverter masks to send them to the Fab to be manufactured.

CMOS Inverter Layout

Mask 1 : N well Mask

Mask 2 : Active Area Mask

Mask 3 : Poly Silicon Mask

Mask 4 : P + Select Mask

Mask 5 : N + Select Mask

Mask 6 : Contacts Mask

Mask 7 : Metal Mask

The Simulation Results

Transient Analysis with no external capacitance at the output

Transient Analysis with 0.05 pf external capacitance at the output

VTC Characteristics of the inverter