Analogowe układy CMOS
Analogowe układy CMOS
w technice VLSI
w technice VLSI
Analogowe układy CMOS
Analogowe układy CMOS
w technice VLSI
w technice VLSI
w
VL
w
VL
w
VL
w
VL
W.Kucewicz
Analogowe układy CMOS w technice VLSI
1
Analogowe układy CMOS
Analogowe układy CMOS
w technice VLSI
w technice VLSI -- part II
part II
Analogowe układy CMOS
Analogowe układy CMOS
w technice VLSI
w technice VLSI -- part II
part II
w
VL
w
VL
pp
w
VL
w
VL
pp
El t ni
El t ni s Hist
s Hist
El t ni
El t ni s Hist
s Hist
Electronic
Electronics History
s History
Electronic
Electronics History
s History
&
& Progress
Progress
&
& Progress
Progress
gggg
W.Kucewicz
Analogowe układy CMOS w technice VLSI
2
Contents
Contents
Contents
Contents
11 Transistor
Transistor S
Story
tory
11 Transistor
Transistor S
Story
tory
1.
1. Transistor
Transistor S
Story
tory
22 Pro ress in the
Pro ress in the
1.
1. Transistor
Transistor S
Story
tory
22 Pro ress in the
Pro ress in the
2.
2. Progress in the
Progress in the
VLSI P d ti
VLSI P d ti
2.
2. Progress in the
Progress in the
VLSI P d ti
VLSI P d ti
VLSI Production
VLSI Production
VLSI Production
VLSI Production
W.Kucewicz
Analogowe układy CMOS w technice VLSI
3
Transistor story
Transistor story
Transistor story
Transistor story
Transistor story
Transistor story
Transistor story
Transistor story
W.Kucewicz
Analogowe układy CMOS w technice VLSI
4
Electron Discovery
Electron Discovery -- 1898
1898
Electron Discovery
Electron Discovery -- 1898
1898
E
D
y
E
D
y
99
E
D
y
E
D
y
99
"To the electron
"To the electron -- may it never be of any use to anybody."
may it never be of any use to anybody."
-- JJ. Thomson's favorite toast
JJ. Thomson's favorite toast
W.Kucewicz
Analogowe układy CMOS w technice VLSI
5
Electron Discovery
Electron Discovery -- 1898
1898
Electron Discovery
Electron Discovery -- 1898
1898
E
D
y
E
D
y
99
E
D
y
E
D
y
99
"To the electron
"To the electron -- may it never be of any use to anybody."
may it never be of any use to anybody."
-- JJ. Thomson's favorite toast
JJ. Thomson's favorite toast
1 Elektron
1 Elektron Î
Î
1 6
1 6 ·10
·10
--19
19
CC
1 Elektron
1 Elektron Î
Î
1 6
1 6 ·10
·10
--19
19
CC
1 Elektron
1 Elektron Î
Î
1,6
1,6 10
10
9
9
CC
1 Elektron
1 Elektron Î
Î
9 1·10
9 1·10
--31
31
kg
kg
1 Elektron
1 Elektron Î
Î
1,6
1,6 10
10
9
9
CC
1 Elektron
1 Elektron Î
Î
9 1·10
9 1·10
--31
31
kg
kg
1 Elektron
1 Elektron Î
Î
9,1 10
9,1 10 kg
kg
1 Elektron
1 Elektron Î
Î
511
511 keV/c
keV/c
22
1 Elektron
1 Elektron Î
Î
9,1 10
9,1 10 kg
kg
1 Elektron
1 Elektron Î
Î
511
511 keV/c
keV/c
22
W.Kucewicz
Analogowe układy CMOS w technice VLSI
6
Mechanical Computer
Mechanical Computer
Mechanical Computer
Mechanical Computer
M
mpu
M
mpu
M
mpu
M
mpu
The Babbage Engine, developed in 1834, was
perceived as a general-purpose computing
perceived as a general purpose computing
machine, with features strikingly close to
modern computers. Besides executing the
basic operations (addition, subtraction,
basic operations (addition, subtraction,
multiplication, and division) in arbitrary
sequences, the machine operated in a two-
cycle sequence, called “store” and “mill”
y
q
,
(execute), similar to current computers.
Unfortunately, the complexity and the cost of
y
p
y
the designs made the concept impractical.
For instance, the design of Difference Engine
I (part of which is shown) required
25,000
25,000
h
l
l
f £
h
l
l
f £
mechanical parts at a total cost of £17,470
mechanical parts at a total cost of £17,470
(in 1834!).
(in 1834!).
Babbage Difference Engine
Babbage Difference Engine
W.Kucewicz
Analogowe układy CMOS w technice VLSI
7
1945
1945 -- First Computer
First Computer –– ENIAC
ENIAC
1945
1945 -- First Computer
First Computer –– ENIAC
ENIAC
99
F
mpu
F
mpu
EN
EN
99
F
mpu
F
mpu
EN
EN
ENIAC
ENIAC
--
EE
lectronic
lectronic
N
N
umerical
umerical
II
ntegrator
ntegrator
A
A
nalyzer and
nalyzer and
CC
omputer
omputer
Filling up a 30 X 50 foot room, ENIAC was made of 17, 468 vacuum
Filling up a 30 X 50 foot room, ENIAC was made of 17, 468 vacuum
tubes,70,000 resistors, and 10,000 capacitors
tubes,70,000 resistors, and 10,000 capacitors --
-- not to mention all
not to mention all
those lights and switches
those lights and switches Most importantly
Most importantly the metal giant could
the metal giant could
W.Kucewicz
Analogowe układy CMOS w technice VLSI
8
those lights and switches.
those lights and switches. Most importantly,
Most importantly, the metal giant could
the metal giant could
add 5,000 numbers in a single second.
add 5,000 numbers in a single second.
Bipolar transistor history
Bipolar transistor history
Bipolar transistor history
Bipolar transistor history
p
y
p
y
p
y
p
y
1940
1940
1940
1940
Ohl develops pn junction (Bell Lab)
Ohl develops pn junction (Bell Lab)
1940
1940
1940
1940
Ohl develops pn junction (Bell Lab)
Ohl develops pn junction (Bell Lab)
1945
1945
1945
1945
1945
1945
1945
1945
Established Shockley’s Lab (Bell Lab)
Established Shockley’s Lab (Bell Lab)
1947
1947
1947
1947
Bardeen and Brattain invent
Bardeen and Brattain invent
1947
1947
1947
1947
Bardeen and Brattain invent
Bardeen and Brattain invent
transistor (Bell Lab)
transistor (Bell Lab)
W.Kucewicz
Analogowe układy CMOS w technice VLSI
9
Bipolar transistor history
Bipolar transistor history
Bipolar transistor history
Bipolar transistor history
The transistor was invented in 1947 at Bell Labs by the team of John Bardeen, Walter
The transistor was invented in 1947 at Bell Labs by the team of John Bardeen, Walter
Brattain, and William Shockley, for which they received the Nobel Prize
Brattain, and William Shockley, for which they received the Nobel Prize (1956)
(1956). .
The first transistor was about half an inch high
p
y
p
y
p
y
p
y
The first transistor was about half an inch high.
W.Kucewicz
Analogowe układy CMOS w technice VLSI
10
Bipolar Transistor History
Bipolar Transistor History
Bipolar Transistor History
Bipolar Transistor History
p
H
y
p
H
y
p
H
y
p
H
y
The first transistor was a germanium point
The first transistor was a germanium point--contact
contact
The first transistor was a germanium point
The first transistor was a germanium point contact
contact
transistor consisting of two thin electrodes in point
transistor consisting of two thin electrodes in point--
contact with the surface of a piece of germanium and
contact with the surface of a piece of germanium and
with a third wire attached to the base.
with a third wire attached to the base.
B
h d
l
f
ld
f l
h
Brattain attached a single strip of gold foil over the
point of a plastic triangle. With a razor blade, he sliced
through the gold right at the tip of the triangle. Voila:
two gold contacts just a hair-width apart
two gold contacts just a hair width apart.
The whole triangle was then held over a crystal of
germanium on a spring, so that the contacts lightly
touched the surface. When a bit of current came
h
h
f h
ld
h
through one of the gold contacts, another even
stronger current came out the other contact.
Emitter
Emitter
Collector
Collector
Ge
Ge
W.Kucewicz
Analogowe układy CMOS w technice VLSI
11
Base
Base
Bipolar Transistor History
Bipolar Transistor History
Bipolar Transistor History
Bipolar Transistor History
p
H
y
p
H
y
p
H
y
p
H
y
J. Bardeen and W. H. Brattain. The transistor, a semiconductor triode.
Physical Review
, 74:230, July 15 1948
Tran
Tran
sient
sient Re
Re
sistor
sistor
Tran
Tran
sient
sient Re
Re
sistor
sistor
T
i
T
i
Transistor
Transistor
W.Kucewicz
Analogowe układy CMOS w technice VLSI
12
Bipolar transistor
Bipolar transistor
Bipolar transistor
Bipolar transistor
pppp
License price 25 k$
License price 25 k$
W.Kucewicz
Analogowe układy CMOS w technice VLSI
13
Bipolar Junction Transistor
Bipolar Junction Transistor -- 195
19511
Bipolar Junction Transistor
Bipolar Junction Transistor -- 195
19511
p
Ju
p
Ju
99
p
Ju
p
Ju
99
Emitter
Emitter
Base
Base
nn
pp
Collector
Collector
nn
Shockley develops junction transistor which can be
Shockley develops junction transistor which can be
maufactured in quantity
maufactured in quantity
W.Kucewicz
Analogowe układy CMOS w technice VLSI
14
1958: Invention of the Integrated
1958: Invention of the Integrated
Circuit
Circuit
1958: Invention of the Integrated
1958: Invention of the Integrated
Circuit
Circuit
uuuu
Jack Kilby from Texas Instruments.
Jack Kilby from Texas Instruments.
N b l P i i 2000
N b l P i i 2000
Nobel Price in 2000
Nobel Price in 2000
1 Transistor and 4
1 Transistor and 4 oother Devices on 1 Chip
ther Devices on 1 Chip
The reason integrated chips are possible
The reason integrated chips are possible
at all is because engineers learned ways to
at all is because engineers learned ways to
build layers, making m
build layers, making miillions of transistors
llions of transistors
W.Kucewicz
Analogowe układy CMOS w technice VLSI
15
build layers, making m
build layers, making miillions of transistors
llions of transistors
across the chip all at the same time.
across the chip all at the same time.
1958: Invention of the Integrated
1958: Invention of the Integrated
Circuit
Circuit
1958: Invention of the Integrated
1958: Invention of the Integrated
Circuit
Circuit
uuuu
Kilby’s notebook pages -summer 1958, Texas Instruments
Integrated circuit patent
Integrated circuit patent
Integrated circuit patent
Integrated circuit patent
W.Kucewicz
Analogowe układy CMOS w technice VLSI
16
195
19533:
: Darlington Patent
Darlington Patent
195
19533:
: Darlington Patent
Darlington Patent
99
D
g
D
g
99
D
g
D
g
“Just after the transistor was invented at Bell Labs, Sidney
checked out for the weekend two of the few existing
transistors from the head of Bell Labs. Transistors were
not generally available and the head of the Labs kept the
few that had been made in his desk. Sidney played with
them at home on the weekend and discovered/invented
the Darlington pair. He realized that they could be put
th Dar ngt n pa r. H r a z that th y c u put
in one package (“on one chip”), and that in fact any
number of transistors could be put in one package. The
next week he was encouraged to have the lawyers draw
up the patent application.
He said it should be written
He said it should be written
for any number in one package but the lawyers only
for any number in one package but the lawyers only
for any number in one package, but the lawyers only
for any number in one package, but the lawyers only
wanted to do it for two
wanted to do it for two—
—which is what was applied for.
which is what was applied for.
As it turned out, if it had not been restricted to two
As it turned out, if it had not been restricted to two
transistors, Bell Labs and Dr. Darlington would receive
transistors, Bell Labs and Dr. Darlington would receive a
a
royalty on every IC chip made today!
royalty on every IC chip made today!
Anyway, that’s
h
h
ll ”
the story he tells.”
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS—I: FUNDAMENTAL
THEORY AND APPLICATIONS, VOL. 46, NO. 1, JANUARY 1999
Darlington’s Contributions to Transistor Circuit Design
D id A H d
F ll
IEEE
W.Kucewicz
Analogowe układy CMOS w technice VLSI
17
David A. Hodges,
Fellow, IEEE
Invention of the Integrated Circuit
Invention of the Integrated Circuit
Invention of the Integrated Circuit
Invention of the Integrated Circuit
f
g
u
f
g
u
f
g
u
f
g
u
1953
1953
1953
1953
Robert Noyce get his PhD at MIT where „few
Robert Noyce get his PhD at MIT where „few
peaple had even heardabout the transistor
peaple had even heardabout the transistor
1953
1953
1953
1953
peaple had even heardabout the transistor
peaple had even heardabout the transistor
1955
1955
1955
1955
He arriving to Shockley’s Lab
He arriving to Shockley’s Lab
1955
1955
1955
1955
1957
1957
1957
1957
He left Shockley’s and forms Fairchild
He left Shockley’s and forms Fairchild
Semiconductor with Jean Hoerni and Gordon
Semiconductor with Jean Hoerni and Gordon
1957
1957
1957
1957
Semiconductor with Jean Hoerni and Gordon
Semiconductor with Jean Hoerni and Gordon
Moore
Moore
H
i i
ts t h i
f diff si i
iti s
H
i i
ts t h i
f diff si i
iti s
1958
1958
1958
1958
Hoerni invents technique for diffusion impurities
Hoerni invents technique for diffusion impurities
into Silicon to build planar transistor and then
into Silicon to build planar transistor and then
using SiO
using SiO
2
2
as insulator
as insulator
1959
1959
1959
1959
Noyce develops first true IC using planar
Noyce develops first true IC using planar
transistor, diode
transistor, diode--isolated silicon resistor, SiO
isolated silicon resistor, SiO
22
W.Kucewicz
Analogowe układy CMOS w technice VLSI
18
1959
1959
1959
1959
insulation and evaporated metal wiring
insulation and evaporated metal wiring
1959: Invention of the Integrated
1959: Invention of the Integrated
Circuit
Circuit
1959: Invention of the Integrated
1959: Invention of the Integrated
Circuit
Circuit
uuuu
Noyce develops first true IC
Noyce develops first true IC
Noyce develops first true IC
Noyce develops first true IC
N y
p f
u
N y
p f
u
using
using
•• planar transistor,
planar transistor,
N y
p f
u
N y
p f
u
using
using
•• planar transistor,
planar transistor,
planar transistor,
planar transistor,
•• diode
diode--isolated silicon resistor,
isolated silicon resistor,
•• SiO
SiO
22
insulation
insulation
planar transistor,
planar transistor,
•• diode
diode--isolated silicon resistor,
isolated silicon resistor,
•• SiO
SiO
22
insulation
insulation
•• and evaporated metal wiring
and evaporated metal wiring
•• and evaporated metal wiring
and evaporated metal wiring
W.Kucewicz
Analogowe układy CMOS w technice VLSI
19
First Commercial Planar IC
First Commercial Planar IC 1961
1961
First Commercial Planar IC
First Commercial Planar IC 1961
1961
Fairchild
Fairchild ––
Dual flip
Dual flip--flop
flop Chip
Chip
4 Transistors and 5
4 Transistors and 5
4 Transistors and 5
4 Transistors and 5
Resistors
Resistors
START OF SMALL SCALE
START OF SMALL SCALE
INTEGRATION
INTEGRATION
INTEGRATION
INTEGRATION
TECHNOLOGY
TECHNOLOGY
W.Kucewicz
Analogowe układy CMOS w technice VLSI
20
First Linear IC
First Linear IC
First Linear IC
First Linear IC
1964
1964
1964
1964
Fairchild
Fairchild
First Linear IC
First Linear IC The
The μμA 702 OPAMP
A 702 OPAMP
1967
1967
1967
1967
Fairchild
Fairchild
First
First IC
IC μμMOSAIC
MOSAIC -- ccreated with Computer
reated with Computer--
Aided Design
Aided Design..
1967
1967
1967
1967
gg
Transistors (organized in
Transistors (organized in columns) could be
columns) could be
easily rewired using a
easily rewired using a two
two--layer interconnect to
layer interconnect to
create different
create different circuits. This circuit
circuits. This circuit contains
contains
W.Kucewicz
Analogowe układy CMOS w technice VLSI
21
~150 logic
~150 logic gates.
gates.
1K bit RAM
1K bit RAM
1K bit RAM
1K bit RAM
Noyce and Moore leave Fairchild and found Intel.
Noyce and Moore leave Fairchild and found Intel.
No
No business plan just a promise
business plan just a promise to specialize in memory chips
to specialize in memory chips
1968
1968
1968
1968
No
No business plan, just a promise
business plan, just a promise to specialize in memory chips.
to specialize in memory chips.
They raise $3M in two days
They raise $3M in two days and move to Santa Clara. By
and move to Santa Clara. By
1971 Intel had 500 employees;
1971 Intel had 500 employees; by 1983 it had 21,500
by 1983 it had 21,500
employees and $1100M in sales
employees and $1100M in sales
employees and $1100M in sales.
employees and $1100M in sales.
1970
1970
1970
1970
Intel
Intel
starts selling a 1K bit RAM, the 1103.
1971
1971
1971
1971
Intel
Intel
starts selling a first EPROM, the 1702.
W.Kucewicz
Analogowe układy CMOS w technice VLSI
22
Microprocessor Intel 4004 (1971)
Microprocessor Intel 4004 (1971)
Microprocessor Intel 4004 (1971)
Microprocessor Intel 4004 (1971)
M
p
( 9 )
M
p
( 9 )
M
p
( 9 )
M
p
( 9 )
2300 transistors
2300 transistors on one chip
on one chip
Freq.
Freq. --108 kHz
108 kHz
Technology
Technology
10
10μμ
Technology
Technology ––
10
10μμ
It was 1/8" by 1/16"
It was 1/8" by 1/16" and
and it was
it was
yy
as powerful as
as powerful as ENIAC
ENIAC
pperform
erforming
ing about 60,000
about 60,000
l l ti s i s
d
l l ti s i s
d
calculations in a second.
calculations in a second.
It was as powerful as ENIAC but was
It was as powerful as ENIAC but was
W.Kucewicz
Analogowe układy CMOS w technice VLSI
23
pp
eclipsed by its more capable brothers
eclipsed by its more capable brothers
Ted Hoff’s invention
Ted Hoff’s invention
Microprocessor Intel
Microprocessor Intel 8800
0088
Microprocessor Intel
Microprocessor Intel 8800
0088
M
p
M
p
M
p
M
p
1972
1972
1972
1972
Intel
Intel -- The microprocessor 8008
It had 3 500 transistors supporting a byte wide data path
1972
1972
1972
1972
It had 3,500 transistors supporting a byte-wide data path.
Despite its limitations, the 8008 was the first
microprocessor capable of playing the role of computer CPU
1974
1974
1974
1974
Intel
Intel -- The microprocessor 8008
h
h d
f ’ d
1974
1974
1974
1974
the 8080 had 6,000 transistors fab’ed in a 6um
process. The clock rate was 2Mhz, more than
enough to ignite the personal computer industry.
At l
t P l All
d hi
t
th
ht h
At least Paul Allen and his partner thought so when
they wrote a BASIC interpreter for the 8080 in
1975.
W.Kucewicz
Analogowe układy CMOS w technice VLSI
24
MOS transistor history
MOS transistor history
MOS transistor history
MOS transistor history
M
y
M
y
M
y
M
y
MOSFET transistor
MOSFET transistor -- Lilienfeld (Canada) in 1925
Lilienfeld (Canada) in 1925
CMOS
CMOS 1960’ b t l
d ith
f t i
1960’ b t l
d ith
f t i
bl
bl
CMOS
CMOS –– 1960’s, but plagued with manufacturing
1960’s, but plagued with manufacturing problems
problems
PMOS in 1960’s (calculators)
PMOS in 1960’s (calculators)
NMOS in 1970’s (4004, 8080)
NMOS in 1970’s (4004, 8080) –– for speed
for speed
CMOS in 1980’s
CMOS in 1980’s –– preferred MOSFET technology
preferred MOSFET technology because of
because of
CMOS in 1980 s
CMOS in 1980 s preferred MOSFET technology
preferred MOSFET technology because of
because of
power benefits
power benefits
BiCMOS
BiCMOS
BiCMOS
BiCMOS
SOI, Copper
SOI, Copper--Low K
Low K
W.Kucewicz
Analogowe układy CMOS w technice VLSI
25
Lilienfeld patent
Lilienfeld patent
Lilienfeld patent
Lilienfeld patent
L
f
p
L
f
p
L
f
p
L
f
p
W.Kucewicz
Analogowe układy CMOS w technice VLSI
26
A New Form of Transistor
A New Form of Transistor --
-- 1962
1962
A New Form of Transistor
A New Form of Transistor --
-- 1962
1962
Metal
Metal--Oxide
Oxide
Semiconductor Field
Semiconductor Field--
Eff
T
i
Eff
T
i
Effect Transistor
Effect Transistor
Radio Corporation of
Radio Corporation of
America (RCA)
America (RCA)
Sarnoff Laboratories
Sarnoff Laboratories
Sarnoff Laboratories
Sarnoff Laboratories
W.Kucewicz
Analogowe układy CMOS w technice VLSI
27
One of the Most Powerful 16
One of the Most Powerful 16--Bit
Bit
Microprocessors
Microprocessors --
-- 1979
1979
One of the Most Powerful 16
One of the Most Powerful 16--Bit
Bit
Microprocessors
Microprocessors --
-- 1979
1979
pppp
The Motorola 68000
The Motorola 68000
WELL INTO THE
WELL INTO THE
WELL INTO THE
WELL INTO THE
LARGE SCALE
LARGE SCALE
INTEGRATION
INTEGRATION
ERA
ERA
ERA
ERA
W.Kucewicz
Analogowe układy CMOS w technice VLSI
28
A Very Early 32
A Very Early 32--Bit Microprocessor
Bit Microprocessor --
1981
1981
A Very Early 32
A Very Early 32--Bit Microprocessor
Bit Microprocessor --
1981
1981
The HP Focus Chip, Hewlett
The HP Focus Chip, Hewlett--
Packard Co
Packard Co 450 000
450 000
Packard Co.
Packard Co. –– 450,000
450,000
Transistors
Transistors
THE VERY L RGE C LE
THE VERY L RGE C LE
THE VERY LARGE SCALE
THE VERY LARGE SCALE
INTEGRATED CIRCUIT
INTEGRATED CIRCUIT
ERA BEGINS
ERA BEGINS
W.Kucewicz
Analogowe układy CMOS w technice VLSI
29
Microprocesor Intel Pentium
Microprocesor Intel Pentium 44 ((200
20011))
Microprocesor Intel Pentium
Microprocesor Intel Pentium 44 ((200
20011))
M
p
um
M
p
um
((
))
M
p
um
M
p
um
((
))
42 millions transistors on
42 millions transistors on
hi
hi
one chip
one chip
Freq.
Freq. –– 1.5 GHz
1.5 GHz
Techn.
Techn. –– 0.18
0.18μμ
Techn.
Techn.
0.18
0.18μμ
Performing billions of
l l
h
d
calculations each second
The density of the
The density of the
transistors is more than
transistors is more than
100k/mm
100k/mm
22
40 transistor will fit on
40 transistor will fit on
W.Kucewicz
Analogowe układy CMOS w technice VLSI
30
40 transistor will fit on
40 transistor will fit on
the cross section of the
the cross section of the
hair.
hair.
Today
Today
Today
Todayyyyy
Many disciplines have contributed to the current
Many disciplines have contributed to the current
state of the art
state of the art in VLSI design:
in VLSI design:
state of the art
state of the art in VLSI design:
in VLSI design:
solid
solid--state physics
state physics
materials science
materials science
materials science
materials science
lithography and fab
lithography and fabrication
rication
device modeling
device modeling
device modeling
device modeling
circuit design & layout
circuit design & layout
architecture
architecture
algorithms
algorithms
CAD tools
CAD tools
W.Kucewicz
Analogowe układy CMOS w technice VLSI
31
Computer Aided Design
Computer Aided Design –– CAD
CAD
Computer Aided Design
Computer Aided Design –– CAD
CAD
p
g
p
g
p
g
p
g
CAD tools are doing following tasks:
CAD tools are doing following tasks:
CAD tools are doing following tasks:
CAD tools are doing following tasks:
•• organize
organize
•• generate
generate
•• organize
organize
•• generate
generate
•• generate
generate
•• verify
verify
•• generate
generate
•• verify
verify
The main g
The main goal
oal of using CAD is
of using CAD is designing circuits with
designing circuits with
The main g
The main goal
oal of using CAD is
of using CAD is designing circuits with
designing circuits with
increasing complexity in
increasing complexity in always shorter times
always shorter times
computer has to take over routine work
computer has to take over routine work
computer has to take over routine work
computer has to take over routine work
deliberate the designer from unnecessary low
deliberate the designer from unnecessary low qualification work
qualification work
shift of design activities to higher level abstract
shift of design activities to higher level abstract work
work
W.Kucewicz
Analogowe układy CMOS w technice VLSI
32
shift of design activities to higher level abstract
shift of design activities to higher level abstract work
work
computer has to support new design methods
computer has to support new design methods
Computer Aided Design
Computer Aided Design –– CAD
CAD
Computer Aided Design
Computer Aided Design –– CAD
CAD
p
g
p
g
p
g
p
g
CAD tools are doing following tasks:
CAD tools are doing following tasks:
CAD tools are doing following tasks:
CAD tools are doing following tasks:
•• organize
organize
•• generate
generate
•• organize
organize
•• generate
generate
•• generate
generate
•• verify
verify
•• generate
generate
•• verify
verify
The main g
The main goal
oal of using CAD is
of using CAD is designing circuits with
designing circuits with
The main g
The main goal
oal of using CAD is
of using CAD is designing circuits with
designing circuits with
increasing complexity in
increasing complexity in always shorter times
always shorter times
computer has to take over routine work
computer has to take over routine work
computer has to take over routine work
computer has to take over routine work
deliberate the designer from unnecessary low
deliberate the designer from unnecessary low qualification work
qualification work
shift of design activities to higher level abstract
shift of design activities to higher level abstract work
work
W.Kucewicz
Analogowe układy CMOS w technice VLSI
33
shift of design activities to higher level abstract
shift of design activities to higher level abstract work
work
computer has to support new design methods
computer has to support new design methods
Gate Array, Semi
Gate Array, Semi--Custom, and Full
Custom, and Full--
Custom ICs
Custom ICs
Gate Array, Semi
Gate Array, Semi--Custom, and Full
Custom, and Full--
Custom ICs
Custom ICs
Custom ICs
Custom ICs
Custom ICs
Custom ICs
VLSI Technology
VLSI Technology
Gate Array
Gate Array
Standard Cell
Chip Area Ratios:
3:2:1
Chip Area Ratios:
3:2:1
Standard Cell
Full Custom
W.Kucewicz
Analogowe układy CMOS w technice VLSI
34
Progress in VLSI
Progress in VLSI
Progress in VLSI
Progress in VLSI
Progress in VLSI
Progress in VLSI
production
production
Progress in VLSI
Progress in VLSI
production
production
production
production
production
production
W.Kucewicz
Analogowe układy CMOS w technice VLSI
35
Prawo Moore’a
Prawo Moore’a
Prawo Moore’a
Prawo Moore’a
w M
w M
w M
w M
W roku 1965 Gordoon Moore
W roku 1965 Gordoon Moore
(założyciel firm Fairchild i Intel)
(założyciel firm Fairchild i Intel)
(założyciel firm Fairchild i Intel)
(założyciel firm Fairchild i Intel)
przewidział, że liczba tranzystorów
przewidział, że liczba tranzystorów
w układzie scalonym będzie
w układzie scalonym będzie
zwiększała się eksponencjalnie w
zwiększała się eksponencjalnie w
funkcji czasu
funkcji czasu ..
W l t h 80
W l t h 80 t h k
b i
t h k
b i
W latach 80
W latach 80--tych przekroczono barierę
tych przekroczono barierę
1 miliona tranzystorów upakowanych w
1 miliona tranzystorów upakowanych w
jednej strukturze
jednej strukturze
jednej stru turze
jednej stru turze
•• 2300 transistors (Intel 4004)
2300 transistors (Intel 4004) –
– 1971
1971
42 Milli
(I t l P4)
42 Milli
(I t l P4) 2001
2001
W.Kucewicz
Analogowe układy CMOS w technice VLSI
36
•• 42 Million (Intel P4)
42 Million (Intel P4) -- 2001
2001
Postęp w produkcji układów scalonych
Postęp w produkcji układów scalonych
Postęp w produkcji układów scalonych
Postęp w produkcji układów scalonych
ęp w p
u j u
w
y
ęp w p
u j u
w
y
ęp w p
u j u
w
y
ęp w p
u j u
w
y
G.Moore - Cramming more
components onto integrated circuits
- Electronics, Volume 38, Number 8,
April 19 1965
April 19, 1965
Ilość tranzystorów w układzie scalonym podwaja się w
Ilość tranzystorów w układzie scalonym podwaja się w
W.Kucewicz
Analogowe układy CMOS w technice VLSI
37
Ilość tranzystorów w układzie scalonym podwaja się w
Ilość tranzystorów w układzie scalonym podwaja się w
ciagu roku
ciagu roku
Technology Influence
Technology Influence
Technology Influence
Technology Influence
gy
f u
gy
f u
gy
f u
gy
f u
0.18 µm
0.5 µm
l
0.12µm
Devices
1999
2002
1993
3 layers
7 layers
8 layers
Interconnects
F
1200 MH
Frequency
W.Kucewicz
Analogowe układy CMOS w technice VLSI
38
120MHz
500MHz
1200 MHz
Technology progress
Technology progress
Technology progress
Technology progress
gy p g
gy p g
gy p g
gy p g
199
19922
200
20022
0 7µm 2 metal layers
0.12µm, 7 metal layers
199
19922
200
20022
0.7µm, 2 metal layers
Up to 100,000 devices on a chip
CPU frequency 50MHz
0. µm, 7 metal layers
Up to 500,000,000 devices
CPU frequency 1,5GHz
IC
1000 pins
1000 pins
40 pins
40 pins
10 years of
10 years of
10 years of
10 years of
W.Kucewicz
Analogowe układy CMOS w technice VLSI
39
pp
evolution
evolution
evolution
evolution
Postęp w produkcji procesorów
Postęp w produkcji procesorów
Postęp w produkcji procesorów
Postęp w produkcji procesorów
ęp w p
u j p
w
ęp w p
u j p
w
ęp w p
u j p
w
ęp w p
u j p
w
P. Gargini – Sailing with the ITRS into nanotechnology - 2004
W.Kucewicz
Analogowe układy CMOS w technice VLSI
40
Ilość tranzystorów w układzie podwaja się w ciagu 2 lat
Ilość tranzystorów w układzie podwaja się w ciagu 2 lat
Postęp w produkcji procesorów
Postęp w produkcji procesorów
Postęp w produkcji procesorów
Postęp w produkcji procesorów
ęp w p
u j p
w
ęp w p
u j p
w
ęp w p
u j p
w
ęp w p
u j p
w
P. Gargini – Sailing with the ITRS into nanotechnology - 2004
W.Kucewicz
Analogowe układy CMOS w technice VLSI
41
Powierzchnia układu scalonego podwaja się w ciagu 6 lat
Powierzchnia układu scalonego podwaja się w ciagu 6 lat
Nanotechnologia
Nanotechnologia
Nanotechnologia
Nanotechnologia
N
g
N
g
N
g
N
g
P. Gargini – Sailing with the ITRS into nanotechnology - 2004
W.Kucewicz
Analogowe układy CMOS w technice VLSI
42
Reasonably familiar Nanotube Different technology
Reasonably familiar Nanotube Different technology
DRAM chip capacity evolution
DRAM chip capacity evolution
DRAM chip capacity evolution
DRAM chip capacity evolution
D
M
p
p
y
u
D
M
p
p
y
u
D
M
p
p
y
u
D
M
p
p
y
u
Memory/chip
Memory/chip
ddoubles every 1.5 year
oubles every 1.5 year
p)p)
ts/chi
p
ts/chi
p
y (kBi
t
y (kBi
t
Year
Year
M
emor
y
M
emor
y
W.Kucewicz
Analogowe układy CMOS w technice VLSI
43
MM
Cost per Transistor
Cost per Transistor
Cost per Transistor
Cost per Transistor
pppp
cost:
cost:
cost:
cost:
¢¢--per
per--transistor
transistor
Fabrication capital cost per
Fabrication capital cost per
transistor (Moore’s law)
transistor (Moore’s law)
0.01
0.01
0.1
0.1
11
transistor (Moore s law)
transistor (Moore s law)
0.0001
0.0001
0.001
0.001
0.000001
0.000001
0.00001
0.00001
0.0000001
0.0000001
1982
1982
1985
1985
1988
1988 1991
1991
1994
1994
1997
1997
2000
2000
2003
2003 2006
2006
2009
2009
2012
2012
W.Kucewicz
Analogowe układy CMOS w technice VLSI
44
Wiring Levels and Mask Levels
Wiring Levels and Mask Levels
Wiring Levels and Mask Levels
Wiring Levels and Mask Levels
W
g L
M
L
W
g L
M
L
W
g L
M
L
W
g L
M
L
29
35
29
29
27
27
27
25
25
25
25
25
25
30
els
20
25
r of
Lev
e
Max. Wiring Levels
7
8
8
8
9
9
9
10 10
10
10
15
Numbe
r
mProcessor Mask Levels
7
0
5
0
2001 2002 2003 2004 2005 2006 2007 2010 2013 2016
Year
W.Kucewicz
Analogowe układy CMOS w technice VLSI
45
Year
On
On--Chip Clock
Chip Clock
φφ
and Power Supply
and Power Supply
V
V
DD
DD
On
On--Chip Clock
Chip Clock
φφ
and Power Supply
and Power Supply
V
V
DD
DD
35
1 2
28,751
0,9
0,9
1
1
1
1,1
30
35
G
Hz)
1
1,2
)
19 348
0 6
0,7
0,9
20
25
uency (
G
0,8
y (Volts
11 511
19,348
0,4
0,5
0,6
15
20
ock Freq
u
0 4
0,6
er Suppl
y
Clock Frequency
VDD
11,511
6,739
5,631
5,173
3,99
3,088
2,317
1,684
5
10
Chip Cl
o
0,2
0,4
Pow
e
VDD
0
2001 200220032004 20052006 20072010 2013 2016
C
0
W.Kucewicz
Analogowe układy CMOS w technice VLSI
46
Year
μμ
Processor and DRAM Fault Density
Processor and DRAM Fault Density
Targets
Targets
μμ
Processor and DRAM Fault Density
Processor and DRAM Fault Density
Targets
Targets
Targets
Targets
Targets
Targets
2493
2748
3000
1963
2493
2148
1752
2236
2000
2500
1464
1426
1356 1356
1500
2000
ts /
m
2
1356 1356 1356 1356 1356 1356 1356
1134
1116
1356
1000
Faul
t
μ
Processor (83% yield)
0
500
μ
Processor (83% yield)
DRAM (89.5% yield)
2001 2002 2003 2004 2005 2006 2007 2010 2013 2016
Year
W.Kucewicz
Analogowe układy CMOS w technice VLSI
47
Long
Long--Term Trends
Term Trends
Long
Long--Term Trends
Term Trends
L g
L g
m
m
L g
L g
m
m
1.
1. IC market growth averaged 17% per year
IC market growth averaged 17% per year
2.
2.
CMOS circuits represent more than 75% of the world’s
CMOS circuits represent more than 75% of the world’s
semiconductors
semiconductors
3.
3. 16% annual chip feature size reduction 1995
16% annual chip feature size reduction 1995--2001
2001
4.
4. 11% annual chip feature size reduction 2002
11% annual chip feature size reduction 2002 –– future
future
5.
5. DRAM chip size increases 12% / year
DRAM chip size increases 12% / year
6.
6. Pins/balls on packages increase 10% / year
Pins/balls on packages increase 10% / year
7.
7. Pin cost decreases 5% / year
Pin cost decreases 5% / year
8.
8. Average package cost increases 5% / year
Average package cost increases 5% / year
9.
9. Packaging share of system cost doubles over 15 years
Packaging share of system cost doubles over 15 years
Drives migration to SoC,
Multi-Chip Modules
(MCMs)
SIP devices
W.Kucewicz
Analogowe układy CMOS w technice VLSI
48
Postęp w produkcji płytek krzemowych
Postęp w produkcji płytek krzemowych
Postęp w produkcji płytek krzemowych
Postęp w produkcji płytek krzemowych
ęp w p
u j p y
z m wy
ęp w p
u j p y
z m wy
ęp w p
u j p y
z m wy
ęp w p
u j p y
z m wy
P. Gargini – Sailing with the ITRS into nanotechnology - 2004
W.Kucewicz
Analogowe układy CMOS w technice VLSI
49
Powierzchnia płytki krzemowej podwaja się w ciagu 8 lat
Powierzchnia płytki krzemowej podwaja się w ciagu 8 lat
Mikroelektronika
Mikroelektronika Î
Î
Nanotechnologia
Nanotechnologia
Mikroelektronika
Mikroelektronika Î
Î
Nanotechnologia
Nanotechnologia
M r
tr n a
M r
tr n a
Nan t chn g a
Nan t chn g a
M r
tr n a
M r
tr n a
Nan t chn g a
Nan t chn g a
P. Gargini – Sailing with the ITRS into nanotechnology - 2004
W.Kucewicz
Analogowe układy CMOS w technice VLSI
50
Powierzchnia płytki krzemowej podwaja się w ciagu 8 lat
Powierzchnia płytki krzemowej podwaja się w ciagu 8 lat
Mikroelektronika
Mikroelektronika Î
Î
Nanotechnologia
Nanotechnologia
Mikroelektronika
Mikroelektronika Î
Î
Nanotechnologia
Nanotechnologia
M r
tr n a
M r
tr n a
Nan t chn g a
Nan t chn g a
M r
tr n a
M r
tr n a
Nan t chn g a
Nan t chn g a
P. Gargini – Sailing with the ITRS into nanotechnology - 2004
W.Kucewicz
Analogowe układy CMOS w technice VLSI
51
Powierzchnia płytki krzemowej podwaja się w ciagu 8 lat
Powierzchnia płytki krzemowej podwaja się w ciagu 8 lat
Technology Directions
Technology Directions
Technology Directions
Technology Directions
gy D
gy D
gy D
gy D
Year
Year
1999
1999
2002
2002
2005
2005
2008
2008
2011
2011
2014
2014
Feature size [nm]
Feature size [nm]
180
180
130
130
100
100
70
70
50
50
35
35
Feature size [nm]
Feature size [nm]
180
180
130
130
100
100
70
70
50
50
35
35
Mtrans/cm
Mtrans/cm
22
77
14
14--26
26
47
47
115
115
284
284
701
701
Chip size [mm
Chip size [mm
22
]]
170
170
170
170--214
214
235
235
269
269
308
308
354
354
Chip size [mm
Chip size [mm ]]
170
170
170
170 214
214
235
235
269
269
308
308
354
354
Signal pins/chip
Signal pins/chip
768
768
1024
1024
1024
1024
1280
1280
1408
1408
1472
1472
Wiring levels
Wiring levels
66--77
77--88
88--99
99
99--10
10
10
10
Power supp [V]
Power supp [V]
1.8
1.8
1.5
1.5
1.2
1.2
0.8
0.8
0.6
0.6
0.6
0.6
Power [W]
Power [W]
90
90
130
130
180
180
170
170
174
174
183
183
Battery power [W]
Battery power [W]
1.4
1.4
2.0
2.0
2.4
2.4
2.0
2.0
2.2
2.2
2.4
2.4
W.Kucewicz
Analogowe układy CMOS w technice VLSI
52
Moore’s Law Predictions
Moore’s Law Predictions –– 2025?
2025?
Moore’s Law Predictions
Moore’s Law Predictions –– 2025?
2025?
M
L w
M
L w
M
L w
M
L w
Assumptions in 1995 based on 1960
Assumptions in 1995 based on 1960--1995
1995
Transistors per chip doubles every 1.5 years
Transistors per chip doubles every 1.5 years
Transistors per chip doubles every 1.5 years
Transistors per chip doubles every 1.5 years
Minimum feature size is cut in half every six years
Minimum feature size is cut in half every six years
Chip area goes up 2.3 times every six years
Chip area goes up 2.3 times every six years
Manufacturing costs remain about constant
Manufacturing costs remain about constant
Manufacturing costs remain about constant
Manufacturing costs remain about constant
This leads us to the following predictions
This leads us to the following predictions
Wafer size will be
32 inches!
32 inches!
A wafer fabrication will cost a tens of G$
3 b 6 i h
3 b 6 i h
The chips will be
3 by 6 inches
3 by 6 inches
The minimum feature size will be
100 Angstroms
100 Angstroms
(which is
about 5 photoresist molecules wide)
about 5 photoresist molecules wide)
A Memory chip will hold
64 TERA bits
64 TERA bits
W.Kucewicz
Analogowe układy CMOS w technice VLSI
53
will they come true?!
will they come true?!