0% found this document useful (0 votes)
87 views454 pages

AVLIS Final Merge

This document appears to be a lecture outline for a course on advanced VLSI design. It includes 14 lecture topics that will be covered throughout the course, ranging from sequential circuits and timing issues to arithmetic circuits, memory circuits, and deep sub-micron device engineering. The course content will also include introductions to VLSI design, clock generation and distribution networks, asynchronous system design, and interfacing circuits. Evaluation will consist of quizzes, assignments, a mid-semester test, and a comprehensive exam.

Uploaded by

Srinu Boddula
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
87 views454 pages

AVLIS Final Merge

This document appears to be a lecture outline for a course on advanced VLSI design. It includes 14 lecture topics that will be covered throughout the course, ranging from sequential circuits and timing issues to arithmetic circuits, memory circuits, and deep sub-micron device engineering. The course content will also include introductions to VLSI design, clock generation and distribution networks, asynchronous system design, and interfacing circuits. Evaluation will consist of quizzes, assignments, a mid-semester test, and a comprehensive exam.

Uploaded by

Srinu Boddula
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 454

Lecture_0_Introduction_to_AVLSID_Course_1642315600440 2

Lec_1_Introduction_1642315645671 9
Lec_2_Sequential_logic_1642923673798 40
Lec_3_Dynamic_Sequential_1643530824129 72
Lec_4A_Pipelined_Registers_1644133424311 95
Lec_4B_Static_Timimg_Analyisis_1644133457649 104
Lec_5_Clock_Generatio_and_Distribution_Part_1_1644752385887 116
Lec_5_Clock_Generatio_and_Distribution_Part_2_1644752403211 129
Lec_6_Latch_based_Clocking_And_Asynchronous_Clocking_1645353703599 146
Lec_7A_Interfacing_Circuits_Part_1_1645962856852_1645962956291 167
Lec_7B_Interfacing_Circuits_Part_2_1645962991634 174
Lec_8_Interfacing_Circuits_Part_3_1647784649397 195
Lec_9_Arithmetic_Circuits_Part_1_1649432767319 221
Lec_10_Arithmetic_Circuits_Part_2_1649597449288 256
Lec_11_Memory_Circuits_1650202929351 287
Lec_12A_LPC_Part_1_1650798675349 337
Lec_12B_LPC_Part_2_1650799616584 363
Lec_12C_LPC_Part_3_Adiabatic_Logic_1651408229137 382
Lec_13_Interconnects_1651408262173 388
Lec_14A_Deep_Submicron_MOSFET_operation_1652025321991 413
Lec_14B_CMOS_Technology_Scaling_1652025348068 444
BITS Pilani Presentation
BITS Pilani Dr. Sanjay Vidhyadharan
EEE WILP
Pilani Campus

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


BITS Pilani
Pilani Campus

MEL ZG623 Advanced VLSI Design


Lecture No. 0

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Introduction to AVLSI Design Course

By Dr. Sanjay Vidhyadharan

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Course Content
1. Introduction to AVLSI Design
2. Sequential Circuits
3. Dynamic Latches
4. Timing Issues in Clocked Systems
5. Clock Generation and Distribution Networks
6. Asynchronous System Design
7. Interfacing Circuits
8. Wire Design Principles
9. High Speed Computer Arithmetic Design
10.Computer Arithmetic
11.Deep Sub-micron Device Engineering

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Textbooks
➢ T1: Digital Integrated Circuits - A design Perspective , Jan M Rabaey, Anantha
Chandrakasan, Borivoje Nikolic, 2nd edition, Prentice Hall, 2005.7

➢ Reference Books:

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Evaluation

Sl.No. Evaluation Component Type Weight Duration Day, Date,


Session, Time

EC - 1 Quiz 1 Online 10% TBA To be announced

Assignment Online 20% To be announced


EC - 2 Mid-Semester Test Open Book 30% 2 hours
EC - 3 Comprehensive Exam Open Book 40% 3 hours

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Thank you

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


VLSI Design / Advanced VLSI Design:
2021-22
Lecture 1
Introduction to VLSI Design

By Dr. Sanjay Vidhyadharan

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


VLSI Design
Integrated circuits are also categorized according to the number of
transistors or other active circuit devices they contain.

➢ An IC is said to use small-scale integration (SSI) if it


contains fewer than 10 transistors.

➢ An IC that contains from 10 to 100 transistors is said to use


medium-scale integration.

➢ A large-scale integration (LSI) IC contains from 100 to 1,000


transistors

➢ And one that uses Very-Large-Scale Integration (VLSI)


contains more than 1,000 transistors.
2

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


ENAIC - The First Electronic Computer

1945

17,468 vacuum tubes and consumed 160 kW 3

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Evolution of Digital Technology

➢ Resistor Transistor Logic : 1961


➢ Diode Transistor Logic : 1962
➢ Transistor Transistor Logic : 1963 Discrete IC
➢ Emitter-coupled Logic : First Microprocessor 360
➢ CMOS :
➢ 1974 Intel 4004 which had 2000 Transistors
Channel Length of 10 µm.
➢ 2021 AMD 7 nm has billions of Transistors
Channel Length of 7 nm.

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


RTL Logic

Advantages
➢ The primary advantage of RTL technology
was that it involved a minimum number of
transistors, which was an important
consideration before integrated circuit
technology

Disadvantages
➢ The obvious disadvantage of RTL is its high
current dissipation when the transistor
conducts to overdrive the output biasing
resistor.
➢ Passive Pull up.
➢ Limited Fanout
➢ No Rail-to-Rail Output

1/15/2022 5

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


DTL Logic

Advantages
➢ Diodes perform the logical function

Disadvantages
➢ The obvious disadvantage of DTL is its high
current dissipation when the transistor
conducts to overdrive the output biasing
resistor.
➢ Passive Pull up.
➢ Limited Fanout
➢ No Rail-to-Rail Output

1/15/2022 6

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


TTL Logic

Advantages
➢ Multi-Emitter perform the logical function
➢ Active Pull-up and Active Pull down

Disadvantages
➢ The obvious disadvantage of TTL is its high
current dissipation
➢ No Rail-to-Rail Output

1/15/2022 7

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


ECL Logic

Advantages
➢ Fast
➢ High Fan-out

Disadvantages
➢ High current dissipation

1/15/2022 8

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Advantages of CMOS Technology

➢ Low Static power consumption VDD


➢ High input impedance
➢ Rail to Rail Output
➢ Active Pull-up and Pull-down of Output
➢ Simple Design Vin Vout
➢ High Packing Density
➢ High noise immunity CL
➢ Temperature Stability
➢ Scalability

1/15/2022 9

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Pentium

1993 : Technology : 0.8 μm (16.7 mm by 17.6 mm)


60–66 MHz
ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION
Latest CPUs and GPUs
Latest Trends :- Intel : 14 nm AMD : 7nm

➢ Billions of Transistors/ Chip


➢ Frequency of Operation in GHz
➢ Complex Computational Capability

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Moore’s Law
In 1965, Gordon E. Moore—co-founder of Intel (NASDAQ: INTC)—postulated that
the number of transistors that can be packed into a given unit of space will double
about every two years.

Million-transistor/chip barrier crossed in the late 1980s


12

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Die Size Growth

100
Die size (mm)

P6
486 Pentium ® proc
10 386
286
8080 8086
8085 ~7% growth per year
8008
4004 ~2X growth in 10 years

1
1970 1980 1990 2000 2010
Year

13

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Frequency
10000
Doubles every
1000
2 years
Frequency (Mhz)

P6
100
Pentium ® proc
486
10 8085 386
8086 286

1 8080
8008
4004
0.1
1970 1980 1990 2000 2010
Year

Microprocessors frequency doubles every 2 years


14

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


ITRS Prediction of Power Consumption

Semiconductor Industry Assoc., International Technology Roadmap for


Semiconductors, 2002 Update; http://public.itrs.net.
1/15/2022 15

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Limitations of Static CMOS
Intel’s Prediction of Power Consumption

Online Available: http://computerscience.chemeketa.edu/cs160Reader/Parallel Processing/


MooresLaw.html
16

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Power Dissipation in CMOS Circuits

Total Power

Dynamic Power Static Power

Switching Short circuit Glitch Leakage

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Static Loss

Static Power Dissipation in CMOS

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Static Loss
➢ Causes for High Static Power Consumption

Effect of Decreasing VDD on Delay

𝐾1 𝐶𝐿 𝐾2𝐶𝐿
𝑃𝑟𝑜𝑝𝑎𝑔𝑎𝑡𝑖𝑜𝑛 𝐷𝑒𝑙𝑎𝑦 (𝑡𝑝𝑑) = = 2`
𝐼𝐷 𝑉𝐷𝐷−𝑉𝑡ℎ
Effect of Decreasing Vth on Power

• Intel estimated leakage power consumption at more than 50W for a


100nm technology node.

• Leakage depends strongly on a Threshold voltage (Vth) of the transistor

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Static Loss

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Static Loss
➢ Reducing Static Loss
• Device Level Techniques
Tunnel Field Effect Transistors
CNFETs
• Circuit Level Techniques when using CMOS

[3] K. Roy, S. Mukhopadhyay, and H. Mahmoodi-Meimand, “Leakage c rre mecha sms


and leakage reduction techniques in deep-submicrometer CMOS c rc s,” Pr ceed gs f he
IEEE, vol. 91, no. 2, pp. 305–327, Feb. 2003.

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Short Circuit Loss

𝑉𝐷𝐷
➢ Use devices with Vth close to & steep switching characteristics
2
➢ Short Circuit loss reduces with Cload

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Short Circuit Loss

➢ Short Circuit loss reduces with Cload

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Switching Loss

𝑉 1 2
Energy stored in CLoad (𝐶𝐿)= ‫׬‬0 𝐷𝐷 𝑉𝐶 . 𝐶𝐿𝑑𝑉𝑐 = . 𝑉𝐷𝐷 ∗ 𝐶𝐿
2
𝑇 2
Energy consumed from power supply= 𝑉𝐷𝐷 ‫׬‬0 𝑖 𝑡 𝑑𝑡 = 𝑉𝐷𝐷 . 𝑄𝐶𝐿 = 𝑉𝐷𝐷 . 𝐶𝐿
21
Energy dissipated in pMOSFET during charging = . 𝑉𝐷𝐷 . 𝐶𝐿
2
2 1
Energy dissipated in nMOSFET during discharging = . 𝑉𝐷𝐷 . 𝐶𝐿
2
2
Power Consumption = Frquency. 𝑉𝐷𝐷 . 𝐶𝐿
ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION
High Dynamic Power Consumption
PDynamic = Freq * VDD2 * CL
➢ VDD & CL : Reduced by 30 % each generation

➢ Frequency : 43 % Increase
▪ Architectural optimizations
▪ Consumer requirements

➢ Transistors per chip : Increased exponentially


▪ Advancement in lithography
▪ Multilayer metallization
▪ Efficient partitioning and routing techniques

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Figure of Merits of a Digital Circuit
1. Noise Margin

26

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Figure of Merits of a Digital Circuit
2. Propagation Delay (Eg. Inverter)

27

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Figure of Merits of a Digital Circuit
3. Fan-in and Fan-Out

1/15/2022 28

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Figure of Merits of a Digital Circuit
4. Power and Energy Consumption
➢ Power consumption of design determine
➢ How much energy is consumed per operation
➢ How much heat the circuit dissipates

➢ Determine critical design decision


➢ Power supply capacity, battery lifetime,
➢ Supply line sizing, packaging and cooling

➢ Peak power is important for supply line sizing

➢ Average power dissipation is important for cooling or


battery requirements
1/15/2022 29

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Figure of Merits of a Digital Circuit
5. Power Delay Product
PDynamic = Freq * VDD2 * CL

𝐾1 𝐶𝐿 𝐾 2 𝐶𝐿
𝑃𝑟𝑜𝑝𝑎𝑔𝑎𝑡𝑖𝑜𝑛 𝐷𝑒𝑙𝑎𝑦 (𝑡𝑝𝑑) = = 2`
𝐼𝐷 𝑉𝐷𝐷−𝑉𝑡ℎ

1/15/2022 30

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Thank you

1/15/2022 31

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Advanced VLSI Design: 2021-22
Lecture 2
Sequential Circuits: Latch and Flip-flops

By Dr. Sanjay Vidhyadharan

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Combinational vs. Sequential Circuits

n-inputs Combinational m-outputs


Circuit (Depend only on inputs)
Combinational Circuit

n-inputs m-outputs
Combinational
Circuit Storage
Next Elements Present
state state

Sequential Circuit

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Latch vs. Flip-flop

1/23/2022 3

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Latch
Storing 1
0 1

1 0

Stores 1 as Long as Power Supply is Provided

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Latch
Storing 0
1 0

0 1

Stores 0 as Long as Power Supply is Provided

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Latch
How to Write Data into a Latch?

1 0
1
0

We are going to use this property of NOR and NAND to


write data into the latch
6

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


SR Latch
S R Q ഥ
𝑸
1
0 0 Q ഥ
𝑸
0 1 0 1 0

1 0 1 0
1 1 0 0

0
1

0
1
0
7

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


SR Latch
S R Q ഥ
𝑸
1
0 0 Q ഥ
𝑸
0 1 0 1 0

1 0 1 0
1 1 0 0
1 0 1
0
1 1

0 1 0
1
0
8

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


SR Latch
S R Q ഥ
𝑸
1
0 0 Q ഥ
𝑸
0 1 0 1 0

1 0 1 0
1 1 0 0

0 1
0 1

1 0 0
1
1 1
9

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


SR Latch

S R Q ഥ
𝑸
0 0 Q ഥ
𝑸 No change

0 1 0 1 Reset Q = 0

1 0 1 0 Set Q = 1

1 1 0 0 Forbidden

10

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


SR Latch
SR Latch with NOR Gates

1/23/2022 11

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


SR Latch
SR Latch with NAND Gates
0
1

S’ R’ Q Q’
0 0 1 1 Forbidden
0 1 1 0 Set
1 0 0 1 Reset
1 1 Q Q’ No change

1/23/2022 12

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


SR Latch
Clocked SR Latch

NAND Implementation AND-NOR Implementation

1/23/2022 13

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


SR Latch
Clocked SR Latch

1/23/2022 14

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


D Latch

1/23/2022 15

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


D Latch

1/23/2022 16

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


D Latch
Mux Based Latch

1/23/2022 17

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


D Latch
Mux Based Latch

1/23/2022 18

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


D Latch
Multiplexer-based NMOS latch using NMOS-only pass transistors.

1/23/2022 19

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Race around in Latches

Xn
D

Yn = Xn . X’n-1

1/23/2022 20

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Latch vs. Flip-flop

Latch – Responds to change in level of clock pulse

The key to the proper operation of a flip-flop is to trigger it


only during signal transition.

1/23/2022 21

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Timing Constraints of a Flip-flop
➢ Setup Time
➢ Hold Time

1/23/2022 22

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Flip-Flop
Master-slave positive edge-triggered register using multiplexers.

The set-up time is equal to 3*tpd_inv + tpd_tx

Hold time is Nil


Propagation delay = tpd_tx + tpd_inv

1/23/2022 23

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Flip-Flop
Set-up time simulation

1/23/2022 24

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Flip-Flop
Simulation of propagation delay

1/23/2022 25

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Flip-Flop
Reduced load clock load static master-slave register.

The penalty for the reduced clock load is increased design complexity. The
transmission gate (T1) and its source driver must overpower the feedback
inverter (I2 ) to switch the state of the cross-coupled inverter.

The transistors of inverter I2 should be made weaker. This can be accomplished by


making their channel-lengths larger than minimum.

1/23/2022 26

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Flip-Flop
Clock Jitter

When the clock goes high, the slave stage should


stop sampling the master stage output and go into a
hold mode. However, since CLK and CLK are both
high for a short period of time (the overlap period),
both sampling pass transistors conduct and there is a
direct path from the D input to the Q output. As a
result, data at the output can change on the rising
edge of the clock, which is undesired for a negative
edge- triggered register.
1/23/2022 27

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Flip-Flop
Two-phase non-overlapping clocks

1/23/2022 28

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Single Phase Global Clock Generation

1/23/2022 29

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Non-overlapping Clock Generation

1/23/2022 30

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Flip-Flop
Solving the leakage problem using multiple-threshold CMOS.

Negative latch
➢ During normal mode of operation, the sleep devices are tuned on.
➢ The shaded inverters and transmission gates are implemented in low-threshold devices.
1/23/2022 31

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Thank you

1/23/2022 32

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Advanced VLSI Design: 2021-22
Lecture 3
Dynamic Registers

By Dr. Sanjay Vidhyadharan

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Static CMOS Latch

➢Area Large
➢Complexity
➢Static Dissipation
➢Dynamic Dissipation
Short Circuit
Switching Loss

1/30/2022 2

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Dynamic Register

Principle of Operation:
➢ Temporary storage of charge on parasitic capacitors.
➢ A stored value can hence only be kept for a limited amount of
time, typically in the range of milliseconds.
➢ To preserve signal integrity, a periodic refresh of its value
➢ Registers are used in computational structures are constantly
clocked such as pipelined datapath.

1/30/2022 3

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Dynamic Register

Dynamic Positive edge-triggered register

➢ Only 8 transistors
➢ Only 6 transistors if NMOS Gates used
➢ Low Power
➢ Low Propagation Delay ( One Pass Transistor Delay + One Inverter Delay)
➢ Set-up Time : (One Pass Transistor Delay + One Inverter Delay)
➢ Hold Time : Nil

1/30/2022 4

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Dynamic Register

Dynamic Positive edge-triggered register

Limitations

During the 0-0 overlap direct path for data from D to Q (T1PMOS – T2PMOS)
During the 1-1 overlap direct path for data from D to Q (T1NMOS – T2NMOS)

1/30/2022 5

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Dynamic Register

Dynamic Positive edge-triggered register

Suppose data changes


here . Node 1 not
effected

Limitations Node 1 salved to D

Q salved to D

Q can change on the falling edge

1/30/2022 6

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Clock-Skew Insensitive C2MOS Register

1/30/2022 7

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Clock-Skew Insensitive C2MOS Register

If the D input changes during the overlap period, node X can make a transition, but cannot
propagate to the output.
1/30/2022 8

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Clock-Skew Insensitive C2MOS Register

If the D input changes during the overlap period, node X can make a transition, but cannot
propagate to the output.
1/30/2022 9

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Clock-Skew Insensitive C2MOS Register

If the D input changes during the overlap period, node X can make a transition, but cannot propagate to the
output. However, as soon as the overlap period is over, the PMOS M8 is turned on and the 0 propagates to
output. This effect is not desirable. The problem is fixed by imposing a hold time constraint on the input
data, D, or, in other words, the data D should be stable during the overlap period.
1/30/2022 10

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Clock-Skew Insensitive C2MOS Register
Clock rise time (or fall time) should be smaller than approximately five times the
propagation delay of the register.

1/30/2022 11

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Dual-edge Registers

➢ Input sampled on both edges


➢ Lower frequency for same functional
throughput
➢ Power savings in the clock distribution
network

1/30/2022 12

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


True Single-Phase Clocked Register (TSPCR)

➢ When Clock is Low :


➢ If data is 0 , Node 1 is Charged to ‘High’. Node Out retains previous value.
➢ If data is 1 , Node 1 is retains previous value. Node Out retains previous value.
➢ When Clock is High :
➢ Chain of two inverters Latch is transparent and Q slaved to D

1/30/2022 13

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


True Single-Phase Clocked Register (TSPCR)

➢ In two-phase clocking schemes, care must be taken in routing the two clock signals to
ensure that overlap is minimized.
➢ A register can be constructed by cascading positive and negative latches.
➢ The main advantage is the use of a single clock phase.
➢ The disadvantage is the slight increase in the number of transistors -12 transistors are
required.
1/30/2022 14

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


True Single-Phase Clocked Register (TSPCR)

➢ TSPC offers an additional advantage: the possibility of embedding logic functionality


into the latches.
➢ This reduces the delay overhead associated with the latches.
➢ This approach of embedding logic into latches has been used extensively in the design of
high-performance processors.

1/30/2022 15

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


True Single-Phase Clocked Register (TSPCR)

In a 0.25 mm, the set-up time of such a circuit using minimum-size devices is 140 psec. A
conventional approach, composed of an AND gate followed by a positive latch has an effective set-up
time of 600 psec (we treat the AND plus latch as a black box that performs both functions). The
embedded logic approach hence results in significant performance improvements.

1/30/2022 16

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


True Single-Phase Clocked Register (TSPCR)

➢ The TSPC latch circuits can be further reduced in complexity, where only the first
inverter is controlled by the clock.
➢ Not all node voltages in the latch experience the full logic swing the voltage at node A
(for Vin = 0 V) for the positive latch maximally equals VDD– VTn, which results in a
reduced drive for the output NMOS transistor and a loss in performance.

1/30/2022 17

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Positive Edge-Triggered Register TSPC

When CLK = 0 :
The input inverter is sampling the inverted D input on node X.
The second (dynamic) inverter is in the precharge mode, Y to VDD.
The third inverter is in the hold mode, since M8 and M9 are off.
On the rising edge of the clock :
The dynamic inverter M4-M6 evaluates. If X is high on the rising edge, Y discharges.
The third inverter M7-M8 is on during the high phase, and the node value on Y is
passed to the output Q.

1/30/2022 18

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Positive Edge-Triggered Register TSPC

Hold Time: On the positive phase of the clock, note that node X transitions to a low if the D
input transitions to a high level. Therefore, the input must be kept stable till the value on node
X before the rising edge of the clock propagates to Y. This represents the hold time of the
register (note that the hold time less than 1 inverter delay since it takes 1 delay for the input to
affect node X).
Set-up Time: The set-up time is the time for node X to be valid, which is one inverter delay.

Propagation Delay : The propagation delay of the register is essentially three inverters since
the value on node X must propagate to the output Q.

1/30/2022 19

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Positive Edge-Triggered Register TSPC

Transistor sizing is critical to avoid Glitch


➢ Case D is 0
➢ Y and Q’ Both pulled to Zero initially and Q’ pulled
up again after Y brough sufficiently low
➢ M4 & M5 needs to be made stronger w.r.t. M7 & M8

1/30/2022 20

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Pulse Register

➢ Propagation Delay : Two Inverter Delay


➢ Set-up Time : Nil
➢ Hold Time : Glitch Time
➢ The disadvantage is a substantial increase in verification complexity
1/30/2022 21

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Sense amplifier Register

Sense amplifier circuits are used extensively in memory cores and in low swing bus drivers
to amplify small voltage swings present in heavily loaded wires.

1/30/2022 22

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Thank you

1/30/2022 23

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Advanced VLSI Design: 2021-22
Lecture 4 : Part-1
Pipelined Registers and
Static Timing Analysis

By Dr. Sanjay Vidhyadharan

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


4-Bit Adder

Performance Figure Value


Function Delay 3* _ ( ) _ ( )

Power 4
Throughput @ (3 * _ ( ) _ ( ))

Gate complexity 4*

2/6/2022 2

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Pipelined 4-Bit Adder
Principle of Operation:

Performance Figure Value


Function Delay Tclock (Tclock > _ + +
Power 4 + 22
Throughput @ TClock
2/6/2022 Gate complexity 4* + 22 * 3

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


4-Bit Multiplier

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


4-Bit Pipelined Multiplier

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Pipelining with Latches

A non-overlapping clock essential for correct operation. Else there will be race around

2/6/2022 6

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Pipelined Logic using C2MOS

Potential race condition during (0-0) overlap in C2MOS-based design


Similar considerations are valid for the (1-1) overlap.
2/6/2022

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


NORA CMOS

It combines C2MOS pipeline registers and NORA dynamic logic functional blocks.

2/6/2022 8

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


NORA CMOS
It combines C2MOS pipeline registers and NORA dynamic logic functional blocks.

NORA offers designers a wide range of design choices. Dynamic and static logic can be mixed freely. A NORA datapath
consists of a chain of alternating CLK and CLK modules. While one class of modules is precharging with its output latch
in hold mode, preserving the pre-vious output value, the other class is evaluating. Data is passed in a pipelined fashion
from module to module.
2/6/2022 9

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Advanced VLSI Design: 2021-22
Lecture 4 - Part 2
Pipelined Registers and
Static Timing Analysis

By Dr. Sanjay Vidhyadharan

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Timing Constraints of a Flip-flop
 Setup Time
 Hold Time

2/6/2022 11

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Clock Skew and Jitter
Clock skew (sometimes called timing skew) is a phenomenon in synchronous digital
circuit systems (such as computer systems) in which the same sourced clock signal arrives
at different components at different times.

Clock Jitter: Sometimes some external sources like noise, voltage variations may cause
to disrupt the natural periodicity or frequency of the clock. This deviation from the natural
location of the clock is termed to be clock jitter.

Clock Uncertainty = Clock Jitter + Clock Skew

2/6/2022 12

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Timing Constraints of a Flip-flop

Data setup violation caused by clock jitter

2/6/2022 13

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Timing Constraints of a Flip-flop

Data hold time violation caused by clock jitter

2/6/2022 14

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Static Time Analysis

There are two main problems that can arise in synchronous logic:
 Max Delay: The data doesn’t have enough time to pass from one register
to the next before the next clock edge.
 Min Delay: The data path is so short that it passes through several
registers during the same clock cycle.
 Max delay violations are a result of a slow data path, including the
registers, tsu therefore it is often called the “Setup” path.
 Min delay violations are a result of a short data path, causing the data to
change before the thold has passed, therefore it is often called the “Hold”
path.

2/6/2022 15

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Static Time Analysis
Setup (Max) Constraint

 After the clock rises, it takes tcq for the data to propagate to point A.
 Then the data goes through the delay of the logic to get to point B.
 The data has to arrive at point B, tsu before the next clock.

Setup Slack = Data Required Time – Data Arrival Time


 Positive Slack : No Timing Violation
2/6/2022
 Negative Slack : Timing Violation 16

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Static Time Analysis
Hold (Min) Constraint
Hold problems occur due to the logic changing before thold has passed.
This is not a function of cycle time – it is relative to a single clock edge!
• The clock rises and the data at A changes after tcq . The data at B changes tpd (logic) later.
• Since the data at B had to stay stable for thold after the clock (for the second register), the
change at B has to be at least thold after the clock edge.

Hold Slack = Data Arrival Time– Data Required Time


 Positive Slack : No Timing Violation
2/6/2022
 Negative Slack : Timing Violation 17

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Static Time Analysis

T= 15 ns

Setup slack = Min. Clock Path Delay - Max. Data Arrival Time
= (15 ns + 2ns + 5 ns + 2 ns - 4 ns) - ( 2 ns + 11 ns +2 ns + 9 ns + 2 ns)
= 20 ns – 26 ns = -6 ns : Setup Time Violation.
Hold time slack = Min. Data Arrival Time - Max. Clock Path Delay
= (1 ns + 9 ns + 1 ns + 6 ns + 1 ns) - (3 ns + 9ns + 3 ns + 2 ns) -
2/6/2022
= 18 ns – 17 ns = + 1ns : No Hold Time Violation. 18

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Timing Constraints of a Sequential Circuit
Maximum Clock Frequency

2/6/2022 19

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Static Time Analysis
Given the data setup time of the flop is 6ns, the hold time of the flop is 2ns, and the clock to
Q delay is given as 10ns.
a. Calculate the minimum clock period required to handle the circuit by drawing a digital
logic circuit for function clock frequency divided by 2.
b. Also determine the status of hold time violation and give a proper reason.

Tminimum >= Tsetup_time + Tclock_Q + dly


Tminimum >= 6ns + 10ns + 0
Tminimum >= 16ns
fmaximum = 1/16ns = 62.5MHz is the maximum possible frequency of operation
Thold_time <= Tclock_Q + delay
2ns <= 10ns + 0
2ns <= 10ns No Hold Time Violation
2/6/2022 20

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Thank you

2/6/2022 21

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Advanced VLSI Design: 2021-22
Lecture 5
Clock Generation and Distribution
Part-1

By Dr. Sanjay Vidhyadharan

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Phase-Locked Loops

Crystal oscillators generate accurate, low-jitter clocks with a frequency range


from 10’s of Megahertz to approximately 200MHz.

To generate a higher frequency required by digital circuits, a phase-locked loop


(PLL) structure is typically used. A PLL takes an external low-frequency
reference crystal frequency signal and multiplies its frequency by a rational
number N

2/13/2022 2

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Phase-Locked Loops

PLL

2/13/2022 3

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Phase-Locked Loops

The XOR as a phase detector

➢ Any deviation in a positive or negative direction from the perfect in-phase condition
produces the same change in duty factor resulting in the same average voltage.
➢ If the local clock is a multiple of the reference clock frequency, the output of the phase
detector will still be a square wave of 50% duty cycle, albeit at a different frequency
2/13/2022 4

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Phase-Locked Loops
Phase-Frequency Detector

2/13/2022 5

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Phase-Locked Loops
Phase-Frequency Detector

2/13/2022 6

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Phase-Locked Loops

Charge Pump

2/13/2022 7

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


VCO

2/13/2022 8

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


VCO

2/13/2022 9

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


VCO
Delay-Locked Loop

2/13/2022 10

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


VCO

2/13/2022 11

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


VCO

2/13/2022 12

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Thank you

2/13/2022 13

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Advanced VLSI Design: 2021-22
Lecture 5
Clock Generation and Distribution
Part-2

By Dr. Sanjay Vidhyadharan

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Clock Distribution

Network Types: H-Tree Model


➢ One large central driver
➢ Recursive H-style structure
to match wire-lengths
➢ Halve wire width at
branching points to reduce
reflections

Flip-flops

2/13/2022 15

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Clock Distribution

Network Types: H-Tree Model

Advantages
Low cost wiring
Low Capacitance
Disadvantages
Difficult to balance path delays due to
asymmetric FF distribution
Sensitive to variations
Flip-flops

2/13/2022 16

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Clock Distribution

Network Types: H-Tree Model

Slew degradation along long RC paths


unrealistically large central driver

Clock drivers can create large temperature A B


gradients (ex. Alpha 21064 ~30° C)

Inherently non-scalable (wire resistance


skyrockets)

Solution to some problems


Introduce intermediate buffers along
the way Specifically at branching points

2/13/2022 17

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Clock Distribution

Network Types: H-Tree Model


Alpha 21264 clock
distribution -- grid + H-
tree approach
Power = 32% of total

Wire usage = 3% of
metals 3 & 4

4 major clock quadrants, each with a large driver connected to local grid
structures
2/13/2022 18

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


CAD for Tree Architecture
• Topology generation
– H-tree: widely used
– Method of means and medians (MMM) [Jackson et al. DAC 90]
• Goal: reduce wirelength while minimizing skew.
• Divide set S of points into Sleft and Sright, based on median.
– | Sleft | = | Sright |
• Connect/route center of mass (CM) of S to CM of Sleft and Sright.
• Recurse on Sleft and Sright.

2/13/2022 19

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Clock Distribution

Grid/Mesh
• Gridded clock distribution was common on
earlier DEC Alpha microprocessors

• Advantages:
– Skew determined by grid density and not
overly sensitive to load position
– Clock signals are available everywhere
– Tolerant to process variations
– Usually yields extremely low skew values

2/13/2022 20

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Clock Distribution

Grid/Mesh
• Huge amounts of wiring & power
– Wire cap large
– Strong drivers needed – pre-driver cap large
– Routing area large
• To minimize all these penalties, make grid pitch coarser
– Skew gets worse
– Losing the main advantage
• Don’t overdesign
– let the skew be as large as tolerable Still
– grids seem non-feasible for SOC’s
2/13/2022 21

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Clock Distribution
Hybrid Architecture: Tree + Cross-links
• Reducing Clock Skew Variability via Cross Links
– [Rajaram et. al., DAC 2004]
• tree + short-circuit some sink pairs => non-tree topology
– clock signal propagates through multiple paths; reduces skew and skew
variability between shorted sinks
• reduces skew variability by 30-70%
• very small wire-length penalty (2%) over tree topology
• Drawback:
source
– does not consider buffering

2/13/2022 22

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Clock Distribution
Hybrid Architecture: Mesh + Trees

• Hybrid Structured Clock Network (Da, CDa)


Construction [Hu & Sapatnekar, ICCAD 01] source

– Hybrid clock topology


a c
• simple top-level global mesh
• zero-skew local trees at bottom
d
b
– Presents wire sizing scheme to achieve
latency and skew reduction..

2/13/2022 23

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Clock Architectures
Clock source

Mesh
-- excellent for low skew, jitter
-- high power, area, capacitance
Flip-flops
flip flops
-- difficult to analyze
-- used in modern processors
Tree
-- low cost (wiring, power, cap)
-- higher skew, jitter than mesh
-- widely used in ASIC designs Best architecture depends on the application
Clock source

Flip flops

crosslink
crosslink
tree

Hybrid: tree + cross-links


-- low cost (wiring, power, cap)
Hybrid: mesh + local trees
-- smaller skew, jitter than tree Local trees
-- suitable for coarse mesh
-- difficult to analyze Flip flops

2/13/2022 24

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Clock Gating

0
1
0/1 0
1

0
0/1 1
1/0
1

[5] M. Pedram and J.M. Rabaey, “Power aware design Methodologies,”


[Online]. Available: https://www.springer.com/gp/book/9781402071522
2/13/2022 25

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Clock Gating

[5] M. Pedram and J.M. Rabaey, “Power aware design Methodologies,”


[Online]. Available: https://www.springer.com/gp/book/9781402071522
2/13/2022 26

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Clock Gating

m Watts 20 % Reduction in Power due to Clock Gating

60 SHIFT REGISTER
SCALER
ALU m Watts

50
ADDER
Powe

FF
r

MUX

EQUAL TO

Before Clock After Clock


Gating Gating
[5] V. Natarajan, A. Nagarajan, N. Pandian, and V. Savithri “Power aware design Methodologies,” DOI: 10.5772/intechopen.73729
[Online]. Available: https://www.intechopen.com/books/very-large-scale-integration/low-power-design-methodology
2/13/2022 27

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Clock Gating

Cl M

S
Conventional
ll dder

S
Modified
ll dder

i e ns

2/13/2022 28

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Current conduction mechanism in CMOS and Tunnel FETs

2/13/2022 29

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Thank you

2/13/2022 30

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Advanced VLSI Design: 2021-22
Lecture 6
Latch Based Clocking & Asynchronous
Clocking

By Dr. Sanjay Vidhyadharan

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Pipelining
т𝑝𝑑 , 𝑃𝑎𝑑𝑑𝑒𝑟
A
+ Reg
B
+ Reg F
C Reg
+ Clock
D Clock
Performance Figure Value
Function Delay 2*TClock
Power 3 ∗ 𝑃𝑎𝑑𝑑𝑒𝑟 + 3 ∗ 𝑃𝑅𝑒𝑔𝑠𝑖𝑡𝑒𝑟
Throughput @ TClock ( т𝑐𝑙𝑘−𝑄 + т𝑝𝑑_𝑎𝑑𝑑𝑒𝑟 + т𝑠𝑒𝑡𝑢𝑝 )
Gate complexity 3 * 𝐺𝑎𝑑𝑑𝑒𝑟 + 2 * 𝐺𝑅𝑒𝑔𝑖𝑠𝑡𝑒𝑟
Functional Flexibility Nil
2/20/2022 Function Expandability Nil 2

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Latch Based Clocking

▪ A stable input is available to the combinational logic block A (CLB_A) on the falling edge of CLK1
(at edge 2 ) and it has a maximum time equal to the T CLK /2 to evaluate (that is, the entire low
phase of CLK1). On the falling edge of CLK2 (at edge 3 ), the output CLB_A is latched and the
computation of CLK_B is launched. CLB_B computes on the low phase of CLK2 and the output is
available on the falling edge of CLK1 (at edge 4 )
➢ It possible for a logic block to utilize time that is left over from the previous logic block
and this is referred to as slack borrowing
2/20/2022 3

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Latch Based Clocking

2/20/2022 4

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Latch Based Clocking

2/20/2022 5

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Synchronous Clocking

➢ Clock period is chosen to be larger than the worst-case delay of each pipeline stage.
➢ Hence, The throughput rate of the pipelined system is directly linked to the worst-case
delay of the slowest element in the pipeline

➢ As all the clocks in a circuit transitions at the same time, significant current flows over a
very short period of time (due to the large capacitance load). This causes significant noise
problems due to package inductance and power supply grid resistance.

2/20/2022 6

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Self-Timed Circuit Design

This avoids all problems and overheads associated with distributing high-speed clocks.

A self-timed circuit proceeds at the average speed of the hardware in contrast to the worst-case model
of synchronous logic.

The automatic shut-down of blocks that are not in use can result in power savings. Additionally, the
power consumption overhead of generating and distributing high-speed clocks can be partially avoided.

Self-timed circuits are by nature robust to variations in manufacturing and operating conditions such as
temperature.
2/20/2022 7

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Completion-Signal Generation

Dual-Rail Coding

Redundant signal representation to include transition state.

2/20/2022 8

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Completion-Signal Generation

Dual-Rail Coding

Dual-rail coding above allows tracking of the signal statistics, it comes at the cost of power
dissipation. Every single gate must transition for every new input vector, regardless of the
value of the data vector

2/20/2022 9

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


The Ripple-Carry Adder

𝐹𝑜𝑟 4 − 𝑏𝑖𝑡 𝐴𝑑𝑑𝑒𝑟 ≈ 12 𝐺𝑎𝑡𝑒 𝐷𝑒𝑙𝑎𝑦𝑠

𝐹𝑜𝑟 8 − 𝑏𝑖𝑡 𝐴𝑑𝑑𝑒𝑟 ≈ 24 𝐺𝑎𝑡𝑒 𝐷𝑒𝑙𝑎𝑦𝑠 𝐹𝑜𝑟 16 − 𝑏𝑖𝑡 𝐴𝑑𝑑𝑒𝑟 ≈ 48 𝐺𝑎𝑡𝑒 𝐷𝑒𝑙𝑎𝑦𝑠
2/20/2022 10

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Manchester-Carry Adder Circuit

𝐹𝑜𝑟 4 − 𝐵𝑖𝑡 𝐴𝑑𝑑𝑒𝑟 ≈ 6 𝐺𝑎𝑡𝑒 𝐷𝑒𝑙𝑎𝑦𝑠 𝐹𝑜𝑟 8 − 𝐵𝑖𝑡 𝐴𝑑𝑑𝑒𝑟 ≈ 12 𝐺𝑎𝑡𝑒 𝐷𝑒𝑙𝑎𝑦𝑠
2/20/2022 𝐹𝑜𝑟 16 − 𝐵𝑖𝑡 𝐴𝑑𝑑𝑒𝑟 ≈ 18 𝐺𝑎𝑡𝑒 𝐷𝑒𝑙𝑎𝑦𝑠 11

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Self-Timed Adder Circuit

2/20/2022 12

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Replica Delay

The advantage of this approach is that the logic can be implemented using a standard non-
redundant circuit style such as complementary CMOS.
Also, if multiple logic units are computing in parallel, it is possible to amortize the
overhead of the delay line over multiple blocks

2/20/2022 13

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Completion-Signal Generation using Current Sensing

2/20/2022 14

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Self-Timed Signaling

The four events, data change, request, data acceptance, and acknowledge, proceed in a cyclic
order. This protocol is called two-phase.
➢ The Req event terminates the active cycle of the sender. The sender is free to change the data
during its active cycle.
➢ The receiver’s cycle is completed by the Ack event. The receiver can only accept data during
its active cycle.
2/20/2022 15

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Muller C-element.

2/20/2022 16

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Self-Timed Signaling

Data Ready Ack Ack’ Req

0 0 1 0
0 -> 1 0 1 0 -> 1
1/1 -> 0 0 1 1
1/0 0 -> 1 1 -> 0 1/1 -> 0
1 -> 0 1 0 1 -> 0
2/20/2022 17

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Two-Phase Self-Timed FIFO

2/20/2022 18

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Four-Phase Self-Timed FIFO

2/20/2022 19

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Four-Phase Self-Timed FIFO

Data Ready Req S S’ Ack Ack’

0 0 0 1 0
0 -> 1 -> 0 0 -> 1 0 1 0
0->1
0 0 1 0 1
0 -> 1 -> 0 0 1 0 1
1 -> 0
0 1

2/20/2022 20

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Thank you

2/20/2022 21

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Advanced VLSI Design: 2021-22
Lecture 7
Interfacing Circuits – Part-1
Synchronizer and Arbiters

By Dr. Sanjay Vidhyadharan

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Synchronizer

PT Switch

2/27/2022 2

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Synchronizer

2/27/2022 3

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Synchronizer

𝐹𝑜𝑟 𝑇 = 0, 𝑅𝑎𝑛𝑔𝑒 = 𝑉𝐼𝐻 − 𝑉𝐼𝐿

𝐹𝑜𝑟 𝑇 = 2𝜏, 𝑅𝑎𝑛𝑔𝑒 = (𝑉𝐼𝐻 −𝑉𝐼𝐿 ) 𝑒 −2

𝐹𝑜𝑟 𝑇 = 4𝜏, 𝑅𝑎𝑛𝑔𝑒 = (𝑉𝐼𝐻 −𝑉𝐼𝐿 ) 𝑒 −4

2/27/2022 4

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Synchronizer
Vin is a periodical waveform with an average period T signal between transitions and
with identical rise and fall times tr .

Average synchronization errors per second with no synchronizer

Average synchronization errors per second with synchronizer

errors/sec

Mean time-to-failure (MTF) = 1/Nsyn(T)

2/27/2022 5

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Arbiters

2/27/2022 6

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Thank you

2/27/2022 7

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Advanced VLSI Design: 2021-22
Lecture 7
Interfacing Circuits – Part-2
Schmitt Triggers
By Dr. Sanjay Vidhyadharan

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Nosie Margin Variation with VDD

2/27/2022 2

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Delay Variation with VDD

2/27/2022 3

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Inverter Switching Threshold

2/27/2022 4

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Inverter Switching Threshold

2/27/2022 5

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Subthreshold Conduction

IDO is a process-dependent parameter that is dependent on VT


.
2/27/2022 6

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Subthreshold Inverter

Analysis of the subthreshold CMOS logic inverter


Sherif M.Sharroush
Ain Shams Engineering Journal Volume 9, Issue 4, December 2018, Pages 1001-1017

2/27/2022 7

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Schmitt Trigger

2/27/2022 8

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Schmitt Trigger

2/27/2022 9

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Schmitt Trigger

With Vin less than the threshold voltage of M1, VX remains at VDD - VTHN3

2/27/2022 10

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Schmitt Trigger Design Example
Design and simulate a Schmitt trigger using the short-channel CMOS process with
VSPL = 400 mV and VSPH = 700 mV. Given Vth = 0.25 V and (W/L) min = 10

2/27/2022 11

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Schmitt Trigger Delays

2/27/2022 12

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Applications of the Schmitt Trigger
Oscillator design using a Schmitt trigger

2/27/2022 13

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Applications of the Schmitt Trigger
Voltage-controlled oscillator using Schmitt trigger

2/27/2022 14

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Schmitt Trigger Power

2/27/2022 15

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Schmitt Trigger Design

Adjustment of Hysteresis Width

2/27/2022 16

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Schmitt Trigger Design

2/27/2022 17

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Schmitt Trigger

Vidhyadharan, A.S. and Vidhyadharan, S. "Improved hetero-junction TFET-based Schmitt


trigger designs for ultra-low-voltage VLSI applications", World Journal of Engineering,
Vol. ahead-of-print No. ahead-of-print. 2021 DOI: 10.1108/WJE-08-2020-0367

2/27/2022 18

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Schmitt Trigger

2/27/2022 19

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Schmitt Trigger

2/27/2022 20

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Thank you

2/27/2022 21

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Advanced VLSI Design: 2021-22
Lecture 8
Interfacing Circuits – Part-3
Level Shifters and IO PADS

By Dr. Sanjay Vidhyadharan

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


TTL to CMOS Level Shifter

2.0V

0.8V

Voltage level of TTL and CMOS The corresponding VTC


3/20/2022 2

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


CMOS Inverter Switching Threshold
The Switching Threshold, VM , is the point where Vin = Vout .
This can be calculated:
» Graphically, at the intersection of the VTC with Vin = Vout

3/20/2022 3

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


TTL to CMOS Level Shifter

Designing the Receiving Inverter Gate


• Adjust the TR ratio such that the Inverter Threshold VM is midpoint between
0.8V and 2.0V

3/20/2022 4

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


TTL to CMOS Level Shifter

𝑉𝐷𝐷 +𝑉𝑇𝑝 +𝑟𝑉𝑇𝑛


𝑉𝑀 =
1+𝑟

0.8+2
For 𝑉𝐷𝐷 = 5 𝑉 𝑎𝑛𝑑 VM = 2
= 1.4 𝑉 , 𝑉𝑇𝑛 = 1 𝑉, 𝑎𝑛𝑑 𝑉𝑇𝑝𝑛 = −1 𝑉

r = 6.5

𝜇𝑛 𝐶𝑜𝑥 𝑊𝑛 /𝐿𝑛
𝑟=
𝜇𝑝 𝐶𝑜𝑥 𝑊𝑝 /𝐿𝑝

𝑊𝑛 /𝐿𝑛 1 2 169
= 6.5 =
𝑊𝑝 /𝐿𝑝 3 12

3/20/2022 5

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


TTL to CMOS Level Shifter

Non-inverting TTL Level-shifting Circuit

3/20/2022 6

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


CMOS Level Shifter

VDDH = 1.8 V VDDH = 1.8 V


VDDH = 1.8 V

VT = 0.75 V VT = 0.75 V VT = 0.75 V

VDDL = 1.2 V
VDDL = 0.9 V VDDL = 0.7 V

Works Fine VDDH and VDDL


are Close.

3/20/2022 7

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


CMOS Level Shifter

Works Fine for VDDL > VT.

P.O. Pouliquen, in Proceedings of 2010 IEEE International Symposium on Circuits and


Systems (2010), pp. 40974100. DOI 10.1109/ISCAS.2010.5537627
3/20/2022 8

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


CMOS Level Shifter

VDDH = 1 V ( 90 nm)

VDDL = 0.18 V

VHVT = 0.535 V

VSVT = 0.360 V

VLVT = 0.230 V

M. Lanuzza, P. Corsonello, S. Perri, IEEE Transactions on Circuits and Systems II:


Express Briefs 59(12), 922 (2012). DOI 10.1109/TCSII.2012.2231037
3/20/2022 9

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Electrostatic Discharge (ESD)

(b) machine model


(a) Human Body 1.5 kV of static voltage stress

(c) charged device model, for ESD testing

Effective protection networks can withstand as high as 8-kV

3/20/2022 10

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Electrostatic Discharge (ESD)
Models for ESD testing

3/20/2022 11

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Electrostatic Discharge (ESD)
ESD Protection Network

-0.7 V < VA < VDD+.7V

1-3kὨ

3/20/2022 12

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Electrostatic Discharge (ESD)
Protection network with thick-oxide transistor

Ml and M2 have threshold values of 20 to 30 V.

3/20/2022 13

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Electrostatic Discharge (ESD)
Typical ESD failure modes.

3/20/2022 14

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Input Circuits

Internal Drivers to Prevent


Voltage drop in bonding wires

3/20/2022 15

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Output Circuits

12 transistors Four Very large Sized transistors

3/20/2022 16

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Output Circuits
High rate of change in the current di/dt and can cause significant on-chip noise problems due
to the L(di/dt)drop across the bonding wire connecting the output pad to the package.

Cload = 100 pF and t = 5 ns, then


𝑄 = 𝐶𝑉

𝐼𝑚𝑎𝑥 𝑡𝑠 𝑉𝐷𝐷
∗ = 𝐶𝑙𝑜𝑎𝑑
2 2 2
for a bonding wire with L = 2 nH,

This voltage drop would be quadrupled if t were reduced by a factor of two.


Trade-off between the delay time and the noise
17

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Output Circuits
In a CMOS chip the current surge can be as high as 1100 mA/ns at power and ground
terminals. For a microprocessor with 32 bits or higher number of data bus lines, the noise
problem can be significantly escalated if all output drivers are driven simultaneously. In such
cases, it is desirable to stagger the switching times with built-in delays

The role of two nMOS transistors controlled by the strobe signal (ST) is to pre-charge
the gate potentials of the last-stage driver transistors at an approximate midpoint
between the initial and final potentials of the load capacitor.
18

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Output Circuits

The circuit sends out only changes in


the data pattern

3/20/2022 19

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Output Circuits

3/20/2022 20

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Packages

Package functions
– Electrical connection of signals and power from chip to board
– Little delay or distortion
– Mechanical connection of chip to board
– Removes heat produced on chip
– Protects chip from mechanical damage
– Compatible with thermal expansion
– Inexpensive to manufacture and test

3/20/2022 21

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Package Types

3/20/2022 22

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Multichip Modules

3/20/2022 23

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Chip-to-Package Bonding

3/20/2022 24

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Heat Dissipation

3/20/2022 25

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Thank you

3/20/2022 26

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Advanced VLSI Design: 2021-22
Lecture 9
Arithmetic Circuits: Part-1

By Dr. Sanjay Vidhyadharan

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Microprocessor Design
Pads for Bus Control, Clock, Reset, Interrupts, Testing and Power Supply

Clock-Phase Bus Controller


Generators

State Instruction
Control Store
Sequencer Decoders

Control word Decoder


Instruction Pre-fetch Register

Internal A Bus

Address Data
out PC R0 R1 Rn Shifter ALU In/out
Buffer Reg
Internal B Bus

4/8/2022 Address Pads Data Pads 2

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


ALU Design

4/8/2022 3

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Half Adder

X Sum = X•Y’ + X’•Y = X ⊕ Y


HS Carry = X•Y
Y

Time Delay for the Half Adder?


1 gate delay
A gate delay of an xor for the half sum
CO A gate delay of an AND gate for the carry out
4

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Full Adder
Full-adder can also implemented with two half adders and one OR gate.

S = A ⊕ B ⊕ Cin
Cout = AB + BCin + ACin
= AB + Cin (A+B)
= AB + Cin (A ⊕ B+ AB)
= AB + Cin (A ⊕ B) + Cin AB = AB(1+ Cin) + Cin (A ⊕ B)
= AB + Cin (A ⊕ B)

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Full Adder

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


CMOS 28T Adder

4/8/2022 7

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


CMOS 28T Mirror Adder

𝐶𝑎𝑟𝑟𝑦 = 𝐴𝐵 + 𝐵𝐶 + 𝐶𝐴 = 𝐴𝐵 + 𝐶 𝐴 + 𝐵

𝑃𝑢𝑙𝑙 − 𝐷𝑜𝑤𝑛 𝑁𝑒𝑡𝑤𝑜𝑟𝑘 𝑓𝑜𝑟 𝐶𝑎𝑟𝑟𝑦 𝐵𝑎𝑟 = 𝐴𝐵 + 𝐶 𝐴 + 𝐵

𝑃𝑢𝑙𝑙 − 𝑈𝑃 𝑁𝑒𝑡𝑤𝑜𝑟𝑘 𝑓𝑜𝑟 𝐶𝑎𝑟𝑟𝑦 𝐵𝑎𝑟 = 𝐴′ 𝐵′ + 𝐵′ 𝐶 ′ + 𝐶 ′ 𝐴′


= 𝐴′ 𝐵′ + 𝐶′(𝐴′ + 𝐵′ )

4/8/2022 8

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


CMOS 28T Mirror Adder

12 12 12

12 12

6 6

6 6 6

4/8/2022 9

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Ripple Carry Adder
This is called Ripple Carry Adder, because of the
construction with full adders are connected in cascade.

Delay= 4 X Full Adder Delay = 8 Gate Delays


Delay= (N-1) tcarry + tsum 10

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Carry Look-Ahead Adder

1-bit CLA

B A

11

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Carry Look-Ahead Adder

CLA

CLLB

C1=G0+P0C0
12

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Carry Look-Ahead Adder

C2=G1+P1C1

C2 = G1+P1(G0+P0C0)
= G1+P1G0+P1P0C0

13

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Carry Look-Ahead Adder

C2= G1+P1G0+P1P0C0 C1=G0+P0C0


14

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Carry Look-Ahead Adder

C1=G0+P0C0

C2= G1+P1G0+P1P0C0

C3=G2+P2G1+ P2P1G0 +P2P1P0C0

C4=G3+P3G2+ P3P2G1 +P3P2P1G0 +P3P2P1P0C0

15

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Carry Look-Ahead Adder

16

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Carry Look-Ahead Adder

8 Bit Full Adder

Delay = 2 X 3 = 6 Gate Delay


17

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Ripple Carry Adder

8 Bit Full Adder

Delay = 8 X 2 Gate Delay


18

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Carry Look-Ahead Adder

19

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Carry Bypass or Carry Skip Adder

4/8/2022 20

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Carry Bypass or Carry Skip Adder

4/8/2022 21

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Carry Ripple versus Carry Bypass

4/8/2022 22

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Manchester Carry Chain

4/8/2022 23

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Linear Carry-Select Adder

4/8/2022 24

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Square Root Carry-Select Adder

𝑡𝑎𝑑𝑑 = 𝑡𝑠𝑒𝑡𝑢𝑝 + 𝑀𝑡𝑐𝑎𝑟𝑟𝑦 + 2𝑁 𝑡𝑚𝑢𝑥 + 𝑡𝑠𝑢𝑚

4/8/2022 25

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Square Root Carry-Select Adder
𝑁 𝐵𝑖𝑡 𝑎𝑑𝑑𝑒𝑟, 𝑀 − 𝐵𝑖𝑡𝑠 𝑖𝑛 𝐹𝑖𝑟𝑠𝑡 𝑆𝑡𝑎𝑔𝑒 , 𝑃 − 𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑆𝑡𝑎𝑔𝑒𝑠

𝑁 = 𝑀 + 𝑀 + 1 + 𝑀 + 2 + 𝑀 + 3 + ⋯ . +(𝑀 + 𝑃 − 1)

Series: a, a+d, a+2d,……,a+(n-1)d


𝑃(𝑃 − 1)
𝑁 = 𝑀𝑃 + sn = n/2(2a + (n-1)d)
2
𝑃 2 1
𝑁= + 𝑃(𝑀 − )
2 2

𝑀 ≪ 𝑁 𝑒. 𝑀 = 2 𝑎𝑛𝑑 𝑁 = 64

𝑃2
𝑁≈
2

4/8/2022 26

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Adder Delays - Comparison

4/8/2022 27

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Serial Adder
Initialize to
0

4/8/2022 28

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


4 Bit-Adder Subtractor
Add 4 & -3 Add -4 & -5 Add -8 & 4 Add 4 & 4

0100 1100 1000 0100


1101 1011 0100 0100
1 0001 1 0111 1100 1000

Overflow
29

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Binary Coded Decimal
General digital systems
User enters decimal → BCD i/p→ Binary i/p → compute in binary
→ Binary o/p → BCD o/p → Decimal output shown to user

30

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Binary Coded Decimal
BCD addition
4+5 4 0100
5 0101
9 1001 Expected Result

4+8 4 0100
8 1000
1100 Is this expected Result ?

Expected answer 0001 0010


is BCD of 12 31

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Binary Coded Decimal
BCD addition
4+8 4 0100
8 1000
Greater than 9 1100
0110 Add correction of +6
00010010
1 2 = To skip 6 invalid
states (10 - 15) BCDs

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Binary Coded Decimal
BCD addition
9+9 9 1001
9 1001
Carry out generated 1 0 0 1 0 Expected result ?
0110 Add correction of +6
00011000
1 8

After addition if carry out is generated or if sum is greater


than 9 there is need for correction

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Binary Coded Decimal
BCD addition 0000
0001
0010
0011
0100
0101
0110
0111
1000
1001

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Thank you

4/8/2022 35

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Advanced VLSI Design: 2021-22
Lecture 10
Arithmetic Circuits: Part-2

By Dr. Sanjay Vidhyadharan

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


The Binary Multiplication

4/10/2022 2

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


The Array Multiplier
A0B3 A0B2 A0B1 A0B0
A1B3 A1B2 A1B1 A1B0
A2B3 A2B2 A2B1 A2B0
A3B3 A3B2 A3B1 A3B0
A1B3 A1B2 A0B3 A1B1 A0B2 A1B0 A0B1 A0B0

HA FA FA HA

A2B3 A2B2 A2B1 A2B0

FA FA FA HA
A3B3 A3B2 A3B1 A3B0

FA FA FA HA

P7 P6 P5 P4 P3 P2 P1 P0
4/10/2022 3
ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION
The Array Multiplier
𝑀 𝐵𝑖𝑡 𝑋 𝑁 𝐵𝑖𝑡 −> 𝑃 = (𝑁 + 𝑀 − 1) A1B3 A1B2 A0B3 A1B1 A0B2 A1B0 A0B1 A0B0

HA FA FA HA

A2B3 A2B2 A1B3 A2B1 A2B0

FA FA FA HA
A3B3 A3B2 A3B1 A3B0

FA FA FA HA

P7 P6 P5 P4 P3 P2 P1 P0
𝑅𝑒𝑞𝑢𝑖𝑟𝑒 𝑁 − 1 𝑀 − 𝐵𝑖𝑡 𝐴𝑑𝑑𝑒𝑟𝑠 𝑅𝑒𝑞𝑢𝑖𝑟𝑒 𝑀 𝑋 𝑁 𝐴𝑁𝐷 𝐺𝑎𝑡𝑒𝑠

𝑡𝑚𝑢𝑙 = 𝑀 − 1 + 𝑁 − 2 𝑡𝑐𝑎𝑟𝑟𝑦 + 𝑁 − 1 𝑡𝑠𝑢𝑚 + 𝑡𝑎𝑛𝑑


4/10/2022 4

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


The Carry-Save Multiplier
A0B0
HA HA HA HA

HA FA FA FA

HA FA FA FA
Vector Merging Adder

HA FA FA HA

P7 P6 P5 P4 P3 P2 P1 P0

𝑅𝑒𝑞𝑢𝑖𝑟𝑒 𝑁 𝑀 − 𝐵𝑖𝑡 𝐴𝑑𝑑𝑒𝑟𝑠 𝑅𝑒𝑞𝑢𝑖𝑟𝑒 𝑀 𝑋 𝑁 𝐴𝑁𝐷 𝐺𝑎𝑡𝑒𝑠

𝑡𝑚𝑢𝑙 = 𝑡𝑎𝑛𝑑 + 𝑁 − 1 𝑡𝑐𝑎𝑟𝑟𝑦 + 𝑡𝑚𝑒𝑟𝑔𝑒


4/10/2022 5

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Multiplier Floorplan

4/10/2022 6

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Transmission Gate XOR

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Adder Cells in Array Multiplier

4/10/2022 8

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Sequential Multiplier

4/10/2022 9

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Booth Algorithm

4/10/2022 10

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Booth Algorithm
Example:

4/10/2022 11

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Comparators

• 0’s detector: A = 00…000


• 1’s detector: A = 11…111
• Equality comparator: A = B
• Magnitude comparator: A < B

4/10/2022 12

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


1’s & 0’s Detectors
• 1’s detector: N-input AND gate
• 0’s detector: NOTs + 1’s detector (N-input NOR)

4/10/2022 13

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Equality Comparator
• Check if each bit is equal (XNOR, aka equality gate)
• 1’s detect on bitwise equality

4/10/2022 14

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Magnitude Comparator

1 0 1

1 1 0

15

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Magnitude Comparator

1 0 1

1 1 0

16

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Magnitude Comparator
1-Bit Comparator A0 B0 G0 L0 E0
A0 0 0 0 0 1
B0 0 1 0 1 0
Let G0 indicate greater, L0 indicate 1 0 1 0 0
Lesser and E0 indicate equal 1 1 0 0 1

G0= A0B0’

L0= A0’B0

E0= (A0 ⊕ B0)’ = (A0 ʘ B0) = (G0 + L0)’

17

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Magnitude Comparator
1-Bit Comparator G0= A0B0’
A0
B L0= A0’B0
0
E0= (G0 + L0)’ = (A0 ⊕ B0)’

B0 A0

1-Bit
Comp

L0 E0 G0

18

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Magnitude Comparator
2-Bit Comparator B1 A1 B0 A0

A A1A0 1-Bit 1-Bit


Comp Comp
B BB
1 0
L1 E1 G1 L0 E0 G0

A = B When A0 = B0 and A1=B1


AND
E = E0 . E1

E (A=B)

19

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Magnitude Comparator
B1 A1 B0 A0
2-Bit Comparator
1-Bit 1-Bit
A A1A0 Comp Comp

B BB
L1 E1 G1 L0 E0 G0
1 0

E1 G0
A > B When A1 > B1 OR
When A1 = B1 and A0 > B0 AND G1

G= G1 + E1 . G0
OR

G
20

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Magnitude Comparator
B1 A1 B0 A0
2-Bit Comparator
1-Bit 1-Bit
A A1A0 Comp Comp

B BB
L1 E1 G1 L0 E0 G0
1 0

E1 L0
A < B When A1 < B1 OR
When A1 = B1 and A0 < B0 AND L1

L = L1 + E1 . L0
OR

21

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Magnitude Comparator
1-Bit Comparator

G =G0 L =L0 E =E0

2-Bit Comparator
G = G1 + E1. G0 L = L1 + E1. L0 E = E1. E0

3-Bit Comparator
G = G2 + E2. G1 + E2. E1. G0
E = E2 .E1. E0
L = L2 + E2. L1 + E2. E1. L0

22

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Magnitude Comparator

L3
4-Bit Comparator
E3 E = E2 .E1. E0
G3
G = G3 + E3.G2 + E3.E2. G1 + E3.E2. E1. G0
L2 L = L3 + E3.L2 + E3.E2. L1 + E3.E2. E1. L0
E2

G2

L1
E1
G1

L0
E0
G0

23

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Shifters

Logical Shift:
– Shifts number left or right and fills with 0’s
• 1011 LSR 1 = 0101, 1011 LSL1 = 0110

• Arithmetic Shift:
– Shifts number left or right. Rt shift sign extends
• 1011 ASR1 = 1101 1011 ASL1 = 0110

• Rotate:
– Shifts number left or right and fills with lost bits
• 1011 ROR1 = 1101 1011 ROL1 = 0111

4/10/2022 24

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


The Binary Shifter

4/10/2022 25

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Multi-bit Shifters

• Cascade one-bit shifters?


• Complex, unwieldy, slow for larger number of shifts
• Two other types of shifters
– Barrel shifter
– Logarithmic shifter

4/10/2022 26

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


The Barrel Shifter

Area dominated by wiring


Propagation delay is theoretically constant

4/10/2022 27

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


The Barrel Shifter

4/10/2022 28

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Logarithmic Shifter

4/10/2022 29

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Shifter Using Mux

4/10/2022 30

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Thank you

4/10/2022 31

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Advanced VLSI Design: 2021-22
Lecture 11: Memory Design

By Dr. Sanjay Vidhyadharan

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Semiconductor Memory Classification

Volatile Non-Volatile Non-Volatile


Read-Write Memory Read-Write Memory Read-Only Memory

Random Access Non-Random Access


EPROM
Mask-Programmed
E2PROM
Programmable (PROM)
FIFO FLASH
SRAM LIFO
DRAM Shift Register
CAM

4/17/2022 2

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Memory Timing: Definitions

4/17/2022 3

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Memory Architecture

Decoder reduces the number of select signals K = log2N


Problem: ASPECT RATIO or HEIGHT >> WIDTH

4/17/2022 4

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Array-Structured Memory Architecture

Better ASPECT RATIO

4/17/2022 5

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Hierarchical Memory Architecture

Advantages:
1. Shorter wires within blocks
2. Block address activates only 1 block => power savings
4/17/2022 6

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Hierarchical Memory Architecture

Column Column Column

Advantages:
1. Shorter wires within blocks
2. Block address activates only 1 block => power savings

4/17/2022 7

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


6T SRAM

Write bit bit_b


word
➢ Drive one bitline high, the other low P1 P2
N2 N4
➢ Then turn on wordline
A A_b
➢ Bitlines overpower cell with new value
N1 N3
➢ Writability
➢Must overpower feedback inverter
A_b
➢N2 >> P1
1.5 A

Ex: A = 0, A_b = 1, bit_b


1.0
bit = 1, bit_b = 0
Force A_b low, then A rises high
0.5
word

0.0
0 100 200 300 400 500 600 700
time (ps)

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


6T SRAM

Read bit bit_b


word
➢ Precharge both bitlines high
P1 P2
➢ Then turn on wordline N2 N4
➢ One of the two bitlines will be pulled A A_b
down by the cell N1 N3
➢ Read stability
➢A must not flip A_b bit_b

➢N1 >> N2 1.5

1.0
bit
Ex: A = 0, A_b = 1 word

bit discharges, bit_b stays high 0.5


But A bumps up slightly A
0.0
0 100 200 300 400 500 600
time (ps)

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


6T SRAM

SRAM Sizing
High bitlines must not overpower inverters during reads
But low bitlines must write new value into cell

bit bit_b
word
weak
med med
A A_b
strong

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


6T SRAM

SRAM Sense Amplifier

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


3-Transistor DRAM Cell

No constraints on device ratios Reads are non-destructive


Value stored at node X when writing a “1” = VWWL-VTn

4/17/2022 12

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


1-Transistor DRAM Cell

Write: CS is charged or discharged by asserting WL and BL.


Read: Charge redistribution takes places between bit line and storage capacitance

Voltage swing is small; typically around 250 mV

4/17/2022 13

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


1-Transistor DRAM Cell

➢ 1T DRAM requires a sense amplifier for each bit line, due to charge redistribution
read-out

➢ DRAM memory cells are single-end in contrast to SRAM cells.

➢ The read-out of the 1T DRAM cell is destructive; read and refresh operations are
necessary for correct operation.

➢ Unlike 3T cell, 1T cell requires presence of an extra capacitance that must be


explicitly included in the design.

➢ When writing a “1” into a DRAM cell, a threshold voltage is lost. This charge loss
can be circumvented by bootstrapping the word lines to a higher value than VDD

4/17/2022 14

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Read-Only Memory
Internal Logic 32 X 8 ROM

In General for 2k X n ROM


K x 2k Decoder and n OR gates
Each OR gate has 2k inputs
15

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Programming the ROM
Address 3 = 10110010 is permanent storage using fuse link

X : means connection

16

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Programming the ROM
1. Masking During Metallization

17

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Programming the ROM
2. Fuse (PROM)

18

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Programming the ROM
3. EPROM

19

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Programming the ROM
3. EPROM

20

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Programming the ROM

4. EEPROM

21

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Programming the ROM

22

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Programming the ROM

Example: Design a combinational circuit using a ROM. The


circuit accepts a 3-bit number and generates an output binary
number equal to the square of the input number.

23

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Programming the ROM
Example: Design a combinational circuit using a ROM. The
circuit accepts a 3-bit number and generates an output binary
number equal to the square of the input number.

24

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


MOS OR ROM

4/17/2022 25

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


MOS NOR ROM

4/17/2022 26

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


MOS NAND ROM

4/17/2022 27

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Pre-charged MOS NOR ROM

4/17/2022 28

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Row Decoders

Example 8 bit Decoder

1. Implementation 8 Inverters + 256 NAND

2. Implementation 8 Inverters + 256 NOR

4/17/2022 29

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Row Decoders
Multi-stage implementation improves performance

4/17/2022

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Row Decoders

4/17/2022 31

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


4-to-1 tree based column decoder

Number of devices drastically reduced. Delay increases quadratically with # of sections;


prohibitive for large decoders
Solution : Buffers
4/17/2022 32

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Programable Logic Array (PLA)

Programmable device capable of implementing


functions expressed in SOP.
➢ Consists of input buffers and inverters followed by:
➢ Programmable AND plane, followed by
➢ Programmable OR plane.
➢ Can implement m logic functions of n variables. Limit is the number of
product terms that can be generated inside of the device.

33

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Programable Logic Array (PLA)

34

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Programable Logic Array (PLA)

35

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Programable Array Logic Array (PAL)

36

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Programable Array Logic Array (PAL)

37

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Programable Array Logic Array (PAL)

38

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Programable Array Logic Array (PAL)
Example

W(A, B, C, D) = ∑ (2, 12, 13)

X(A, B, C, D) = ∑ (7,8,9,10,11,12,13,14,15)

Y(A, B, C, D) = ∑ (1,2,8,12,13)

W(A, B, C, D) = ABC’ + A’B’CD’

X(A, B, C, D) = A + BCD
Y(A, B, C, D) = ABC’ + A’B’CD’+ AC’D’ + A’B’C’D

= W + AC’D’ + A’B’C’D

39

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Programable Array Logic Array (PAL)
Example
W(A, B, C, D) = ABC’ + A’B’CD’
X(A, B, C, D) = A + BCD
Y(A, B, C, D) = W + AC’D’ + A’B’C’D

Product AND Inputs Outputs


A B C D w
1 1 1 0 -- -- W = ABC’ + A’B’CD’
2 0 0 1 0 -- Section 1
3 -- -- -- -- --
4 1 -- -- -- -- X = A + BCD
5 -- 1 1 1 -- Section 2
6 -- -- -- -- --
7 -- -- -- -- 1 Y = W + AC’D’ + A’B’C’D
8 1 -- 0 0 -- Section 3
9 0 0 0 1 --

40

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Programable Array Logic Array (PAL)

Example

W(A, B, C, D) = ABC’ + A’B’CD’

X(A, B, C, D) = A + BCD

Y(A, B, C, D) = W + AC’D’ + A’B’C’D

41

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Flash Storage

• Most prominent solid state storage technology


– No other technology is available at scale
• NAND- and NOR- flash types available
– NOR-flash can be byte-addressed, expensive
– NAND-flash is page addressed, cheap
– Except in very special circumstances, all flash-storage we see are NAND-flash

4/17/2022 42

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Flash Storage
Flash memories store information in memory cells made from floating gate transistors.

NOR flash is faster to read than NAND flash, but it's also more expensive.
NAND has a higher memory capacity than NOR.
4/17/2022 43

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Flash Storage
NOR flash

4/17/2022 44

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Flash Storage
NOR flash

NOR FLASH memories are very fast to program and read. Erasure through tunneling is
much slower. However, this kind of array suffers from low density due to the same
reason that impacts NOR ROM density the need for multiple grounds.

4/17/2022 45

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Flash Storage
NAND flash

4/17/2022 46

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Content Addressable Memory (CAM)

4/17/2022 47

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


CAMs

10T CAM Cell

4/17/2022 48

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


CAMs

4/17/2022 49

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Thank you

4/17/2022 50

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Advanced VLSI Design: 2021-22
Lecture 12-A: Low Power VLSI Design
Part-1: Gate Level Optimization

By Dr. Sanjay Vidhyadharan

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Overview of Power Consumption

4/24/2022 2

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Overview of Power Consumption

4/24/2022 3

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Overview of Power Consumption

4/24/2022 4

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Transition Activity

4/24/2022 5

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Transition Activity

4/24/2022 6

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Transition Activity

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Transition Activity

• Low-Power Gate-Level Design


• Low-Power Architecture-Level Design
• Algorithmic-Level Power Reduction
• RTL Techniques for Optimizing Power

4/24/2022 8

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Gate-Level Design – Technology Mapping

4/24/2022 9

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Gate-Level Design – Technology Mapping

𝛼1 = (1 − 0.2 ∗ 0.2 ) 0.2 ∗ 0.2 = 0.0384


𝛼2 = (1 − 0.04 ∗ 05 ) 0.04 ∗ 0.5 = 0.0196

4/24/2022 10

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Gate-Level Design – Technology Mapping

Switching Activity Minimization in


Combinational Logic Design
R. V. Menon, S. Chennupati, N. K.
Samala, D. Radhakrishnan and B. Izadi

4/24/2022 11

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Gate-Level Design – Phase Assignment

4/24/2022 12

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Gate-Level Design – Pin Swapping

➢ An input signal to a gate is called critical if it is the last signal of all inputs to assume a
stable value.
➢ The path through the logic which determines the ultimate speed of the structure is called
the critical path.
➢ Putting the critical-path transistors closer to the output of the gate can result in a speed-up.

4/24/2022 13

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Gate-Level Design – Pin Swapping

4/24/2022 14

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Gate-Level Design – Glitching Power

4/24/2022 15

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Gate-Level Design – Glitching Power

4/24/2022 16

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Static Glitch Example

4/24/2022 17

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Static Glitch Elimination

18

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Static Glitch Elimination

19

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Gate-Level Design – Precomputation

4/24/2022 20

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Gate-Level Design – Precomputation

4/24/2022 21

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Gate-Level Design – Clock Gating

4/24/2022 22

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Gate-Level Design – Clock Gating

4/24/2022 23

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Gate-Level Design – Input Gating

4/24/2022 24

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Reduced-Power Shift Register

4/24/2022 25

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Thank you

4/24/2022 26

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Advanced VLSI Design: 2021-22
Lecture 12-B: Low Power VLSI Design
Part-2: Architecture, Algorithmic &
RTL Level Optimization

By Dr. Sanjay Vidhyadharan

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Architecture-Level Design – Parallelism

4/24/2022 2

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Architecture-Level Design – Pipelining

Performance Figure Circuit (a) Circuit (b)


Function Delay in ns 2 ns 2 ns
Throughput @ 2ns @ 1 ns

4/24/2022 3

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Architecture-Level Design – Retiming

4/24/2022 4

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Architecture-Level Design – Retiming

4/24/2022 5

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Architecture-Level Design – Bus Segmentation

• Avoid the sharing of resources


Reduce the switched capacitance
• For example: a global system bus
A single shared bus is connected to all modules, this structure results in
a large bus capacitance due to
The large number of drivers and receivers sharing the same bus
The parasitic capacitance of the long bus line
• A segmented bus structure
Switched capacitance during each bus access is significantly reduced
Overall routing area may be increased

4/24/2022 6

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Architecture-Level Design – Bus Segmentation

4/24/2022 7

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Algorithmic-Level Design – factivity Reduction

• Two-bit binary counter:


State sequence, 00 → 01 → 10 → 11 → 00
Six bit transitions in four clock cycles
6/4 = 1.5 transitions per clock

• Two-bit Gray-code counter


State sequence, 00 → 01 → 11 → 10 → 00
Four bit transitions in four clock cycles
4/4 = 1.0 transitions per clock

4/24/2022 8

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Algorithmic-Level Design – factivity Reduction
Binary Counter

Gray Counter

4/24/2022 9

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Algorithmic-Level Design – factivity Reduction

4/24/2022 10

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Algorithmic-Level Design – factivity Reduction

4/24/2022 11

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


FSM State Encoding

4/24/2022 12

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Bus Encoding for Reduced Power

• Example: Four bit bus


0000 → 1110 has three transitions.
If bits of second pattern are inverted then 0000 ,
then 0000 → 0001 will have only one transition.
• Bit-inversion encoding for N-bit bus

4/24/2022 13

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Bus Encoding for Reduced Power

4/24/2022 14

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


RTL-Level Design – Datapath Reordering

4/24/2022 15

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


RTL-Level Design – Memory Partition

BHE

BHE

BHE

4/24/2022 16

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


RTL-Level Design – Memory Partition

4/24/2022 17

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


RTL-Level Design – Memory Partition

4/24/2022 18

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Thank you

4/24/2022 19

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Advanced VLSI Design : 2021-22
Lecture 12 C
Low Power VLSI Design – Part-3
Adiabatic Logic

By Dr. Sanjay Vidhyadharan

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Adiabatic Logic
1. Adiabatic Operation

2
Adiabatic Logic
CMOS Symmetric Pass Gate Adiabatic Logic

3
Adiabatic Logic

4
Adiabatic Logic vs. Static CMOS

Sanjay Vidhyadharan et all “An advanced adiabatic logic using Gate Overlap Tunnel FET
(GOTFET) devices for ultra-low power VLSI sensor applications”
5
Thank you

5/1/2022 6

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Advanced VLSI Design : 2021-22
Lecture 13
Interconnects

By Dr. Sanjay Vidhyadharan

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Interconnects

• Chips are mostly made of wires called interconnect


– In stick diagram, wires set size
– Transistors are little things under the wires
– Many layers of wires
• Wires are as important as transistors
– Speed
– Power
– Noise
– Area
• Alternating layers run orthogonally

2
Wire Geometry

3
Layer Stack

4
Choice of Metals

5
Metal Layers 45 nm Technology

6
Wire Resistance

7
Sheet Resistance

8
Contacts Resistance

9
Wire Capacitance

10
Capacitance Trends

11
Capacitance Trends

12
Capacitance Trends

13
Polysilicon

• Diffusion capacitance is very high (about 2 fF/mm)


– Comparable to gate capacitance
– Diffusion also has high resistance
– Avoid using diffusion runners for wires!

• Polysilicon has lower C but high R


– Use for transistor gates
– Occasionally for very short wires between gates

14
Lumped Element Models

15
Lumped Element Models
L Model

The delay of a wire is a quadratic function of its length! This means that doubling
the length of the wire quadruples its delay
16
Lumped Element Models
Example

17
Lumped Element Models

Elmore delay model

18
Repeaters

19
Repeaters

20
Crosstalk

A capacitor does not like to change its voltage instantaneously

A wire has high capacitance to its neighbor


When the neighbor switches from 1→0 or 0→1, the wire tends
to switch too Called capacitive coupling or crosstalk

Crosstalk effects
Noise on nonswitching wires
Increased delay on switching wires

21
Crosstalk Delay

22
Crosstalk

𝐶𝑎𝑑𝑗 ∆𝑉𝑎𝑔𝑔 − ∆𝑉𝑣𝑖𝑐 = 𝐶𝑔𝑛𝑑 (∆𝑉𝑣𝑖𝑐 )

23
Crosstalk
Driven Victims

If the noise is less than the noise margin, nothing happens Static CMOS logic will eventually
settle to correct output even if disturbed by large noise spikes

But glitches cause extra delay, also cause extra power from false transitions

Dynamic logic never recovers from glitches.

Memories and other sensitive circuits also can produce the wrong answer
24
Thank you

5/1/2022 25

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Advanced VLSI Design : 2021-22
Lecture 14-A
Deep-Submicron MOSFET operation

By Dr. Sanjay Vidhyadharan

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


MOSFET Operation

Cut-off Region

𝐹𝑜𝑟 𝑉𝐺𝑆 < 𝑉𝑇 𝐼𝐷 = 0

5/8/2022 2

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


The Threshold Voltage

As a practical definition, The threshold Voltage VT is that gate voltage when


the surface is said to be inverted, i.e. the density of mobile electrons on the
surface becomes equal to the density of holes in the bulk (p-type) substrate.
5/8/2022 3

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


The Threshold Voltage

1. The work function difference TGC between the gate


and the channel reflects the built-in potential of the MOS
system, which consists of the
p-type substrate, the thin silicon dioxide layer, and the
gate electrode. Depending on the gate material, the work
function difference is
∅𝐺𝐶 = ∅𝐹_𝑆𝑢𝑏𝑠𝑡𝑟𝑎𝑡𝑒 - ∅𝐹𝑚𝑒𝑡𝑎𝑙 𝐹𝑜𝑟 𝑀𝑒𝑡𝑎𝑙 𝐺𝑎𝑡𝑒

∅𝐺𝐶 = ∅𝐹_𝑆𝑢𝑏𝑠𝑡𝑟𝑎𝑡𝑒 - ∅𝐹𝑃𝑜𝑙𝑦𝑠𝑖𝑙𝑖𝑐𝑜𝑛 𝐹𝑜𝑟𝑃𝑜𝑙𝑦𝑠𝑖𝑙𝑖𝑐𝑜𝑛 𝐺𝑎𝑡𝑒


2. The externally applied gate voltage must be changed to
achieve surface inversion, i.e., to change the surface
potential by - 2 ∅ F. This will be the second component of
the threshold voltage.

5/8/2022 4

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


The Threshold Voltage

3. Another component of the applied gate voltage is necessary


to offset the depletion region charge, which is due to the fixed
acceptor ions located in the depletion region near the surface.

4. There always exists a fixed positive charge density Qox at


the interface between the gate oxide and the silicon
substrate, due to impurities and/or lattice imperfections at
the interface. The gate voltage component that is necessary
to offset this positive charge at the interface is - QOX/Cox.

5/8/2022 5

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


The Threshold Voltage

The body effect occurs in a MOSFET when the source is not tied to the substrate
(which is always connected to the most negative power supply in the integrated circuit
for n-channel devices and to the most positive for p-channel devices). The substrate
then acts as a “second gate” or a back-gate for the MOSFET
5/8/2022 6

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


MOSFET Current
Linear Region- Small VDS
𝑉𝑜𝑣 = 𝑉𝐺𝑆 - 𝑉𝑇

𝑄
𝐶𝑜𝑥 𝑊𝐿 =
𝑉𝑜𝑣
𝑄
𝐶ℎ𝑎𝑟𝑔𝑒 𝑝𝑒𝑟 𝑢𝑛𝑖𝑡 𝐿𝑒𝑛𝑔𝑡ℎ = = 𝐶𝑜𝑥 𝑊𝑉𝑜𝑣
𝐿

𝑉𝐷𝑆
𝐸𝑙𝑒𝑐𝑡𝑟𝑖𝑐 𝐹𝑖𝑒𝑙𝑑 𝑖𝑛 𝐶ℎ𝑎𝑛𝑛𝑒𝑙 =
𝐿
𝑉𝐷𝑆 µ𝑛
𝑉𝑒𝑙𝑜𝑐𝑖𝑡𝑦 𝑜𝑓 𝐶ℎ𝑎𝑟𝑔𝑒 𝑖𝑛 𝐶ℎ𝑎𝑛𝑛𝑒𝑙(𝑣) = µ𝑛 𝐸 =
𝐿
𝑄 𝑉𝐷𝑆 µ𝑛
𝐶𝑢𝑟𝑟𝑒𝑛𝑡 𝑖𝑛 𝐶ℎ𝑎𝑛𝑛𝑒𝑙 (𝐼𝐷 ) = 𝑣 ∗ 𝐿 = * 𝐶𝑜𝑥 𝑊𝑉𝑜𝑣
𝐿

µ𝑛𝐶𝑜𝑥 𝑊(𝑉𝐺𝑆 − 𝑉𝑇 )𝑉𝐷𝑆


𝐼𝐷 =
𝐿
5/8/2022 7

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


MOSFET Current
Linear Region- Small VDS

𝑉𝑜𝑣

µ𝑛𝐶𝑜𝑥 𝑊(𝑉𝐺𝑆 − 𝑉𝑇 )𝑉𝐷𝑆


𝐼𝐷 =
𝐿
µ𝑛𝐶𝑜𝑥 𝑊𝑉𝑜𝑣
𝑇𝑟𝑎𝑛𝑠𝑐𝑜𝑛𝑑𝑢𝑐𝑡𝑎𝑛𝑐𝑒 𝑜𝑓 𝐶ℎ𝑎𝑛𝑛𝑒𝑙 𝑔𝐷𝑆 =
𝐿
𝑘𝑛′ 𝑊𝑉𝑜𝑣 𝑉𝐷𝑆
𝑃𝑟𝑜𝑐𝑒𝑠𝑠 𝑡𝑟𝑎𝑛𝑠𝑐𝑜𝑛𝑑𝑢𝑐𝑡𝑎𝑛𝑐𝑒 𝑝𝑎𝑟𝑎𝑚𝑒𝑡𝑒𝑟 𝑘𝑛′ = µ𝑛𝐶𝑜𝑥 𝐼𝐷 =
𝐿

5/8/2022 8

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


MOSFET Current
Linear Region as VDS is Increased

Charge in the tapered channel is proportional to the


channel cross-sectional area
𝑉𝐷𝑆
𝑘𝑛′ 𝑊(𝑉𝑜𝑣 − 2 )𝑉𝐷𝑆
𝐼𝐷 =
𝐿
5/8/2022 9

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


MOSFET Current
Saturation Region

𝑉𝐷𝑆 = 𝑉𝑜𝑣 = 𝑉𝐺𝑆 - 𝑉𝑇


𝑉𝐷𝑆
𝑘𝑛′ 𝑊(𝑉𝑜𝑣 − 2 )𝑉𝐷𝑆
𝐼𝐷 =
𝐿
𝑘𝑛′ 𝑊(𝑉𝐺𝑆 − 𝑉𝑇 )2
𝐼𝐷 =
2𝐿
5/8/2022 10

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


MOSFET Current

𝑘𝑛′ 𝑊(𝑉𝐺𝑆 − 𝑉𝑇 )2
𝐼𝐷 =
2𝐿

5/8/2022 11

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


MOSFET BODY CONNECTION

➢ NMOS Body to Lowest Possible Potential (Gnd)


➢PMOS Body to Highest Possible Potential (VDD)

5/8/2022 12

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


MOSFET BODY CONNECTION

5/8/2022 13

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


MOSFET BODY CONNECTION

5/8/2022 14

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


MOSFET BODY CONNECTION

➢ NMOS Double well : Body not same as Substrate

5/8/2022 15

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


MOSFET LATCH

5/8/2022 16

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


MOSFET LATCH

5/8/2022 17

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


MOSFET LATCH

5/8/2022 18

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Deep-submicron MOSFET operation

5/8/2022 19

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Deep-submicron MOSFET operation

➢ Threshold voltage reduction


➢ VT Roll Off
➢ Drain-induced barrier lowering (DIBL)
➢ Mobility degradation due to a vertical field
➢ Velocity saturation effects
➢ Channel length modulation
➢ Subthreshold (weak inversion) conduction
➢ Hot-electron effects on output resistance

5/8/2022 20

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


VT Variation

➢ VT Roll Off
➢ Drain-induced barrier lowering (DIBL)

5/8/2022 21

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Mobility Degradation
There also exists a normal (vertical) field originating from the gate voltage that further
inhibits channel carrier mobility. This effect, which is called mobility degradation, reduces
the surface mobility with respect to the bulk mobility.

5/8/2022 22

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Velocity Saturation Effect
When the electric field reaches a critical value EC the velocity of the carriers tends to
saturate.

5/8/2022 23

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Velocity Saturation Effect

5/8/2022 24

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Velocity Saturation Effect

5/8/2022 25

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Channel Length Modulation

When the VDS is increased beyond VOV , the pinch-off point is moved slightly away from the drain,
toward the source. The additional voltage applied to the drain appears as a voltage drop across the narrow
depletion region between the end of the channel and the drain region. This voltage accelerates the
electrons that reach the drain end of the channel and sweeps them across the depletion region into the
drain.

5/8/2022 26

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Channel Length Modulation

5/8/2022 27

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Channel Length Modulation

𝑉𝐴 𝐸𝑎𝑟𝑙𝑦 𝑉𝑜𝑙𝑡𝑎𝑔𝑒
1
𝜆= λ ∝ 1/L
𝑉𝐴

5/8/2022 28

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Subthreshold Conduction

where IS and n are empirical parameters, with


n > 1 and typically ranging around 1.5.

Subthreshold current has some important repercussions. In general, we want the


current through the transistor to be as close as possible to zero at VGS = 0. This is
especially important in the so-called dynamic circuits, which rely on the storage of
charge on a capacitor and whose operation can be severely degraded by subthreshold
leakage.
5/8/2022 29

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Hot Carrier Effects

Increase in the electric field strength causes an increasing energy


of the electrons.
➢ Some electrons are able to leave the silicon and tunnel into the gate oxide.
➢ Such electrons are called “Hot carriers”.
➢ Electrons trapped in the oxide change the VT of the transistors.
➢ This leads to a long term reliability problem.
➢ For an electron to become hot an electric field of 104 V/cm is necessary.
➢ This condition is easily met with channel lengths below 1µm.

5/8/2022 30

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Thank you

5/8/2022 31

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Advanced VLSI Design : 2021-22
Lecture 14-B
CMOS Scaling

By Dr. Sanjay Vidhyadharan

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION


Technology Scaling

• Currently, technology scaling has a threefold objective:


– Reduce the gate delay by 30% (43% increase in frequency)
– Double the transistor density
– Saving 50% of power (at 43% increase in frequency)
• How is scaling achieved?
– All the device dimensions (lateral and vertical) are reduced by 1/
– Concentration densities are increased by 
– Device voltages reduced by 1/ (not in all scaling methods)
– Typically 1/ = 0.7 (30% reduction in the dimensions)

Trieste, 8-10
CMOS technology 2
November 1999
ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION
Technology Scaling
• The scaling variables are:
– Supply voltage: Vdd → Vdd / 
– Gate length: L → L/
– Gate width: W → W/
– Gate-oxide thickness: tox → tox / 
– Junction depth: Xj → Xj / 
– Substrate doping: NA → NA × α

This is called constant field scaling because the electric field across
the gate-oxide does not change when the technology is scaled
If the power supply voltage is maintained constant the scaling is called
constant voltage. In this case, the electric field across the gate-oxide
increases as the technology is scaled down.
Due to gate-oxide breakdown, below 0.8µm only “constant field”
scaling is used.
3
ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION
Technology Scaling

Some consequences 30% scaling in the


constant field regime ( = 1.43, 1/ = 0.7):

• Device/die area:
W  L → (1/)2 = 0.49
– In practice, microprocessor die size grows about 25% per technology
generation! This is a result of added functionality.
• Transistor density:
(unit area) /(W  L) → 2 = 2.04
– In practice, memory density has been scaling as expected.
(not true for microprocessors…)

4
ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION
Technology Scaling
• Gate capacitance:
W  L / tox → 1/ = 0.7
• Drain current:
(W/L)  (V2/tox) → 1/ = 0.7
• Gate delay:
(C  V) / I → 1/ = 0.7
Frequency →  = 1.43
– In practice, microprocessor frequency has doubled every
technology generation (2 to 3 years)! This faster increase rate is
due to two factors:
• the number of gate delays in a clock cycle decreases with time (the
designs become highly pipelined)
• advanced circuit techniques reduce the average gate delay beyond
30% per generation.

5
ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION
Technology Scaling
• Power:
C  V2  f → (1/)2 = 0.49
• Power density:
1/tox  V2  f → 1
• Active capacitance/unit-area:

Power dissipation is a function of the operation frequency, the power
supply voltage and of the circuit size (number of devices).
If we normalize the power density to V2  f we obtain the active
capacitance per unit area for a given circuit. This parameter can be
compared with the oxide capacitance per unit area:
1/tox →  = 1.43
– In practice, for microprocessors, the active capacitance/unit-area
only increases between 30% and 35%. Thus, the twofold
improvement in logic density between technologies is not
achieved.
6
ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION
Technology Scaling

• Interconnects scaling:
– Higher densities are only possible if the interconnects
also scale.
– Reduced width → increased resistance
– Denser interconnects → higher capacitance
– To account for increased parasitics and integration
complexity more interconnection layers are added:
• thinner and tighter layers → local interconnections
• thicker and sparser layers → global interconnections and
power
Interconnects are scaling as expected

7
ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION
Technology Scaling
Quantity Sensitivity Constant Field Constant Voltage
Scaling Parameters
Length L 1/S 1/S
Width W 1/S 1/S
Gate Oxide Thickness tox 1/S 1/S
Supply Voltage Vdd 1/S 1
Threshold Voltage VT0 1/S 1
Doping Density NA, ND S S2
Device Characteristics
Area (A) WL 1/S2 1/S2
 W/Ltox S S
D-S Current (IDS) (Vdd - vT)2 1/S S
Gate Capacitance (Cg) WL/tox 1/S 1/S
Transistor On-Resistance (Rtr) Vdd/IDS 1 S
Intrinsic Gate Delay () RtrCg 1/S 1/S
Clock Frequency f f f
Power Dissipation (P) IDSVdd 1/S2 S
Power Dissipation Density (P/A) P/A 1 S3 8
ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION
Technology Scaling

Lithography:
Optics technology Technology node
248nm mercury-xenon lamp 180 - 250nm
248nm krypton-fluoride laser 130 - 180nm
193nm argon-fluoride laser 100 - 130nm
157nm fluorine laser 70 - 100nm
13.4nm extreme UV 50 - 70nm

9
ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION
Technology Scaling

Lithography:
• Electron Beam Lithography (EBL)
– Patterns are derived directly from digital data
– The process can be direct: no masks
– Pattern changes can be implemented quickly
– However:
• Equipment cost is high
• Large amount of time required to access all the points
on the wafer

CMOS technology 10
ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION
Thank you

5/8/2022 11

ELECTRICAL ELECTRONICS COMMUNICATION INSTRUMENTATION

You might also like