General Instructions To The Candidate

BIRLA INSTITUTE OF TECHNOLOGY AND SCIENCE, PILANI
EEE C443/ EEE F313/ INSTR F313: Analog and Digital VLSI Design
I- Semester 2017-2018
Comprehensive Exam (Open Book)
DATE: 11.Dec.2017 M.M-100
TIME: 3 hrs
General Instructions to the candidate
Please write your answers legibly and neatly.

Symbols, constants, terminology used have their usual meaning unless specified specially.
There are a total of FIVE questions and all are compulsory. Each carries marks as indicated.
Answer the sub-parts of a question at one place and in the order in which they appear.
Preferably start each question on a fresh page however sub parts can be written in continuation.
Common data: Use the following common data if not mentioned specifically in the question
For 0.5µm Technology node
• Use long channel approximation and lambda based design rules unless specially mentioned.
• Use square law current equation unless specifically mentioned.
• Neglect body effect, channel length modulation if not mentioned specifically.
• The body of all PMOS is tied to VDD while all NMOS is tied to ground unless specially mentioned.
Q1 Answer the following-

(I). For Fig. Q1, each flip flop has:
• Setup time of 3 ns
• Clock-to-Q maximum delay of 4ns
• Clock-to-Q minimum delay of 2ns
• Each AND/OR gate has:
Propagation delay of 5ns
Contamination delay of 3ns.
• NOT gate has min delay of 1ns
and max 2ns, while XOR gate has
min delay of 2ns and maximum
delay of 8ns.
a) What is the highest clock frequency that could be safely used on this circuit? Clearly show your
calculations at one place.
b) In order for this circuit to work correctly, what would be the acceptable value for the hold time
requirement of the D flip-flops? Clearly show your calculations at one place.
c) Now, assume a clock skew of 6ns between Clk-1 and Clk-2, Now, identify whether there is any
timing violation. If yes, then propose the remedy and justify your calculations without changing
clock frequency.
(13M)
(II). In order to drive a large capacitance (CL = 20 pF) from a minimum size inverter gate (with input
capacitance CIN = 10fF), you have decided to introduce a two-staged buffer. Assume that the
propagation delay of a minimum size inverter is 70 ps. Also, assume that the input capacitance of a
gate is proportional to its size.
a) Determine the sizing of the two additional buffer stages that will minimize the propagation delay.
b) If you could add any number of stages to achieve the minimum delay, how many stages would
you insert? What is the propagation delay in this case?
(7M)
(I). A CPU design consumes 0.3mW/MHz when fabricated using a 0.13 µm process. The area of the
design is 0.7 mm2. Assume a 200 MHz clock frequency, and 1.2 V power supply. Take long channel
devices.
a) If the supply voltage of the CPU design (scaled to 90 nm) is reduced to 1V, What will be area,
frequency, power consumption and power density.
b) What supply voltage should be fixed to keep the power density equal to the original power
density (without scaling). Also find new frequency and power consumption? Mention the type
of scaling.
(II). An NMOS transistor is fabricated with the following physical dimensions and dopant
concentrations: tox = 200 Å, W = 10 µm, Lm ( Mask length of channel) = 1.5 µm, LD (length of drain
region) = 5µm, xd (overlap) = 0.25 µm, xj = 0.4 µm, ND = 1020 cm3, Substrate Doping NA = 1016 cm3,
Channel Stop Implant Doping NA (side wall) = 1019cm3, Φo (junction built-in potential under zero
bias) =0.933 V, Φswo (sidewall junction built-in potential under zero bias) = 1.111V
Cjo = 2.98*10-8 F/cm2, Cjswo = 8.23*10-7 F/cm2. Assume an abrupt junction.
a) Determine the capacitance at drain junction for VDB = 5V
b) Calculate the overlap capacitance between gate and drain.
[20M]

a) Implement the following function using 3 stage TSPC logic style such that each stage performs
some logic computation (where, first stage implements A+BC). Inputs available are A, B, D, F, and
H.
Y = [ F + ( A + BC ). D]H
b) What are the limitations for each input transition for proper operation of each stage?
c) For circuit of part (a) If VDD= 2.5 V, VIH=1.6 V, VIL=0.8 V, Ileakage=0.9 pA (maximum leakage current in
each stage), pre-charge and pre-discharge time of each stage is 0.8 ns and worst evaluation time
of each stage is 4 ns.
d) (Given total capacitance at each output node is 6 fF (i.e. before each tristate inverter and after
each tristate inverter present in each stage). Now calculate maximum and minimum frequency of
operation? Assume 50 % Duty Cycle of clock.
[20M]
Q4 Consider a CMOS inverter of (fig. 4) with following specifications where switching threshold (VM) is
1.25 V.
VDD=2.5 V, VT0N = 0.4 V, VT0P = -0.4 V, µnCOX=120 µA/V2, µpCOX=40 µA/V2, Ln=Lp=1 µm
a) Now, VM must be reduced from 1.5 V to 1 V. Due to layout

constraints, only adjustable parameter is width of n-MOS
(Wn). If, Wn=2 µm when VM =1.25 V, what is the new Wn for
VM=1 V.
b) Voltage transfer characteristics (VTC) of the above CMOS

Fig. 4
inverter (VM =1.25 V and Wn=2 µm) is shown in Fig. 4.
Calculate the gain Av (Fig. 4) of the CMOS inverter considering
λn=λp=0.235 V-1
c) Calculate VIL and VIH from Fig. 4 using the gain (Av) of CMOS
inverter. Also, calculate the value of NML and NMH.
[20 M]
Q5 For the OPAMP circuits shown in Fig.5 (a), and (b). ro. . Take load capacitance at node Vout1 and
Vout2= CL. Iss is implemented with basic current mirror circuit. Given Vdd=3V, Vin (dc)=1V, Vgs of all
transistors =0.9V, Vtn= lVtpl =0.7V, λp,n=0.01 V-1,
Do the following parts (i to v) for fig. 5(a), and fig. 5(b) separately. And fill the results in Table1. (make
this table in your answer sheet)
a) Determine D.C. voltage at nodes x, m, Vb1. Hence determine ICMR.
b) Identify the amplifier configuration with following transistors--(M1, M2, M3, M4), (M5, M7), (M9,
M11)
c) Intuitively, determine the expression for overall differential small signal gain 'Adm'= [(Vout1-
Vout2)/ vin].
d) Intuitively, determine the expression for pole frequency at node x, node n, and node Vout1.
Neglect parasitic capacitances. Identify dominant pole/ s
e) Neglecting zero/s and non-dominant poles, qualitatively find phase margin and comment on
stability of OPAMP in feedback mode.
f) Intuitively, derive the expression for zero frequency due to mirror node n. In fig. 5a, what will
happen to zero frequency if gate connection of M9, and M10 transistor is connected to x, and y
nodes respectively.
g) Sketch and label a circuit to generate Vdd and temperature variation compensated Iss.
[20]
Table-1
Fig. 5(a) Fig. 5(b)
D C VOLTAGE:
• X
• Y
• M
• N
• Vout1
• Vout2
ICMR
Amplifier Configuration:
• (M1, M2, M3, M4)
• (M5, M7)
• (M9, M11)
Adm= [(Vout1-Vout2)/ vin]
Pole frequency:
• Node x
• Node m
• Node vout1
Phase margin
Stability
Zero frequency due to
mirror node n (wz)
-------------------------------------------------------------------------------------------------------------

General Instructions To The Candidate

Uploaded by

Copyright:

Available Formats

General Instructions To The Candidate

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

General Instructions To The Candidate

Uploaded by

Copyright:

Available Formats

BIRLA INSTITUTE OF TECHNOLOGY AND SCIENCE, PILANI

General Instructions to the candidate

Please write your answers legibly and neatly.

For 0.5µm Technology node

Q1 Answer the following-

Cjo = 2.9810-8 F/cm2, Cjswo = 8.2310-7 F/cm2. Assume an abrupt junction.

a) Determine the capacitance at drain junction for VDB = 5V

b) Calculate the overlap capacitance between gate and drain.

Q3 Answer the following-

a) Now, VM must be reduced from 1.5 V to 1 V. Due to layout

b) Voltage transfer characteristics (VTC) of the above CMOS

You might also like