AICD Lab5 OpAmp 180tsmc 3
AICD Lab5 OpAmp 180tsmc 3
Objectives:
To design an un-buffered two stage CMOS Operational Amplifier with an n-channel input
pair
To study the large signal dc characteristics of the Op-Amp
To study the small signal characteristics of the Op-Amp
Lab 5-1. To write the specification of a two stage CMOS Op-Amp and to fix the
transistor sizes and bias currents from the level 2 model parameters of the NMOS
and PMOS transistors.
The design procedures of the Two-stage CMOS Op-Amp are described in great details in the book :
CMOS Analog Circuit Design, Second Edition by Philips E. Allen and Douglas R. Holberg,
OXFORD University Press. You are required to study section 6.3 Design procedure for the Two-
stage CMOS Op-Amps (Page 269-280) carefully and design the op-amp using typical NMOS and
typical PMOS transistors of the Analog library provided by Cadence.
In this lab we will use tsmc 180nm pdk obtained under a Non-disclosure agreement signed between
BUET and TSMC. You must sign the student portion of the NDA and must not copy it or use it for
any purpose other than this lab and your project.
Create a new library named opamp and attach tsmc 180 nm process from the following library.
/home/cad/euromixed/33v/mixed33v/tsmc18
All NMOS are nmos3v and all PMOS are pmos3v used in the design.
simulator lang=spice
* VTI-derived Level=2 nominal model
.model nmos4 nmos level=2 vto = 0.775 gamma=0.4 tox = 400e-10 nsub = 8e+15 xj = 0.15U ld =
0.20U u0 = 650 ucrit = 0.62e+5 uexp = 0.125 vmax = 5.1e+4 neff = 4.0 delta = 1.4 rsh = 36 cgso =
1.95e-10 cgdo = 1.95e-10 cj = 195U cjsw = 500P mj = 0.76 mjsw = 0.30 pb = 0.8
.model pmos4 pmos level=2 vto = -0.75 gamma=0.57 tox = 400e-10 nsub = 6e+15 xj = 0.05U ld =
0.20U u0 = 255 ucrit = 0.86e+5 uexp = 0.29 vmax = 3.0e+4 neff = 2.65 delta = 1.0 rsh = 101 cgso =
1.90e-10 cgdo = 1.90e-10 cj = 250U cjsw = 350P mj = 0.535 mjsw = 0.34 pb = 0.8
PRE_LAB : From the model parameters and with reference to Figure 6.3-2 of the Book CMOS
Analog Circuit Design by Philips E Allen, and Douglas R. Holberg design the various
components of the Op-Amp and fill up the table below. Please fill up the form before coming
to the Lab for this experiment. You will not be allowed to proceed further without completing
the PRE-LAB.
1.8 V
W=7.5u W=7.5u W=67u
L=1u L=1u L=1u
30uA 3p
10p
W=3u W=3u
L=1u L=1u
1.8 V
W=4.5u W=4.5u W=14u
L=1u L=1u L=1u
Vss
Find the transfer characteristics of the Op-amp and determine the output offset voltage of the
amplifier. For this purpose set the voltage of the V+ terminal as a variable and then make a
coarse sweep of this variable form -1.8 Volts to +1.8 Volts and run a dc analysis to find the
values of Vin where the output makes the transition from VDD to VSS. Once the transition
range is found, Vin is swept over values that only include the transition region. Use the
following voltage specification for the V+ and the V- terminal.
For this purpose we will perform the ac analysis in spectre. In the ac sweep use frequency as
a sweep variable, and vary the frequency from 10 kHz to 1 GHz. Set sweep type as automatic.
Now a plot of output voltage magnitude in V versus frequency in Hz will be plotted out.
However, we want to plot the output voltage in dB. For this purpose in the graph window
execute Tools → Calculator. The Virtuoso Visualization and Analysis Calculator window
will appear as follows. In the calculator window execute Options → Algebric. Now double
click the dB20 function and it will appear in the evaluation window.
The objective of this simulation is to get an output that is equal to CMRR or can be
related to CMRR. The following figure shows a method that can accomplish this objective.
Two identical voltage sources designated as Vcm are placed in series with both op amp inputs
where the op amp is connected in the unity gain configuration. To accomplish this first create
a symbol of the Op-Amp and then connect it as shown below. For this circuit it can be shown
V 1
that out
Vcm CMRR
The amplifier CMRR is measured by observing how the voltage offset changes as the input
ommon-mode voltage at the amplifier’s input stage changes. Similarly you can measure the
PSRR by changing the power supply voltages and how the voltage offset changes.
The equivalent circuit of the op-amp is as shown below from where we can derive CMRR.
To simulate PSRR we could use the xf analysis in spectre. With the xf analysis, you'd specify the
frequency sweep and the output of the circuit. Then after simulation, you could use the direct plot
form and find the transfer function from Vdd to the output for PSRR+ and Vss to the output for
PSRR- from your single simulation run.
The circuit diagram of the test cell is shown below. The output Vout is shorted to V+ with a 0.1 ohm
resistor in series to avoid SPICE connection problem.
4. The initial pin and transistor placement in layout will look like this:
After rearranging the transistor we can come out with a convenient layout like below:
6 Parasitic Extraction.
Now perform Parasitic extraction from the layout and run SPECTR SPICE with the parasitic
included in the net list. For this purpose you need to execute QRC Setup Quantus QRC
Report :
1. Provide the circuit diagram and performance curves of the circuits
2. Compare in a table the design specifications with the simulated specifications.
3. Discuss about the discrepancy between the design specifications and the simulated
specifications.
4. Simulate the design at fast fast and slow slow corner and compare the specification again.