University of Pennsylvania

Department of Electrical and Systems Engineering

ESE206: Electrical Circuits and Systems II – Lab

MOSFET Amplifier Mini project

1. Objectives

The objective of this mini-project is to design and build a NMOS common-source amplifier.
The design goal is to maximize the amplification and to explore the limits of voltage gain
using a single transistor. In addition you will design the amplifier according to specifications
of lower 3-dB cut-off frequency and input impedance. Specifically you will:

1. Learn to design a common-source amplifier using a NMOS transistor

2. Learn what parameters determine the voltage gain and how to maximize it.
3. Learn the low-frequency response of the amplifier
4. Verify the design with PSpice
5. Build the amplifier
6. Measure the amplifier's characteristics

2. Background

The common-source amplifier is shown in Figure 1 below. It is a similar amplifier as the one
you used in the previous lab. During this lab you analyzed the biasing circuit and measured
the voltage-gain. The goal of the new lab assignment is to design the amplifier, build and test
it.

Figure 1: Common-Source Amplifier with a NMOS transistor.

JVdS/2006 1 MOSFET Amplifier design /ESE216

2.1. Voltage Gain, Input and Output resistance

We already know that voltage gain GV is given by

RG RG
Gv   g m (R D || R L || ro )   g (R || RL ) (1)
R G R sig RG R sig m D
The open-circuit overall voltage gain is given by

RG
Gvo   g R (2)
RG R sig m D
in which gm is the transconducance given by

W ' W
g m k n (VGS V t )  2I D k n
'
(3)
L L

The input resistor is RG= R G1||RG2 and output resistor R out=RD.

2.2 Low-Frequency Response

The frequency response at low frequencies is limited by the coupling capacitor CC1 and CC2 ,
and the bypass capacitor CS . Assuming that the poles associated with each capacitor does not
influence each other one can easily find the poles. The expression of the pole associated with
capacitor Ci can be written as,

 3dB 1 1
f 3dB    (4)
2  2  2 C i Rieq

in which Rieq is the equivalent resistor seen over the terminals of capacitor Ci when the other
capacitors are short circuited. Let us apply this for each of the three capacitors CC1, CC2 and
CS:

1
f pCc1  (5)
2C C 1 ( R G R s i g )

1
f pCc 2  (6)
2C C 2 (RD R L )
1 1 gm
f pCs    (7)
2 C S (RS || 1 ) 2C R S 2C S
gm
1 g m R S
S

JVdS/2006 2 MOSFET Amplifier design /ESE216

The simplification is expression (7) is valid as long as R S >> 1/gm. These three formulas can
be used to find the values of the three capacitors, as explained below.
2.3 Selecting the capacitors

One notices that the resistor seen by the bypass capacitor is 1/gm which is a relatively small
value. As a result, we will usually make the pole associated with C S to be the 3-dB frequency
fL.

1 1 gm
f L f p C s   
2 C S (R S || 1 ) RS 2 C S
gm 2C S
1 g m R S

The poles associated with the coupling capacitors are usually made about 10 times smaller
than the one caused by the bypass capacitor. Since the resistors RG and R D are pretty large,
the capacitors are not as large as the bypass capacitor.

fpC1 = f pC2 = f L/10

3. Pre-lab Assignment

Your task is to design a NMOS common-source amplifier, as shown in Figure 1 with

maximum gain. The amplifier has the following specification:

- Maximize the open-circuit voltage gain Gvo; you should get an amplification of at
least -25 V/V but not larger than -50 V/V without a load resistance R L.
- Input impedance 10Kohm or larger, the
- Lower3-dB frequency fL should be maximum 50 Hz and
- The output swing should be at least 5Vpp.

The NMOS transistor has the following characteristics: Vt =1.2V; kn 'W/L=0.7mA/V2, =

0.004V. The signal resistor Rsig = 50 Ohm.

3.1 Reading

Review the background material of the previous lab (MOSFET Lab2 - Biasing and the
Common Source Amplifier) or read the sections "Common-Source Amplifier" in the
textbook Microelectronic Circuits by Sedra-Smith (5th ed), section 4.7.3, and "Frequency
Response of the CS Amplifier", sections 4.9.1 and 4.9.3.

3.2 Hand calculations: Design

a. Find: the values of the resistors, and DC current ID. There are many possible
solutions that will fulfill the spec for the mid-frequency amplification. Make
judicious choices and use the rule of thumbs when appropriate.

JVdS/2006 3 MOSFET Amplifier design /ESE216

 Hint: Write down the expression for the amplification as a function of ID and R D.
This gives one relationship between ID and RD. Is there another relationship you can
come up with that will help you find the values of ID and RD? You should select
values for the resistors that are available in the lab: see list of available components.

b. Once you have determined the DC current and all resistors find the transconductance
gm and small signal output resistance ro of the amplifier. Also, verify that the
amplification Gvo is what you intended it to be.

c. What is the minimum drain voltage and corresponding output swing at the drain?
Also calculate the DC power dissipation.

d. What is the input Rin and output resistance Rout of the amplifier? Now, add a load
resistor RL of 15kOhm. What is the overall voltage gain GV? Calculate the DC
power dissipation. Notice the drastic effect of adding a load resistance on the
amplification? Is this drastic dependence on the load resistance related to the fact
that your open-circuit amplification is very large? Is there a way to reduce the
influence of the load resistance?

e. Determine the values of the coupling capacitor and the bypass capacitor. You can
use the procedure outlined in the background section. Select values of the capacitors
that are available in the lab.

3.3 PSpice simulations

a. Enter the schematic. Do not include the load resistor RL yet. Use for the NMOS
transistor one of the CD4007 transistors (same as the one used in the lab). The
model can be found in the ese216lib library. This library can be downloaded from
the ESE216 website. Save both the cd4007.lib and CD4007.OLB in your directory.
Add the library path to the PSpice and the Simulator (see Adding Vendor Libraries
in the PSPICE Primer).
b. Do a Bias simulation and verify all DC voltages and DC currents. Compare these
values with the calculated one.
c. Next apply a sinusoidal input signal of 0.05 Vampl and 5kHz. Do a transient
simulation and display the signal at the drain. Verify that the voltage amplification is
close to the specified one.
d. Now add a 15 kOhm load and find the amplification Gv. Compare it with the hand
calculations.
e. Do an AC analysis and plot the magnitude of the voltage gain Gvo and GV. Do a
sweep from 0.1 Hz up to 1MHz (note: Spice uses MEG for Mega and not M).
Determine fL . Where are the lower poles located? Do they correspond to the one you
used to design the circuit?

JVdS/2006 4 MOSFET Amplifier design /ESE216

3.4 Summarize

Summarize your results in table format shown below

Specification/Calculation Simulated results

GVo (V/V)
GV (V/V)

Rin (kOhm) ---

Rout (kOhm) ---
V Dmin (V) ---
Max. output swing (Vpp) ---
Power dissipation (mW)
fL (Hz)

4. In-Lab Experiments

4.1 Parts

HFC4007UB MOS transistor array (data sheet from ST Microelectronics)

Resistors
Capacitors
Oscilloscope with FFT module
Digital multimeter (Voltage and Current meter)

4.2 Procedure

a. Build the amplifier you designed as part of the prelab shown in Figure 3 below. Use
the HFC4007 (or CD4007) MOSFET transistor array, shown in the Figure 3 below.
It may be convenient to use a 50kOhm potentiometer for RG2 . Do not add RL.

JVdS/2006 5 MOSFET Amplifier design /ESE216

 Figure 2: Common source amplifier

Figure 3: The HFC4007UB

MOSFET array. Pin 7 is connected
to the substrate of the NMOS and
should be connected to the most
negative voltage of the circuit (0r if
one uses only one NMOS transistor it
can be shorted to the source); pin 14
is the bulk of the PMOS and should
be connected to the most positive
voltage in the circuit. (Source:
National Semiconductor CD4007
Datasheet)

b. Measure the DC voltages (VD, VG and VS) and the current ID. Compare these values
with the one of your design. Measure also the total current supplied by the power
supply. What is the corresponding power dissipation?

c. Do not connect a load resistor RL yet. Apply a sinusoidal input signal of amplitude
of 50mV and frequency of 5 kHz. Measure the output voltage. If the output is
distorted you need to decrease the input amplitude (use a voltage divider if needed).
Verify the open-circuit voltage gain Gvo. It is possible that the amplification is
somewhat off from the calculated one. This is due to the fact that the actual
transistor is different from the one used in the calculations and the simulations. As
long as you get an amplification about 40V/V it should be ok. Also, take the FFT of
the output signal and determine the total harmonic distortion THD.

d. Increase the input signal to determine the maximum output signal swing before
noticeable distortion occurs? Take the FFT of the maximum output signal and
determine its total harmonic distortion THD. How much has the THD increased as
compared to the one measured in the previous section?

e. Adjust the input amplitude to 50mV and vary the frequency of the input signal
between 1Hz and 1MHz. These measurements are done without a load resistor RL.
Take about 10 measurement points per decade around the low frequency poles.
Determine fL Ones the frequency response is flat you can take one or tow
measurements per decade (e.g. 1kHz, 4kHz, 10kHz, 40kHz, 100kHz, etc) up to the
point where the amplitude starts to decrease. Measure the high frequency 3dB point,
fH. Take several measurements around this frequency so that you can draw the Body
plot. What is the bandwidth and the gain-bandwidth product of your amplifier?

JVdS/2006 6 MOSFET Amplifier design /ESE216

 f. Apply a load resistor of 15kOhm and determine the voltage gain Gv. Compare the
result with the one from the pre-lab.

Follow the guidelines for writing the report. In addition, the report should include:

a. Summary of the design process and results.

b. Experimental results and a comparison with the design values. If there are
differences explain what may have caused these differences between the calculated
and measured values. Is there an disadvantage of having such alarge amplification
(consider bandwidth, gain-bandwidth, output impedance, load dependency).
c. Table summarizing the calculation, PSpice and measured results.
d. Discussion.

Created by Jan Van der Spiegel

jan_at_seas.upenn.edu

April 17, 2006

JVdS/2006 7 MOSFET Amplifier design /ESE216