Sheet 2

Alexandria University
Faculty of Engineering
Electrical Engineering Department
EE431: Digital Integrated Circuits
Sheet 2: MOS Dynamic Characteristics
1) Consider switching delays for 1 pF in a 10-k resistive-load inverter circuit, where
a. Find PHL (50% high-to-low transition delay) by using the average-current

method. Assume that the input signal is an ideal rectangular pulse switching
between 0 and 5 V with zero rise/fall times. You will have to calculate VOL to
solve this problem.
b. By using an appropriate differential equation and the proper initial voltage across
the capacitor (when the input voltage is at VOH) which is VOL and not 0 V,
calculate PLH. Use the same input voltage as in part (a).
2) Consider a CMOS ring oscillator consisting of an odd number (n) of identical inverters
connected in a ring configuration as shown in Figure 1. The layout of the ring
oscillator is such that the interconnection (wiring) parasitics can be assumed to be
zero. Therefore, the delay of each stage is the same and average delay is called
intrinsic delay (p) as long as identical gates are used. The ring oscillator circuit is
often used to quote the circuit speed of a particular technology using ring oscillator
frequency (f).
a. Derive an expression for the intrinsic delay (p) in terms of number of stages n.
b. Show that (p) is independent of the transistor sizing (it remains the same when all
gates are scaled uniformly up or down).
Figure 1: ring oscillator
3) Suppose that the resistive-load inverter examined in Exercise 1 is connected to a load

capacitance of 1.0 pF which is initially discharged. The gate of the NMOS transistor is
driven by a rectangular pulse which changes from high to low at t=0. As a result, the
nMOS transistor begins to charge up the capacitor. Solve the following two parts by
using the differential equations and not by using the average-current methods.
a. Determine the 50% low-to-high delay time, which is defined as the time difference
between 50% points of input and output waveforms when the output waveform
switches from low to high.
b. Determine the 50% high-to-low delay time, which is defined as the time difference
between 50% points of input and output waveforms when the output waveform
switches from high to low when the capacitor is initially charged to 5.0 V.
4) A resistive-load inverter with RL=50 k has the following device parameters:
A ring oscillator is formed by connecting nine identical inverters in a closed loop. We

are interested in finding the resulting oscillation frequency.
a. Find the delay times of the inverter for an ideal step pulse whose voltage swing is
between VOL and VOH, i.e., PHL and PLH. It should be noted that the loading
capacitance of each inverter is strictly due to the drain parasitic capacitance and the
gate capacitance of the following stage. For simplicity, neglect the drain parasitics
and assume Cload is equal to gate capacitance.
b. The rise and fall times are defined between. 10% and 90% of the full voltage
swing. But for simplicity, we will assume that fall = 2 PHL and rise= 2 PLH.
Estimate the actual propagation delays PHL and PLH by using the rise and fall
times of the inverter and the ideal delays found in (a).
c. Find the oscillation frequency from the information in part (b).
5) Consider a CMOS inverter, with the following device parameters:
nMOS
VT0,n = 0.8 V
n Cox = 50 A/V2
pMOS
VT0,p = - 1.0 V
p Cox = 20 A/V2
The power supply voltage is VDD=5V. Both transistors have a channel length of
Ln=Lp=1m. The total output load capacitance of this circuit is Cout=2pF, which is
independent of transistor dimensions.
a. Determine the channel width of the nMOS and the pMOS transistors such that the
switching threshold voltage is equal to 2.2 V and the output rise time is rise = 5 ns.
b. Calculate the average propagation delay time p for the circuit designed in (a).
c. How do the switching threshold Vth and the delay times change if the power supply
voltage is dropped from 5 V to 3.3 V. Provide an interpretation of the results.
6) Consider a CMOS inverter with the same process parameters as in Problem 3. The
switching threshold is designed to be equal to 2.4 V. A simplified expression of the
total output load capacitance is given as:
Cout = 500 fF + Cdb,n + Cdb,p
Furthermore, we know that the drain-to-substrate parasitic capacitances of the NMOS and
the PMOS transistors are functions of the channel width. A set of simplified capacitance
expressions are given below.
Cdb,n = 100 fF + 9 Wn
Cdb,P = 80 fF + 7 Wp
Where Wn and Wp are expressed in m.
a. Determine the channel width of both transistors such that the propagation delay
pHL is smaller than 0.825 ns.
b. Assume now that the CMOS inverter has been designed with (W/L)n = 6 and
(W/L)p = 15, and that the total output load capacitance is 250 fF. Calculate the
output rise time and fall time using the average current method.
7) Consider a CMOS inverter with the following parameters:
nMOS VT0,n = 1.0 V n Cox = 45 A/V2 (W/L)n = 10
pMOS VT0,p = -1.2 V p Cox = 25 A/V2 (W/L)p = 20
The power supply voltage is 5 V, and the output load capacitance is 1.5 pF.
a. Calculate the rise time and the fall time of the output signal using
i.
exact method (differential equations)
ii.
average current method
b. Determine the maximum frequency of a periodic square-wave input signal so that
the output voltage can still exhibit a full logic swing from 0 V to 5 V in each cycle.
c. Calculate the dynamic power dissipation at this frequency.
d. Assume that the output load capacitance is mainly dominated by fixed fanout
components (which are independent of Wn and Wp). We want to re-design the
inverter so that the propagation delay times are reduced by 25%.
Determine the required channel dimensions of the NMOS and the PMOS
transistors. How does this re-design influence the switching (inversion) threshold?

Sheet 2

Uploaded by

Copyright:

Available Formats

Sheet 2

Uploaded by

Document Information

Original Description:

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

Sheet 2

Uploaded by

Copyright:

Available Formats

Alexandria University

1) Consider switching delays for 1 pF in a 10-k resistive-load inverter circuit, where

a. Find PHL (50% high-to-low transition delay) by using the average-current

Figure 1: ring oscillator

3) Suppose that the resistive-load inverter examined in Exercise 1 is connected to a load

A ring oscillator is formed by connecting nine identical inverters in a closed loop. We

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.