IEEE Solid-State Circuits Magazine - Fall 2014 - 45

The key differentiating feature of his
work in this area is that it preserves
the rigor of the underlying device
physics while at the same time focusing on relevance for circuit design.
Here I will highlight the extent of his
influence in this area, with emphasis
on his book Operation and Modeling
of the MOS Transistor [1], 2], see Figs.
1 and 2, which for brevity I shall refer
to as OMMT.

One of the reasons I liked the
book so much when I bought my
first copy was the honesty in discussion of imprecise quantities.

III Influence on Students

II Style
In two aspects of style OMMT is head
and shoulders above other text books
on MOS transistors: clarity of presentation, and clarity of concepts.
One of the reasons I became a disciple of OMMT was that it is so easy to
read: All concepts are explained clearly
and succinctly, and re-emphasized a
few times at different places to make
sure they sink in. The way material
is presented overall, from 2- to 3- to
4-terminal MOS structures and from 5to 9-capacitance element small-signal
models, is designed to lead students
along a logical path to full understanding of MOS transistors while avoiding
scaring them off by presenting the full
complexity as a fait accompli in step
1. Moreover, it is not just the organization and structure that are easy to
follow, the English prose is clear and
flows beautifully, so learning and absorbing the material is almost effortless. It is not sufficient for an author
to be an expert in the subject material
being presented, he or she must be
able to communicate effectively, and
I always found it extremely easy to
read OMMT.
In almost all expositions on MOS
transistors you will find terms such
as "threshold voltage" and "subthreshold" used as if they are definitive. Indeed, I used to use such
language myself. Fig. 3 shows MOS
transistor I D ^VGSh characteristics at
VDS = 1V, on linear and log-linear
scales. You can see that there is a
region (to the left, marked as weak
inversion) where I D varies approximately exponentially with VGS and a
region (to the right, marked as strong
inversion) where I D varies roughly

Figure 1: My copy of the first edition,
with original stickies that marked the sections I referred to most often showing.

linearly with VGS . Can you pick out
a single, exact point (the threshold
voltage VT) that defines the boundary between near exponential and
near linear behavior?2 Most descriptions of MOS transistor behavior
explicitly assume that such a point
exists, and the vast majority of MOS
transistor models are constructed
based on the concept of this fictitious point. That unfortunately alltoo-common "mental model" is an
impediment to proper understanding
of how MOS transistors behave.
One of the most important aspects
of OMMT is that is does not, with
little explanation or justification,
assume that VT exists, and continue from there. It clearly explains
the "muddiness" that surrounds VT,
and then rather than considering
MOS transistor operation as being
split into below and above VT operation delineates weak, moderate, and
strong inversion regions, and defines what these mean. More important, it is completely above-board in
describing the imprecisions in the
boundaries between these regions,
especially between moderate and
strong inversion, and presenting
practical ways to determine them
from experimental data.
2) If you can, patent how you do so, or at
least write a paper on it!

The potential "customer base" for
OMMT is much greater on the circuit
design side than on the modeling
engineer side, for the simple reason
that only one suite of simulation
models (provided by a few modeling engineers) is required per technology but they are reused for the
development (by many design engineers) of all of the circuit blocks
implemented for that technology.
Therefore the main focus of OMMT
is to educate circuit designers, not
modeling engineers. In this role
OMMT has become the most widely
used text to train university students about MOS transistors.
In large part I believe this is because,
as noted above, of the clear and simple
prose style, which makes it easy to
understand, and the logical structure
and flow of the material, which makes
it easy to follow and enables students
to step-by-step assimilate the details
of how MOS transistors operate.
But it is also because the material is presented by an expert circuit
designer, rather than a device engineer, who understands what aspects
are best emphasized to convey what
is most important for circuit design.

Figure 2: The second edition.

IEEE SOLID-STATE CIRCUITS MAGAZINE

fa l l 2 0 14

Table of Contents for the Digital Edition of IEEE Solid-State Circuits Magazine - Fall 2014