IEEE Solid-State Circuits Magazine - Winter 2014 - 19

Meyer's style of work unites design, analysis,
simulation, and measurement (four section
headings that appear in almost all
his publications).

work, over their entire career. They
believed, rightly so at the time, that
if their role was to produce the next
generation of technical leaders, their
teaching and research should be
informed by industry practice, and
by connection to the marketplace. I
am of a later generation of practicing
academics who, although educated
in that tradition found it hard to follow as religiously, because of the
quickening pace of competition in
the now swollen ranks of academia.
Encouraged by the culture prevailing in the EECS Department at
Berkeley in the 1970s and 80s, Bob
consulted one day a week for semiconductor
companies.
Through
these consulting engagements he
designed many commercial products, with the understanding that he
would publish this work when justified. Coupled with his style of work
that unites design, analysis, simulation, and measurement (four section
headings that appear in almost all
his publications), this allowed an
easy transition of the cutting edge
design trends from company into
classroom, and naturally enough, a
transmission of this style of work
to his graduate student advisees,
many of whom are now leaders
themselves.
Here, as I see them, are Bob Meyer's abiding research themes.

Wideband Amplifier Circuits
Newly arrived on the Berkeley faculty
after completing his PhD at the University of Melbourne under the guidance of Daryl Hooper (1930-1985),
Bob Meyer must have been under the
sway of what remains, even today,
the definitive textbook on the design
of wideband amplifiers: Cherry and

Hooper's Amplifying Devices and
Low-Pass Amplifier Design, published in 1968. "Low-Pass" in its
title dates the book, by setting
apart its systematic design methodology from the much better
understood bandpass amplifiers
of the day used for radio communication and telephony. Today
we would understand "lowpass" to
mean "wideband".
Bob first associated himself as
a consultant with Hewlett Packard,
where he put to use his expertise
in designing wideband amplifiers
and analyzing their distortion for
the then new market of receivers
for analog television over cable.
While it was hard enough to realize
a low-cost amplifier with a bandwidth extending from essentially DC
to 300 MHz, it was especially hard
to meet the specifications on intermodulation and cross-modulation
distortions when the amplifier input
comprises many TV channels of
equal strength. For some perspective, 5% cross-modulation will badly
distort a TV picture, although it is
not noticeable in an analog voice
channel. We see for the first time in
[1] the multistage combined seriesshunt feedback stages that Meyer
will use for many years to come.
Here it is realized as a hybrid IC
prototype. Particularly interesting
in the appendix are simple expressions for 2nd and 3rd order intermodulation; distortion analysis in
those pre-SPICE days was not for
the faint-hearted, particularly when
at the upper limits of the passband
it might be caused by nonlinear
delays and capacitance. But Meyer
was one of the first to seize the work
at Bell Labs [2] that showed how to

use Volterra series to analyze
dynamic distortion, and in an early
publication [3] he had laid the foundations on which the cable TV amplifiers were to be designed to exacting
specifications.
While feedback lowers distortion
by the loop gain, at high frequencies
the loop gain fall offs and distortion
rises. In subsequent work [4] Meyer
shows the use of feedforward error
cancellation to improve distortion at high frequencies across the
300 MHz band.
All
communications
receivers need variable gain amplifiers
(VGAs). These circuits bring a set
of challenges not seen in amplifiers
with fixed gain, such as how to vary
gain over a large range while holding frequency response relatively
unchanged, and how to control the
amplifier's noise figure to maintain
a certain dynamic range over the full
range. Two papers by Sansen and
Meyer address these problems [5, 6],
building on basic analysis of distortion and noise which leads to new
circuit methods [7].
A 1977 publication [8] shows the
wideband amplifier for cable TV,

IEEE SOLID-STATE CIRCUITS MAGAZINE

w I n t E r 2 0 14

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