IEEE Power Electronics Magazine - June 2019 - 97

White Hot

(continued from page 100)

include all electromagnetic interference (EMI) filtering, monitoring and
protection circuits, fans, connectors,
and enclosures, while meeting worldwide safety and EMI standards. With
per for ma nce at this level, there
is little that the universities can offer
for improvement.
The answer to why power electronics research in renewable energy and
the electrification of transportation is
so common is simple-this is where
the money is. It seems that most university research these days is funded
by government grants. Most governments are interested in reducing the
use of and dependence on fossil fuels
to fight climate change, to improve
economic security, or for political
motives. This leads to funding in
research in renewable energy and the
electrification of transportation. So
why does the industry not invest more
in university research? Before I answer
that question, I do want to say that my
viewpoint is somewhat U.S. centric. If
you have a different perspective, I
invite you to write to me at the email
address at the end of this article.
I have been involved in industrysponsored university research several times in my career. In the mid1980s, I worked in the Corporate
Power Technology Group at the
Di gital Equipment Corporation. In
that group, I was part of a team overseeing high-frequency conver ter
research at the Massachusetts Institute of Technology, quasi-resonant
converter research at the Virginia
Polytechnic Institute and State University, and magnetics research at
Duke University. Later, while working
in the Technology Group at Artesyn
Technologies, I was a liaison to the
University of Colorado, Boulder,
where we sponsored research in dc-
dc converters. This experience has
given me a good look at how industry-sponsored research works-and
how it doesn't work.
One of the reasons that industry
and academia do not collaborate is a

difference in expectations. Too many
times when a company sponsors
research at a university, it expects
to receive a ready-to-manufacture
product. However, what academia
expects as the outcome of a research
project is publications. This mismatch leads to frustration on both
sides. Another barrier to industrysponsored university research is the
ownership of intellectual property
(patents). The sponsoring company
genera l ly bel ieves t hat si nce it
paid for the research, it should own
a ny patents that come from that
research-just as it owns any patents created by its employees. Universities tend to think that since the
idea originated with their faculty
and students, they should own the
patents. These barriers are unfortunate because both the industry
and academia would benefit from
more industry-sponsored university research.
When it comes to expectations,
the industry needs to be realistic. It
c a n not con sider s pon sor i ng a
research project at a university as an
inexpensive way to do product development. Instead, companies should
sponsor research as an inexpensive
way to get basic ideas evaluated.
First, it is simply less expensive to
pay a graduate student for a year than
a senior engineer on staff, particularly considering the overhead costs of a
regular employee (vacation, health
insurance, office space, lab equipment, etc.). There is also the lost
opportunity cost of having an experienced engineer exploring something
that might not work versus having
that engineer work on a product that
will eventually produce revenue for
the company. Of course, it is not that
simple. For a company, having a
senior engineer work on research
instead of development can be good
for the engineer's career. Such projects can increase employee satisfaction and provide long-term gains to
the company.

June 2019

What companies need to expect
from university research is, at best,
a proof-of-concept-not a product.
Companies need to expect that they
will get the verification of ideas that
can be introduced to the product
development lab with the good confidence of a successful project-
and not a design ready for the manufacturing f loor. Companies also
must be prepared to learn what
does not work. It may be that ideas
evaluated in the university research
lab may not be worth developing into
products or even further research.
Companies need to understand this
as a possible outcome of sponsored
research and recognize that this is
still money well spent. After all,
lear ning what does not work is
valuable, and it costs a lot less to
learn this lesson in a university
resea rch lab tha n the compa ny
development lab.
But here is a real problem. All of
that is not easily quantified by the
finance. The finance department
looks at the money spent with no tangible, quantifiable return that can be
put on the company balance sheet as
a loss-money wasted. This attitude
will need to be tempered or overcome
if companies are to sponsor more
research at universities.
Universities also have a change to
make. They must be ready to give up
ownership of patents if a company
pays for the research. University
intellectual property departments
often think that all patents are
extremely valuable and that they can
reap a fortune in royalties. That
might be true if you are working on
cancer drugs, but it is not true if you
are working in power electronics. In
my 40 years in this field, I can count
the number of patents that made
their owners a lot of money on the
fingers of one hand. And for some of
those patents, the earned money came
in the form of jury awards after winning a lawsuit rather than in profits on
the sale of products.
z IEEE POWER ELECTRONICS MAGAZINE

IEEE Power Electronics Magazine - June 2019

Table of Contents for the Digital Edition of IEEE Power Electronics Magazine - June 2019