Corrections or additions?

This article by Barbara Fox was prepared for the August

7, 2001 edition of U.S. 1 Newspaper. All rights reserved.

Fine Tuning Fiber Optics

It was a sad day for Chuni Ghosh when Tachonics, an

integrated circuit firm, closed its doors at Morgan Lane. His company

had 80 percent market share of an industry that was expected to grow

and prosper, yet it was controlled by a much larger firm, and the

owners had decided that Tachonics "just didn’t fit."

After Tachonics closed in March, 1990, Ghosh (pronounced with a long

"O" worked at a salaried job for three years. Then he started

another company, Princeton Electronic Systems, which developed and

manufactured products for telecommunications firms (

This time around, he did not depend on outside investors. Instead

he developed two areas simultaneously, one familiar technology that

could quickly bring in cash plus a couple of leading edge technologies

that required both time and capital.

Now Ghosh’s firm is ready to hit the big time with the longer-term

technology. To signal this, he has changed its name from Princeton

Electronic Systems to Princeton Optronics. He has hired a public


firm, commissioned a new logo, and moved from Princeton-Hightstown

Road to a high profile location, the former Base 10 building at 1

Electronics Drive. With 67 employees in 48,000 square feet now, he

hopes to expand the workforce to 90 and also build a clean room. Ghosh

is taking the big step from research and development into


he is shipping products in sample quantities now, and is prepared

to ship in volume by September. A second round of funding is expected

to come through later this month.

An avid soccer player — he had considered turning professional

— Ghosh says that his soccer experiences helped him learn to


temporary defeat. "We won many times, and we lost many times,

and when we lost we cried, but we picked up and got ready for the

next match. That has been a tremendous experience, to learn to pick

up after a tremendous loss and get ready for the next game."

Princeton Optronics works in fiber optics to solve the bandwidth


As Internet users demand more and more bandwidth, fiber optic


providers are under pressure to come up with tools to increase


Ghosh says he has leveraged success in familiar radio frequency (RF)

and microwave technologies to gain purity and efficiency for optics

uses. Now he is using leading edge technologies to solve the power


What is the power problem? Fiber carries data and voice

messages as light signals, in contrast to copper wire, which sends

messages with an electric current. The messages are converted at the

sender’s end from an electrical impulse into packets of 1s (where

the light is bright) and 0s (where the light is dim), and then they

are reconverted at the receiver’s end of the line.

This conversion process is called optoelectronics, and it requires

lasers. Other successful Princeton area companies also work with


Greg Olsen’s Sensors Unlimited, JDS Uniphase Epitaxx division (founded

by Olsen), Vladimir Ban’s PD-LD, and Princeton Lightwave (a Sarnoff

spinoff), (U.S. 1, May 9). But Ghosh has the distinction of early

entry into one of the hottest new areas: VCSEL (Vertical Cavity


Emitting) technology for tunable lasers, a low-cost high-power


for telecom carriers. Tunable lasers sell in the range of $3,000 to


"We have a major technology for getting high power out of lasers,

and we are the forerunners for this kind of technology," says

Ghosh. "We think we have a major advantage compared to other ways

of producing this high power for current and next generation


systems. Our products are being evaluated by a number of


Here is the problem he is helping to solve: Telecom carriers are


with onerous networking costs. Until recently they relied on the


force" approach using legacy technology: fixed wavelength lasers.

With these, the carriers pay for a channel whether it has traffic

or not — one conversation requires one channel. To get more


they buy more channels, yet sometimes the channels are empty.

The newer tunable lasers can be "tuned" over various wave

lengths for flexibility in inventory and placement. They can send

more than one conversation on the same channel because each


travels on a different wave length — a different color of light

— which eliminates gaps. It’s more elegant and scalable solution.

But until now, tunable lasers have not had enough power to be


says Ghosh.

Ghosh solves part of the power problem by incorporating MEMs (Micro

Electro-Mechanical Systems) technology in a tunable laser. Lasers

with MEMs are electrically, rather than thermally, tuned and this

improves power efficiencies. One of his tunable lasers, for instance,

operates at 10 gigabits and 40 gigabits and targets the long haul

market (hundreds of miles) as well as the ultra long-haul market


of miles), and metro markets (tens of miles). "This combination

provides the widest tuning range possible with unprecedented power

output, at a fraction of the cost," says Ghosh.

Ghosh is an early adopter of MEMS technology, and he is also a pioneer

in another technology used in tunable lasers, Vertical Cavity Surface

Emitting Lasers (VCSEL, pronounced vi-xel). They may eventually


the legacy technology, described as "edge emitting lasers"

or "edge emitters" Early in the history of Ghosh’s company,

the department of defense asked for power levels substantially higher

than what could be attained with edge emitters, so he decided to


with VCSELs.

The difference between edge emitters and VCSELs: Think of trying to

fit an elliptical beam of light into a circular fiber. Some of the

light generated would be wasted. In contrast to the edge emitter’s

elliptical beam, the VCSEL has the cylindrical beam, which fits into

a circular fiber much more efficiently.

Ghosh says that because his firm entered the VCSEL field early, he

can cost effectively expand the power output to 20 miliwatts and up.

This compares with the 4-10 miliwatts that is being offered by later

entrants to the VCSEL field.

All the lasers discussed so far, the source lasers, send the initial

signals directly from the handset. Both the traditional fixed lasers

and the newer tunable lasers convert the electrical signal to an


signal. Fixed wave length lasers cost about $1,000, in contrast to

$5,000 for tunable lasers.

Ghosh is also working with pump lasers that are situated along the

path of the signal; they amplify the signal so that it can go for

longer distances. Pump lasers can be compared to gas pumps along the

highway. When a signal runs out of energy, it can get amplified or

re-energized by the pump laser.

Ghosh can use his VCSELs in both the tunable lasers and the pump


He compares the VCSEL to an engine that can be put into a Cadillac

or a Jeep. The tunable laser with the fancy MEMS technology goes into

the Cadillac. The pump laser has more raw power and like the Jeep

it can go more places. "The VCSEL laser, when we put in a package

with MEMS, becomes a tunable laser. By itself, the VCSEL laser is

a pump laser," says Ghosh.

The traditional pump laser, with an Erbium Doped Fiber Amplifier


can use either the old technology or the VCSEL method. The newer Raman

pump lasers amplify the signal continuously — like switching on

the power in an electric train set. They are more efficient but also

power hungry and too expensive to use in quantities at the present

time. In the meantime, says Ghosh, "we can offer more economical

high powered pump lasers using VCSELs than incumbent technology."

Chuni L. Ghosh says his business inspiration comes from

his grandmother, an intelligent and highly-organized matriarch who

was married to a retail store owner near Calcutta. "She was a

great leader; she could make people follow what she wanted to do.

`Our fortunes are up and down’ she said. `Get the best education you

can. That, nobody can take away from you.’ She impressed this on all

of her children and grandchildren."

His father was one of the few Indian natives who volunteered to fight

in World War II. He served in Burma under an American commander,


Stillwell, and spent the rest of his career in the retail hardware


Ghosh was born in 1948, the year after India won its independence.

He is the oldest of three boys and the only family member to move

to the United States. He and his Indian wife — who is also an

engineer — have two grown daughters, both chemical engineers.

Ghosh went to BN College in Bengal and has a PhD in physics from


University. At the Bhabha Institute in India he worked on the first

night vision image intensifier device and won the "Young Scientist

of the Year" award from India’s prime minister. In 1980 he founded

a III-V (tables) chip and subsystems company, ITT GaAsTeK, that


Gallium Arsenide (GaAs) technologies yielding high speed electronic

systems, including low-cost radar systems.

Then in 1985 he founded Tachonics, a Grumman-funded company that


gallium arsenide-based high-speed integrated circuits. He managed

to acquire an impressive 80 percent market share in GaAs MMIC


and Grumman planned to move Tachonics to Long Island. But when Grumman

was sold to Northrop, everything was shut down.

It was a big and disappointing loss. "We were the best in an


that has become a multibillion business today. We were the leading

company with an 80 percent market share. Somebody at the very high

level decided Grumman did not want to be in that business. It was

not my decision. It was a Wall Street kind of decision," says

Ghosh. Northrop now owns the patents. "That influenced me not

to seek financing in the short term from anybody else," says


After just three years on salary at Sarnoff, as director of the


and parts of the RF laboratory, he was back out on his own, founding

Princeton Electronic Systems. He structured this company so it would

be self sufficient for the first half-dozen years. "But now we

are taking venture capital money and have just closed a second VC

round." The initial investors for a first $5 million round of

financing included Intel Capital and Novak-Biddle Venture Partners,

and second round funding, to be announced later this month, will be

used to bring the tunable lasers to market.

From the beginning Ghosh leveraged knowledge gained in academic labs

at Rutgers, NJIT, the University of Michigan, and Princeton,


at Princeton’s Photonics and OptoElectronic Materials Center (POEM).

Joe Montemarano is the director of industrial liaison at the POEM

center — and is also the featured speaker at U.S. 1’s Technology

Forum on Thursday, August 30, at 4 p.m., at the Doral Forrestal.


has been a long time member of POEM and has had a number of scientists

appointed to work in our clean room facilities," says Montemarano,

"and soon we expect to sign a collaborative research


The name change from Princeton Electronic Systems to

Princeton Optronics focuses attention on the optoelectronics side

of the business. "Right from the beginning we were playing on

both sides, optoelectronics and RF," says Ghosh. "It took

much longer to develop the optoelectronics technology, and we would

have needed outside investment, so RF provided the cash flow."

The synergy between the two technologies helps differentiate his firm

from competitors, he believes. "Most of our optical competitors

do not have the RF technology. All of telecommunications is going

to 10 and 40 gigabits, and once you try to make things at that high

speed, somebody who understands RF can do that very well."

He has changed the name, hired a PR firm, fashioned a new logo, and

moved to a high profile location — all in an attempt to raise

the profile of the company to the level of his competition. Jerry

Fennelly of NAI Fennelly found the space for Ghosh. Ghosh has also

hired a human resources specialist, which saves him, he says, two

hours a night.

"Now all the products are coming to fruition, and we are going

to the customers now. Many of our customers our in the Fortune 500,

and we are competing against major public companies with a high


We need to play in a different league and think big to win big,"

he says.

Ghosh is not worried about the turmoil that telecom companies are

going through, nor that analysts are gloomy about the growth of the

optical long-haul market — his market. After all, tunable lasers

are a technology for the future. "We feel that the tunable laser

is a cost saver and revenue enhancer," he told a trade magazine.

"If you use a tunable laser, you don’t have to stock up on many

fixed-wavelength laser parts. By using the tunable laser boards, you

can provision the wavelength very easily. That results in cost


He is, nevertheless, concerned about promising too much too soon.

"We need to set up manufacturing in the right way, not only in

terms of the performance of the product, but in terms of ramp ups.

If you take the orders and cannot deliver, people don’t forget these


That he did not go to business school, he says now, is probably a

mistake. "I learned things the hard way. And I had a lot of good

managers in ITT and in Grumman, a very fine manager. We learned by

seeing them."

"I am not risk averse," says Ghosh. "I learned that when

you make a decision, you will regret either way. If you do not take

the risk, you will regret it. What do I have to lose? is my motto.

I am a professional and good at what I do. If worse comes to worse

I can always work for someone else. I am not hung up on a particular

lifestyle. I can live any lifestyle."

Princeton Optronics, 1 Electronics Drive,


08619, Box 8627, Princeton 08540. Chuni Ghosh, CEO. 609-584-9696;

fax, 609-584-2448. Home page:

Top Of Page

For stories on ITXC go to and and

Next Story

Corrections or additions?

This page is published by

— the web site for U.S. 1 Newspaper in Princeton, New Jersey.

Facebook Comments