Optical Gyros

A brief personal history of the fiber-optic gyro on its 40th anniversary
01 October 2016
Eric Udd

In 1976, Victor Vali and Richard Shorthill demonstrated an operational fiber-optic gyro for the first time. In that same year, McDonnell Douglas Astronautics Co. in Huntington Beach, CA, (MDAC-HB) completed a project to redesign a new, lower-cost inertial measurement unit (IMU) for the Delta rocket based on dry-tuned mechanical gyros.

NASA agreed to split the profit with McDonnell Douglas on future Delta launches, in exchange for the company funding development costs for the IMU. But NASA had an option in the contract to cancel profit-sharing with McDonnell Douglas if another company created a cheaper, high-performance gyro for rocket guidance.

Knowing that this was a possibility, the guidance and control group at MDAC-HB funded a small project with the Electro-Optics Laboratory at MDAC to investigate optical gyros.

That small project led to a fiber-optic innovation explosion over the next 10 years, beginning with the development at MDAC-HB of the closed-loop fiber-optic gyro, originally called the dispersive fiber gyro and/or the phase-nulling optical gyro.

The closed-loop fiber-optic gyro, in turn, was a key development that enabled the commercial realization of high-performance fiber-optic gyros, secure fiber-optic communications, and a series of Sagnac acoustic, strain, and distributed sensors for the new field of fiber-optic smart structures. The Sagnac interferometer continues to be a useful tool for a variety of sensing and communication applications.

Fiber-optic gyros enabled this self-portrait of NASA’s Curiosity Mars rover at the “Big Sky” site. The scene combines dozens of images taken 6 October 2015 by the Mars Hand Lens Imager (MAHLI) camera at the end of the rover’s robotic arm.
Photo courtesy NASA

When I started work at MDAC-HB on 6 September 1977, Richard Cahill managed the Electro-Optics Lab. My first assignment involved devising an optical inspection tool for cryogenic foam used in the Delta rocket and in liquid natural gas tanks. After designing and demonstrating a “breadboard” unit, a technician built additional units and Cahill assigned me the optical gyro investigation.

In late September 1977, while discussing the relative merits of ring laser versus fiber-optic gyros, Cahill remarked that, “the fiber gyro could be an all solid-state device. Too bad it has a sinusoidal output and is nonlinear.”

Shortly thereafter, I came up with the “dispersive fiber gyro.” The idea involved placing a frequency shifter in the Sagnac loop, so that the wavelengths of the counter-propagating light beams would be identical after a complete circuit. But the frequencies of the counter-propagating light beams would be different in most of the fiber loop. This effect generated an optical path difference controlled by input frequency.

Cahill and I soon realized that the dispersion effect was much smaller than the net phase shift induced by the frequency difference. On 29 September 1977, the first complete written and witnessed description of the closed-loop fiber gyro occurred.

Realizing the importance of the invention, we filed formal disclosure statements with the McDonnell Douglas corporate patent department. We also began extending and refining the initial ideas and working on a hardware demonstration since McDonnell Douglas would not file a patent without the demonstration.

We ordered a Tropel HeNe laser with a long coherence length; an acousto-optic modulator designed to operate at 50 MHz; and 100 m of Valtec optical fiber with a 2-micron core designed to be single mode at 633 nm.

The Delta Rocket program at MDAC-HB sponsored the first closed-loop fiber-optic gyro demonstration on 28 July 1978.

On 28 July 1978, the first signals were obtained from the demonstration unit (above).

The Delta rocket program monitored progress closely and demanded a 10-cm-diameter unit for the next phase of the project. Our lab countered with 15 cm. We compromised on a 12.5-cm unit, and work began in mid-1978.

The patent application was filed 7 December 1978 and was followed by publication of the first experimental results.


The 12.5-cm fiber-optic gyro represented the first solid-state fiber-optic gyro. It utilized one of the first Hitachi single-mode laser diodes and some of the first low-loss single-mode optical fiber produced by Corning. The original design included two balanced acousto-optic (AO) modulators, but one failed and a single AO modulator configuration, (below) was built and tested extensively.

The early phase-nulling optical gyro was 12.5 cm in diameter, built and demonstrated in 1979.

Alignment of the 12.5-cm unit took about four weeks to complete. Bulk optic lenses were used with all housings epoxied in place. The ends of the four-micron core optical fiber coil were polished at an angle and we enabled final alignment with a set of three rotatable optical wedges placed in front of each fiber end.

A redesigned, 6.3 cm-diameter optical gyro was built and demonstrated in 1980.

Important work conducted by Reinhard Ulrich on the need to use a second beamsplitter and polarizer led to design improvements, which were included in the 6.3-cm unit of 1980 (above). This 6.3-cm unit responded to requests by MDAC sponsors for smaller demonstration models and improved alignment stations.

In an effort to make even smaller packages, a 2.5-cm-wide, 7.5-cm-long open-loop fiber-optic gyro was constructed shortly after the completion of the 6.3-cm circular design. This inspired the interest of an oil and gas service company that sponsored work on a tool to be used for navigation during the drilling process.

MDAC used its expertise in molding electronics to support high vibrations and acoustic levels during launch to build orthogonal, oval, open-loop fiber gyros capable of measuring less than 1 degree per hour stably.

Other improvements in packaging the closed-loop fiber gyros included 9.5-cm-diameter units built for Eglin Air Force Base and 11-cm long x 4.5-cm high x 5.5-cm wide units (below) that could be stacked into a cube with accelerometers.

Closed-loop fiber gyros built in 1986.

In 1987, MDAC began licensing its closed-loop fiber gyro patents worldwide, first with US-based companies and later in Europe and Asia.

I was not happy with the decision. I felt that MDAC could establish itself in the inertial navigation field and I had a plan to move toward production prototypes. MDAC management, in an effort to make the decision less painful, assured me that all internal funding for fiber gyros would now be applied to “fiber-optic smart structures” and “secure fiber-optic communication,” two other areas I was exploring. This allowed the fiber sensor group at MDAC-HB to remain intact.

The technology supporting early efforts in these fields were fiber sensors derived from the Sagnac interferometer. The first sensor of this type was the Sagnac acoustic sensor. It has a number of unique features that include low sensitivity in the center of the coil and higher sensitivity on fibers located near the central beamsplitter.

Its response increases with frequency and it can be used to create filters that are optimized for specific frequency ranges. This sensor has been used to support commercial site-security systems.

Another derivative sensor is the Sagnac strain sensor. The early closed-loop fiber gyro that used acousto-optic modulators relies on a fixed-frequency offset that can be tuned to offset rotation-induced phase shifts. This fixed-frequency offset results in a net phase shift between the counter-propagating light beams that is proportional to the length of the optical fiber. This causes the entire Sagnac loop to be a strain sensor.

The Sagnac strain sensor supported demonstrations of strain measurement in composite materials around 1985 and was considered for long-gauge-length strain sensors to monitor earthquake fault lines before GPS technology deployed.

A Sagnac distributed sensor to locate and measure time varying events formed by interleaving two Sagnac loops operating independently that may be at 1300 and 1550 nm.

By interlacing multiple Sagnac interferometers, distributed sensors intended to localize and measure time-varying events can be realized. The schematic above shows a configuration that was constructed using broadband light sources at 1300 and 1550 nm with standard 1300/1550 nm biconical taper wavelength division multiplexing (WDM). The sensitivity of each loop is zero in the center and increases as it moves away from the center position. By taking the ratio of a signature, a location can be identified and by taking the sum, the amplitude may be measured.

One principal advantage of the Sagnac acoustic and distributed sensors is that they can be supported by very low-cost single-mode optical fiber. This opens up a number of interesting applications, including identifying leaks in pressurized pipes and containers, identifying the location of insects in grain storage facilities, and locating termites in wood.

This article is a condensed and edited version of a paper presented at SPIE Defense + Commercial Sensing in April 2016 at a session on the 40th anniversary of the fiber-optic gyro. The full proceedings paper, "The early history of the closed loop fiber optic gyro and derivative sensors at McDonnell Douglas, Blue Road Research and Columbia Gorge Research," can be accessed in the SPIE Digital Library: dx.doi.org/10.1117/12.2228277

SPIE Fellow Eric Udd

SPIE Fellow Eric Udd is founder and president of Columbia Gorge Research and has worked in the fiber-sensor field continuously since 1977. He has authored and/or presented more than 200 papers on fiber-optic sensor technology, edited two books on the topic, and chaired more than 30 technical conferences. The recipient of 52 US patents, he was the founder of Blue Road Research and worked at McDonnell Douglas for 16 years as an engineer, manager, and Fellow. He has an MSE from Princeton University and a BS from University of Washington. Udd wrote about starting a high-tech business in the January 2006 issue of SPIE Professional

Recent News
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research