Changes in strain are correlated with changes in speckle patterns.

Stennis Space Center, Mississippi

An improved fiber-optic strain gauge is capable of measuring strains in the approximate range of 0 to 50 microstrains with a resolution of 0.1 microstrain. (To some extent, the resolution of the strain gauge can be tailored and may be extensible to 0.01 microstrain.) The total cost of the hardware components of this strain gauge is less than $100 at 2006 prices. In comparison with prior strain gauges capable of measurement of such low strains, this strain gauge is more accurate, more economical, and more robust, and it operates at a higher update rate. Strain gauges like this one are useful mainly for measuring small strains (including those associated with vibrations) in such structures as rocket test stands, buildings, oilrigs, bridges, and dams. The technology was inspired by the need to measure very small strains on structures supporting liquid oxygen tanks, as a way to measure accurately mass of liquid oxygen during rocket engine testing.

This improved fiber-optic strain gauge was developed to overcome some of the deficiencies of both traditional foil strain gauges and prior fiber-optic strain gauges. Traditional foil strain gages do not have adequate signal-to-noise ratios at such small strains. Fiber-optic strain gauges have been shown to be potentially useful for measuring such small strains, but heretofore, the use of fiber-optic strain gauges has been inhibited, variously, by complexity, cost, or low update rate.

The improved fiber-optic strain gauge is partially composed of a multimode fiber optic which is wound in an elliptical pattern and bonded to the structure of interest. A laser is fixed within an adjustable cylindrical steel enclosure and aimed at one end of the optical fiber. The laser light emerging from the other end of the fiber forms a speckle pattern that changes as strain is applied to the structure. The speckle pattern is intercepted by an array of photocells, so that any change in the speckle pattern manifests itself in changes in the intensities of light measured by the individual photocells. The outputs of the photocells are collected by a customized data-acquisition system that includes a signal-conditioning subsystem. The photocell outputs are then fed to a neural network that recognizes the correlation between changes in the outputs and changes in strain.

Inasmuch as the changes in the intensities of light incident on the photocells are repeatable for a given amount of change in strain, the neural network can be quickly trained by use of speckle patterns associated with known levels of strain. For measurement of temporally varying strain (for example, when vibrations are present), the update rate and, hence, the dynamic analysis rate depends on the data-acquisition rate.

This work was done by Fernando Figueroa of Stennis Space Center and Ajay Mahajan, Mohammad Sayeh, and Bradley Regez of Southern Illinois University, Carbondale.

Inquiries concerning rights for the commercial use of this invention should be addressed to Intellectual Property Manager, NASA Stennis Space Center; (228) 688-1929. Refer to SSC-00243.

The U.S. Government does not endorse any commercial product, process, or activity identified on this web site.