Calibration of optical fiber power meters: the effect of connectors

Connector-induced offsets in optical fiber powermeters

We discuss issues related to calibration of optical fiber powermeters. The power readings may vary when using various types of fiber connectors, or even the same type of connectors made by different vendors. We study the connector effects by calibrating four types of optical fiber powermeters using six different fiber connectors from four vendors at three wavelengths. We found variations of as much as 12% due to the different reflection properties of the detector windows, connectors, and detectors. We also compared connector measurement results with the data obtained by using open beams.

Realization of the scale of high fiber optic power at three national standards laboratories

Nowadays the transmission powers in optical telecommunication networks are often hundreds of milliwatts. Such high power levels are known to cause several nonlinear effects, thus affecting data transfer. Therefore, accurate measurements of such high power levels are required. The general issues that are to be considered when one is realizing a scale for high fiber optic power are discussed. The scales of the national standards laboratories in Finland, Sweden, and Denmark are described, and the results of a trilateral comparison of these scales are presented. The power range of the comparison was 1-200 mW. The results show that the stated measurement uncertainties of the three laboratories (1.3%-2.9%, k = 2) are applicable over this power range.

Optical trap detector for calibration of optical fiber powermeters: coupling efficiency

The optical trap detector is based on two, 1 cm x 1 cm silicon photodiodes and a spherical mirror contained in a package that is highly efficient for measuring light diverging from the end of an optical fiber. The mathematical derivation of the coupling efficiency relies on the integral directional response weighted by the angular intensity distribution of an idealized parabolic optical beam. Results of directional-uniformity measurements, acquired with the aid of a six-axis industrial robotic arm, indicate that the trap has a collection efficiency greater than 99.9% for a fiber numerical aperture of 0.24. Spatial uniformity measurements indicate that the variation of detector response as a function of position is less than 0.1%. The detector's absolute responsivity at 672.3, 851.7, and 986.1 nm is also documented by comparison with other optical detectors and various input conditions and indicates that the design is well suited for laser and optical fiber power measurements.

International intercomparison of detector responsivity at 1300 and 1550 nm

An intercomparison of spectral responsivity measurements has been carried out among the national laboratories of 13 countries. Measurements were made at wavelengths of 1300 and 1550 nm, which are important in optical communication systems. Three germanium photodiodes were circulated in each of three separate paths, beginning and ending at the National Institute of Standards and Technology as the coordinating laboratory. The results show agreement within 1%.