Much attention has been paid to the development of transfer standards for fiber optic power measurements. The NIST of the United States in Boulder, Colorado and the standards organizations of most other countries have worked to provide good standards to work with. Now we can ensure traceability for our calibrations, but still, the errors that occur when making measurements cannot be ignored. Even when fiber optic power meters are calibrated within the specifications, the uncertainty of a measurement can be as much as +/- 5% (about 0.2 dB) compared to the standards.
The first source of error is the optical coupling. The fiber light expands in a cone. It is important that the fiber geometry detector is such that all the fiber light hits the detector, otherwise, the measurement will be less than the actual value. But every time the light passes through an air-glass interface, such as the window in the detector, a small amount of light is reflected and lost. Finally, cleaning the optical surfaces involved can cause absorption and dispersion. The total sum of these potential errors will depend on the type of connector, wavelength, fiber size and numerical aperture.
Beyond coupling errors, there are also errors associated with wavelength calibration. Semiconductor detectors used in fiber optic instruments (and also systems) have a sensitivity that is wavelength dependent. Since the wavelength of the actual source is poorly understood, there is an error associated with the spectral sensitivity of the detector. By industrial convention, the three essential wavelengths (850, 1300 and 1550 nm) are used for all power measurements, and not the wavelength of the exact source.
There is also another source of error for measurements of high and low levels. At high levels, the optical power can overload and saturate the detector, which will cause the measurement to be wrong. At low levels, the inherent noise of the detector adds to the signal and becomes an error. If the signal is 10 dB above the minimum noise threshold (10 times the noise), the offset error is 10% or 0.4 dB.
Instrument resolution vs. Measurement uncertainty
If the uncertainty of most fiber optic measurements is considered, instrument manufacturers have provided loss and power meters with a measurement resolution that is usually much larger than necessary. The uncertainty of optical power measurements is around 0.2 dB (5%), loss measurements are likely to present uncertainties of 0.2-0.5 dB or more, and optical return loss measurements have an uncertainty of 1 dB.
Instruments that have reading screens with a resolution of 0.01 dB are generally only suitable for laboratory measurements of very low component losses or changes caused by environmental variations. Within the laboratory, a resolution of 0.01 dB can be extremely useful, since the loss of connectors or joints that are below 0.10 dB or changes in a loss under environmental stress that are below 0.1 dB are usually measured. The stability of the sources and the physical tension in the cables limits the uncertainty of measurement to approximately 0.02 to 0.05 dB per day, but a resolution of 0.
Field measurements have a greater uncertainty because more components are measured at a time and losses are greater. Practically, the measurements are better when the resolution of the instrument is limited to 0.1 dB. The readings will be more likely to be stable when read, and more indicative of the uncertainty of the measurement.
No comments:
Post a Comment