Distributed fiber optic temperature sensing based on Rayleigh scattering is a relatively new technique offering data density unachievable with point sensors such as thermocouples and RTDs. Thousands of temperature measurements can be generated by a single fiber optic cable suspended within a flow field. And unlike imaging techniques such as laser induced fluorescence, fiber optic sensors are suitable for applications involving opaque fluids. But verifying measurement accuracy along a distributed temperature sensor (DTS) can be problematic. Unlike traditional sensors such as thermocouples, DTS calibration shifts can accompany sensor handling or movement because they respond to strain as well as temperature. This paper describes an assessment of a Rayleigh scattering-based sensing system used to measure air temperature within a 1 × 1 × 1.7 m tank used for thermal mixing experiments. Two 40 m-long DTSs were strung across the tank midplane at 16 levels. Stability in stagnant air was examined over seven days and found to be generally better than ± 0.5°C with local regions of drift up to 1.5°C. DTSs were also tested in isothermal flow to assess signal degradation associated with flow-induced vibration. Noise increased with flow velocity, inducing data loss that grew with distance along the fiber. Despite data losses >50% in high noise regions, mean temperatures after simple filtering agreed with low noise regions to within ∼4°C.

This content is only available via PDF.
You do not currently have access to this content.