High-speed tunable diode laser absorption spectroscopy for sampling-free in-cylinder water vapor concentration measurements in an optical IC engine

Abstract

A novel, fiber-optic in situ laser hygrometer was developed to measure water vapor with microsecond time resolution directly inside an internal combustion (IC) engine. The instrument is intended for sampling-free quantification of recirculated exhaust gas in combustion engines. Direct tunable diode laser absorption spectroscopy was employed to allow absolute and self-calibrating H2O measurements. The compact and user-friendly instrument combines a fiber-coupled, 1.37 μm distributed feedback diode laser with kHz-fast, continuous wavelength scanning. Only small, typically 10 mm, optical access ports in the engine are needed. The new in situ hygrometer was tested via measurements in a motored optical research engine operated on ambient air, without any artificial humidification. Scanning the laser at 4 kHz resulted in a time resolution of 250 μs (i.e., 3° crank angle at 2,000 rpm), while the DC-coupled detector signals are digitized with a 4MSamples/s 16-bit data acquisition system. Absolute water vapor concentrations around 1 vol.% could be measured and quantified during the full compression stroke, i.e., over a pressure/temperature range of 0.07–0.52 MPa/300–500 K. Without any scan averaging or bandwidth filtering we could demonstrate signal-to-noise ratios between 51 (at p = 0.1 MPa) and 33 (at p = 0.4 MPa), which corresponds to H2O detection limits between 0.02 and 0.035 vol.% or length and bandwidth normalized detectivities of 285 and 477 ppb m Hz−½, respectively. Comparison of the dynamic H2O behavior during the compression stroke across several engine cycles and different operating conditions showed good reproducibility and absolute accuracy of the results, consistent with the boundary conditions, i.e., motored air operation. This new sensor therefore opens up new possibilities for engine cycle-resolved, calibration-free in situ AGR quantification and optimization in engine applications.