Drawing of the optomechanical set-up of QUALITAS, designed to fit within the horizontal space restrictions of a rack. The optics itself uses only 3 HU. We abandoned the use of an optical breadboard, and use only two plates, orthogonally mounted to each other, to fix the optical components. From the angle of view shown, the cold station for the laser and the detectors is almost fully hidden by the the horizontal mounting plate, the mirror objective (R1,…,R3), and the Herriott-type multipass cell. After passing focus F1, which is used for adjustment purposes, mirror R4 guides the beam into the cell. The outcoming beam traverses via mirrors R5–R7 (R6 is hidden by the horizontal mounting plate) into the detector. For an explanation of the mirror arrangement, see the text. The back reflection of the cell window is used to install a reference path, guided via R5r and an IR-Fresnel lens onto the reference detector after passing the reference cuvette.
2D-sketch of the optical path, which approximates the geometrical set-up. The optical pathway outside the cell for the measurement signal is shown as a dashed line, whereas the path of the reference signal, which is produced by the reflection from cell input window, is shown as a dashed–dotted line. indicates the number of the reflection outside the multipass cell from the laser to the respective signal (Sig) or reference (Ref) detector. marks the foci before and after the cell.
Photograph of QUALITAS, including the optics, the electronics and the gas flow system (without the pump). In the upper row from left to right, the electronics consists of the ADC/DAC-unit, combined with the DSP/FPGA-board, the embedded PC, the power supply and some analog circuits. The lower row shows two lock-in amplifiers (for the measurement, and the reference signal), analog circuitry for the preparation of the laser modulation signal, the laser temperature and current controller, and a display unit. The upper two thirds of the rack, containing the electronics and the gas flow system can be easily removed, giving full access to the lower third of the housing, carrying the optics. Compared to Fig. 1, the optics is seen from the side of the reference cell, which can be recognized quite clearly on the photograph. The mirrors R5 and R5r are left of the cell, just behind the corner of the housing.
The calculated (a) absorption spectrum and a measured -signal (b) for the calibration gas during flight (see Sec. VI). Based on the data of HITRAN (Ref. 15), the optical density at the line center is for the calibration gas containing of CO. Besides the calibration signal, a signal from CO-free air (also in-flight measurements) and its variation within is shown. The frequency range from to approx. is used to calculate the concentration via a linear fit algorithm. The detection limit and the signal-to-noise ratio are estimated using the peak-to-peak calibration signal relative to the rms of the background variation. We find a normalized detection limit of , which is an excellent value compared to our former measurements (Ref. 16) and other values given in literature.
Allan plot of a CO time series of bottled gas , measured in the laboratory with the QCL. The inset shows the underlying -time series ( time bins with a measurement duty cycle of 96%). For an interpretation of the Allan plot, see the text.
Time Series ( averages) of carbon monoxide (solid symbols), measured with QUALITAS, and ozone (open symbols) of a flight in March 2004 during the airborne measurement campaign UTOPIHAN-ACT.
Characteristic figures for different types of measurements with QUALITAS. The column “Lab” indicates measurements of bottled gas at a concentration of within the laboratory and the QCL as the light source, see Fig. 5. “QCL” and “LC-TDL” indicates data from a flight in March 2004 with a QCL, and a flight in July 2003 with a LC-TDL. The characteristic figures are defined similar to Kormann et al. (Ref. 8). Shortly summarized, the detection limit is the rms of the measurements of CO-free gas, the reproducibility of calibration the rms of the calibrations, compared to its precessor. The noise level is the rms of ambient measurements at time periods of low concentration change, and the rms of the laboratory measurements of the bottled gas over a time interval. The calibration gas uncertainty is estimated from intercomparison measurements within the laboratory, tracing back the used standards during the flights to NOAA approved standards. The (normalized) detectable o.d. is estimated from the noise levels using the line parameters found in HITRAN (Ref. 15) and the operation conditions of QUALITAS.
Article metrics loading...
Full text loading...