CCD array based spectrometers are widely used in radiometric measurements. Ambient temperature and nonlinearity effects are significant factors for high accuracy measurement in the field. Here, a temperature correction method for the CCD array spectrometers was developed, which calculated the spectrometer response at each pixel. The deviation between measured and calculated spectrometer responses at a randomly selected temperature is less than 1%. In addition, the radiant power nonlinearity effects were investigated by supplementary-light methods. The gain settings nonlinearity effect was evaluated using FEL-type transfer standard lamps. The nonlinearity correction coefficients were calculated and analyzed based on the experiment, respectively.
Deuterium lamp is used as the transfer standard of air-UV spectral irradiance (200nm to 400nm). The CCPRK1. b comparison of spectral irradiance 200nm to 350nm took deuterium lamp as transfer standard lamp. Spectral irradiance is measured by a spectroradiometer with finite bandwidth. The bandwidth can cause measurement error. In order to correct the measurement error, we apply SS and DO bandwidth correction methods to the spectrum of Deuterium lamp. We obtain the correction effect preliminarily.
In 2011, new primary standard apparatus of spectral radiance was setup at Changping campus of NIM based on high temperature blackbody BB3500M and double-grating monochromator of M207D. The temperature of the BB3500M was measured by a LP4 thermometer with uncertainty of 0.64 K at the temperature of 2980 K, which was calibrated by the Pt-C and Re-C fixed point blackbodies, and checked by a WC-C fixed point blackbody. The consistency of the temperature at 3021 K was better than 70 mK between NIM and VNIIOFI. The image of the measuring source was focused on the entrance slit of the monochromator with magnification 1:1. A mask was put in front of the entrance slit to limit the target spot size of the tungsten strip and the water-cooled aperture was 0.6 mm wide by 0.8 mm tall rectangle. The solid angle of spectral radiance measurement was approximately 0.008 sr. Uncertainty of spectral radiance scale realization was analyzed in this paper. The source of the uncertainty scale includes repeatability of the signal ratio of the blackbody and the transfer lamp, lamp alignment, temperature measurement of HTBB, non-uniformity of HTBB source, instability of HTBB source, correction of different size of source (BB and lamp), nonlinearity of the measurement system, current passed through the transfer lamp, wavelength error, polarization effects, bandwidth etc. The measurement uncertainty (k=2) of spectral radiance was 1.8 % at 250 nm, 0.90 % at 400 nm, 0.64 % at 800 nm, and 1.3 % at 2500 nm respectively.
In photometry and radiometry, photodetectors such as silicon detector and PMT detector are widely used. In precision metrology, the uncertainty of the nonlinearity should be considered. Superposition method is used to analyse the linearity. The silicon trap detector is measured using both nonmochromator light and monochromator light. First, integrating sphere with broadband light is used to test the linearity. The result shows that the nonlinearity is (1-3)×10<sup>-4</sup> from 1uA to 1mA. The monchromator light result shows that the nonlinearity is below 3×10<sup>-4</sup> from 1uA to 1mA, which is consistent with the integrating sphere method. For the PMT detector, the linearity is measured only using monochromator light. Experiment shows that the nonlinearity is less than 1×10<sup>-3</sup> through three orders of magnitude.
Stray light due to the array spectroradiometer characteristic can’t be ignored in the ultraviolet region. In order to obtain a true spectral power distribution, stray light correction must be considered. Array spectraradiometer covering 200nm- 460nm is investigated using lasers and filters. First, several lasers are measured using the array spectroradiometer. Due to the fact that the wavelengths of the lasers are beyond the capabilities of the spectroradiometer, the response in the UV region is originated from stray light. Results show that the stray light contribution is at the level around 2×10<sup>-5</sup>. In order to correct the stray light, filters with different bandpass wavelength are used to correct the stray light from different wavelength region. Results show stray light consistency using lasers and filters.
CCD based array spectrometers are widely applied in remote sensing, earth observation, and other industries. However, the signals of ultraviolet region are very weak. Thus, the stray light is one of the most important factors on accurate measurements. In this work, the in-range stray light of commercial UV/VIS CCD array spectrometer and VIS/NIR CCD spectrometer were corrected by mathematical correction method. The measured stray light value at any pixel is of the order of 10<sup>−3</sup> ~10<sup>−5</sup> of the true in-range. A reduction of the stray light effect by 1-2 orders of magnitude can be achieved using a correction matrix based on line-spread functions (LSFs), which can be determined with the help of spectrally tunable lasers. On the other hand, the bandwidth of the commercial CCD array spectrometer was corrected due to the increasing needs for high accurate calibration and measurement of spectral radiometry. The correction outcome is in good agreement with the measured results by monochromator spectroradiometer.