The camera-target method is an indirect measurement of the flux profile, but offers a high level of spatial resolution. It is widely applied in the CST R&D community.
A diffusely scattering target which either transmits, or more commonly, reflects incident light, is generally placed parallel to the receiver aperture or perpendicular to the axis of the concentrator, as near as possible to the focal plane. The light reflected off this target is then recorded with the use of a camera, generally with a charge-coupled device (CCD) sensor (Figs 18.6 and 18.7). The camera image files must have a highly linear response, which is not the case with images produced by common consumer grade digital cameras[38]. Analysis of this recorded image gives a relative intensity distribution of the flux profile at the plane of the target, which can be calibrated using an absolute measurement device such as a radiometer. The camera-target method offers fast data retrieval and a large number of data points captured simultaneously.
The surface of the target used in the camera-target method should ideally be perfectly diffuse, i. e. having zero specular reflectivity, and diffuse reflectivity which obeys Lambert’s cosine law. Radiation reflected from such a surface has the same observed radiance regardless of the observation angle, so that light arriving at different incidence angles on the target (i. e., from
|
Paraboloidal reflector 18.6 Camera-target method. |
|
18.7 Image created on a diffuse cooled target by the Australian National University SG4 dish concentrator, with most of the dish reflective area covered in order to reduce the flux on the target (dark filters on the camera lens are used to make some of the focal region detail visible). |
different regions of the collector) is imaged with equal weighting. For reliable flux determination using the camera-target method, a surface with diffuse reflectance within ±5% of ideal Lambertian reflectance, for all incident angles up to the rim angle of the collector, should be used. Reflectivity profiles for thermally-stable Lambertian surfaces are given by Neumann and Schmitt (2003). For accurate quantitative analysis, the camera CCD array must be calibrated in order to obtain accurate relative flux intensities from the image.
In order to determine the absolute flux across the recorded distribution, a calibrated radiometer can be embedded in the target surface at a given point in the flux profile to provide a reference to which the rest of the profile can be compared. This will be observed in the image as a dark spot, and interpolation of the image grey values is required over this region, for calibration to the flux recorded by the radiometer. Such interpolation can, however, introduce significant uncertainty, as flux distributions at the focal region can exhibit very steep gradients (Ulmer et al., 2002). Alternatively, the image intensity can be scaled by equating the integrated sum of the greyscale levels on the CCD array to the predicted total power incident on the target. The total power is given by the product of the solar irradiance, collector mirror aperture area, mirror reflectivity, and target intercept factor. The maximum total error of measurement for experiments conducted on the PSA solar dishes (Ulmer et al. 2002), using this calibration technique, has been determined at -6.2%, +10.6%.
A key issue when using this method with high concentration ratio systems is that the concentrated flux has the potential to overheat and damage the target. Water cooling of the target is a standard approach. Another technique is to move an uncooled target through the flux whilst a series of images are captured. In this approach a radiometer can be located just behind the plane passed over by the moving target. Another possibility is the use of Jupiter or the full moon as the light source. A detailed description of a camera-target system is given in Section 18.3.2.
