3.2.2.1 Specular reflectance

The specular reflectance is defined as the ratio of the energy flux reflected by a surface in the specular direction to the radiation incident on it [7]. It describes the case of reflection on a perfectly flat or planar surface (i.e., with roughness much less than the radiation wavelength, for instance, highly polished), where parallel incident beams are reflected as parallel beams, according to the laws of reflection [10] (see Fig. 3.2 left).

Although this is the ideal case, reflection at real surfaces is always mixed with a certain amount of scattering [11,12]. For highly specular surfaces, this scattering is close to the specular direction. The specular reflectance is written as ρ(λ, θ_i, θ_r, φ, T_s) (where θ_r can be omitted) and depends on the acceptance angle, φ, which is the angle associated to the detector aperture of the measurement instrument. In this text, φ means the half acceptance angle. This parameter is also defined as the angle between the specular direction and the direction of the admissible dispersion on the surface for the specific application [7].

Strictly speaking, the ideal specular reflectance is related to φ=0, whereas the measurement at any small offset φ is more appropriately termed near specular reflectance [10]. To simplify, near specular reflectance is typically known as specular reflectance. If a specular surface element irradiated by an incident beam is viewed, the element appears brighter from the specular viewing direction—that is, the mirror-like effect is produced.

In solar thermal concentrating applications, only the solar radiation reflected by the concentrator in the specular direction with a certain φ reaches the receiver and is converted in thermal energy. As a consequence, the specular reflectance is the representative parameter of a solar reflector.

3.2.2.2 Diffuse reflectance

When a set of parallel beams impacts on a body with a rough or microscopically structured surface of a size equal or greater than the λ of the incident radiation, each individual beam meets a different surface slope and the law of reflection takes effect for a different θ_r to the surface normal at the contact point (see Fig. 3.3).

As a result, the radiation is diffusely scattered in all directions in the plane of incidence (see Fig. 3.2 center). If a diffuse surface element irradiated by an incident beam is viewed, the element appears equally bright from all viewing directions. The amount of diffuse scattering depends on the level of roughness or microscopic surface structuring. The ideal diffuse reflector is a Lambertian surface, where an equal amount of radiation is scattered in each direction of the hemisphere. The goal in reflector materials for concentrating solar applications is that the surface structure is flat enough to minimize the scattering. Real surfaces are not only specular or diffuse, but generally have an intermediate behavior, as indicated in Fig. 3.2 right.

3.2.2.3 Hemispherical reflectance

The hemispherical reflectance is the ratio of the energy flux reflected by a surface within the complete hemisphere over that surface to the radiation incident on it [7]. Therefore, it comprises the total amount of reflected intensity (both specular and diffuse reflection) over the entire hemisphere. The hemispherical reflectance is written as ρ(λ, θ_i, h, T_s). In the hemispherical reflectance the angle θ_r is substituted by “h,” which means hemispherical.

3.2.2.4 Specularity

The term specularity describes the beam spread caused by scattering and is defined as the ratio between specular and hemispherical reflectance. A surface with perfect specularity produces no scattering. In that case, specular and hemispherical reflectance values are the same and the specularity is equal to the unit.

3.2.2.5 Spectral reflectance

The spectral reflectance is the reflectance measured at a given T_s and λ, within a small wavelength interval, Δλ, centered at λ [7]. Spectral reflectance is typically named monochromatic reflectance and may be specular, diffuse, or hemispherical. This parameter is indicated with the subscript “λ” after the ρ symbol. Consequently, ρ_λ(λ, θ_i, θ_r, φ, T_s) is the spectral specular reflectance and ρ_λ(λ, θ_i, h, T_s) is the spectral hemispherical reflectance. A graph representing the specular reflectance over a λ range is called a reflectance spectrum.

3.2.2.6 Solar reflectance

The solar (also called solar-weighted) reflectance is a weighted average of the spectral reflectance over the solar λ range, with a standard solar spectral irradiance distribution as the weighting function. To express that a value is the solar-weighted reflectance, a subscript “s” is included after the ρ symbol. Accordingly, ρ_s([λ_a, λ_b], θ_i, θ_r, φ, T_s) is the solar specular reflectance and ρ_s([λ_a, λ_b], θ_i, h, T_s) is the solar hemispherical reflectance. In this nomenclature, λ_a and λ_b are the lower and upper limits of the λ range for the evaluation. According to the standards ASTM E903-82 [13] and ISO 9050 [14], solar weighting is calculated as in Eq. (3.3):

${\rho }_s\left(\left[{\lambda }_a,{\lambda }_b\right],{\theta }_i,{\theta }_r,{\displaystyle T_s}\right)=\frac{{\displaystyle \underset{{\lambda }_a}{\overset{{\lambda }_b}{\int }}}{\rho }_{\lambda }\left(\lambda ,{\theta }_i,{\theta }_r,T_s\right)G_b\left(\lambda \right)d\lambda }{{\displaystyle \underset{{\lambda }_a}{\overset{{\lambda }_b}{\int }}}{G}_b\left(\lambda \right)d\lambda }$

where G_b(λ) is the direct solar irradiance. It is recommended to use the direct solar irradiance spectrum ASTM G173-03 [15] for direct irradiance and the appropriate Air Mass (i.e., for Europe and United States it is AM 1.5) [7]. If another solar irradiance spectrum is used, it must be indicated when expressing the solar reflectance to avoid confusion in the interpretation of the results. Fig. 3.4 shows both a typical hemispherical reflectance spectrum of a silvered second-surface mirror and the solar irradiance spectrum from ASTM G173-03 [15], plotted in the same λ range.

Concerning the proper wavelength range, the following recommendations are included in Ref. [10]:

• The λ range of the solar spectral irradiance provided in Ref. [15] is λ=[280, 4000] nm. However, as the near-infrared range has low impact, it is practically appropriate to evaluate the range λ=[280, 2500] nm, using standard laboratory equipment.

• Some instruments require the use of an extra ultraviolet (UV)-lamp to measure from 280 nm. Therefore, the starting range of the measurement can be changed to 300 nm. This is allowed since the UV tail has low impact on the final average value.

• The wavelength interval of the measurement shall be the maximum of 5 nm.

According to Ref. [7], unless otherwise specified, T_s taken into consideration is the ambient temperature (25°C). Consequently, in the rest of this work, T_s is omitted from the nomenclature except when it is different from 25°C.

3.3.1.1 Silvered thick-glass mirrors

The most widespread type of reflectors is the monolithic-glass-based one. In general, it is composed by a reflecting silver layer coated with a glass layer on the front side. Although there is no strict classification between thick and thin glass reflectors, the distinction level is considered at a glass-layer thickness of about 1 mm [21]. The glass thickness for thick-glass reflector typically is 3 or 4 mm. It is well known that glass may absorb little of the solar energy spectrum if its Fe₂O₃ content is low (Fe₂O₃ content around 0.02%) [22]. The type of glass for solar applications is called “clear white,” or low-iron glass.

The silver layer (with a typical density of between 0.8 and 1.2 g/m² [2]) is extremely vulnerable to ambient contaminants, moisture, or salty atmospheres. Thus, although in this configuration silver is properly protected on the front side by the glass layer, the silver must be conveniently protected also on the back side. This protection is usually achieved by a copper layer and two or three protective paints, as shown in Fig. 3.7.

These mirrors are manufactured following the conventional wet chemistry process: the clean glass is sensitized with SnCl₂, the Ag layer is applied by chemical reductive processes, the Cu layer is applied by chemical processes, the mirror-backing paint is applied by various techniques, and the applied paint is force cured by heating [23]. The copper layer has a significant protective role to retard tarnishing of the silver layer and provides a surface for the mirror paint to adhere to [21].

To obtain more environmentally friendly mirrors, the copper layer should be avoided (i.e., substituting it with a cation-ion based solution [21,23]) and the back paints should reduce the lead content. Currently, the lead content of back protective paints vary between 0.5% and 2.5% by weight [21] and the goal is to achieve lead-free paints, with <0.15% lead by weight [2].

In the manufacturing process of a curved thick-glass mirror, the flat glass panel is heated on precise parabolic molds in ovens to achieve the parabolic shape; the glass is then silvered on the back and covered with the back coatings (see Fig. 3.8 left). Finally, ceramic pads used for mounting the mirrors to the collector structure are attached with a special adhesive [24]. When reflectors are used for solar tower heliostats, they are manufactured flat and are cold-conformed in a process called canting (see Fig. 3.8 right). Therefore, due to the increased cost of the manufacturing process, mirrors for PTC are more expensive than those for heliostats. In both cases, recent market developments have promoted a significant cost reduction.

Concerning reflectance properties, high reflectance is obtained due to the silver intrinsic behavior (see Fig. 3.6) and small scattering can be expected because glass is a highly smooth substrate. The version of these reflectors with copper layer and paints with a high lead content has demonstrated excellent durability [24]. However, new reflectors without lead and copper as corrosion inhibitors still have to prove their durability.

The maximum tensile strength of the glass mirrors depends on their thermal treatment (annealing, hardening) in the production process. Depending on the cooling speed, the resistance to mechanical breakage and the size of the fragments may vary significantly, having an impact on breakage mechanisms with the highest wind-loads and handling safety.

3.3.1.2 Silvered thin-glass mirrors

The composition of silvered thin-glass reflectors is basically the same as the previous one. The only difference lies in the glass thickness, in this case, less than around 1 mm. To obtain the concentrator shape, reflectors are bonded to metal, composite, or glass back structures. These mirrors have been widely used in those solar thermal concentrating technologies were no standard geometry and size is preestablished, like it is the case of the Eurotrough-type PTC. Examples are found on small-sized PTCs (see Fig. 3.9 left), parabolic dishes (see Fig. 3.9 right), and secondary concentrators [25].

Reflectance is even higher than in thick-glass reflectors because the glass topside layer is smaller and, as a consequence, the optical path taken by the solar beams is shorter. The durability is similar to that in thick-glass reflectors, also depending on the copper and lead content, but silvered thin-glass reflectors are relatively lightweight in comparison to thick-glass reflectors. They are cheaper than shaped thick-glass reflectors in PTC because they are originally flat, but the cost of the back structure must be added.

3.3.1.3 Laminated silvered glass mirrors

Another type of glass reflector is the laminated silvered glass mirror which uses silver as reflector layer, protected by a glass layer both on the topside and also on the backside, like windshields in automotive application. The reflectance is as high as the thin-glass reflectors, because the topside layer typically has a thickness of 1–2 mm. The total thickness is similar to the thick-glass reflectors. In principle, excellent durability is expected because the hard glass layer is placed on both sides. The shape is given in the thermal manufacturing process. Such mirrors are on the market and in commercial use, although production cost is higher than for monolithic thick glass.

3.3.2.1 Aluminum mirrors

Depending on the top coating, several types of aluminum reflectors are available. The most extended is built of a pure aluminum deposition onto a polished aluminum substrate with a protective aluminum-oxide layer in between (anodized) and some transparent coatings on the top (see Fig. 3.10). For instance, an alumina (SiO₂) layer has been proven to protect against abrasion and corrosion, but its suitability is still subject to demonstration by economic considerations in order to develop this type of reflector [26]. Other types of aluminized reflectors are based on organic or anodized aluminum top coatings.

Aluminum reflectors are increasingly being used in concentrated solar applications because of their light weight, high ductility (they support high wind loads without damages), and flexibility in the design, construction, and assembly of new collectors [27]. Also, the manufacturing process is well suited for mass production because of the coil coating process. Consequently, they offer a significant cost reduction potential compared to glass mirrors.

On the other hand, in addition to the lower reflectance of the aluminum (see Fig. 3.6), the rolling marks that result from the production gives them a rougher surface than glass mirrors have, which leads to higher scattering [28]. The lower reflectance of the aluminum in comparison to silver makes them more applicable for low-temperature concentrators, such as industrial process heat or water heating systems. The lower durability of aluminum reflectors compared to glass in urban, industrialized, and polluted locations is sometimes solved by incorporating a glass cover on the aperture plane of small-sized PTCs [29] (see Fig. 3.11). In addition, manufacturers are continuously adapting the product to improve the top coatings.

3.3.2.2 Silvered polymer film mirrors

Silvered polymer film mirrors have been produced for several decades [24]. They basically use a silver reflective layer protected on the top by several polymer layers and deposited on a back substrate (sometimes also a polymer). Since the polymeric substrate limits the temperature during the deposition of silver, temperatures exceeding 60–80°C are not recommended, even though silver can bear temperatures up to 100°C with no agglomeration and therefore shows no losses in specular reflectance [30].

These materials are lightweight and flexible, thus they have the potential to be formed into any concentrator shape without limitations. The bonding process of the film to the substrate (e.g., glass, polycarbonate, methacrylate) must be performed with great care to avoid air bubbles being trapped between the film and the substrate, which could introduce deviations in the reflected beams. In addition, any surface deviation of the substrate is reproduced on the reflector. Regarding durability, manufacturers are still improving the resistance of silvered polymer film mirrors to abrasion. Some examples of developed products are included in Refs. [21,24,31,32].

3.4.1.1 Reflectance measurement

Suitable instruments to measure specular reflectance over an appropriate range of λ, θ_i, and φ are required to assess solar reflector materials properly. A suitable measurement protocol is required not only to provide all the information to assess this kind of materials, but also to guarantee that values given by different manufacturers and independent institutions are comparable among them.

An international group of experts has been created under the framework of the International Energy Agency (IEA) SolarPACES Task III, collaborating on the project “Development of guidelines for standards for CSP components.” The goal is to collect and prepare recommendations of procedures and protocols to be presented to the standardization organizations and further translated into national and international standards. Concerning the reflectance measurement topic, guidelines titled “Parameters and method to evaluate the solar reflectance properties of reflector materials for concentrating solar power technology” were published [10]. A new version of the document is currently under preparation.

These guidelines contain the properties of the optimum instrumentation needed to evaluate adequately all solar reflector materials for CSP applications. As this ideal instrument does not currently exist, the minimum requirements for selecting existing and future instruments to best characterize the minimum relevant reflector parameters were also specified. These ideal requirements are summarized as follows [10]:

• λ range between 280 and 2500 nm, according to the range of the solar spectrum [15], and λ intervals which allow accurate solar weighted calculations, with a maximum interval of 5 nm. In some instruments cost can be greatly reduced by avoiding the use of an extra UV-lamp and starting the measurement at 300 nm. This is allowed since the UV and NIR tails have low impact on the final average value. As a minimum requirement, the specular reflectance must be measured at several (at least three) narrow λ range appropriately spaced along the solar spectrum or at a specified range that accounts for differing λ dependent scattering properties.

• Various θ_i ranging from near normal (θ_i≤15 degree) to at least 50 degree or an innovative approach to obtain the reflectance as a function of θ. When this is not possible, the reflectance must be measured at least at near normal incidence.

• For specular reflectance, several precise and selectable acceptance apertures or an innovative approach that covers a range of φ appropriate to measure the specular reflectance as a function of φ. Specular reflectance measurements must be carried out at least from φ>0 to φ≤20 mrad.

• For hemispherical reflectance, if an integrating sphere is part of the setup, its size needs to be adequate to minimize errors. A diameter of at least 150 mm is recommended.

• Absolute measurements are taken, thus no reference mirror is necessary. If it is not possible to achieve this requirement, measurement must be done with a stable, calibrated reference mirror.

• Adjustment options for correcting beam alignment, to account for different mirror thicknesses and surface curvature.

• Measurement spot size as large as possible, with the option of reducing size in case of interest.

• All measurements included in the process of obtaining the desired result should be done at the same spot on the sample and with the same instrument.

• Noncontact measurement to avoid damaging the surface.

• Portable for field measurements, which implies high battery autonomy, battery status display, light-weight, small compact size, easy to handle, robust, and digital data storage.

• A coordinate system that is incorporated in the instrument to identify the position of defect points on a mirror and study their evolution over time would be a useful addition for monitoring quality.

• Low and stable type-B uncertainties [33].

• No influence by external stray light.

• No risk to human health or environmental hazard involved.

Although the ideal instrument accomplishing all desired requirements to measure reflectance of solar mirrors properly does not exist on the market, several items of marketed equipment are commonly employed. A brief description of these devices is given next.

Spectrophotometers

A spectrophotometer is an instrument consisting of a spectrometer for producing light of any selected λ, and a photometer for measuring the light intensity. The use of these instruments is widely extended in the industry (mainly in chemistry) to measure optical properties (transmittance, absorptance, and reflectance) of both liquids and solid materials.

Concerning the list of requirements for instrumentation presented above, the main advantage of a conventional spectrophotometer is that the reflection can be measured over the entire solar spectrum because these instruments usually utilize two light sources and a monochromator to scan the near UV, visible (vis), and NIR spectrum in desired wavelength intervals. The main disadvantages are that they are not portable and thus can only be used in laboratories. They do not offer beam adjustment, adaptation to mirror curvature, or different mirror thicknesses. Measurements are typically taken at a near normal incidence angle of θ_i=8–15 degrees.

Both specular and hemispherical reflectance can be measured with spectrophotometers depending on the accessory coupled. Integrating spheres are widely employed to measure hemispherical reflectance accurately. Although several accessories to measure specular reflectance with a spectrophotometer are available, most of them still have important disadvantages. The simplest method is to use the integrating sphere specular opening port (or light trap) to measure diffuse reflectance and later calculate specular reflectance by the difference between hemispherical and diffuse components. However, this method is not recommended for CSP applications because the resulting φ is too large (around 80 mrad). Another example of specular accessory is the Universal Reflectance Accessory, which measures at several θ_i but with unknown φ (estimated to be around 100 mrad).

The spectrophotometer commonly used is the Perkin Elmer (PE) Lambda 950 or Lambda 1050 (see Fig. 3.15) [34], or similar. The PE scanning spectrophotometers models Lambda 950 and 1050 are UV-vis-NIR double beam and double monochromator ones with two light sources: a deuterium lamp for the UV range and a halogen lamp for the vis/NIR range. The beam source has a spectral range from 175 to 3300 nm, θ_i=8 and about 12.0×4.5 mm² size.

Reflectometers

Reflectometers are measurement devices that measure the intensity of the light source after reflection on a sample at a few λ or even at only one. After calibration with a known reflectance reference standard, the measured intensity is automatically transformed into a reflectance reading by correlating the measured flux intensities of the reference standard and the sample. Most reflectometers give readings of specular reflectance at near normal incidence and one or several selectable φ. In addition, there are some instruments that incorporate a small integrating sphere to measure hemispherical and diffuse reflectance thanks to a specular opening port.

The main advantage in this case is portability, which makes reflectometers useful for quality control and/or cleanliness assessment in the solar field. In this sense, the following requirements are important when choosing a suitable instrument for on-site measurements:

• size and weight that allow a single operator to use it even overhead

• easy handling and robustness

• short measuring time

• mechanism for beam adjustment and positioning easily accessible and controllable

• stable and durable calibration mirror

• capable of measuring on curved and/or second-surface mirrors

• ideally the possibility of storing the digital data in addition to a display reading

• high battery autonomy, indication of battery status, and no influence of the battery level on the readings

• stability in the typical operating environment of a CSP plant (this includes high, low, and variable ambient temperatures, relative humidity, wind, and dust)

• capable of working in strong sunlight without stray light problems

The portable reflectometer most commonly mentioned for reflectance measurements of solar reflectors is model 15R-USB manufactured by Devices and Services (D&S; see Fig. 3.16) [35], μScanTM manufactured by Schmitt Measurement Systems [36], Condor by Abengoa [37], CM-700d/600d by Konica Minolta [38], and 410 Solar by Surface Optics [39]. Also, the PFlex developed by Fraunhofer-ISE has recently been marketed by PSE [40]. Among them, the reflectometer most commonly used to measure monochromatic specular reflectance is the 15R-USB by D&S. It has a light emitting diode (LED) source of λ=[635, 685] nm, with a peak at 660 nm. A new version of the instrument is capable of measuring at λ={460, 550, 650, 720} nm. φ can be selected from φ={3.5, 7.5, 12.5, 23} mrad. First- and second-surface mirrors can be measured by modifying a screw located on the central side that controls the mirror-glass thickness. In addition, it allows measuring curved mirrors using two adjusting screws situated on its base. With these two screws, the optical measuring system is aligned and adapted to the curved surface [35].

3.4.1.2 Soiling evaluation

Reflectance of mirrors continuously suffers a decrease due to outdoor exposure to the environment in the solar field, which can provoke a significant decay in the power production of a CSP plant. The soiling rate in a CSP plant depends on the type of reflector and the plant location. The type of reflector affects the soiling rate because particle adhesion to the top surface is influenced by the material properties [41]. With respect to the location, several aspects must be considered, such as: weather (rain profile, humidity, wind speed and direction, temperature, dew formation, etc.); closeness to coasts, industrial areas, cities, roads, and zones with high pollution; fauna and flora (specially the presence of seeds); and particle (dust and sand) concentration, size, and chemical composition.

A deep review about the impact of dust accumulation on the use of solar energy (including photovoltaics, concentrating photovoltaic and CSP systems) was published in 2013 [42] and has recently been updated [43]. Several research studies about reflectance losses of solar reflectors due to dust accumulation in outdoor exposing were conducted in different locations. Table 3.2 includes a summary of some representative studies.

3.4.1.3 Durability assessment

Degradation of the mirror surface can be cost-intensive. NREL has measured an annual degradation rate of 0.2% (pp) in hemispherical reflectance for silvered-glass mirrors exposed in different climate conditions in the United States [54]. The examined mirrors contained high contents of lead in the protective paint layers for enhanced durability. For environmental reasons, the lead content of the prime and intermediate coat has been reduced from 20 to 2.5 wt% and 10 to 1 wt%, respectively. Furthermore, a tendency towards lead-free mirrors (<5 ppm Pb) is desirable, but the durability of these products has not yet been proven. Higher degradation rates than 0.2% (pp) are expected for newer materials with a low Pb-content. Additional losses may occur during the exposure in desert sites with high wind speeds and dust content due to glass erosion and the associated reduction in specularity caused by wind blown particles.

For example, if we assume an annual degradation rate of 0.3% (pp) in specular reflectance in a 200 MW_el power plant with an annual capacity factor of 40% (like the Noor-2 plant in Morocco), the loss in reflectance equals 0.45 less efficiency per year. On average, over the lifetime of the plant (20 years), the electricity production will be reduced by around 4.5%. For a revenue of 0.14 US$/kWh, the cost of degradation can then be calculated to≈88 million US$, which could be at least partially saved by using durable reflector surfaces.

The high impact of the component durability on the revenue over the plant lifetime raise the importance of durability assessment. For this purpose, solar reflectors are tested in accelerated aging tests, at increased levels of humidity, radiation and in interaction with corrosive or erosive agents, etc. Predicting the lifetime of the material in service based on the results of accelerated tests is a complex task. Correlations between three outdoor reference sites and a specifically developed accelerated aging testing sequence have been derived for aluminum mirrors as described in Chapter 6 (accelerated aging methodology). Developing such correlations requires to monitor the degradation outdoors of exposed samples in different climatic environments as reference for the accelerated tests. The research for a proper accelerated aging testing sequence for silvered-glass mirrors is still ongoing and standardization committees (e.g., AENOR GT2) are currently working on this issue [55]. For more details, see Chapter 6 (accelerated aging methodology).

3.4.2.1 3D measurement with photogrammetry

Close range photogrammetry methodology [56–58] builds a 3D model of a point pattern attached on the mirror surface analyzing the information of several digital images captured from a wide range of viewing positions (Fig. 3.17). The points of interest have to be highlighted by special markers. The application of these markers to the collector surface or structure means a high preparation effort. However, results of the photogrammetry consist of 3D coordinates of the point pattern. By applying geometrical methods (Delaunay triangulation, Sobel filters, etc.) to the coordinates obtained by photogrammetry, local slope vectors between adjacent points can be obtained in order to calculate SDx and SDy values.

Spatial resolution of photogrammetric measurements is limited by the point pattern density, ranging from more than 1000 points per square meter for single facet sizes (1–3 m²) to around 100 points per square meter for whole heliostats or PTC (more than 50 m²) [59].

As photogrammetry can be applied in any mirror position, it is very helpful not only to compute mirror slope deviation parameters SDx and SDy, but also to measure for example, deformations at different positions due to gravity or different loads conditions (see Fig. 3.18), or time-dependent deformations using video-photogrammetry (e.g., wind effect on mirror shape). More examples of photogrammetry are given in Sections 5.2.2.2 and 5.2.3.1.

3.4.2.2 Slope measurement with deflectometry

Deflectometry is performed with an optical measurement system for high resolution and high precision quality assurance measurements of the shape of mirror panels or larger concentrator units. It uses a non-contact optical measurement and digital image processing technique based on the deflectometric measurement principle (Fig. 3.19). This measurement technique was developed at DLR and specifically optimized for the measurement of shape deviations of concentrating solar reflector panels [60,61]. It has been adapted to industrial requirements and is applied in industrial production quality control. Regular line patterns of different resolution, phase-shift, and orientation are projected on a flat white target of sufficient dimensions; a digital camera takes images of the reflected line patterns seen in the mirror object (Fig. 3.20); and a control unit performs the image processing and evaluation for calculating the local normal vectors of the reflecting surface. Fig. 3.21 shows resulting maps of RP3 type parabolic trough mirror panels in local slope deviation of the normal vector, and in focus deviation of the reflected rays. The measurement duration has been below 10 seconds for automated inline systems.

The spatial resolution of the result depends on the camera resolution and mirror size and is typically 1–3 mm on mirror panels. Accuracy and reproducibility are around 0.1 mrad and are backed by reference measurements on transfer standards and on a water surface. Ongoing work in research aims at reducing the influence of mirror shape deformations due to the tracking angle or the assembly accuracy [62]. Applications and specific deflectometry methods for large scale concentrators applicable in the solar field are presented in Section 5.2.2.1.

3.4.2.3 Slope measurement with laser reflection scanning

The Video Scanning Hartmann Optical Test System (VSHOT) is a laser ray-trace system to characterize both point and line-focusing concentrators [63]. Such scanning approaches are rather time-consuming when high spatial resolutions are required. A computer-controlled laser scanner and digital-camera image acquisition are used to provide optical surface contour data. The incident laser beam is parallel to the optical axis of the system. The laser scans a mirror in a pattern predefined by the user/programmer. At each scanned position, the laser beam is reflected back to a target and the location is imaged using a camera. The surface slope is calculated at each scanned position using the laser output angle and return-spot location.

The laser reflection scanning method can be used in the forward configuration as described above with the incident laser beam parallel to the optical axis of the system and the target at the focal point or line. But a reverse configuration is also possible, where a point light source is placed in the focus and the camera scans the mirror in a predefined pattern.

Laser reflection scanning techniques have been adapted to measure reflector panel shape accuracy in laboratory quality control as well as in field setups [64–66], as can be seen in Fig. 3.22.

3.4.2.4 Laser radar

Recent development of commercial laser-scanner has improved very significantly their accuracy, allowing the use of those systems as an alternative to photogrammetry for shape assessment of mirrors and facets. Since laser scanners also offer 3D coordinates of the mirror surface, the procedure will be the same as photogrammetry methods, with the advantage that point density is higher with laser-scanners, allowing the measuring of more than 100,000 points per square meter. On the contrary, the reflecting surface must be painted to avoid specular laser reflection, or measurements must be done on the rear side of the mirror. Fig. 3.23 shows the measurement setup of a heliostat facet using a laser-scanner.