WO2012014236A1

WO2012014236A1 - Quadrant photodetector and related method for sun tracking

Info

Publication number: WO2012014236A1
Application number: PCT/IT2010/000343
Authority: WO
Inventors: Alessandro Rossi
Original assignee: Alitec S.R.L.
Priority date: 2010-07-30
Filing date: 2010-07-30
Publication date: 2012-02-02

Abstract

A quadrant photodetector has four photosensor arranged in quadrature and an opaque mask arranged at a predefined distance from the photosensor and it extends from the center of the photodetector until covering, in a top view, about half the area of each photosensor. Such a quadrant photodetector is very cost effective and it allows a high spatial resolution in a specific angle range and it is able to give useful information in a wider angle range. A very efficient method for sun tracking can be associated to the above photodetector.

Description

TITLE

QUADRANT PHOTODETECTOR AND RELATED METHOD FOR SUN TRACKING

TECHNICAL FIELD

The present invention concerns a photodetector, to be used in particular as a sun position detector.

The present invention also concerns a method for sun tracking, that is for aligning the orientation (azimuth and altitude) of a solar tracking system comprising a photodetector with the direction of the light coming from the sun.

STATE OF THE ART

Photodetectors are sensors of light and they are widely used as for instance in sun tracking systems. In this last field one of the most common types of photodetectors is quadrant photodetector, which is schematically shown in fig. 1 of the accompanying drawings, that is an optical detector providing four photosensitive areas, PD1, PD2, PD3 and PD4, in quadrature. The relationship of the output from each of the quadratures can be used to determine the centroid location of a incident light source. The light source can be directly illuminating the quadrant photodetector, with little or no additional elements in the optical path, or it can be reflected light in which light is incident onto a surface and the reflected light is directed onto the quadrant photodetector. Optical elements such as lenses and filters can be used to make sure that the desire optical flux is incident onto the photodetector. The output from each of the four photosensitive areas is measured and with appropriate circuitry the value of the incident flux can be obtained. The magnitude of the response from each photosensitive area is used to determine the relationship of the x and y centroid. As shown in fig. 2 conventional quadrant photodetectors have associated a shield, S, provided with a pinhole, P, that defines the size of the diameter of the incident light spot, LS. Readout per pair is normally used so that the output of the photosensitive areas are properly used for determining the X position (azimuth) and the Y position (elevation).

The size of the incident light spot LS, and the distance between the pinhole and the photosensitive areas effect the spatial resolution, that is the ability to resolve minute changes in the position of the light beam, and the tracking angle range of a conventional quadrant photodetector. In fact, a smaller incident light spot will result in a greater spatial resolution, while a greater incident light spot will result in a greater tracking range. As it can be deduced in fig. 2, a lower distance of the pinhole will result in a faster moving incident light beam so that a greater spatial resolution but a lower tracking angle range are obtained. Therefore, in quadrant photodetectors there is a trade-off between spatial resolution and tracking angle range.

However, there are factors which influence just one between spatial resolution and tracking angle range. In conventional quadrant photodetectors the size of the photosensitive areas affects the tracking angle range. In fact, as it is easily understandable, the more is the size of the photosensitive area, the greater will be the tracking angle range of the photodetector. On the other hand, as it is obvious, the spatial resolution is influenced by intrinsic performance parameters (spectral range, sensitivity, dark current, noise-equivalent power, gain, etc) of the photosensitive elements which form the photosensitive areas.

Therefore, in conventional quadrant photodetectors with pinhole the size of the photosensitive areas has to be increased and the performance parameters of the photosensitive elements have to be improved in order to have better tracking angle range and spatial resolution. Undoubtedly, the above cannot be achieved without increasing the costs of the photodetector.

In addiction, conventional quadrant photodetectors with pinhole have a great limit in the fact that due to the pinhole the incident light beam illuminates only a small portion of the photosensitive areas so that little photocurrents are generated and most photosensitive elements remain idle.

Another conventional quadrant photodetector is known, which is specifically dedicated to be used as a sun sensor and it has been developed to overcome the above limit, in which, as shown in fig. 3 and 4, a center pole, CP, replaces the pinhole and it is able to cast a shadow on the photosensitive areas depending on the position of the photodetector relative to the sun. In this case almost all photosensitive areas are illuminated. The main benefit of center pole quadrant photodetectors, compared with pinhole quadrant photodetectors, is that the photodetector has a simplified structure as there is no need to realize a calibrated pinhole which has to be properly aligned with the photosensitive areas. Another apparent benefit could be the fact that all the photosensitive elements are available for power generation but that also leads to the need of managing the subtraction of large photocurrents.

In some systems a big sized photosensitive area with a good spatial resolution is obtained by providing arrays of discrete photosensitive elements. However due to, among other things, the large amount of photosensitive elements involved, and the electrical system that processes the positional information outputted by the photosensitive elements, such systems are quite expensive and complex.

The available sun tracking systems uses two exclusive strategies in order to track the solar position: a closed loop feedback, based on the signal of a solar sensor; a geo-referenced tracking algorithm, based on ephemeris calculation. Both strategies present their peculiar benefits and disadvantages: the closed loop systems suffer in cloudy condition, due to the direct sun light lacking. The geo-referenced systems maintains the perfect alignment in all the weather conditions, but requires a very stable and heavy foundation in order to guarantees the alignment reliability during the time. Furthermore the initial tracker alignment should be made with a strong accuracy, and this activity should be repeated for each tracker.

SUMMARY OF THE INVENTION

It is object of the present invention to propose a quadrant photodetector which is simple and cost effective and allows at the same time a great accuracy in a specific tracking angle range.

Another object of the present invention is to propose a new and improved method for sun tracking.

According to an aspect of the present invention, the above objects are attained by providing a photodetector comprising at least two photosensitive areas placed on a surface, each photosensitive area comprising at least one photosensitive element adapted to produce a signal in response to a light beam incident thereon with said signals being useful to calculate a direction of said incident light beam, and comprising an opaque mask for casting a shadow on said photosensitive elements, said photodetector being characterized in that said opaque mask is arranged in order to mantain a predetermined distance away from said surface , said opaque mask being shaped and sized so that, in a top view, it occludes a predetermined percentage of each photosensitive area.

As it is known by a skilled person the output signals of two photosensors (photosensitive areas) arranged on a flat surface can be used to track a light beam incident thereon along one direction; an array of three or more photosensors anyway arranged on a surface can be used, by providing a proper calculation algorithm, to track a light beam incident thereon along two (perpendicular) directions in the plane. A photodetector according to the present invention can be arranged according to both the first or the second case in order to have a minute spatial resolution in a very cost effective photodetector.

Advantageously, in a particularly efficient and cost effective layout for performing tracking of a light source along two directions, the photodetector is a quadrant photodetector comprising four photosensitive areas arranged in quadrature on a flat surface, each photosensitive area comprising at least one photosensitive element adapted to produce a signal in response to a light beam incident thereon with said signals being read in pairs to calculate a X and Y directions of said incident light beam, and said opaque mask extends from the center of said quadrant photodetector up to occlude a predetermined same percentage of each photosensitive area.

The quadrant photodetector preferably comprises at least an opaque baffle extending from said flat surface to said opaque mask, said baffle dividing said flat surface in four identical regions each comprising one of said photosensitive areas centred therein.

The particular design of the sensor architecture allows a great accuracy in a little angle range but it provides sun direction information in a wide angle range: ± 90° on both directions with respect to the photodetector central axis, in order to find the solar position starting from any initial tracker position. This feature is useful in the first tracking system initialization or after an energy blackout. In fact, the above allows a quicker installation process as there is no need of performing an accurate alignment of the sun tracking system during installation. According to another aspect of the present invention, the above objects are attained by a method for aligning the orientation (azimuth and elevation) of a solar tracking system with the direction of the light coming from the sun, said tracking system comprising a support member where a quadrant photodetector is firmly fastened, a processing unit for processing data from the quadrant photodetector and controlling a drive mechanism comprising a first actuator for horizontal sweeping (azimuth) and a second actuator for vertical sweeping (elevation), the processing unit controlling said first and second actuators according to X and Y position values calculated based upon the outputs of said quadrant photodetector, said quadrant photodetector comprising at least one photosensitive element per quadrant placed on a flat surface and an opaque mask arranged parallel to the flat surface spaced a predetermined distance away from it, said opaque mask being shaped and sized so that, in a top view, it extends from the center of said quadrant photodetector up to occlude a predetermined same percentage of each photosensitive area, said X and Y position values being each calculated by said processing unit according to output values from photosensitive elements of at least two different quadrants with a readout per pair, said method being characterized in that when said processing unit is not able to perform the calculation of said X and Y position because of a too great misalignment of said quadrant photodetector relative to the sun, it uses output information from the illuminated photosensitive areas for determining which is the direction where said first and/or second actuators have to be operated in order to realign said quadrant photodetector, and it consequently operates said first and/or second actuators until it is able again to perform the calculation of said X and Y positions. Basically the sun tracking algorithm is based on the ephemeris calculation, but each time the tracker alignment is out of the acceptable tolerance, the four quadrant photodetector gives the right signal in order to correct the future trajectory.

Thanks to this hybrid method the tracking system can correct all initial or future misalignments, allowing a faster, lighter and cheaper installation, without suffering the peculiar deficiencies of the closed loop tracking method.

In addiction, once the photodetector's simmetry axis has been aligned to the sun, the power produced by an associated PV system is monitored while a predetermined pseudorandom path around the photodetector's symmetry axis is run by operating said first and second actuators, in order to find the Xp_max and Yp_max values which maximize the energy produced by said PV system- Thanks to the above, the method of the invention allows the best tracking alignment which is the one which maximizes the power generated by the PV system which, in general, can differ from the photodetector symmetric axis, in which X = Y = 0. Furthermore this method allow the photodetector to be mounted in the tracker system without any calibration, being the best alignment direction calculated by the said pseudorandom path algorithm.

BRIEF DESCRIPTION OF DRAWINGS

These and other characteristics of the invention will be clear from the following description of preferred forms of embodiment, given as a non-restrictive example, with reference to the attached drawings wherein:

- figures 1 and 2 show schematic representations in top view and side view of the working principle of a conventional pinhole quadrant photodetector; - figures 3 and 4 show schematic representations in top view and side view of the working principle of a conventional center pole quadrant photodetector;

- figures 5A to 5D show schematic representations in side view of the working principle of a quadrant photodetector according to the present invention;

- figure 6 shows a top view of an embodiment of a quadrant photodetector according to the present invention;

- figure 7 shows a section view of the embodiment of fig. 6.

DESCRIPTION OF PREFERRED EMBODIMENTS

With reference to figs. 6 and 7 an embodiment of a quadrant photodetector according to an aspect of the present invention is pointed as a whole with 10.

The quadrant photodetector 10 includes a PCB (printed circuit board), 11, that has housed on a top surface, 12, four photosensors, 13, 14, 15 and 16 arranged in quadrature with respect to a central point of the top surface 12 of the PCB 11. Each photosensitive element 13, 14, 15 and 16 generates an electrical signal based on the light intensity of a light beam incident on it. The photosensitive elements 13, 14, 15 and 16 preferably are commercially available photosensors, for instance PIN photodiodes with a relatively small responsive area. From the bottom surface, 18, of the PCB 11 a set of pin connections, 19, extend which are an anode connection for each photosensitive element and a cathode connection common to all photosensitive elements. The PCB 11 is housed in a base, 20 which has holes, 21, for connecting to a support member, for instance of a sun tracking system. The base 20 could be of a plastic or metallic material.

A central stem, 22, extends from the top surface 12 of the PCB 11, and supports an opaque mask, 23, which rests above the PCB 11 spaced a predetermined distance away from the PCB 11. The opaque mask 23 is preferably an opaque thin member with a circular cross- section which, in a top view, occludes the central area of the PCB 11 up to substantially the centre of the photosensitive elements 13, 14, 15 and 16. The opaque mask 23 could have a different cross section shape such a square shape or other shape which is able to shadow symmetrically the photosensitive elements, at least with respect to a X axis and a Y axis, in fact could be useful to have different sensibility along the two axis.. The space between the PCB 11 and the opaque mask 23 is divided into four separate quadrants, 24, 25, 26 and 27 by a baffle assembly, 28. Each quadrant 24, 25, 26 and 27 comprises a photosensitive element respectively 13, 14, 15 and 16. A transparent dome, 29, is mounted on the base 20 and it encloses the PCB 11, the photosensors 13, 14, 15 and 16, the mask 23 and the baffle assembly 28.

A specific design of the above described embodiment of the present invention provides that the photosensors 13, 14, 15 and 16 have a lens diameter of 1,8 mm and two opposite photosensors are arranged with a centre distance of 13 mm. The opaque mask 23 is a disk of 13 mm of diameter so that, in a top view, it covers substantially half each photosensor. The mask is spaced 16 mm away from the top surface 12 of the PCB 11. Each vertical wall of the baffle assembly 28 extends 12 mm from the centre pole 22. Such an arrangements allows a spatial resolution of 0,01° within a solid angle of ± 2° in both X e and Y directions. Within the above 16 square degree angle range all the photosensors are contemporaneously illuminated by a directly incident light beam coming from a sun-like light source. In a solid angle range of about 2π at least one photosensors is illuminated by a directly incident light beam and it is able to provide not quantitative but useful information on which is the direction the photo-detector's support has to move in order to recover the alignment with the light source. The lens diameter of the photosensors 13, 14, 15 and 16 and the distance between the opaque mask 23 and the PCB 11 are the most important parameters in optimizing the accuracy of the quadrant photo-detector of the present invention and the amplitude of the solid angle in which the sensor return the X and Y measurement with high accuracy. The above described solution is aimed to obtain a quadrant photo-detector having very low overall production costs which, at the same time, has a very good accuracy in a specific angle range. Still maintaining a similar ratio between the diameter of the photosensitive elements 13, 14, 15 and 16 and the distance between the opaque mask 23 and the PCB 11, a quadrant photo-detector of the present invention could be designed for having a much smaller global size. In this case the photosensitive elements 13, 14, 15 and 16, the opaque mask 23 and the baffle assembly 28 could be directly obtained by known techniques on the top surface 12 of the PCB 11. It can be demonstrated that the best possible accuracy is obtained when the opaque mask 23 covers, in top view, exactly half the area of each photosensitive element, in fact with this configuration the measurement linearity is maximized.. However, the opaque mask 23 could be arranged for covering a different percentage of the area of the photosensors and the measurement could be linearized thanks to a normalization curve..

Different embodiments of a photodetetor according to the present invention could be provided still maintaining the same peculiar target of the opaque mask covering substantially a suitable percentage of the photosensitive areas. In fact, a photodetector according to the present invention could have just two photosensitive areas arranged on a same flat surface and the photodetector can be used in this case for tracking a light source along a single direction. Still a photodetector of the present invention could have an array of photosensors or photosensitive areas anyway arranged on a surface so that by a proper calculation algorithm a light source could be tracked along two (perpendicular) directions in a plane.

With reference to figs. 5A to 5D the working principle of a quadrant photo-detector of the present invention and a method for sun tracking employing a quadrant photo-detector according to the present invention will be now explained. Figs 5A to 5D show different angular positions of a light source relative to the X axis, but the same applies to the Y axis. For simplicity of exposure in the following a description is given in which it is assumed that the relationship between the X and Y signals and the quadrant signals is given by the following formulas (in which quadrants are labelled as in fig.l with photosensor 13 corresponding to PD1 and photosensor 14 corresponding to PD2):

X=(PD1-PD2)/(PD1+PD2)

Y=(PD3-PD4)/(PD3+PD4)

Nevertheless, the X and Y axis could be rotated of any other angle with respect to the position shown in figures, and, for a rotation of 45° as it happens in most quadrant photo-detectors, the above relationship could be given, for instance, by the formulas:

X=(PD 1 +PD3)-(PD2+PD4)/(PD 1 +PD2+PD3+PD4)

Y=(PD1+PD4)-(PD2+PD3)/(PD1+PD2+PD3+PD4)

In fig. 5A the quadrant photo-detector of the invention is perfectly aligned with the light source, the opaque masque 23 casts a shadow on about half the area of each photosensitive element and consequently the incident light beam LS illuminates exactly the same area of the photosensitive elements 13 and 14. In fig. 5B the quadrant photo-detector of the invention is slightly misaligned relative to the light source and the incident light beam illuminates both the photosensitive elements 13 and 14 but it illuminates a greater area of the photosensitive element 14 and a lower area of the photosensitive element 13. Until the incident light beam directly illuminates both the photosensitive elements 13 and 14 the X position of the quadrant photodetector relative to the light source can be conventionally obtained by comparing the outputs of the two photosensitive elements 13 and 14.

In figs. 5C and 5D the misalignment of the quadrant photodetector relative to the light source increases up to a misalignment of about 90°. In the above angle range only the photosensitive element 14 is directly illuminated by the incident light beam LS while the photosensitive element 13 is completely shadowed by the opaque mask 23 together with the baffle assembly 28. In this case no quantitative information is furnished about the position of the quadrant photodetector relative to the light source but a useful information is obtained about which is the direction on the X axis the quadrant photodetector has to rotate in order to recover the alignment with the light source.

A quadrant photodetector according to the invention which employs the above defined working principle can be advantageously used in sun tracking systems following a method for sun tracking according to the invention which will be now explained with reference to the X axis but the same applies to the Y axis.

A sun tracking systems comprises a support member where a quadrant photodetector is firmly fastened, a processing unit for processing data from the quadrant photodetector and controlling a drive mechanism comprising a first actuator for horizontal sweeping (azimuth) and a second actuator for vertical sweeping (elevation). The processing unit controls the first and second actuators according to X and Y position values calculated based upon the outputs of the quadrant photodetector. The quadrant photodetector comprises at least one photosensitive element per quadrant placed on a flat surface 12 and an opaque mask 23 arranged parallel to the flat surface 12 spaced a predetermined distance away from it, said opaque mask being shaped and sized so that, in a top view, it extends from the center of said quadrant photodetector up to occlude a predetermined same percentage of each photosensitive area. In the above sun tracking system when the quadrant photodetector is aligned with the position of the sun the opaque mask 23 casts a shadow on a same percentage of the photosensitive areas of each quadrant and the first and second actuators are not operated. With reference to the X position detection shown in figs. 5A to 5D , when the quadrant photodetector is misaligned of a relative little entity, fig. 5B, such that both photosensors 13 and 14 are directly illuminated by the incident light beam LS the processing unit calculates the X position based upon the outputs of both photosensors 13 and 14 and controls the first and/or second actuators in order to align the position of the quadrant photodetector relative to the sun. When the quadrant photodetector is misaligned of a greater entity, figs. 5C and 5D, such that only one photosensor 14 is directly illuminated by the incident light beam the processing unit, taking into account which is the photosensor/s illuminated, controls the first and/or second actuators for sweeping the support member of the quadrant photodetector until the photosensor 13 becomes directly illuminated by the incident light beam. More generally speaking, the above . means that when said processing unit is not able to perform the calculation of said X and Y position because of a too great misalignment of said quadrant photodetector relative to the sun, the processing unit itself uses output information from the illuminated photosensitive areas for determining which is the direction where said first and/or second actuators have to be operated in order to realign said quadrant photodetector, and it consequently operates said first and/or second actuators until it is able again to perform the calculation of said X and Y positions. When the misalignment has been reduced to an amount where X and Y position calculation is possible, such calculation is performed according to conventional procedures and the first and/or second actuators are controlled accordingly.

The possibility to calculate with high precision the X and Y positions in the solid angle of ± 2°, allow the processing unit to align the sun tracking system along any direction laying in this solid angle. This feature permits to align the sun tracker not only along the photodetector symmetric axis, the one having both X = 0 and Y = 0, but along any axis laying between the solid angle of ± 2°. In fact the best tracking alignment is the one which maximize the power generated by the PV system which, in general, can differ from the photodetector symmetric axis. The processing unit can calculate the maximum power point direction, monitoring the power produced by the PV system during a pseudorandom path run around the old best direction, which, in the first tracking system initialization, will be the photodetector symmetric axis. Accordingly, the method of the invention provides the step of, once the photodetector' s symmetry axis has been aligned to the direction of the sun, monitoring the power produced by an associated PV system while running a predetermined pseudorandom path (by operating the first and second actuators) around the photodetector' s symmetry axis.

This method deals with an hybrid control system, in order to offer the benefits of both the geo-referenced and closed loop tracking strategies. Basically the sun tracking algorithm is based on the ephemeris calculation, but each time the tracker alignment is out of the acceptable tolerance, the four quadrant photodetector gives the right signal in order to correct the future trajectory.

Thanks to this hybrid method the tracking system can correct all initial or future misalignments, allowing a faster and lighter installation, without suffering the peculiar deficiencies of the closed loop tracking method.

Furthermore, during the cloudy days the system can manage the possibility of choosing among different tracking strategies that is to follow the sky point in which the photodetector measures the maximum light intensity, to follow barely the ephemeris algorithm or to choose a compromise tracking strategy which maximises the energy produced and minimises the energy used by the actuators. A great tracking efficiency is obtained by using a simple and low cost photodetector.

The advantages and characteristics of the present invention are still valid also in different embodiments of the photodetector and in different embodiments of the method for sun tracking above described.

In fact, the above disclosure of specific embodiments is useful to understand the field and scope of the present invention, so that people skilled in the art is able to put it into practice by modifying and adapting the above disclosed embodiments; such modifications will be then considered as equivalent of the disclosed embodiments. The numbers and terminology used are to be intended merely for disclosure and for comprehending the scope of the invention and then they do not limit the invention.

Claims

Photodetector comprising at least two photosensitive areas on a flat surface, each photosensitive area comprising at least one photosensitive element adapted to produce a signal in response to a light beam incident thereon with said signals being useful to calculate a direction of said incident light beam, and comprising an opaque mask for casting a shadow on said photosensitive elements characterized in that said opaque mask is arranged parallel to said flat surface spaced a predetermined distance away from it, said opaque mask being shaped and sized so that, in a top view, it occludes a predetermined same percentage of each photosensitive area.

Photodetector according to claim 1 characterized in that it is a quadrant photodetector comprising four photosensitive areas arranged in quadrature , each photosensitive area comprising at least one photosensitive element adapted to produce a signal in response to a light beam incident thereon with said signals being read in pairs to calculate a X and Y directions of said incident light beam, said opaque mask extending from the center of said quadrant photodetector up to occlude a predetermined same percentage of each photosensitive area.

Photodetector according to the previous claim characterized in that said opaque mask is a thin member having a circular shape with a diameter which is substantially equal to the distance between the center points of two opposite photosensitive areas. Photodetector according to claim 2 or 3 characterized in that it comprises at least an opaque baffle assembly extending from said flat surface to said opaque mask, said baffle dividing said flat surface in four identical regions each comprising one of said photosensitive areas centred therein.

5. Photodetector according to any preceding claim characterized in that each photosensitive area comprises one photosensitive element.

6. Photodetector according to the previous claim characterized in that each photosensitive area is composed of a single PIN photodiode.

7. Photodetector according to any preceding claim characterized in that said opaque mask is supported by a central stem extending from said flat surface.

8. Method for aligning the orientation (azimuth and elevation) of a solar tracking system with the direction of the light coming from the sun, said tracking system comprising a support member where a quadrant photodetector is firmly fastened, a processing unit for processing data from the quadrant photodetector and controlling a drive mechanism comprising a first actuator for horizontal sweeping (azimuth) and a second actuator for vertical sweeping (elevation), the processing unit controlling said first and second actuators according to X and Y position values calculated based upon the outputs of said quadrant photodetector, said quadrant photodetector comprising at least one photosensitive element per quadrant placed on a flat surface and an opaque mask arranged parallel to the flat surface spaced a predetermined distance away from it, said opaque mask being shaped and sized so that, in a top view, it extends from the center of said quadrant photodetector up to occlude a predetermined same percentage of each photosensitive area, said X and Y position values being each calculated by said processing unit according to output values from photosensitive elements of at least two different quadrants characterized in that when said processing unit is not able to perform the calculation of said X and Y position because of a too great misalignment of said quadrant photodetector relative to the sun, it uses output information from the illuminated photosensitive areas for determining which is the direction where said first and/or second actuators have to be operated in order to realign said quadrant photodetector, and it consequently operates said first and/or second actuators until it is able again to perform the calculation of said X and Y positions.

9. Method according to the previous claim wherein once the photodetector' s simmetry axis has been aligned to the sun, the power produced by an associated PV system is monitored while a predetermined pseudorandom path around the photodetector' s symmetry axis is run by operating said first and second actuators, in order to find the X and Y values which maximize the energy produced by said PV system.