Ermanno has been working on the board and designed it such, that the sensor may either be an imaging sensor by AZUR SPACE solar power or it may consist of 4 LEDs as photocells in a grid with 10.92mm spacing. Now I had to come up with a shading beam arrangement that fits these sensors.
This is my solution. 3mm LEDs will be set into holes within the front plate with a central hole for a countersunk screw. A 15x15mm aluminum beam will be screwed on top as a shading beam. Please see the sectional drawing through the top surface of the aluminum plate and the isometric side view below.
The shading beam is secured to the mounting plate of the controller with a countersunk screw. The LEDs are ground off such that they are still protruding from the plate by about 1mm. This arrangement will ensure that water and dirt can pass over the LEDs and dirt cannot build up. Only a leaf or a bird dropping will disable the sensor. I will need to make sure, that such an event is reported to field control over RS485.
Here is an earlier version of the shading beam sensor on the desktop tracker I used for "living room table software development". It has a 3d-printed shading beam. That remains an option too, but I prefer a simple, sturdy metal part for extended outdoor application.
The advantage of this shading beam sensor will be its availability and simplicity. On the other hand, the AZUR SPACE sun sensor will be more sensitive and much easier to clean. It is basically a 4 pixel camera and I am describing it in detail here:
In particular, it will be beneficial in a tracker field where cleaning trucks with rotating brushes run along the CPV sun trackers to crean the modules. The shading beam may get damaged but the flat front glass of the "4 pixel camera" will just be cleaned. Have a look here, to see what I mean:
In any case, Ermanno found a PCB layout which accomodates both versions of the tracking sensor, so you can choose. We are rapidly approaching our final design.