Solar SCADA ™ Universal Tracker Controller

Lots of people are working on tracked solar arrays. My core competancy is not in the actual panel and their magical internals, but in the data and control of things which produce power and follow the sun. As such, I've developed a Universal Tracker Controller, which is based on a PLC running code written by me in C which includes solar position data from the NREL-Solar Position Algorithm (SPA)

This is a brief outline of my small test rig I built at Clarkson to test the accuracy, not having money to use a Trac-Stat SL1 . Turns out the unit performed admirably. Unfortunately, the Thesis intervened, and I haven't yet had a chance to substantially improve this design. For more information and features, including a video of the device operating a rickety old tracker in Phoenix, see Solar SCADA ™ .

This photo shows the internals of the 'little red box', my demo kit. In designing this, I was keeping to industry standard, ruggedized components. Of course, as a demo, I needed to add a lot of extra things, seeing as I had to be able to drive motors from 5V for my test stand to 36V and higher for field demos. The only connections necessary are a single set of wires going to the motors, a sensor cable, and an HMI (Human-Machine Interface) cable. The controller is configurable via Modbus, but also supports DNP3 or any other protocol that the controller on which the code is implemented can deal with.

This is the device that controlled the tracker depicted at . I've learned a bit since this device, and am working when I find time to improve this design.

Oh the joy of testing solar equipment in the middle of January. This shows the sensor mounted on top of a camcorder PTZ (Pan-Tilt-Zoom) head, which was the cheapest and most portable elevation / azimuth device I could find that was easily transportable. That's my Toyta Celica with 240k miles in the background, set up in a lovely part of Clarksons' TAC center on a nice snowy Christmas-break day.

Here's the same device, from a different view. As I mentioned before, this was meant to test the accuracy of my control and motor drive algorithms, and this device was set up several days while I was editing or writing the thesis sitting on the floor next to it.

Seeing as I didn't have access to a device designed to measure tracker accuracy, I set up my digital camera to take photos every few minutes, using a target and pinhole a known distance away to measure the divergence from being "on sun". This picture shows one of the hundreds of photos taken this way. With the distances used, each dotted circle represents 0.5 degrees. Therefore, so long as the dot stays within the inner-circle for the entire pass of the day, all is well. The prototype performed well, at least within the bounds measurable by my Hi-Tech Optical Tracker Accuracy Measurement System.