Speck Design is a Palo Alto base design and engineering firm which performs top-notch work. In the late 2000's they brought me in as a consultant to assist on an large scale engineering project. The client was Amonix CPV systems, now defunct, who had a novel idea for commercial scale solar energy collection. Speck Design was contacted to help design a structure that was as cost effective as possible.
​
We went designed and fabricated a single panel (60ft by 15ft) and deployed it for real-world testing. As completion of the construction and testing was the end of the consulting contract, I do not know what the final outcome was. I can surmise, however, that based on the fact that they failed as a startup, the financial goals were not met somewhere along the line.
A full array of the CPV system, 105' long by 60' tall. The array is comprised of 7 vertical panels each consisting of 12 by 3 PV panels, each panel being approximately 5ft on a side. Each panel is comprised of a 5 x 4 grid of PV elements
​
Each vertical panel is 60' by 15' comprised of 36 sub-arrays. Each sub-array is comprised of 20 Fresnel lenses for concentration of light onto the PV elements.
​​
Being a Mechanical Engineer, I was working on the structure, not the optics or electronics. However, as with most systems outside of the digital world, Mechanical Engineers were responsible for ultimate fit and function, so overall understanding of the individual components was necessary
​
The Speck Design team was brought in to cost reduce the steel structure and maintain integrity. To the left you can see the truss structure that was developed, shown as a single vertical module (to be lined up 5 in a row).
​
In the video below, you can see a test we performed with a partially constructed module. As you can imagine, a complete PV system would be subject to some teriffic wind loading. To verify the updated structure we designed, we actually ran a finite element analysis and verified the results structurally. The analysis showed ample margin in the design, and the actual failure point was well in excess of the predicted failure from the numerical analysis.