The following post has been submitted by Michael Lammert. It is a follow-up to this article that was published here yesterday.
Correction to an Assumption in Engineering Challenges for Developing the
SunCell’s Concentrator Photovoltaic Cells and Geodesic Dome Array
Michael Lammert (AKA Dr. Mike)
One of the assumptions that was made for the development of the Masimo CPV cells for the geodesic array Brilliant Light Power’s SunCell was pointed out to be incorrect by e-catworld reader “tip” in that large area CPV cells will not be used to form the geodesic dome array. Masimo plans on building triangular array modules of series connected small cells, then connecting the triangular modules to form the geodesic dome, instead of developing large area, triangular CPV cells. Masimo’s development plan can be found at: http://disq.us/url?url=http%3A%2F%2Fwww.brilliantlightpower.com%2Fwp-content%2Fuploads%2Fpresentations%2FMasimo-BrLPIndustryDay-102616.pdf%3ADVcN4biGAI-vuJwSlyEYgys0GHc&cuid=2168707
Masimo’s plan consists of:
- Demonstrate the array module called TRU (Triangle Receiver Unit) using 209 single junction cells with dimensions of 0.267cm by 0.366cm.
- Increase the CPV cell size to 0.389cm by 0.564cm (0.219cm2) and build single junction InAlGaAs on InP substrate cells to form modules that will output 15.5KW when configured in the geodesic dome array.
- At this same cell size build double junction InAlGaAs on InP substrate cells to form modules that will output 28.5KW when configured in the geodesic dome array.
- No plans are presented for the bifacial growth triple junction cells.
Without increasing the size of the CPV cells there should be no issues with fabricating the individual CPV cells, including no need to increase the metal thickness. By series connecting a lot of small devices in a module, the output voltage of the module will be high and the current will be low (the output current of a single cell). The only issues with building the geodesic dome array are acquiring the technology to build the “TRU” modules then developing the method to connect the modules to form the geodesic dome. There may be some technology issues with the initial fabrication of the “TRU” modules, but these issues should be easy to resolve before moving to triple junction devices. It should be noted that there is considerable loss of cell area fitting small cells into triangular modules. The triangular modules have an area of 23.1cm2, whereas eighty 0.219cm2 cells have a total area of only 17.6cm2, or a ratio of 0.76 (and this doesn’t include the bus bar area loss). I would conclude that BLP will have a CPV cell array available to integrate with the SunCell system within the first half of 2017 even if it is only a single junction based array that outputs 15.5KW. There shouldn’t be much difficulty having the double junction array that will produce an output of 28.5KW available by the second half of 2017.
While the triple junction cell was not part of the presented development plan, I did some calculations to determine the expected output from triple junction CPV cells with these assumptions:
- The cell size is the same as for the prototype- 0.389cm by 0.564cm (0.219cm2)
- Although Masimo did not list the number of this size cell that would fit into their 23.1cm2 “TRU” module, I calculated 78-88 cells would fit, depending on the spacing between cells. It will be assumed that 80 cells fit into the module.
- The maximum current density for the triple layer cell is the same as the BFG3-2BB-1010-A10 cell, about 16.1A/cm2.
- Based on the data from the Masimo presentation I’m not sure if the geodesic dome will be 40 modules (a hemisphere) or 60 modules (3/4 of a sphere). I will still assume 60 modules.
The total output power can be calculated as:
Pout = 60 modules x 80 cells/module x 3V/cell x 16.1A/cm2 x 0.219 cm2 = 50,865KW
Note that he BFG3-2BB-1010-A10 cell has a spec of maximum solar concentration of 1200X or about 120W/cm2. The triple junction CPV cells will be located at a position in the SunCell where the light intensity is about 125W/cm2. Therefore, the array output may be 52-53KW, but the CPV cells may be running at a few percent above spec. Bus bar areas were not included in this calculation so the actual output power may be a little less than 50KW and the true current density may be 10-15% higher than the spec for the BFG3-2BB-1010-A10 cell. The only issue with the triple junction cell is that it requires a backside contact which means the modules will have to be modified to accommodate this contact. There shouldn’t be any major difficulties demonstrating an output power of about 50KW with triple junction CPV cells.
To increase the blackbody radiation to 3500ºK modifications will need to be done to the triple junction cells, including doubling the metal thickness and modifying the semiconductor layers to raise the maximum current density to about 30A/cm2. Assuming that the triple junction cell can be modified to handle higher current densities, the key issue will be verifying that the modified CPV cells are still reliable.
Masimo appears to have a good solution to avoid having to operate CPV cells at high currents and make high current connections in the geodesic dome array by building the geodesic domes of modules of small cells. The only drawback to modules of small area cells is loss of 20-25% of the potential CPV cell area. There should not be any major issues with the development plans for the CPV cells and arrays up to the point of increasing the operating temperature to 3500ºK and thereby about doubling the current density in the triple junction cells. It may not be possible to improve the existing triple junction CPV cell sufficiently to operate it reliably at a factor of 2X higher current density. The CPV cell geodesic dome array should not be a limiting factor in demonstrating a SunCell prototype.