3 ms·

The current state of the art for commercialized multi-junction cells, used almost exclusively in space applications, is 30-32% cell efficiency: http://www.spec

The current state of the art for commercialized multi-junction cells, used almost exclusively in space applications, is 30-32% cell efficiency: http://www.spectrolab.com/solarcells.htm http://www.spectrolab.com/solarcells.htm http://www.azurspace.com/index.php/en/products/products-space/space-solar-cells http://www.azurspace.com/index.php/en/products/products-spac... https://solaerotech.com/products/space-solar-cells-coverglass-interconnected-cells-cic/ https://solaerotech.com/products/space-solar-cells-coverglas... Multi-junction cells similar to these have so far reached efficiencies up into the mid-40% range with optically concentrated sunlight under experimental conditions, though such high efficiencies have not been reached in mass production. Multi-junction cells represent less than 0.01% of annual PV cell production. About 95% of annual PV cell production is based on different variants of single-junction crystalline silicon, with current commercial cell efficiencies in the 16-25% range. The 30% cells for space use also sell for more than 1000x the price-per-watt of a simple 16% silicon cell. A theoretical "infinite-junction" solar cell could achieve up to 68.7% efficiency under full-power illumination from the sun on a clear day, or up to 86.8% efficiency on a clear day under optically concentrated sunlight: https://en.wikipedia.org/wiki/Solar_cell_efficiency#Thermodynamic_efficiency_limit_and_infinite-stack_limit https://en.wikipedia.org/wiki/Solar_cell_efficiency#Thermody... The theoretical limit for single-junction silicon cells is 29.4%: http://ieeexplore.ieee.org/document/6557081/ http://ieeexplore.ieee.org/document/6557081/ SunPower has reached just over 25% efficiency industrially for its best cells, which are single junction silicon using advanced cell structures. The highest conversion ever achieved experimentally for single-junction silicon is 26.6%, announced just a couple of weeks ago: http://www.sciencedirect.com/science/article/pii/S092702481730332X http://www.sciencedirect.com/science/article/pii/S0927024817... The experimenters combined two leading industrial techniques for high efficiency silicon cells (interdigitated back contact structure, like SunPower uses, and heterojunction cells using amorphous silicon, like Panasonic/Tesla use) to reach the 26.6% record.

k__ 9y ago
Cool, thanks :D What does an increase of 10% mean? That I could get 10% more Watts from the same area of panels? How much do we need? Would a houshold in Canada need much more panels than one in Mexico? Is 15% or 20% already enough to power a whole 1-family-house? Are batteries or PV panels the bigger problem here?
- philipkglass 9y ago
  A household in Canada would typically need more panels than one in Mexico. Canadian homes tend to consume more electricity and the Canadian rooftop will get less sunlight. If you are space-constrained, like on a rooftop, 10% more power from the same area of panels could be significant. It's a step function. It goes from "not enough room for desired wattage" to "enough room for desired wattage" very abruptly. Once you've reached "enough room for desired wattage," further efficiency improvements matter much less. You don't care a lot whether you have 25% or 40% leftover roof space after installing the panels. Paying for further efficiency improvement is worthwhile only if it offsets enough costs elsewhere to lower systemic costs (from using e.g. less mounting hardware, less connecting wire, less installation labor.) The higher the industry-average efficiency, the less of a constrained-space advantage there is for expensive, higher-efficiency modules. That is, SunPower had a more compelling advantage when they were selling expensive 19% panels and cheap Chinese panels were at 12%. But now that SunPower offers expensive 22% panels and cheap Chinese panels are at 15%, more households can now reach their desired wattage-per-roof targets without paying for premium-efficiency. If you just want to zero out your household's annualized average electricity consumption, because e.g. you have a net-metered grid connection, adding batteries to a PV system will increase the expected financial time-to-payback. If you are off-grid you need batteries to use solar power after dark but should make every effort to cut your non-daylight power needs before sizing the battery system. Batteries are currently a lot more expensive per watt-hour delivered than the instantaneous output of solar panels. In an intermediate case, where you don't get full net metering but are still grid connected, there are a few but not many cases where household batteries yield positive financial ROI at current prices. You could make the numbers work for a modest fraction of Hawaiian households and smaller fractions of households in Germany, California, and Australia.
- dimman 9y ago
  Just to clarify a little: If your efficiency increases by 10% you get 10% more output. With a current efficiency of 10%, increasing it 10% would mean your now at 11% efficiency. If we talk about an increase of 10 percentage points, we end up with with 20% efficiency effectively doubling the output (100% increase).

k__ 9y ago

Cool, thanks :D What does an increase of 10% mean? That I could get 10% more Watts from the same area of panels? How much do we need? Would a houshold in Canada need much more panels than one in Mexico? Is 15% or 20% already enough to power a whole 1-family-house? Are batteries or PV panels the bigger problem here?

philipkglass 9y ago
A household in Canada would typically need more panels than one in Mexico. Canadian homes tend to consume more electricity and the Canadian rooftop will get less sunlight. If you are space-constrained, like on a rooftop, 10% more power from the same area of panels could be significant. It's a step function. It goes from "not enough room for desired wattage" to "enough room for desired wattage" very abruptly. Once you've reached "enough room for desired wattage," further efficiency improvements matter much less. You don't care a lot whether you have 25% or 40% leftover roof space after installing the panels. Paying for further efficiency improvement is worthwhile only if it offsets enough costs elsewhere to lower systemic costs (from using e.g. less mounting hardware, less connecting wire, less installation labor.) The higher the industry-average efficiency, the less of a constrained-space advantage there is for expensive, higher-efficiency modules. That is, SunPower had a more compelling advantage when they were selling expensive 19% panels and cheap Chinese panels were at 12%. But now that SunPower offers expensive 22% panels and cheap Chinese panels are at 15%, more households can now reach their desired wattage-per-roof targets without paying for premium-efficiency. If you just want to zero out your household's annualized average electricity consumption, because e.g. you have a net-metered grid connection, adding batteries to a PV system will increase the expected financial time-to-payback. If you are off-grid you need batteries to use solar power after dark but should make every effort to cut your non-daylight power needs before sizing the battery system. Batteries are currently a lot more expensive per watt-hour delivered than the instantaneous output of solar panels. In an intermediate case, where you don't get full net metering but are still grid connected, there are a few but not many cases where household batteries yield positive financial ROI at current prices. You could make the numbers work for a modest fraction of Hawaiian households and smaller fractions of households in Germany, California, and Australia.
dimman 9y ago
Just to clarify a little: If your efficiency increases by 10% you get 10% more output. With a current efficiency of 10%, increasing it 10% would mean your now at 11% efficiency. If we talk about an increase of 10 percentage points, we end up with with 20% efficiency effectively doubling the output (100% increase).