Really good solar-cell news

I don't think this will sell versus current PV technolgy where chips are bellow $1/watt and being driven down. The idea that you could use it to create a more efficient multi layer chip, that is not much more expensive has some viabiity. Installation is the biggest cost, and more efficient panels, may mean fewer square meters, which may drive down costs.

The other idea is to use cheap mirrors that can be built in manufacturing plant, plus efficient chips and liquid cooling to drive down system cost per kw.
High-Concentration PV Thermal System From IBM -- 80% Efficiency, Potable Water, Air Conditioning −

 

While nice to have more efficient cells, for the most part the market is being driven by $/watt. Additionally, we are at the point where balance of system is costing more than the panels themselves. Cheaper mounting systems are where I expect the important improvements to come from.

 

That sounds rather optimistic. If combined to sum the open circuit voltages, then the two junctions are being connected in series. This is great if the junctions have (nearly) matching short circuit currents (Isc). But absent more detail, there is no reason to expect the Isc's to match, so each junction will be operating at less than its potential solo efficiency. That is OK if the mismatch is small. But if it is large, then the two junctions will need to be isolated and converted separately.

Also, when summing component efficiencies, beware of double counting. The low band junction efficiency figure could be including the capture of some photons that really belong to the higher band junction. If so, then the low band junction's efficiency will drop a bit when the layers are stacked and the high band junction gets first crack at the those photons.

 

Where installation is the largest cost, the incentive for cheaper transducers seems small. Is there room for better designs of pre-fabricated modules that can be installed more rapidly? Not that all locations would accommodate the same modules.

 

Solyndra's big advantage was supposed to be lower installation cost, but clearly they were not low enough.

Solar cell efficiency a few years ago, for general purpose cells, were typically 13%-20%. The 15% efficiency noted here falls right in line, which means there is no reason to really switch manufacturing to that technology. The Chinese are appearantly driving down the cost to about 36 cents/watt in 2017, add in mark up if the market is no longer flooded and shipping, but there is not much space. Someone that has looked more recently, please corect these numbers.

They have 45% efficient chips, but they are much more expensive. But if you could add this technology to the current technology and get 30% efficient chips for $1/watt in 2017, there may be a reason to go that way. You would only need about 60% of the installation area.

Other costs in the US, are permitting, finding the right contractor, getting the right loan, etc. These costs are fixed no matter which installation goes in, but are much higher in the US than in germany. The govrnment could definitely help here.

 

Anybody know the effy of cells currently used in orbiting satellites? I assume that they are quite high. Maybe not highest, given their stressful operating environment.

 

Lightweight and higher efficiency cells should help bring mounting hardware and installation costs down too.
Look at the original wall mounts for the first 16x9 big screen TV's- the mounts alone cost several hundred dollars because the TV's were so heavy. Now equivalent size modern LED TV's weigh one quarter that of the original plasma sets and the mounts can be had for $25 vs. $200+ for the old TV's.
A 25lb solar panel will be much easier/cheaper to mount than a 75lb panel- and you'll need less of them make the equivalent current which will help home owners that have limited roof space.

 

Here are some triple-junction Spectrolab cell product lines, from 26.8 to 29.5%: Space Products: cells

Note that space cells are intended to make use of the UV spectrum blocked by the terrestrial atmosphere. Apparently some triple-junction lab units are running about 38% without concentration, over 40% with concentration. This paper mentions up to 6 layers: Raising the Efficiency Ceiling in
Multijunction Solar Cells

 

Nothing in that article made me think the weight of a panel would be significantly lighter.

 

It is my understanding that the robustness of mounting hardware is driven by UPWARD forces (i.e. wind) and LOADS (e.g. snow), lighter panels don't help with either of those things. Mounting directly to a steel roof, for example, would.

 

+1

A friends roof was destroyed last year, and it was very inexpensive when using the insurance money to put in a steel roof and solar.