First Solar is making PV panels for less than China’s biggest producer | Hacker News Reader

First Solar is making PV panels for less than China’s biggest producer

bloomberg.com

139 pointsOsiris3028 comments

Cost of solar panel is dropping every year: https://en.wikipedia.org/wiki/Swanson%27s_law

The question is not whether you can get cheaper once, but whether you can be cheaper and maintain that advantage by improving each year.

Strange parallels to Bitcoin mining where you're worried your rig might not pay for itself before faster mining hardware comes out that pays better.

With exponentially increasing efficiency, when is the ideal to invest? This is solved by a "wait calculation"[0].

[0] https://en.wikipedia.org/wiki/Wait_Calculation

Swanson's Law is much more realistic than the simplistic linear extrapolation from the linked article which shows negative prices by 2030.

The response of the Chinese company Trina spokesperson is quite off the mark. First Solar expects to ship around 3GW of modules in 2016 so that is not exactly lab scale ;) FSLR expects to improve their shipped modules efficiency by more than 10% over last year. 15.4 to 17.

FSLR also will also introduce their new larger modules that are easier to install. Reducing the cost of building a power plant.

I really regret not investing in FSLR stock when I could have :(

Off course when looking at energy supplies the physical efficiency is not important, one must look at the economic efficiency. Numbers that are hard to find :(

Yup, I invested in them a year or so ago, unfortunately I also invested in SunEdison at the same time. Took a hit to get out of that position. Happy with my portfolio now though: Accel Group, First Solar, Tesla, Vestas. They look good for the long run.

FSLR was at 140 in 2011, and now they're at 60.36. Perhaps it's not too late to buy?

No longer have the spare cash to play on the stock market :(

It's a funny kind of regret to wish you had something that is worth half as much as it used to be worth.

Stock was just below 12 at one point in the dumping Solyndra failure time range.

Exactly although for me it would have been just after IPO.

Part of the reason they survived to now is likely due to all the anti-dumping laws the US imposed. http://www.bloomberg.com/news/articles/2015-07-09/u-s-impose...

Interesting: CdTe has a bandgap of 1.5eV, closer to the ideal for solar efficiency than Si, which has a bandgap of 1.1eV. Multijunction cells can achieve efficiencies of over 80% since each junction can address a different part of the solar spectrum. However, they're not cost-effective yet.

1. Excellent

2. Significant costs now are building plants. Next cost dropping wave will be with robotic solar plant building (Also liked that rollable solar panel for the military)

I used to imagine the future to have self assembling robots that build their own power sources,etc.

Its good to know we are nearing that future.

Another step to self-sufficient AI I guess. At some point along the way we will have to stop and think about what we're doing there.

Have we ever done that? Maybe the AI will remember their grandparents if they don't commit suicide from existential angst.

Nothing solves existential angst like the drive to reproduce.

Right now there are 2 different articles on Hacker News front page stating the opposite (almost).

Bloomberg article (http://www.bloomberg.com/news/articles/2016-04-14/first-sola...) states the following: "First Solar is producing panels for as little as 40 cents a watt, or about 15 percent less than China’s Trina Solar Ltd. In 2019, First Solar’s module cost could be as low as 25 cents a watt, according to analysts’ models.".

VOX article (http://www.vox.com/2016/4/18/11415510/solar-power-costs-inno...) states the following: "How cheap? Sivaram and Kann argue that the industry should set a goal of pushing the installed price of solar to 25 cents per watt by 2050 — down from around $3 per watt today."

Perhaps there's distinction between panel cost and installed cost?

dollars/cents per watt should be a standard measure. Plus, the VOX article states that "...we'd also likely need to rethink how we install solar panels. After all, the panel itself is only half the cost.".

The key difference here is a single word in the VOX article: installed solar.

The 40cents/watt in Bloomberg article is just the panel (because it's an article about costs of producing panels). This would mean even if the price of a panel dropped to $0.01/watt you'd still need $2.60 per watt to install it on your roof.

That's why the VOX article in the next sentence says

> Current approaches to cutting costs won't necessarily get us there. We may need experimental new technologies. Or novel ways of integrating solar into our walls and windows. Or robot installers.

Or we just realize that current installation costs are acceptable when your system is paid back within ~11 years, and the power is free after that.

That realization doesn't do anything to alter the calculations that the bulk of investment dollars will do.

(the point being that changes to the cost do impact those calculations)

See my reply below. According to VOX article, "installed" means 2x of money/watt value. So, according to Bloomberg article, if First Solar panel is 40 cents/watt and Trina panel is 45 (40+15%), installed should be 90 (45x2). But then again, according to VOX, 3$ per watt is far away from 90¢. In my opinion, even taking in account the cost of installation, the values expressed in the two articles are completely different.

Installation cost isn't proportional to panel cost. On the contrary, I would expect there to be little to no correlation at all.

>> “A discussion of cost in the lab is meaningless."

Agree. If it read American solar company out produces, sells, etc. that would be meaningful.

Does the US even have the raw materials & production capacity to use the research? If not, sounds like a complete waste of money.

If this is true, then why are the First Solar salesmen at my local Home Depot still pushing a $40K price tag at me?

Reply on news.ycombinator.com