Energy Return on Energy Investment - Solar PV

[Sparweb]

This question has been an irritating source of misinformation, I know, but maybe it's not dead: How much energy does a solar PV panel produce in its lifetime, compared to the energy required to fabricate and install it?

I found an interesting article today, following a link that led me to a magazine called "Energy Policy":

https://ac.els-cdn.com/S03014215163...t=1515897937_a165529c8912dce33c0ca0ea6137a12f

Following up on it, I soon found another article debunking most of the claims in the first:

https://ac.els-cdn.com/S03014215163...t=1515891817_424b9d51149ca1e275140da2d4f62e5c

Which brings me right back to what I originally believed, which was that a typical PV panel can be expected to generate 7 times the energy in its lifetime than it took to make it in the first place. Even in a region of moderate sunlight like Switzerland, the home of the article authors and source of much of their PV usage and lifetime data.

Neither of these articles seem to be peer-reviewed. The second article has authors whose work was cited in the first, so this may have become "personal" to some of them. I don't know much about energy policy myself, and never questioned the received wisdom until now. I thought I'd just share these with the forum to see if anyone still thinks, despite the rebuttal, that there's still some merit in the first paper's point of view, if not their methodology. I don't have any more facts to bring to it, myself.

STF

 

[GregLocock]

I've had a modern PV installation on my new house for almost exactly a year. The performance has been in line with the modeller I used, but I don't know if this year has been typical for insolation.

Anyway, it's a 5kW system with E and W facing panels, so it does not produce the maximum energy, but it produces more energy in the morning and evening, maximising the amount of 'free' electricity I use.

The economics are totally distorted by various government policies, but basicallY I spent $7000 and effectively make a slight net profit each year from my grid connection. That's a crude payback of around 7 years. The nice thing is that it is one of the few investments where I don't pay tax on the profits.

So the number you want to know is how much energy does a 5 kW system generate in a year? 5963 kWh

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[Sparweb]

So if I estimate the rating was based on 800W/m^2, and 14% efficient, then you would have installed about 45 m^2 of panels.
The 5963 kWh your panels generated means your production is 134 kWh/m^2/year.
That's better than the rate the paper's authors were using, 120 kWh/m^2/year, but they are in Switzerland and getting better output in Australia doesn't surprise me.

Any comments on the energy invested in the manufacture of panels, or the figures used by the papers' authors?

STF

 

[IRstuff]

Seems like a no brainer.

6000 kWh/yr * 20 yr * $0.16/kWh = $19200, which was about the retail price of my panels, which included the installation cost, which was about 1/2. That puts the produced electricity at least 16x the production electricity, assuming 100% markup and production electricity at 25% of production cost.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[moltenmetal]

If a panel gives an economic payback shorter than infinity, it is a CERTAINTY that it will give a substantial net energetic payback over its lifetime. It cannot be otherwise if you think about it- even if all the energy used to make the panel was cheap heat and no electricity (and it isn't), the embodied energy cost would dominate the price of the finished panel and make it unaffordable if it came anywhere close to the amount of electricity the panel could generate over its lifetime when installed.

As to accurate figures of the embodied energy of solar panels associated with manufacture, I'm sure it has been studied to death, and no doubt it has been studied by both competent teams such as the ANL GREET modellers, and by countless incompetents as well. That has definitely been the case with vehicle and fuels lifecycle estimates, particularly in relation to EV batteries where I have read the studies and compared them against one another. I'm also sure that the embodied energy/emissions of panels is steadily improving, or else panel prices wouldn't keep dropping they way they have.

 

[GregLocock]

I'll get the system specs when I'm back at that house, but there is no way it's 45 m^2, I think the nominal efficiency was 18%.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[IRstuff]

My solar system is 18 panels rated at 275 W/panel. Based on 65.5 in x 39 in spec area --> 29.7 m^2

I just checked my numbers, mine was installed in April of last year, so I only have about 9 months of production at 6440 kWh; latitude 33.906081N

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[Sparweb]

Yes, I do agree that PV is now cheap enough to pay back in just a few years here, too. I've run the numbers myself.
That isn't what I meant.
Maybe I wasn't clear, maybe you didn't read the papers, maybe you did but think the first one is bunk and won't engage on it, but I'll rephrase the issue I was actually trying to get at.

The authors (of the first paper) contend that the cost of solar panel manufacture is artificially low due to low material and wage costs in the countries they are made in. If those costs were accounted for in western currencies and at western wages and at western environmental standards, then solar panels would be so expensive to manufacture, one would not see a payback in the lifetime of the panel. There are no solar panels manufactured in Europe or North America, I believe, except for a few on government life-support. Now, the authors of the second paper's rebuttal is that they went on a "fishing trip" for their data, but I'm not putting my blinders back on, just yet.

This wouldn't be an issue but I'm starting to hear from government policy makers who imagine that PV could "someday" be a substantial portion of the province's/country's energy mixture. I'm still having trouble believing that. First from the enormous surface area required to do it, but also from the scale of the economics. Installing PV up to 0.1% of a country's energy already means an awful lot of panels. Imagining a country getting up to 25% power implies that other countries will be doing the same, too, which then implies that world production of PV panels will grow by maybe 100 times. So I'm trying to figure out if that's even possible.


PS
If I need to plant a flag in the ground so that you know what "side" I'm on, well I have a wind turbine and many PV panels at my place. I want to use this stuff. While it's getting cheaper all the time to use PV, is a point of inflection coming? It's just not clear to me yet that there aren't hidden costs that will grow as the PV industry grows, hidden now because they are small, but impossible to ignore once they become large, just like there are hidden costs in oil, gas and even hydro.

STF

 

[IRstuff]

"The authors (of the first paper) contend that the cost of solar panel manufacture is artificially low due to low material and wage costs in the countries they are made in. If those costs were accounted for in western currencies and at western wages and at western environmental standards, then solar panels would be so expensive to manufacture, one would not see a payback in the lifetime of the panel. "

Well, that's a bit of a silly exercise, isn't it? That could be said of pretty much everything we buy in the West. Additionally, I think that's mostly crock anyway, relative to solar cells; you can't make them with a bunch of $1/day workers in the sweat shop, so the wage argument doesn't really hold much water. Even China is looking to outsourcing, due to the skyrocketing of wages in the last 20 years or so. REC Solar's panels are manufactured in Singapore, so not exactly 3rd world anymore, these days.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[Sparweb]

That's why I put the question out.
When faced with an analysis like that, I have to admit I live in a too small industry bubble: working on airplanes surrounded by a gang of oil pumpers.
Sure, I do try to stay informed, but attempting to understand a quantitative analysis of world industrial production... I'm just not an economist.
Thanks IRStuff

STF

 

[moltenmetal]

Again, the EROEI MUST be significantly positive, or else the cost of the panel would NEVER PAY BACK- the energy inherent in its production would overwhelm the value of the energy it could generate over its lifetime.

Panels pay back their embodied energy in a comparatively short period, and pay back a MULTIPLE of their embodied emissions over their guaranteed lifetime. And that net payback increases every day as the cost of panels drops- it cannot drop unless the energy input per kW of panel nameplate capacity also dropped.

 

[Sparweb]

But what if (playing devil's advocate here) all of the minerals mined and labor hours worked to make those panels happened in North America or Europe? I'm sure the cost in dollars would be higher, but we westerners are also more energy intensive in our industrial activity. So would there be a higher net investment in energy in any given solar panel if it was made in a 1st world country? It would cost more - that's a given, but would it also have required more energy to produce?

Anyway, that turns out to be moot.
IRStuff's point is valid for most things we consume nowadays, so to single out PV panels is unfair. The answer to my question is that we westerners are living at a standard of living made artificially high by cheap labour in other countries and cheap world transport to our shores. Which is stating the obvious but I had my head down looking at the silicon cells too closely. Once I was reminded of that then the rest became clear.

STF

 

[IRstuff]

"But what if (playing devil's advocate here) all of the minerals mined and labor hours worked to make those panels happened in North America or Europe? I'm sure the cost in dollars would be higher, but we westerners are also more energy intensive in our industrial activity. So would there be a higher net investment in energy in any given solar panel if it was made in a 1st world country? It would cost more - that's a given, but would it also have required more energy to produce?"

We're energy intensive partly because of automation, and partly because we possibly have A/C in our factory sweat shops. One of the benefits of no longer being that far ahead of everyone else is that their labor and energy costs are continually climbing, and the transport costs now make it possible to be relatively competitive in overall production cost. There was a recent article mentioning that immigrant Chinese computer science students are returning to China, because the wages are sufficiently attractive to return home.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[rterickson]

Real time numbers from the California Independent System Operator CAISO: Link

AS OF 14:50 01/26/2018
24,194 MW
Current demand
7,264 MW
Current solar

That's 30% solar.

 

[GregLocock]

That's nice. On a recent hot day in Australia the renewables made very little contribution to the grid, despite being well above 10% in nameplate of the total generating capacity. I'm struggling to find the data in question. The reason is that most mainland renewables are wind, and the wind didn't blow. So, you have to have 100% redundancy, in other words if you want a 30% renewables target, you still need to build 100% despatchable power, in Australia. Or start switching customers/voters off. Or as our little greeny idealist mendicant state did, fire up the diesel powered generators.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[GregLocock]

That's funny, here's the live data dashboard today, a nice sunny day. The mendicant state (SA) has a nominal 57% renewables, as you can see, they ain't doing nothing today

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[berkshire]

Ir stuff
I am a bit south of you at latitude 33.08* I have had a nominal 3kw solar panel set on a west facing roof for 10 years. in that time they have generated 35513.0KWhours The initial cost was 19000 dollars so by my estimation depending upon what Kwhr rate you use they are about half paid for. One thing that happened around here was that the utility, SGDE, rapidly changed out the analogue meters to smart meters so they did not have to buy electricity back at retail rates this just about doubled my payback period .
B.E.

You are judged not by what you know, but by what you can do.