Talk:Net energy gain

Photovoltaics
Has anyone done the calculation to see if there is an actual power gain over the life time of a photovoltaic cell,when applied against the whole installation including batteries, control electronics etc.Confusedenviron 08:09, 5 April 2006 (UTC)
 * Yes. Several somebodies have done such calculations over the last decade or two and have always come up with the same answer: Yes, there is a positive NEG for photovoltaics. The typical range is currently (2008) two to three years, but this utterly depends on location and cell orientation (e.g. on-house facade plants are worse than rooftop inclined panels). For a recent study see: Fthenakis V., Alsema E. Photovoltaics Energy payback times, Greenhouse Gas Emissions and External Costs: 2004 – early 2005 Status. Progress in Photovoltaics: Research and Applications, 2006, Vol. 14, p. 275-280 http://www.clca.columbia.edu/papers/Photovoltaic_Energy_Payback_Times.pdf. For some reason it has almost become an urban myth that PV has negative NEG. This misinformation seems very long-lived, and I imagine it must be welcome news for some people, otherwise it would die out. At most, I could imagine (but have never seen information to back this up) that in the very beginning of solar cells, i.e. space exploration in 1958, the cells were fabricated in such small numbers and without regard for production economy, that negative could have NEG resulted. For current cells and terrestrial use in well-designed power plants this is utterly not true.--83.77.173.240 (talk) 18:56, 30 May 2008 (UTC)


 * Yes, many times. NREL has a discussion of energy payback on their website. Geoffrey.landis (talk) 18:53, 28 August 2008 (UTC)


 * Net energy gain is a factor without any unit. It is a multiplier for the invested energy that can be used at the end. Any timeframe such as 2-3 years is not a net energy gain, but the energy payback time. This parameter is not very precise, as it does not tell you how long the whole installation will provide energy after the payback time has been reached. There is a difference if you have a payback time of 4 years and two different generators: one with an operating time of 5 year in average and another one with a lifetime of 50 years. The best concept I have seen so far in energy assessment is Bruce Hannon's energy discounting = energetic net present value. --Gunnar (talk) 16:40, 21 December 2018 (UTC)

I've moved the discussion around new types of photovoltaic cells and monetary cost to the discussion page, as they don't seem to belong in the article.--Mephistopheles 23:32, 2 February 2006 (UTC)

Non-renewable fuels
I find this article extremely confusing. Can anyone clarify why non-renewable fuels always have a NEG below 1? Why does taking just the extraction cost raise the NEG above 1?
 * If you count the energy contained in non-renewable fuels, its NEG can't be above 1--you can't extract more than there is in it. (There is some debate, whether nuclear energy is renewable, since possible supplies are so big.) If you don't count it, its NEG can be well above 1, and an energy source will probably not be economically viable, if its NEG isn't significantly higher than 1.
 * Of course, that's a silly way of defining net energy gain, and not the way that it is ever defined in the real world. Taken to the limit, using that definition says that every energy source has a net energy gain less than 1, since energy is conservedGeoffrey.landis (talk) 18:53, 28 August 2008 (UTC)

Table of net energy gains
Also, it would be good to have a table comparing NEGs of various energy sources.

The payback table seems to be the inverse of NEG: the lower the payback value, I would assume the higher the NEG.

Given the enormous energy costs to locate, mine, refine, process, transport, store, guard, and dispose of uranium, the energy costs to build and decommission nuclear power plants, and the extremely low total efficiency rates of modern fission technology (which is used to heat water to create steam to produce electricity to deliver over vast distances), it's extremely surprising to see coal and nuclear having equivalent payback periods in the German study cited. I would imagine co-generation plants using coal have a much lower payback period than nuclear power plants. Is there a more recent study?


 * I don't know of one, but the energy density of U is fantastically high compared to that of coal, so this isn't that far-fetched Megalophias (talk) 22:49, 29 November 2010 (UTC)

Are transmission and delivery costs included in NEG?
 * I don't know, but probably not, since the distance would have to be estimated

Solar constant
"Given that the solar constant (which assumes rays perpendicular to the surface) is 1367 W / m2, one should probably scale up the payback time by a factor of 5 or so for real-life photovoltaic systems."

This looks wrong not only because the payback time ought to be scaled down but because a major part of those 1367 W/m2 are radiated in the low infrared that can, due to fundamentel laws of physics, not be efficiently used for electricity generation. This might be entirely different at locations where temperature is very low, such as for a satellite in (outer) space.

Does anyone has a firmer and more detailed knowlege about this? 84.160.246.75 10:32, 25 September 2005 (UTC)

From PNA/Physics

 * Net energy gain A lot of dubious statements and confusion, I don't even know where to start with this one. --Sinus 20:41, 10 November 2005 (UTC)

Disputed
"As an example, during the 1920s 50 barrels of crude oil were extracted for every barrel of crude used in the extraction and refining process. Today only 5 barrels are harvested for every barrel used. ..."

Is there a source for this? Could this be explained by a slower production rate, use of horses rather than modern, less energy efficient (but faster) equipment? Is there a source for the other example as well? The solar example doesn't even seem to be an example, but an only partially relevant anecdote.--Mephistopheles 22:14, 2 February 2006 (UTC)

"...but no-one knows by how much as the effective productive life of a PV cell is not known." Incorrect, the average life of a PV cell can and is estimated with high accuracy.
 * If you have better information, please include it on the main page. If the life of a PV cell can be accuratly estimated, state so.
 * Solar array lifetimes are typically 35 years and longer-- there is a large body of work on subjecting solar arrays to accelerated life testing, and 35 years represents about the longest warranty you get on modules. 20 to 30 year warranties are a bit more common in the industry; but in fact a module doesn't explode the day its warranty runs out.  Modern glass encapsulation technologies are very good; there's basically nothing exposed to wear out.  Antique modules from the mid-70s, before the era of glass encapsulation bonded with  EVA-- look pretty bad- but they're still working; and modules from the late 70s and early 80s still look pretty good.  When you inspect them-- and I have-- the EVA encapsulant on the older modules has yellowed a bit from UV exposure, and they're down a few percent in power.  UV darkening of the encapsulant, though, is an issue that people have worked on a lot in the last ten years or so, and in the newer modules you'll see a lot less power degradation.

"if the energy content of non-renewables is taken into account, they will always have a NEG below one; If only the extraction energy is counted, it can be higher." Also incorrect. From a physical standpoint, there is no difference between renewable and non-renewable fuels and sources of energy. If fuel is taken from a forest or a coal mine doesn't matter, both have a conversion efficiency less than 1. Same goes for solar and wind, a PV cell cannot convert all incident light and a wind turbine cannot convert all incoming wind.
 * Certainly, you can't extract more energy, than there is. However, in renewable energy, sources that are "virtually inexhaustible" such as the sunlight or radiation heat from the earth core are not counted. That's the whole point calling the one renewable, the other fossile. Therefore, renewables can have a NEG of higher than one.
 * As do coal and oil. Net energy gain is the (useful) energy you get out of the system minus the energy you have to put into it. What you described was the quota Eout/Ein and is >1 for all sources of energy.--Sinus 22:25, 27 October 2005 (UTC)

Furthermore, the comparison between different technologies looks weird, more sources is needed. --Sinus 10:32, 19 October 2005 (UTC)
 * please add sources or a stub tag, but the disputed tag is inappropriate and should be removed.--J heisenberg 18:16, 19 October 2005 (UTC)
 * I could have used dubious tags instead, but since that would have made 3-4 of them, I used the disputed tag.--Sinus 12:18, 24 October 2005 (UTC)

In Eroi http://en.wikipedia.org/wiki/EROEI the number is different: "For example, when oil was originally discovered, it took on average one barrel of oil to find, extract, and process about 100 barrels of oil. That ratio has declined steadily over the last century to about three barrels gained for one barrel used up in the U.S. (and about ten for one in Saudi Arabia). Currently (2006) the EROEI of wind energy in North America and Europe is about 20:1 which has driven its adoption." —Preceding unsigned comment added by 76.200.102.147 (talk) 02:01, 13 May 2008 (UTC)

Highly inacurate table
A first glance would lead me to believe that the table of "Type of Power Plant vs. Payback" is either highly misinterpreted or completely inacurate. Based solely on economic payback, and considering nuclear vs. thermal generating stations, it has been established for some time that the capitol costs of nuclear stations are exceptionally high in relation.


 * The table really doesn't do anything for the article - it describes "how quickly you get your energy back" for each of the technologies, which would be net energy gain over time. Since the two relevant pieces of data are missing (energy consumed and energy produced) the table imparts no information relating to this article. --Mephistopheles 20:38, 2 February 2006 (UTC)

My only possible interpretation of this table with respect to the nuclear payback is that 7 months continuous capacity might pay off the first fuel load installed in the plant at market prices. Indeed, the construction of these plants is most commonly financed for their lifespan of 40+ years, meaning that they might never be paid off.

A quick analysis of Nuclear power yields us a capitol cost per kW/h for an AECL CANDU model 6 double reactor of $2972/kWnet (all prices in CAD) for a gross output of 1456MW (1346MWnet). Assuming current market prices at approximately $0.055/kWh, such a reactor station would yield $74030/h and $648.5million per year. We can extrapolate this to a payback period of 6.7 years or 81 months (assuming 30 day months). Even this analysis makes some generous simplifying assumptions as such a reactor would cost approx $136.9million in O&M (at 90% capacity factor) and financing costs for a private generating station can be exorbitant. In addition, as mentioned above, nuclear stations require refurbishment after some years of operation, which in Canada at least, has proven to be more expensive than building new stand alone thermal stations. Fuel costs haven't been included because they run about 10% of the O&M costs (with fuel disposal costs around 7% of O&M) and are therefore negligeable in a simple analysis.

All data above is from the CERI Levelised Unit Electricity Cost report of August 2004.

Again, I would hope that the existing table can be either corrected or properly explained to improve its accuracy.

A similar analysis of thermal (pulverized coal fuel) generating station yilds a payback period of approximately 40 months, with thermal (natural gas fired) yileding approximately 18 months, again, O&M and fuel costs not included in the analysis.
 * I think there might be a misunderstanding in concepts here. We're dealing with energy payback, and not monetary payback. Energy payback is the measure of energy invested in extracting the ore, making the steel, the mortar, the cooling towers, building the plant, and then considering the energy gained from the whole deal, it has a rate of return of energy, you could say energy-cash. The fact that it also requires expensive workers, expensive safety measures, that's relatively irrelevant, what is meant here is the energy payback not the monetary payback. As energy is cheap these days, it may not be the major monetary cost factor in construction, but instead the lawyers, regulators, engineers coming up with the plant, the designers, human resource people, health monitoring systems, all these intangible costs to get a plant going could be the real monetary expense. One thing is certain: there is no energy source that can be profitable in a monetary dollar-sense if it's not profitable at least in the energy-cash sense.
 * A windmill that has a long energy-payback time, but because it does not require too many on-site skilled workers to operate, except a maintenance crew that's common for both windmills and nuclear plants, you could say the financial profitability of such a windmill is better than the financial profitability of a nuclear plant, even if the energy profitability is the opposite, a nuclear plant's rate of energy return being higher, but financial profitability lower, because of the necessary management, safety and human labor cost. Perhaps you'd like to make a corresponding financial-payback table to contrast with an energy-payback table. It all depends on how expensive energy really gets - if it's really expensive, then all the human-sweat costs are negligible, but if it's really cheap, then only the human-sweat cost matters. But energy is a must, it's simply a must to function at all, and its monetary price will go as high as necessary, so the monetary price discussion is almost moot point if you don't have the energy-cash profitablity straight in the first place. If it takes more energy to make something than the energy it creates over its lifetime, it can never be profitable, energy-cash-wise of dollar-cash-wise. (Having some kilojoule-dollar pegged at the cost of energy instead of the cost of gold could be an interesting financial measure, insensitive to energy price inflations.)
 * Solar panels may not even require much of a maintenance crew, because they don't have moving parts, they just sit there and churn out juice, and last for decades without much happening, but unfortunately they take a lot of energy to make. A major factor of the current financial cost of a solar panel is technology, having to pay people working in full-body clothing in zero-dust high tech expensive clean-rooms, while the monetary cost of energy obtained by burning cheap coal or natural gas or oil, might be relatively minuscule, other than the distillation-purification part that consumes quite a bit. Problem is what happens when you're fresh out of cheap fossil fuels? If it takes 7 years for a photovoltaic solar panel to provide the energy required making it, you're far from talking about monetary profitability, because even if you work for free, if the plant, lawyers, engineers, everyone works for free to make these things, they still take 7 years to collect enough energy for us to build another one of them. That's a pretty sobering thought there. If a nuclar plant returns the energy required to build the darn thing and extract the ore in under a month, now you have room to talk about how much you can afford to spend on people, what's an acceptable financial rate of return, 8 years financial payback for 0.7 months energy payback? If everyone works for free, then the financial payback could also be 0.7 months. If a solar panel takes 7 years to recover the energy vested into it, how much should it take financially, when there are no cheap fossil fuels to waste to make them? 20 years financial payback time to pay all the people too, perhaps that would be acceptable? That's another pretty sobering thought there. We're living on borrowed time, while we churn through these fossils. To provide for all our energy needs from organic-plant sources, we'd need areas over what's currently used for all agrigultural production. And by the way world population is growing, and we'll need more areas dedicated to food growing too. I'd rather have the administration focus its attention on this topic to solve the energy problems, instead of just going over to take the conventional energy away from people that still have some, because when you churn through even their stuff, and there is no more to take from anyone, what happens then? Ok, so we still got 300 years worth of coal that's convertible to gasoline at $35/barrel longterm, so it's all good, except for the greenhouse gas problem. Nuclear plants have no greenhouse gas issues, and properly reprocessed fuel used in breeder reactors instead of the conventional ones would generate a lot less radioactive waste too, per unit energy produced.Sillybilly 02:07, 17 December 2005 (UTC)

Looking around the internet, I realize that it seems hard to find reliable information on the energy payback for solar cells. Here's another source, that conflicts with the source and calculations mentioned in this Wiki article: Pearce and Lau, 2002  It suggests that the energy payoff, in real-world conditions, is 1-5 years across the US. I think these alternate calculations should also be mentioned in this Wiki, to represent the entire spectrum of known possibilities.69.37.138.145 03:12, 16 January 2006 (UTC)

The actual solar constant is is neither 1000 nor 2200 W/m², but 1367 W/m². This is the unhindered radiation, with rays perpendicular to the surface, which would be continuously available to a solar power satellite system never in the Earth's shadow, but down on this planet's surface, because of the day-night cycles, clouds, precipitation, atmospheric reflection and haze, the overall average number is less. In the US, for instance, instead of 1367 W/m², the overall average is 150 to 375 W/m²[], depending on location (Note: the panels are still perpendicular to the solar rays because they have a latitude tilt. Also, this is the radiated solar dose, and not the final day-night-sunny-cloudy-average power output from the solar panel, which is 15% of these numbers, or 15-60 W/m² average (of course, peak electric output during a bright sunny day can be even as high as 150-200 W/m², but close to 0 during a winter night.)) Given these facts, one should scale up the payback time by a factor of 5 or so for real-life photovoltaic systems, compared to what's given in the above table.

The financial payback can be different than energy payback, because, for instance one can use cheap coal with high energy content to create silicon solar panels that generate electricity. Electricity commands a higher price per kilojoule than coal does. But if one would have to use the electricity provided by the solar panels to generate new solar panels, the above energy-payback scenario would stand for financial-payback too. Also note that coal is only 35% efficiently convertible to electricity in current power plants, at least until fuel cell technology can bring this number into the 45-85% range, because fuel cells are not limited by heat engine laws.

A newer study cites energy-payback times on the order of 2 to 4 years for most silicon based solar technologies, with correct, real life (US Detroit) irradiation doses. As about half the energy cost of solar panels is the energy embodied in the silicon raw material, the other half required for manufacturing operations, there are ways to significantly affect these numbers either by using less silicon, or efficiency and energy conservation improvements in manufacturing operations. For instance, using less silicon by micromachining silicon into ultrathin transparent 50 micron layers allows producing 140 W solar panels from 2 silicon wafer's worth of material, as opposed to 60 needed for conventionally constructed panels.

An alternative strategy
Wouldn't it be better to tackle this subject from the standpoint of macro-energy/economics, in other words from the standpoint of physics? Given that there are no accounting mechanisms in place to keep track of, say, BTU costs accrued in all of the processes that lead up to extraction and distribution of energy sources it seems ridiculous to try to infer claims about net gain (or loss for that matter).

I would suggest an article on Net Free Energy would be useful to guide thinking about what we should be counting as income and costs in order to determine if we actually have free energy left over to run the rest of the economy. It would act to establish principles of macroenergetics that would then establish a framework for determining microenergetics in specific energy sectors such as oil or coal.

The bottom line here is that it takes some capital cost BTUs to get an energy flow up and running, then it takes on-going expense BTUs to keep the energy flowing. You deliver net BTUs to the point of use and those are counted as free energy (in a general sense), that is, energy available to do useful work. This mirrors the methods of cash-flow accounting. And so a framework based on this approach might actually be useful. With a little more thought we might even see that an accrual-based accounting system would be appropriate.

Rather than get hung up on specific cases for which there will, at this juncture in time, be disputes over data and interpretation. I think we should produce a fairly straightforward model of net free energy and not try to prove that we may have a problem or that alternative energy sources are better, or any of that. —Preceding unsigned comment added by Gmobus (talk • contribs)

NEG-ratio
Is there any such thing as an NEG-ratio? Shouldn't it just be energy ratio?, a search on google showed the only place on the net with the term NEG-ratio was wikipedia, kind of suspicious eh? If NEG-ratio exists what is the formula? It should be stated on the page, or referenced.sbandrews 17:43, 23 September 2006 (UTC)

Wind
My understanding is that large modern wind generator systems pay back the energy used to build them in about 3 months, best among general-purpose expandable renewable sources -- seems like they deserve special mention.-69.87.199.117 13:03, 18 June 2007 (UTC)

http://www.sciencedaily.com/releases/2007/08/070808211555.htm —Preceding unsigned comment added by 192.250.34.161 (talk) 21:15, 20 September 2007 (UTC)
 * The newest commercial-scale wind turbines have an excellent energy payback, but as with any form of electrical power generation technology, much of the energy consumed in their construction is in the form of petroleum, a form of energy which happens to be getting particularly scarce, expensive, and insecure. The basic problem is that all forms of electric power generation require transportation in their manufacture and construction, but at the moment we lack practical large-scale methods to use electricity to power transportation. Transportation remains overwhelmingly dependent on liquid hydrocarbon fuels from petroleum (with a small but growing fraction from biofuels). Wind power, hydropower, nuclear power, and just about every other electric power generating technology remain net consumers of petroleum. There are some schemes to address this problem, including the Pickens Plan, the hydrogen economy, plug-in hybrids, electric vehicles, and telecommuting. Large-scale implementation of one or more of these schemes might allow electric power generation technology to actually pay back its petroleum investment. Another way to look at the problem is to ask whether we can use wind power to build more wind power (or nuclear power to build more nuclear power, etc.). A form of energy which consumes a different and unsustainable form of energy is itself not yet sustainable. This is not an objection to wind power by any means, but rather it points out that until we electrify our economies in general and transportation in particular, "net energy gain" analysis may be a rather crude and potentially misleading first-order analysis that ignores the relative scarcity of limiting resources such as petroleum. As an extreme analogy, imagine burning paper money to stay warm in winter - considering only energy would overlook some other important aspects of that exchange. Not all forms of energy are equal. Energy is a way to equate different forms of energy, but only certain kinds of energy conversions are practical due to thermodynamic and engineering constraints. --Teratornis (talk) 18:29, 21 September 2008 (UTC)
 * When we refer to a process as producing a gain, we imply that the output is in some way equivalent to the input. If we cannot actually substitute the output for the input, then our notion of "gain" is somewhat hypothetical. --Teratornis (talk) 18:43, 21 September 2008 (UTC)

Propose deleting a misleading line
I propose deleting this line: From a theoretical perspective, if the energy content of non-sustainables is taken into account, they will always have a NEG-ratio below one;  and also deleting the several sentences later which attempt to clarify and explain this.

While this is technically correct if you use the terms as defined by whoever wrote this, it is misleading enough (as noted by the discussions above) that I think including it in the article is counterproductive, and I don't think that the "theoretical" definition here is actually representative of the way the term is actually used. Geoffrey.landis (talk) 19:25, 28 August 2008 (UTC)

--In the absence of objections, I made the change. Geoffrey.landis (talk) 14:07, 29 August 2008 (UTC)

External links modified (February 2018)
Hello fellow Wikipedians,

I have just modified one external link on Net energy gain. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20081121090310/http://www.greenworldsolutionsltd.com/biofuelvsfossilfuel.html to http://www.greenworldsolutionsltd.com/biofuelvsfossilfuel.html

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 08:33, 16 February 2018 (UTC)