Solar Production

[NickCoons]

A friend of mine and I were discussing solar panels and its efficiency.. not the efficiency in terms of how much solar power can be turned into electricity, but in terms of how much energy it takes into creating a solar panel versus how much energy it generates over its life.  He had said that it takes more energy to create a solar panel then it could ever put out over its life, which he estimated at 30 years.  After doing a little math, I found that hard to believe.

The sun puts out, but the time it hits the surface of the earth, about 1000 watts per square meter.  Let's work with a mildly efficient panel, somewhere in the 10% range.  So a one square meter panel will put out 100wh over the period of one hour, or 2.4KWh over the period of one day.  In one year, that's 876KWh per year, and then 26.28MWh over a 30-year life.

However, a solar panel doesn't operate at night, so let's cut that in half.  And then let's also say that it's only put into use during half of the day, so let's cut it in half again.  So now we can say that this one square meter solar panel will generate 6.57MWh of electricity over a 30-year period.  Someone let me know if my math is wrong.

So I have two unknowns at this point.  The first is the life-cycle.  I don't know how accurate my friend is in his 30-year statement, but I used it in my calculations.  Actually, I wasn't aware that a solar panel wore out, as long as it wasn't damaged or misused, but I could be wrong.

The second thing is that I can't believe in my wildest dreams, from start to finish (mining the raw materials all the way to creating the final product), that it takes 6.57MWh of energy to produce a one square meter solar panel.

Someone please enlighten me.

[pyrocasto]

I think that is a common myth about solar panels.

Here's a website to look at.

http://www.eere.energy.gov/solar/myths.html

A study done:

http://www.solarbus.org/documents/pvpayback.pdf

And the average joe's knowledge of RE:

(WARNING:site contains bad language)

Link

[Ungrounded Lightning Rod]

Wrong comparison.

Solar panels aren't about net generation of energy.  Solar panels are about delivering energy where it is needed.  Power grids consume a lot more energy than they eventually deliver to the load - starting with the carnot cycle losses: about three watts of heat "wasted" for every watt generated just from running a fuel-driven power plant, which is a heat engine.  But you don't hear people complaining about efficiencies that don't beat 100%, do you?

The right comparison is between the building and instalation of the panel vs. stringing a power line (cutting down trees, mining ores, processing them into poles, wire, transformers, insulators, a power meter, and the load's fraction of a power plant, shipping them to the site, clearing the right-of-way, and driving workers to-and-from the site until the install is done) PLUS all the fuel burned to generate the grid power.  Any bets on whether just melting the steel and copper for the pole transformer uses more energy than manufacturing a house's worth of panels?

The right comparison is also not energy, but the COST of construction (which takes into account energy, other resources (such as land tied up in rights-of-way), pollution, lost opportunity, generation and transmission efficiency, and everything else of value to people).  It also takes into account things like energy being used more efficiently in large plants than in small ones, near mines than far from them, and so on.

And of course you don't use expensive high-quality electrical energy, such as that produced by solar panels, to melt and refine the glass and metal used to make them.  Using that for heating (the main energy use in manufacturing) is horribly inefficient.  If you were going to solar-power such fabrication steps you'd use mirrors - just as you use fuel directly rather than wasting most of it to generate electrical power and using the power for heating.  (Except for the final zone refining of silicon, of course, where control and avoidance of conatmination trumps generation inefficiencies.)

Having said that, I hear the claim of energy costs in excess of generation are from earlier generations of panels and that modern panels, over their lifetime, generate significantly more energy than it took to make them.

(I've done a post like this once before.  I should file it for the next time the question comes up.  B-)  )

[Ungrounded Lightning Rod]

However, a solar panel doesn't operate at night, so let's cut that in half.  And then let's also say that it's only put into use during half of the day, so let's cut it in half again.  So now we can say that this one square meter solar panel will generate 6.57MWh of electricity over a 30-year period.

What you're looking for here is "solar hours".  The panels are rated as if they were face-on to the noonday sun.  Solar hours are the amount of sun a panel mounted facing south and tilted by its location's latitude would get in an average day, expressed as if the sun suddenly appeared at the noontime position in a cloudless sky, hung there for that time, then disappeared.  They take into account cosine error from the light's daily and seasonal slant, shadows from nearby mountains, and average weather for the location.  (It's up to you to keep the snow swept off.  B-)  )

Values for locations in the US mid latitudes run in the vicinity of 5.  (A web search for "solar hours" "map" will find maps.)  Your approximation comes out to 6, which is close.

[pyrocasto]

Even if you dont compare everything together like that.

If you already have power from the grid, and made solar panels using that power to replace the grid, would it pay itself off, or not.(yes)

If it takes 10kw to make a square meter panel, and that panel will generate megawatts before it dies, then yes it is cost/energy effective unless you pay a few cents per megawatt.(That would be nice)

Then you include everything you talked about and there is no more argument even. The only good thing I can think of that the power companies give would be jobs. If we switched every plant to solar it would also create jobs, so I guess it's about money/lazyness.

[Ungrounded Lightning Rod]

Then you include everything you talked about and there is no more argument even. The only good thing I can think of that the power companies give would be jobs. If we switched every plant to solar it would also create jobs, so I guess it's about money/lazyness.

It's about money - the money you spend for the panels, inverters, batteries, and their ongoing maintainence vs. the money you spend for the grid hookup and juice.  Also about when you spend it - early money is worth more than later money.  There are a lot of costs other than the energy needed to make the panels.

Rule of thumb:  Compute the price P of your proposed solar project.  Compute its expected lifetime L.  Find the monthly payment for an L year fixed-rate mortgage for P dollars (or adding P dollars to your existing mortgage's running balance if that mortgage is for about L years).  Add the monthly cost for maintainence, saving for battery replacement, etc.  If it comes to less than your monthly electric bill it's a win.  (Or your monthly electric bill plus a similar mortgage for the grid hookup costs if you're building a new home.)

Solar is currently a win for new instalations in some rural areas (where hookup costs are high) and for small loads even in urban areas.  A factor of maybe 2 to 4 improvement in price and it will become a win even in many urban areas and for upgrading already grid-tied houses.

It also can become a win if you would need a battery backup system to cover for grid outages.  In such a situation the only extra cost is the panels - the batteries, inverter/charge controller, and other hookup items are already covered.

[monte350c]

Unfortunately most of the solar insolation charts give the rating in kwh PER DAY so in most areas in the US a square meter will give about 3 to 5 kwh per day. If the panel is 10% efficient then it would output about 300 to 500 watt hours per day.

maps: http://www.wattsun.com/resources/insolation_maps/flat_plate.html

Still being able to make power from the sun is a great example of distributed power production and with some of the newer developments like http://www.cnn.com/2003/TECH/biztech/10/02/solar.cells.reut/

[wooferhound]

If it took 6 megawatts to make the panel, don't you think the cost of that power would be added to the panel before it is sold. Here in North Alabama at 6¢ per kwh it would add $360 to the cost.

[ghurd]

You pay 6 cents?  Send me a skid of it!  I pay 16!

G-

[troy]

This is a common myth, and that's all it is.  It has been thoroughly debunked a number of times.  www.homepower.com did a nice job of it a year or so ago.  Fairly sophisticated analysis showed that a thin film (amorphous) panel paid back it's energy debt in less than two years, and a monocrystaline panel in something like three years (I'm working from memory here, so don't quote me to ten decimal places...).

A typical monocrystaline panel will be happily making 85% of it's original output in 50 years if nothing catastrophic happens to shorten its life.  100 years is not out of the question either.  Most name brand mono or polycrystaline panels come with a 20 or 25 year guarantee that says the same thing.

How fast does a natural gas or coal fired power plant pay back its energy debt.  Oh yeah, never.  It's energy negative from the first day and only gets worse after that.

Good luck, have fun,

From the Don't believe everything you hear dept.

troy

[wooferhound]

The power company here in the North Alabama area -TVA- charges 6¢ a kwh up to 1500 kwh in a month, any power you use after that point is charged at 8¢ per kwh

On my last bill I used 1195 kwh and was charged 74.88 for that power, my calculator shows me 6.3¢ per kwh. almost all the power here is Hydro but there is one nuclear plant with 3 reactors that's about 20 miles from here called Browns Ferry Nuclear Plant

[NickCoons]

Ungrounded Lightning Rod,

<Wrong comparison.

Solar panels aren't about net generation of energy.  Solar panels are about delivering energy where it is needed.>

My comparison was in terms of a society that uses nothing but solar panel, in which case solar panels manufacturered would have to be able to generate enough electricity to manufacture more solar panels, plus provide other power.  Had what my friend told me been true, that scenario couldn't exist.

[NickCoons]

Ungrounded Lightning Rod,

<Values for locations in the US mid latitudes run in the vicinity of 5.  (A web search for "solar hours" "map" will find maps.)  Your approximation comes out to 6, which is close.>

I live in Arizona, so my sun-exposure figures tend to be overestimates for everyone else :-).