The Instigator
Pro (for)
7 Points
The Contender
Con (against)
0 Points

Solar panel farms will be economical in 10 years

Do you like this debate?NoYes+2
Add this debate to Google Add this debate to Delicious Add this debate to FaceBook Add this debate to Digg  
Post Voting Period
The voting period for this debate has ended.
after 1 vote the winner is...
Voting Style: Open Point System: 7 Point
Started: 9/21/2011 Category: Technology
Updated: 5 years ago Status: Post Voting Period
Viewed: 1,585 times Debate No: 18419
Debate Rounds (4)
Comments (22)
Votes (1)




This is partially inspired by the wind generation in 10 years debate that recently finished.

I will propose a solar panel farm to be built in 2021 as a private company, not as a "cover an entire state, or nation" but merely to serve a particular community. My model will likely be based around providing electricity for a few thousand people, though I reserve the right to go larger or smaller.

As such, I may also have back up generation for nights/overcast or choose to use a battery power supply.

Since we are dealing with 10 years in the future, costs will be determined based on their historical trends. This applies to cost of materials, labor, electricity, etc. For this debate, it is agreed that we use historic trends, not what some analyst predicts (since people can be on payrolls, history, not so much).

By "economical," we are refering to it's ability to make money at market values. As seeing as it will be a small scale farm, it will not have a significant impact upon the market value of electricity, and so it will charge market rate and attempt to make a profit. The measuring stick will be against the Utility Bonds annual rate of 4% (this is kind of becoming the standard measuring stick thanks to RoyLatham).

If you have any questions regarding this, or something that I may have forgotten, please ask in the comments.

Also, my opponent should allow me to go first (as I need to setup my model of them to attack), though if they feel the need to make some statements in their R1, they are more than free to.

Thank you.


I shall accept the challenge to show that it cannot be done so soon. Begin.
Debate Round No. 1


I thank my opponent for accepting this debate. I will divide this debate up into 3 parts. First, I will identify what needs to be factored into the costs. Second, I will identify what the numbers will be for 10 years into the future. Third, I will run the calculations to see if money can be made.

Things that need to be factored in. Cost of electricity, maintenance, land, panels, labor to install, backup systems, power delivery, inflation, replacements when needed. We also need to figure in the rate of efficiency decay on the panels. If my opponent has more that he can think of that needs to be factored in, he may bring them up in his round.

Moving on to figuring out what these will be in 10 years. Let's look at each individually.
1)Cost of electricity. From 2000 to 2011, we've observed a 3.8% increase in the national average of electricity [1], this is also matched up with Arizona's average growth over the same time frame [1]. AZ is currently at 10.98 cents per kWH (as of the end of 2010), and so in 2021 it will likely be 16.55 cents per kWH. Now, it could be argued that we are going to be experiencing heavy inflation sometime in the not too distant future, because of massive government debt and money printing, which would speed up the rate of cost increases, but that is too much speculation for me to dive into.
2)Panels. The cost of solar panels has historically been dropping at a rate of 10.1% a year. This is the average from the first commercially available solar cells in 1955 [2]. Since 2005, we've seen price decreases on the average of 12.3%. 11% price decreases per year is a good estimate between what we are currently seeing and the historic. This means that panels that cost $1.44 per watt today [3], will likely cost $0.45 per watt. We also need to look at how their efficiency is increasing. Solar panels have increased an average of 6% a year in efficiency from 1955 to 2010 [2]. Since the current model that we based the price on was getting 143W per square meter [3], it would be safe to assume that in 2021, the models would be getting 256W per square meter. It should also be noted that based on the warranty, these panels deteriorate at about 0.75% per year, which will need to be factored in.
3)Land. Open land in Arizona can run as low as $450 an acre [4], or in the area that I'm particularly looking at, around $1,800 an acre [5]. The value of this land has not changed much over the years, so we'll just apply inflation to it. Average inflation over the last 15 years has been 42.63%, or 2.4% per year. This will put the value of our land at $2,300 per acre for the close land, and $570 per acre for the distant land.
4)Electricity Transmission. Transmission lines cost around $940,000 per mile to install [7]. Since we are not going to be traveling a huge distance, the high power lines are not needed. These too will also be increased based on inflation, since I cannot find a direct history of their costs. So they will come to about $1,190,000 per mile.
5)Labor costs. Labor per solar panel to install has been sitting roughly around $360 per panel [8]. While new technology is bring that done greatly, such as the FlexRack, I have been unable to get a number for the cost of the actual FlexRack, and so we will have to use the standard labor costs.
6)Maintenance. This greatly depends on the number of panels, so this will go more in depth in my actual plan. When it comes to generally cleaning the panels (removal of dust), cheap labor is readily available, however, there will need to be individuals with EE degrees on hand for when more serious issues come up.
7)Backup system. There are two options for this. A massive battery system that the solar panels will charge during the day to power at night, or a separate generator to work at night. For this, I will go with the generator. Natural gas power plants currently cost a little under $1,100 per kW to construct [9], meaning a 10 MW plant would cost around $11 million to build. This labor cost has been pretty level over the years, even with inflation. Natural Gas is the cheapest of all energy sources to construct. The cost of fuel has been pretty level, though it is coming down from a recent high (it has been elevated over the past 8 years and has come back down just recently), it is at about $5.00 per 1,000 cubic feet [10] and 1,000 cubic feet will be 351.7 kWh, since the power plants of today are about 60% efficient [11], so 211 kWh. This can also show that $23.50 of fuel is needed per MWh.

Okay, let's move on to the system. First, the NW part of AZ get 8 to 8.5 average sun hours a day, but we'll use 8 sun hours [12] (these are the average hours the panels are fully operating, really, you get about 8 hours of partial use and 4 hours of full, but it averages to 8). We are going to make the solar farm � of a square km, or about 123 acres. Now since I showed that land can be found in 77 acre chunks, finding 123 acres is not an unreasonable stretch. I will also need land beyond that for the backup and offices, so I'll round to a nice 150 acres. On this, we'll put 311,000 solar panels to get a nice 128.05 MW of solar. These panels will cost about $57.6 million, the labor to install will be about $111.96 million. The land will be only $345,000 (as we can see, land is not an important factor). Since the land is nice and flat, and solar has no noise or pollution emitted (and the backup is going to be small to keep emissions at a minimum), the facility can be close to town. I set the power lines at 20 miles away with 20 miles to spread to reach the homes. This gives a power line cost of $47.6 million. That makes the total for the solar section $217.5 million to build.

On a national average, we use 35% less energy at night [13]. So the backup needs to be able to handle 65% of peak demand and the average home has a peak demand of 1.7 kW [14], so the average night time is going to be 1.1 kW per home. If we let the solar provide the 1.7 kW per home per day, that means that 50,200 homes can be hooked up. This also means that they'll have a night time need of 55.2 MW backup at night. This will cost $60.7 million, which brings our total to $278.2 million to construct. This will also allow the back up to not need to be on at all during the day so that it can be cleaned regularly and can fire up its full load when needed on cloudy days when the solar is not performing at 100%.

With 1/2 the day being at 1.7 kW and 1/2 at 1.1 kW, this puts the total monthly consumption at 1,042 kWh, which is fairly close to the state average. This means that there is a total of 761,280 MWh produced a year for a gross income of $126 million.



Plato_ATODT forfeited this round.
Debate Round No. 2


SOB! I put a lot into that round and it isn't even done, but if you're not going to even make a post I'm not putting more effort into the next rounds.


Plato_ATODT forfeited this round.
Debate Round No. 3


Well, another forfeit. Let this misery end.


Plato_ATODT forfeited this round.
Debate Round No. 4
22 comments have been posted on this debate. Showing 1 through 10 records.
Posted by F-16_Fighting_Falcon 5 years ago
Awesome. I'll let you know if I disagree with your position.
Posted by Ore_Ele 5 years ago
Posted by F-16_Fighting_Falcon 5 years ago
Yeah, I see your point about the price dropping at the same rate now as before. I really have no idea whether a Solar Panel farm will be economical in 10 years, but I'll research that and I will debate you on it if I disagree, you up for that?
Posted by Ore_Ele 5 years ago
Actually, I'm not going to finish this one up. There is a lot of research that went into the first part, and a lot of research to put all the remaining links together and I'm not going through all that since he's gonna forfiet again.
Posted by Ore_Ele 5 years ago
Programming robots would be cheaper, because the reverse is trainning employees. You only need one expensive code that can be copied for all robots, but with manual labor, you have to train each and every employee. And every time you get a new employee, you have to retrain them. I had to go through 3 weeks of training just to work at a call center for $9.50 an hour.

Robots shift the cost away from labor and towards energy (though there is still labor to maintain and programm, it is less labor than the robots replace).

However, I showed that the rate of price dropping over the 58 years has been roughly consistent with that of the last 6 years. That means we have not hit the floor yet. Even in the last month, the price on the CSI panels has dropped $0.15 per watt (that was a 9% drop). So clearly, we are not at the floor yet, and there is no sign that we are close. Typically, when you get close to the floor, you start to level out. It is highly unlikely that we are going to hit that in the next 10 or even 20 years.
Posted by F-16_Fighting_Falcon 5 years ago
You could make a 5 round debate: Round1 - acceptance
Round2 - your plan, your opponent forfeits or just says "continue"
Round2 - continue your plan - your opponent starts arguing
Round3 - you forfeit, or write a line saying "extend", your opponent continues arguing
Round4 &5 - both argue
In reality, there would only be 3 rounds of actual debating.

As for the prices, yes the labor costs would decrease but you can't know that haven't already hit the floor without doing further analysis on the price needed for it to be done by robots. I think it would be fair for your opponent to take issue with it since it works to your advantage.

You would need the capital costs of building and buying those robots, costs of computer engineers programming those robots to carry out a specific task, so that could be another obstacle to the decrease in price.
Posted by Ore_Ele 5 years ago
I'm going to finish it up (since all the outlining takes way more than 8,000 characters) so I can restart it and just point to this debate. That will give an extra round actually argue. I hate when my opponent starts arguing when I'm only 1/2 way done presenting my plan (which is what the 8,000 character limit causes).

As for dropping prices. The floor is really determined by the labor and energy used to produce them. If they make energy cheaper, then they can make themselves cheaper. If they can be made by robots (robots historically have a bad union), which run on the cheap electricity, that minimizes labor costs.

When you dig down to it, everything can be measured in 3 units. 1) Labor (in $, it would be hours, but different jobs pay different amounts). 2) Resources (how many mols of Si and Cu or Ag for wires). And 3) Energy, let's convert to kWh.

Of course, the resources, breaks down into the two other categories (time and energy used to mine them). So really, it is only Labor and Energy.
Posted by F-16_Fighting_Falcon 5 years ago
The problem I have is with the declining price of the Solar Panels. Don't you think it is going to plateau after reaching a certain critical price. It can't keep going lower and lower until it costs nothing. I am not sure historical price figures are accurate in this particular case.
Posted by F-16_Fighting_Falcon 5 years ago
Damn! All that work from Ore_Ele for nothing! I was really looking forward to reading this at its conclusion.
Posted by quarterexchange 5 years ago
It'd be funny if Con's mother or grandmother recently passed away.
1 votes has been placed for this debate.
Vote Placed by F-16_Fighting_Falcon 5 years ago
Agreed with before the debate:--Vote Checkmark0 points
Agreed with after the debate:--Vote Checkmark0 points
Who had better conduct:Vote Checkmark--1 point
Had better spelling and grammar:Vote Checkmark--1 point
Made more convincing arguments:Vote Checkmark--3 points
Used the most reliable sources:Vote Checkmark--2 points
Total points awarded:70 
Reasons for voting decision: Pro made a wonderful case drawing from many reliable sources, and outlined the details on how the plan would work. Con did nothing.