US approves world's biggest solar energy project

[johnsmith]

I'm going to lunch now so that I can figure out a way to gouge the GC I'm working with on a change order that I'm going to charge him 50% more for than I normally would because I already have a contract in place.



See how that works?

[RandomGuy]

And why is it that do you suppose that the overruns regularly reach 400% on nuclear?

I have a theory rooted in doing 100's of etimates on project that rage from power, to highways, to commercial work. By the time these projects get approved, materials and equipment prices have gone up exponentially because the red tape that is needed to cut through takes a ing eternity.

Is that the construction processes fault? The engineers? I'm asking here.

That is the reason for a large majority of nuclear plants cost overruns to my understanding.

They are unpopular and often end up having to fight off a lot of lawsuits. I have followed the local SA utility's nuclear project, and that has taken place.

The fact that they are complex construction projects also leads people to grossly underestimate the time involved in construction. That is the rest of it.

One might use the sword of eminent (sp?) domain to cut through the gordian knot of litigation, at the cost of individual rights, but the complexity of the projects would still drive overruns.

That could be overcome by standardized blue prints though. The problem with THAT would be picking what design to use, and how to enforce that.

[Wild Cobra]

All your solar plants are belong to us.

Seriously though, "40 years"? What basis are you using to calculate that?

The approximate sale price of electricity to the consumer at 10/kwh and estimated average output.

Did I do my math wrong?

[RandomGuy]

The approximate sale price of electricity to the consumer at 10/kwh and estimated average output.

Did I do my math wrong?

To answer that I would need more information.

Show your work.


Would also need to source the data given to evaluate whether the information is current and reliable.

My gut says that 40 year payback is probably wrong for some reason. It would be an interesting calculation to make though.

[MannyIsGod]

If the plan is going to be up for at least forty years I don't care if it takes that long to get paid back.

[Wild Cobra]

Let's revisit the math.

The nameplate capacity will be 1,000,000 kwatts.

The cost will be $4,000,000,000

The equates to $4,000 per kwatt, but if we assign a 25% actual average output because of the earths rotation, angle, clouds, etc. then we only end up with an average usable 250,000 kwatts.

Over the course of a year, this is 2,191,500 kwh. (250,000 x 24 x 265.25)

If we sell this power at $0.07 per kwh wholesale, our annual revenue is $153,405,000, but... that doesn't work. We have power losses in transmission.

If we assign a 35% loss of power in inverter conversion, transmission lines, transformers, etc, we now only have an annual revenue of $99,713,250.

At that rate, it takes 40.12 years to recover the initial start-up costs.

[Wild Cobra]

If the plan is going to be up for at least forty years I don't care if it takes that long to get paid back.

That is only the gross revenue from the sale of electricity. Maintenance personnel, spare parts, downtime, and other operating costs are not factored in yet.

I'll bet it doesn't cover the transmission lines to carry the power either.

[MannyIsGod]

LOL @ you cutting the rated production by 75%

[Wild Cobra]

LOL @ you cutting the rated production by 75%

Yes. The sun is only shining half the time. Under impossibly best cir stances, that is a 50% loss already. The solar intensity is highest at noon. If they are fixed cells, then at best, they get about 70% average daytime exposure. Basically 0% at sunrise and sunset 50% at about 8 AM and 4 PM, about 86% at 10 AM and 2 PM, during the equinox. More in the summer, less in the winter. This averages to about 35% before any atmospheric conditions and dust reduce the power levels converted.

[MannyIsGod]

Your calculation sucks considering the future rated max of the plant is going to be much higher than 1000 MW but keep cutting it down.

[Wild Cobra]

Your calculation sucks considering the future rated max of the plant is going to be much higher than 1000 MW but keep cutting it down.

Keep dreaming our national wealth away.

I give what I believe to be realistic numbers. You challenge, when I show you my reasoning, you split off on a tangent.

Nice debating you, but you have obviously resigned from the issue. Sure, the rating will be higher in the future. After how much more cost though?

Is this a subsidy we should consider? I don't think so. If solar is the way to go, let power companies pay for it when they see profit in it. Not our tax dollars.

[coyotes_geek]

If we sell this power at $0.07 per kwh wholesale, our annual revenue is $153,405,000, but... that doesn't work. We have power losses in transmission.

Looks to me like you're really lowballing the price which the power could be sold. According to this, retail residential electric prices in California are $0.15/kwh. Differences in retail versus wholesale noted, but I have a hard time believing utilities are putting a 100% markup on electricity.

http://www.eia.doe.gov/cneaf/electri...able5_6_b.html

[TeyshaBlue]

Is this a subsidy we should consider? I don't think so. If solar is the way to go, let power companies pay for it when they see profit in it. Not our tax dollars.

It would seem to me that infrastructure costs and issues should be pretty much the prime target for spending taxes.....spending for the demonstrable betterment of citizens and all that.

[RandomGuy]

25% actual average output because of the earths rotation, angle, clouds, etc.

built on public lands in the sun-drenched Mojave desert

Tell me more about the cloudcover at the location.

Then tell me how the production peaks at the same time demand does, and how that might effect your calculations.

[Wild Cobra]

Looks to me like you're really lowballing the price which the power could be sold. According to this, retail residential electric prices in California are $0.15/kwh. Differences in retail versus wholesale noted, but I have a hard time believing utilities are putting a 100% markup on electricity.

http://www.eia.doe.gov/cneaf/electri...able5_6_b.html

OK, maybe the return is better. They pay so much in electricity because of transmission costs, losses, taxes, etc. What would it sell for then? Another 50% higher? That's still about 27 years.

I'm flexible on the costs. My numbers are arbitrary. I think I'm lowballing the transmission loss myself. I'll settle for around 30 years. Sound better? If you disagree, what numbers would you use?

[coyotes_geek]

OK, maybe the return is better. They pay so much in electricity because of transmission costs, losses, taxes, etc. What would it sell for then? Another 50% higher? That's still about 27 years.

I'm flexible on the costs. My numbers are arbitrary. I think I'm lowballing the transmission loss myself. I'll settle for around 30 years. Sound better? If you disagree, what numbers would you use?

Your assumed sale price is the only number that jumped out at me. The 35% capacity number passes the smell test with me and I don't know nearly enough about transmission losses to question that value.

Whether it's a 30 year payoff or even a 40 year payoff I don't have a problem with it. As I said earlier in the thread, it's not uncommon to see infrastructure projects financed out over 40 years.

[Wild Cobra]

Your assumed sale price is the only number that jumped out at me. The 35% capacity number passes the smell test with me and I don't know nearly enough about transmission losses to question that value.

Whether it's a 30 year payoff or even a 40 year payoff I don't have a problem with it. As I said earlier in the thread, it's not uncommon to see infrastructure projects financed out over 40 years.

That 30 year payoff would assume that everyone worked for free, and there were no maintenance costs. It could be more like 60 years when you factor in all the operational costs. Who knows. Costs could exceed revenue, and it may never pay for itself.

[RandomGuy]

Let's revisit the math.

If we sell this power at $0.07 per kwh wholesale,

http://www.eia.doe.gov/electricity/epm/table5_6_b.html
While these aren't wholesale, I would question that.
Retail prices:

Residential 15.22 cents per kwh.
Industrial 13.55 cents per kwh.
All sectors (presumed weighted average) 13.58

Now if one assumes wholesale California:

California exchange website:
http://www.caiso.com/2777/277789c42ac70.html
Yields this report that states on page 2 figure e-1 of executive summary:
http://www.caiso.com/27c7/27c792e4cc50.pdf


Seems an average of 40 bucks per megawatt hour for 2009, do some fancy schmancy cal alatin' 40.00/1,000 = .04

Given though, that is an average monthly price and most defintely NOT peak, we skip ahead to E.6 on page 9 to get a better figure of...
just about the same. .04

It must be noted that the report notes that peak prices are strongly influenced by the cost of natural gas, since that form largely supplies most peak loads.

If one really wants to estimate pay-back for solar though, one would have to fully encapsulate future cost increases in natural gas.

If one factors in the time value of money, and guesses that energy prices will go up a tad faster than inflation due to natural gas depletion over the next 40 years, that would probably be a bit better.

-------------------
Edit:
Assuming 4% price inflation of natural gas (and coal for that matter) would impact the overall price considerably.
Here are the factors for the next 40 years.
year 4% 5%
1 1.040 1.05
2 1.082 1.103
3 1.125 1.158
4 1.170 1.216
5 1.217 1.276
6 1.265 1.340
7 1.316 1.407
8 1.369 1.477
9 1.423 1.551
10 1.480 1.629
11 1.539 1.710
12 1.601 1.796
13 1.665 1.886
14 1.732 1.980
15 1.801 2.079
16 1.873 2.183
17 1.948 2.292
18 2.026 2.407
19 2.107 2.527
20 2.191 2.653
21 2.279 2.786
22 2.370 2.925
23 2.465 3.072
24 2.563 3.225
25 2.666 3.386
26 2.772 3.556
27 2.883 3.733
28 2.999 3.920
29 3.119 4.116
30 3.243 4.322
31 3.373 4.538
32 3.508 4.765
33 3.648 5.003
34 3.794 5.253
35 3.946 5.516
36 4.104 5.792
37 4.268 6.081
38 4.439 6.385
39 4.616 6.705
40 4.801 7.040

[RandomGuy]

If we assign a 35% loss of power in inverter conversion, transmission lines, transformers,

You are also assuming that the producer eats the entire transmission loss. This is unlikely.

Based on the mark-up between wholesale and retail, I would guess that the producer eats almost none of that loss, but that would require some reading of the utility financials to determine.

[Wild Cobra]

Given though, that is an average monthly price and most defintely NOT peak, we skip ahead to E.6 on page 9 to get a better figure of...
just about the same. .04

Do you think it would be safe to assume that the $0.04 is the number before transmission line losses, taxes, operational costs of the PUD, etc.

Wow...

At only $0.04 per generated kwh, that brings the return time to 45 years assuming zero operational costs. So with transmission costs, line losses, operational expenses, etc.... Californians are paying more than 3 times the cost...

Doesn't surprise me. They generate very little of their own power, so they have huge line losses. The farther you go, the larger the losses. I think the 35% loss I calculated is typical for local use. Most the loss is in the transformer stages, but would be more yet at low voltages.

[RandomGuy]

Do you think it would be safe to assume that the $0.04 is the number before transmission line losses, taxes, operational costs of the PUD, etc.

Wow...

At only $0.04 per generated kwh, that brings the return time to 45 years assuming zero operational costs. So with transmission costs, line losses, operational expenses, etc.... Californians are paying more than 3 times the cost...

Doesn't surprise me. They generate very little of their own power, so they have huge line losses. The farther you go, the larger the losses. I think the 35% loss I calculated is typical for local use. Most the loss is in the transformer stages, but would be more yet at low voltages.

Given 4% inflation of electrical prices over a 40 year period starting at .04 cents per kwh, that would make the average price of electricity over that period to be almost exactly 10 cents.

Given a 5% inflation of electrical prices that would make the average price over the same time period to be almost 16 cents.

Your calculations in determining payback period will be heavily influenced by the time value of money.

You can plug in the columns I gave above into a spreadsheet to help.

[RandomGuy]

Other factors:
PV output declines over time.
Few mechanical parts means a much more reliable source, assume wind base of about 2% build cost per month for maintenance per year. (if memory serves, found data in wind turbine company's financials)
This must be indexed for overall inflation/cost of capital.
Also note that it is likely that the PV installation will have a lifespan that far exceeds any equivalent competing source of energy.

[RandomGuy]

They generate very little of their own power, so they have huge line losses.

By the by, the total imported Mwh was also in that pdf file I gave. Fascinating do ent.

[Wild Cobra]

You can plug in the columns I gave above into a spreadsheet to help.

Not really. The number I generated already has a large unknown error factor. Starting at $0.07 per kwh was my guesstimate of what the solar facility would sell the power for. I don't think power prices have gone up by that much, and I missed the year somehow by a decade at first in your linked material. If I assume a 2% annual increase, that $0.04 becomes only about $0.049. $0.054 at 3%, $0.059 at 4%, and $0.065 at 5%. My $0.07 number would require a 5.76% annual increase, so it appears I may have been lowballing the price.

Just how expensive will solar actually be?

[coyotes_geek]

That 30 year payoff would assume that everyone worked for free, and there were no maintenance costs. It could be more like 60 years when you factor in all the operational costs. Who knows. Costs could exceed revenue, and it may never pay for itself.

O&M is going to turn a 30 year payoff into a 60 year payoff? For that to be the case O&M would have to be running running in the +$100mil/yr range. Sorry, I just don't see it.