Will solar power be over 1% of U.S. electricity production in 2016?

[russ]

There's the plant in Spain which came online last fall, with 280MW and thermal storage.

I think I read something recently about this plant actually managing to deliver power 24 hours/day for some substantial fraction of the time, but I can't put my finger on the link right now.

Solana has heat storage for maybe 6 hours. The one in Spain they claim 24 hour operation for though that is the first time I have read of that much heat storage.

[ButchDeal]

Solana has heat storage for maybe 6 hours. The one in Spain they claim 24 hour operation for though that is the first time I have read of that much heat storage.

You also have systems like Ivanpah Solar Power which use natural gas in off hours.

[russ]

You also have systems like Ivanpah Solar Power which use natural gas in off hours.

Different story altogether.

[ButchDeal]

Different story altogether.

Different from storage yes but goes to the point of night usage and original question of solar power being higher percentage of US production. Why build an all NG power plant when you can build a hybrid plant and have cheaper operating costs?

[russ]

Different from storage yes but goes to the point of night usage and original question of solar power being higher percentage of US production. Why build an all NG power plant when you can build a hybrid plant and have cheaper operating costs?

A second boiler would be required? One that you cycle daily? Better to just have the natural gas turbine if that is the case. Excessive cycling of a boiler is a very bad thing.

You would have to effectively build two plants to get one output - the fuel for one would be the sun but the capital costs would be great.

[ButchDeal]

A second boiler would be required? One that you cycle daily? Better to just have the natural gas turbine if that is the case. Excessive cycling of a boiler is a very bad thing.

You would have to effectively build two plants to get one output - the fuel for one would be the sun but the capital costs would be great.

You would need an NG boiler, but the turbine, condensers, generators, substation, and all related systems are shared.

NRG and Google built Ivanpah as a hybrid system. It isn't intended for full night production, just to extend the hours to cover peak billing periods in the evening and keep the system warm for earlier startup in the morning.

[J.P.M.]

Let's run the numbers there for a PV based plant.

100 sq mi is about 258x10^6 square meters. Making a conservative assumption of 150 watts/sq m (about average for PV) then we would see generation of 38 gigawatts peak, or about 233 gigawatt-hours per day. Averaged over 24 hours (to compare it to a base load plant) that would be equivalent to 9 gigawatts, or about 9 1GW power plants (a common size for base load plants.)



Agreed, although thermal storage solar plants will help significantly with that.

First off, the Brightsource plant has nothing to do w/PV - It operates on a steam cycle, just like conventional power plants. The turbine design, feed pumps desuperheters, etc. are all of a modified design, mostly to handle varying input, but it's a steam cycle plant.

After that, according to Brightsource, the plant's builder, the rated capacity is 377 mW (e) with a capacity factor of .31. Furthermore, Brightsource estimates yearly output at 1,000 gW/yr. once all 3 units are operational. Their #'s, not mine.

As a check: (377 X 10^6 X 10^3) X 8,760 X .31 = 1,024 gWhrs./yr. Close enough in my book.

A 2,000 mW(e) conventional power plant will conservatively have a capacity factor of about .90 or greater. If so, the yearly output from such a plant would be:

2,000 X 10^6 X 8760 X .9 = 15,768 gWhrs./yr.

The planned (estimated) yearly output of the Ivanpah facility as a % of the capacity of the 2,000 mW (e) conventional power plant would then be : 1,024/15,768 = 6.49%

I believe my back of envelope analysis suggested 5% - 6%. Brightesource calls the plant footprint "approximately " 3,500 acres ~ 5.47 sq. miles. which would give ~ 86 sq. miles, so, my "about 100 sq. miles" (my words), when scaled up to 15,768 gWhrs./yr." is a bit high. Most everyone else calls it closer to 4,000 acres which would give 98 sq. miles. Either way, I'd suggest a plant of that capacity would require more infra structure and maybe get to 100 sq. miles, but I'll ignore that.

Going the other way, using Brightsource's 3,500 acres = ~ 5.47 sq. miles, each sq. mile will produce 1,000/5.47 = 183 gWhrs/yr.-mile^2 --->>> 15768/183 = 86 sq. miles approximately equal the output of 2,000 mW (e) conventionally generated power, pretty much regardless of individual plant capacity.

To be slightly less conservative than my last post on this thread, I'll change my 5% - 6% # to "less than 7% and lower the required area to "something greater than 85 sq. miles", stick a fork min it and call it a modified Q.E.D. Still pretty good size in my book.

[russ]

You would need an NG boiler, but the turbine, condensers, generators, substation, and all related systems are shared.

NRG and Google built Ivanpah as a hybrid system. It isn't intended for full night production, just to extend the hours to cover peak billing periods in the evening and keep the system warm for earlier startup in the morning.

Agreed

[russ]

To be slightly less conservative than my last post on this thread, I'll change my 5% - 6% # to "less than 7% and lower the required area to "something greater than 85 sq. miles", stick a fork min it and call it a modified Q.E.D. Still pretty good size in my book.

But it makes the greens so happy! Doesn't even have to make sense!

[TxSolarPro]

You guys have posted very cool information about the numbers behind utility scale projects. I'm not an expert on utility-scale PV but it seems that the only viable place for solar in the near future is behind the customer meter. Here in Texas a utility scale plant would have to compete with $.02-.03 wholesale electricity costs, but if you install solar on the customer side of the meter you save the .06 retail rate plus the .04-.05 for delivery and fees actually saving $.10 plus per KWH. That's a huge difference.

[bcroe]

You guys have posted very cool information about the numbers behind utility scale projects. I'm not an expert on utility-scale PV but it seems that the only viable place for solar in the near future is behind the customer meter. Here in Texas a utility scale plant would have to compete with $.02-.03 wholesale electricity costs, but if you install solar on the customer side of the meter you save the .06 retail rate plus the .04-.05 for delivery and fees actually saving $.10 plus per KWH. That's a huge difference.

When I figure the KWH rate, I find every item on the bill that is multiplied by
the number of KWH used. Add those items up, and that's the cost per KWH.
Here that has been around 8.2 cents per KWH (I can see the nuke plant from
my property). I thought TX had really cheap rates, but you are saying 10 cents?

Bruce Roe

[TxSolarPro]

Yes .10 is average in our area. We see from .09-.12 commonly. 8 cents is pretty cheap!