Solar Energy Pt 3

In part 1, I talked about the basic of solar power.

In part 2, I went into detail about thin-film technologies that stand to dramatically bring the price down.

In this part, I’ll talk about Solar Thermal, focusing in on Concentrated Solar Power, or CSP. This is using the heat energy from the sun to generate power.

Before I begin, some housekeeping. I found an article on west Texas wind power interesting because of the information provided as well as the graph that shows how much power is consumed and how it was generated (coal, natural gas, etc). Just about all the power generated for peak demand (between baseload at 35GW to 61GW at peak demand) as well as an additional 15GW 24/7 is all natural gas. Now regardless of which solar mechanism you use (PV, thermal, etc), that is a lot of natural gas and the resultant pollutants that would be removed if solar gets enough traction to replace a large part of the peak demand.

Likewise, oil tycoon T. Boone Pickens is even supporting west Texas wind power. Wind could help abate the usage of natural gas he says, to be used in vehicles as a replacement for gasoline. While I’m not necessarily in favor of that aspect of his plan, it might be a preferable alternative to high gas prices, and made useful in range-extended electric vehicles (REEV – first x miles on electricity, further range on some other fuel).

Also, Ars Technica tries to calm everyone down about supposed shortages of Indium and Gallium (two key components to the production of CIGS thin film solar cells).

On to the issue of solar thermal energy. I’ll start with a very basic way to use heat to reduce the use of power (either electric or natural gas) – rooftop water heaters. By circulating water with a small 1HP pump up to your home’s roof in a black painted tube or bladder, it will heat up. This is a good way to provide for hot water during the day for either your house or your swimming pool (which is common in Las Vegas).

I’ve even seen alternate implementations of this tactic. Back a few years ago on a camping trip in the Utah high desert, folks on the campsite next to us had black bags full of water they left in the sun while they went on a hike. When they came back the water was hot and ready for them to mix with some colder water to take a shower.

OK, so onto solar thermal power generation. There are several types – parabolic trough and power tower are the two most common. After that, there is dish stirling (using a stirling engine as the focal point for an array of mirrors), Fresnel concentrators and reflectors and a recent invention from MIT that can flash boil water for any number of applications.

Lets go over each type, along with an example.

Parabolic trough is where you have mirrors around a tube containing a heat transfer liquid to focus sunlight on the liquid to heat it. There are several of these types of power stations online, including the 64MW Nevada Solar One. I thought I had heard rumors about troubles, but after some googling, I found this PDF that showed some of their production from June 2007, and it was peaking between 55 and 60MW from 10AM until about 4PM, and from there it trailed off until just before 7:30PM. One of the interesting items also noted in that PDF was the future goal of developing and commercializing thermal storage. Doing this, the plant can siphon off power early in the day when the peak demand hasn’t materialized, and then use that energy at the peak (and sell it for more $) as well as after the sun sets until about midnight the peak finally subsides. The trick is to see if they can figure out

Next is power tower – these designs use an array of mirrors on two-axis trackers on the ground, which reflect the sun’s rays onto a tower holding either water or other heat transfer material to ultimately generate power. There is a lot of momentum behind power tower designs – PG&E in California has agreed to buy power from up to 500MW of power tower plants in CA built by BrightSource Energy. There is an optional expansion of 400MW, with the first 100MW scheduled to go online around 2011.

Next is an interesting technology called Linear Fresnel Reflector, or Fresnel Reflectors. This is where the mirrors, either slightly curved or flat, are mounted just above the ground and then rotate along one axis to heat an elevated conveyance filled with water or some other heat transfer fluid. A company called Ausra has just opened a 700MW/yr Linear Fresnel Reflector manufacturing plant in Las Vegas. The manufacturing plant may only employ 50 people, but the resulting construction jobs, as well as the permanent Operations and Maintenance (O&M) jobs for that 700MW/yr will have a large impact.

Dish/Stirling systems are quite unique. A Stirling engine is an engine that generates energy through concentrated heat. Basically, the heat is focused on an area that contains a gas, and when its heated, it expands. That pushes a piston and generates the movement. After the gas has expanded, its allowed to cool through transfer to another piston, and from there it starts all over again. So the application of this is to use of a dish to focus tremendous amounts of energy into the Stirling engine and generate energy that way. Currently, Stirling Energy Systems should be building an array in Southern California called Solar One (not to be confused with the now-shuttered power tower-based Solar One that is also in Southern California), which is expected to eventually grow to 500MW over 4 years, and possibly 850MW. Power production was supposed to start in 2009, however no announcements have been made as to the progress of a 1MW test array or BLM environmental approval. They also were supposed to build a 300MW as well, but no announcements have been made for that either.

Finally, the MIT project is interesting, they purport that the unit cost per dish is very cheap yet it provides steam for whatever purposes – either steam for a building or turning a turbine for electricity generation. Though for the purposes of electricity generation, you’d have to set the dishes up in parallel, which would require more hardware and that can raise the per unit costs dramatically. The students involved have formed a company and are working to productize the dish.

And thats it for solar thermal.

Solar Energy Pt 2

In part 1, I covered the two different types of solar power and the basic on cost and ROI.

This is part 2. I’ll cover some future developments, specifically thin-film technologies which could revolutionize solar power. But TF has its drawbacks too, specifically low power efficiency per sq meter.

This entry I wanted to cover thin-film based technologies. This is a branch of photovoltaic (PV) solar energy.

There are numerous entrants into the thin-film market. I cant cover them all or I’ll be here all day writing.

I’ll start with the one with the biggest brand recognition, mostly due to who is investing in it (Google), and that is Nanosolar. They are producing a type of cell called CIGS (Copper indium gallium selenide). A few weeks ago they posted a video on their blog about their new roll-to-roll printer that would allow them to produce 1GW of solar cells per year. Now granted, that’s not the only piece in the manufacturing pipeline and they didn’t disclose the efficiency of the cells produced from the unit, but the impact is huge if it can really produce that volume of cells. (some back of the envelope math says that if they run that unit 8 hrs/day, 7 days a week, 52 wks/yr, the solar cells that are made will be 12% efficient, for them to produce cells less than 8% efficient would be a waste because it wouldn’t be cost effective in terms of installation)

The most amazing thing is the cost – Nanosolar claims that tool only cost them $1.65M. At $1/Watt, the unit is printing revenue of $1B/yr. Even at 1c/Watt (an unheard of price for solar power, considering the going rate today is 400x that amount) the unit would still be printing $10M/yr in revenue (6x capex). Beyond that, they state that there is the possibility that they could run the tool faster – up to 2,000ft/minute from the current 100ft/minute. This would result in 20GW of solar cells per year from the single tool. To place that in a useful context, 20GW of solar panels installed on rooftops in California could accommodate 75% of the difference between baseload and peak power demand on the sunniest, hottest days of the year. It would certainly be the end of rolling blackouts.

Next up is First Solar. They are producing a type of thin film cell using CdTe (cadmium-telluride). They are further along than Nanosolar, since they are producing panels and implementing projects like a 40MW utility scale project in Germany. And their cost is cheap too – about 3.25Eur/Watt installed ($5/W at today’s exchange rates, stupid weak dollar), compared to the standard cost of $7/Watt. The problem is land – the efficiency of thin film cells is lower and the installation is using 75W panels for a total of 550,000 panels and roughly 300 acres. So while the price looks great, putting 18 panels on your roof will only generate 1.3kW of energy, or 13 100W light bulbs. More efficient panels are needed for smaller installations.

Another CIGS producer is Global Solar in Arizona. They recently expanded their factory in AZ and opened a factory in Germany.

Not everything is peaches and cream however. A few thin film companies have had to delay, scale back or retool their production plans. Miasole had to retool their commercial production line after seeing low efficiencies, even though their research and development production line saw the 8-10% target they were hoping for.

As I mentioned at the beginning of part 1, the problem isn’t the research of the technology. Its the mass production milestone. To go from 1MW to 10MW to 100MW and 1000MW in annual production capacity over the next five years to provide the energy we need.

Even with the 10% efficiency targeted, panels currently in production are only rated at 75W versus 224W silicon PV panels that are available today. The lower rating or the panel increases the amount of land needed and makes rooftop installations less cost effective due to the lack of space and limited amount of panels that can be installed. A 1500 sq ft ranch home might be able to put 20 200W panels on their roof for a 4kW installation, but at 75W the output would only be 1.5kW. Due to the way the economics work with small installations, the payback time could be much longer because of the increased $/Watt cost as well as a decreased power output (only 30% cheaper for providing a smaller part of your electric consumption).

The sweet spot for rooftop installations is about 3kW-5kW, this provides enough power to be worthwhile and still be around only 20-25 panels, so we’ll need to get thin film panels up around 135-150W. I’m confident we’ll get there, but it might take a few years. In the mean time, the panels will be targeted towards utility scale projects.

With solar power there are three key aspects: price, efficiency, and quantity available. Right now we have efficiency in the bulk PV market, and in the thin film we will soon have price and quantity, with questionable efficiency.

Solar energy pt 1

This is my first diary on Solar energy of many.

In this entry I’ll cover the solar basics, and cover some of the economics.

I’ve seen many, many comparisons of the energy independence issue to the space race. If we could land on the moon, we can certainly overcome this problem, right? Well its not that simple. When it came to landing on the moon, only 34 Apollo Command/Service Modules were built, even fewer Lunar Modules. While still a difficult task to research, design and build these units, the issue was manufacturing them fast enough to meet the ambitious launch schedule – and not so much on making them cheaply and in large quantities.

Which is the problem we face when it comes to energy. We have the technology. We just need the ability to manufacture it in massive quantities and to do it cheaply enough to make a difference. In 2007, the United States had about 1TW (1 terawatt, or 1 trillion watts) of generation capacity. So how do we go about replacing that with renewable energy? Solar is one avenue, for those in the south and southwest.

The best property of solar power is that it matches peak demand so well. Peak demand is largest in the summer months (May through September) where the grid’s power demand increases dramatically over the “base load”, or the amount of power the grid is always demanding. For example, on July 1, 2008, the peak demand in the areas of California controlled by CAISO was 39GW (GW = 1 billion watts) at around 5PM local time, and the base load was 24GW. So not only is solar power fuel-cost free, it also generates the most expensive power – peak power. Wholesale rates can approach $340/MWh (34c/kWh) during the highest period of demands in California, and even $130/MWh during summer peak periods, whereas off-peak power can cost as little as $35/MWh (3.5c/kWh).

On to the technology…

There are two general forms of solar power – PV or photovoltaic, which converts energy from photons directly into energy, and thermal, which uses the sun’s heat, concentrated to heat liquids to transfer that heat, boil water and turn a turbine. There are variations on these themes, such as concentrated PV, where a lens can focus more sunlight onto a PV cell, and Stirling-engine based solar thermal which uses a Stirling engine to generate the power through a large temperature gradient (e.g. concentrated sunlight and ambient air temperature).

Traditional PV solar modules (or panels) come in a range of sizes and power ratings. Some of the most recent advances include a 224 Watt panel from Sharp, and a double sided solar panel from Sanyo useful for carports and other elevated installations that can produce up to 215 Watts. SunPower Corp. has even announced a 300W solar panel that measures the same 3.5’ by 5’ industry standard panel size.

The problem is that a PV panel is expensive, about $4/Watt. After other factors like additional parts needed – like inverters to convert the DC energy the panel puts out into AC energy your appliances use, and a new power meter than can spin backwards to track your production – and installation of the panels, the cost of the entire system will be about $7-8/Watt. At this price, a reasonable 5kW system that you might install on your roof is about $35,000-40,000, though the cost can be reduced through various rebates and tax incentives. In a sunny environment like my hometown of Las Vegas, the system would pay for itself in about 12-15 years after rebates, the cost of the loan and escalating energy prices.

So you can see that a rooftop PV system is fairly expensive. Even large-scale PV systems on the order of MWs (Megawatts, or millions of watts) are in the same $7/Watt price range.

Solar thermal is also expensive as well, though less expensive than PV-based solar power. The Mojave Desert in the southwest is home to the largest solar thermal system in the United States, SEGS or Solar Energy Generation System, which has a total capacity of 354MW. The most recent solar thermal installation is Nevada Solar One at 64MW and was constructed for about $266M, or just over $4/Watt. Of course, you cant put one of these systems on your roof, and to make the project sufficiently cost effective you’ll need a large swath of land to build on. Finally, solar thermal can also address the biggest problem with solar – it only works when the sun shines. There are projects to store the thermal energy in various heat transfer mechanisms (molten salt) to provide for power generation after the sun sets, until about midnight.

Which brings me to the next point – late last month, the BLM put all solar projects on hold in lieu of a Programmatic EIS (Environmental Impact Survey) to asses the environmental impacts of large scale solar impacts on the desert. Well early this week, they reversed course and said they wlll continue to accept and process applications. Now whether or not its just words on paper or if they really intend to process these applications wont be known for a while since they take so long to process in the first place.

So I think that’s enough for part 1. I went over the costs and the technology. Next I’ll talk about what gets me interested in solar – what developments are coming down the pike, for traditional PV, thermal, and thin-film technologies. And what problems and opportunities those developments unlock. After that, what it could mean for transportation technologies.