This Friday I am scheduled to spend two hours with MIT’s Future of Solar Energy Study. As the May announcement of the study said:
Under the direction of Institute Professor John Deutch, the group will spend about a year and a half analyzing the prospects for solar photovoltaics, solar thermal generating systems, solar water heating, and the use of solar energy to produce fuels ["such as hydrogen from water"].
The nine faculty members on the team [named here] include specialties in chemistry, mechanical engineering, electrical engineering and computer science, materials science and engineering, physics, economics, and management.
Yes, I’ll be suggesting that splitting water is not the most productive use of solar, and, of course, I will be talking up solar thermal baseload.
But since this is close to the beginning of their study, I think the more interesting issue to raise is — what are the key questions the study should answer?
In particular, I will be spending this week trying to come up with a list of questions I personally would like the answer to, such as:
- Just how much total solar power could the nation and world generate by 2025 and 2050 if we were very serious?
- How much could we generate once we get appropriately desperate about avoiding catastrophic global warming and adopt a WWII- style strategy?
I find most studies in this area simply fail to get to the scale issues involved. And the few that do merely say the scale is unattainable, which of course it is if one maintains conventional thinking.
In my 450 ppm “solution,” I pose some 3 wedges of solar thermal baseload (~5000 GW peak) and up to one of solar PV (2000 GW peak) by 2050 (see “Is 450 ppm (or less) politically possible? Part 2: The Solution“). The 350 ppm-ers probably need 2 to 3 solar wedges by 2030 alone (see “the real truth about stabilizing at 350 ppm“).
I’m not terribly interested in a study explaining the political and logistical impracticalities of stabilizing at 450 ppm or below. I already know them. I am interested in a study that answers the question:
- What are the most important material, technology, and skilled labor constraints to achieving multiple solar wedges by mid-century and how can we address them?
And here’s another question people haven’t been thinking about:
- Given the massive influx of venture capital funding into solar energy, how should the Department of Energy best redesign its R&D strategy so as not to duplicate that effort but rather to leverage it (see “Despite market downturn, cleantech venture investment hits record $2.6B in 3rd quarter“)?
What questions would you like answered of MIT’s Future of Solar Energy Study? I’ll take the best ones and pose them to the committee.
Related Posts:
RSS
Subscribe by Email
Follow Climate Progress on Twitter
Mr. Romm,
I think a key issue for large scale solar is the difference between distributed and concentrated generation. There are groups that will fight tooth and nail to save the Mojave. No one cares what the top of warehouses in L.A. look like.
I know this thought goes against many very good reasons for preferring solar baseload. But it should be considered.
How do transmission losses affect the balance between big plants versus distributed power generation? How much unused area exists on large rooftops in areas with sufficient sunlight? Are there secondary benefits to rooftop solar, for example, insulation effects requiring less air conditioning? What are the funding obstacles to each approach?
we know a lot about habitat impacts of utility solar and wind. with possibly explosive growth in local solar for electricity and heat, i’d like to know a little more about the various mass-market-candidate equipments’ lifecycles — what will be the byproducts, how long will the equipment live, and where will what can’t be reclaimed get dumped?
i’m sure a little system thinking wouldn’t hurt the brains of pure engineers.…
I agree with Tm R. comments. We need to look at roof top solar. We have acres and acres of flat roofs on warehouse and industrial building in the Dallas area where I live. I can see some of these roofs bake all summer long in the hot Texas sun when electricity demand is at its peak. We need economical roof top solar solutions. We also need to understand what distributed power means for the grip. Our roof tops are one of our best solar resources.
Just putting an elegant, well engineered solution on the table does not mean it will be adopted.
They are engineering the interface to the energy… not really the energy itself.
It all has to come up against politics, human nature and our capacity for decisive change. Architects have to respect the desire line. MIT engineers should discover the solar desire line.
Roof tops are fine for PV/thin film. Thin film bonded to roofing materials is a wonderful way to harvest a lot of sunlight between “10am and 4pm”.
Roof tops don’t give us power from 4pm until people turn off their TV and go to bed. That’s where thermal solar comes in to play with its ability to store heat and use it later for generation. You’re pretty unlikely to put a thermal solar plant on your roof….
—
People who are worried about ruining the desert might be well served by spending some time in the desert. Not in the scenic parts that we should preserve, but it the “wastelands”, and there’s a lot of that.
Sure, it would be nice to cover none of the desert with solar plants, but the alternative is pretty nasty. More nukes? Continue burning coal?
We’re going to have to give up some land to solar and wind unless someone invents the “magic electricity box”.
How can solar power plants serve as winter baseload?
How far north (south) are they of really good use? (E.g. what is the potential in Sweden
)
Dear Joe,
I have one hopefully minor concern about thermal baseload power that I think should be discussed. This question is: Has enough thought been given to account for the water usage of thermal baseload plants? I had the pleasure of visiting the SEGS plant in Kramer junction last year and the operator tells me that they use water from the Colorado River for the cooling towers, which roughly translates to, “we buy it from LA.” While this is fine up to a point, it seems like we will need a better solution if we expect many GW of thermal baseload power. Obviously we free some water up globally from shutting down dirty power plants in other locations, but this can’t account for the touble moving water from one site to another or the relative thermodynamic inefficiency of thermal baseload compared to say a coal plant, which I assume works at a higher temperature. It seems like the only good solution here is water free cooling towers, which I’m sure can exist, but will undoubtedly cost more to construct and design. Is research currently going on to deal with this problem? If not, should we get started?
Cheers,
Dwight
[JR: The plants will have to use air cooling, and take the efficiency hit.]
A few months ago, I attended a GCEP conference at Stanford. GCEP = Global Climate and Energy Project. These yearly conferences are open to the public, and are well worth spending the few days.
There were many good talks, but one of the most interesting (in my and many others’ opinions) was
Artificial Photosynthesis: Membrane Supported Assemblies That Use Sunlight to Split Water, by Nate Lewis & co of Caltech.
Look first at page 52.
People know that solar->electricity->hydrogen is an inefficient path, and of course, Joe (and many others, including a bunch of us at the conference) have rather strong reservations about “the hydrogen economy”, any time soon.
Nate’s talk was the first I’ve seen that seemed to re-open the issue, although it is still early in research (as noted on p. 52).
To summarize the *goals* and approach:
a) Use (mostly) silicon, (mostly) avoid expensive/toxic/exotic elements, use methods capable of fabbing large volumes, so that costs can get low. Use silicon nanorods and ideas adapted from plant membranes. (page 8, 12 especially)
b) Convert water -> oxygen & hydrogen *directly* rather than via electricity. This of course lends itself to distributed generation, using hydrogen as battery.
Nate was very convincing that the goals were right, and the approach interesting … but this presentation combines biology, chemistry, quantum physics, nanofabrication manufacturing … i.e., nontrivial for a non-expert to assess! Nate was *very* convincing, but then I’ve heard a lot of talks that sounded very convincing…
SO: I’d ask: understanding that this is early research (R1/R2 in my usual scale), what is their opinion of this sort of approach? Are there reasons obvious to experts why this doesn’t make sense?
I expect if you didn’t insist on taking up space there, Magnus W, Sweden could be totally paved with mirrors and heat engines and power the world, provided only some way of six months’ load shifting were in hand. In the Stanford-natgas-booster thread I mentioned my best guess at that.
Coal and nuclear plants have both been built with air cooling, Dwight. On hot days it reduces efficiency.
Joe, as you know, solar PV experts like Larry Kazmerski and Ken Zweibel have long been pointing out that roughly 7% of the U.S.’s 50 million hectares of urban land area covered with solar photovoltaic panels at today’s 10%-efficient systems could hypothetically provide 100% of U.S. electricity. Brownfields could provide most of the land area. Over the longer time period building-integrated photovoltaic systems (BIPVs) could provide half of the power (sources: Kazmerski, L.L. (2002), “Photovoltaic myths–The seven deadly sins,” Solar Today, July/August
16(4):40-43. Zweibel, K. (2004), “ 2nd Generation (2G) PV: CdTe, CIS, and a-Si thin films, and some reflections on the Federal R&D Program,” presentation at Rice University,
http://www.nrel.gov/ pv/ thin_film/ docs/ zweibel_rice_2004_2g_thin_films.ppt). Also, as you know, research calculations suggest that vehicle-to-grid could stabilize large-scale (one-half of U.S. electricity) wind or solar power with 3% of the fleet dedicated to regulation for wind or solar, plus 8% to 38% of the fleet providing operating reserves or storage for wind or solar (Kempton, W. and J. Tomić (2005), “Vehicle to grid implementation: From stabilizing the grid to supporting large-scale renewable energy,” Journal of Power Sources 144(1):280-294, June 10, http://www.udel.edu/V2G). The water benefits would also be considerable, given that solar PVs use nearly two orders of magnitude less water than thermal power plants (which now use 40% of total U.S. water). And, certainly there are REAL homeland security benefits from onsite and locally distributed solar PV systems being more resilience to malicious disruption threats or acts of nature, as well as the several hundred other distributed benefits documented in Lovins et al Small is Profitable, the Hidden Economics Benefits of Making Electrical Resources the Right Size (2002).
Given where solar PV is today in cost relative to other options, looking at innovative financing mechanisms seems important (i.e., mortgage-lenth financing that enables solar PV to be cash-positive fromt the day of installation), as well as bundling of financial options (private, muncipal, utility, state and federal components given the social benefts and values noted above). We also know that nano-scale production R&D indicates potential solar PV efficiencies several-fold higher than today, and what does this suggest for designing balance-of-system components that can be relatively easily upgraded over time with higher efficiency components, as well as cradle-to-cradle recycling operations of salvaged and end-of-life components.
Along the lines of the last point in your post:
Is there a specific way / how should government invest in solar energy to provide a catalyst for commercial solar energy? Are there specific barriers to deployment, or is it mostly price? How will we know when government money is not being spent wisely on solar, and how do we know where to redirect that money?
Joe:
I assume you know the MIT press office has declared that one of the panel members, Prof. Nocera, has invented a new hydrolysis anode material and electrolyte that will save the planet. http://www.realitybase.org/ journal/ 2008/ 8/ 6/ mits-press-office-saves-humankind.html Instead if just expressing your skepticism about that and moving on, why not draw them out on why they think this is a good way to store solar energy for later use. Even if it allowed for electricity generation to be extended into the evening, what about cloudy days? What’s the storage plan for that? (Same question applies of course to solar thermal.) Do we have to have standby generators or a grid (to be mostly a standby facility) extending outside the shady region? Maybe this would lead to an analysis of different complementary mixes of power sources that would work in several selected regions–a systems analysis instead of heroically inventing the one device that displaces all others. And maybe they could develop some comforting responses to legitimate utility company concerns about dispatchability of solar.
Winter thermal solar might need to be shipped in from Baja. I spent a couple of winters on the Sea of Cortez and I don’t remember many cloudy days.
And remember that the wind tends to blow a lot in the winter, waves are a lot stronger, streams have more water….
Take a look at the European model of how they are considering hooking together a grid from Scandinavia south to North Africa in order bring all sorts of sources together. Winter sun in Morocco and pump-up storage in the northern mountains.
http://www.dailykos.com/ storyonly/ 2008/ 12/ 9/ 63137/ 5623
And we hang on to all our existing natural gas turbines. They can be brought on line quickly if needed.
We’d even be ahead if we could park our existing coal plants for most of the year as well build out new sources.
Joe – my solar questions are pretty much the same as yours.
The big question that I don’t see addressed is how can be best bring all this stuff together – solar, wind, geothermal, existing hydro and nuclear, …?
I’d love to see a “SimCity” type model something like the first picture in the dailykos link above. But interactive.
Let knowledgeable people sketch in their versions of where the HDVC lines need to run, where the wind and solar farms are best situated.
And then let us crank up the production from each of those sources and see what each change would cost. Perhaps provide readouts for how long it would take to construct each component, how many jobs would be created (construction vs. long term jobs).
Give us a total electricity goal that must be met. Include a negawatt slider and ways/costs to produce the negawatts.
I think the country could benefit from a “plan”. Something that we could get our brains around.
Even in Redmond, cloudier and rainier than Seattle, a solar hot water heater, using but a small fraction of the roof, does most of the hot water heating.
In a recent tlak by a VP (engineer) of the local power company, he mentioned favoring distributed generation: “improves reliabity”.
dwight: “I have one hopefully minor concern about thermal baseload power that I think should be discussed. This question is: Has enough thought been given to account for the water usage of thermal baseload plants? … Obviously we free some water up globally from shutting down dirty power plants in other locations, but this can’t account for the touble moving water from one site to another or the relative thermodynamic inefficiency of thermal baseload compared to say a coal plant, which I assume works at a higher temperature. …”
Clean, Green Solar Power Falls Short in Achieving Water Efficiency
“… Department of Energy: a coal fired plant uses 110 to 300 gallons per megawatt hour; a nuclear plant uses between 500 and 1100 gallons/MWh; and a solar parabolic trough plant uses 760 -920 gallons/MWh.”
“… The one solar plant in Arizona the Sierra Club is supporting is the Solana plant, a 280-megawatts plant proposed for the Gila Bend area. Bahr says, ‘It is on private land, not public land, and is currently agricultural land. Depending on how you calculate it [the power plant] will use 75 to 85 percent less water than the current agricultural use. It is still a fair amount of water but it is much less than it takes to grow alfalfa.’”
My question would be:
What is your assessment of the expectable cost curve for high temperature thermal energy storage with the following technologies:
1) molten salt only
2) thermocline (with variety of filler materials)
3) concrete or other solid
Cost of replacement of fossil generation by solar thermal electric depends in part on (updating) these numbers.
There’s a least one thermal solar plant that’s being design/built using graphite for heat storage.
http://www.aussierenewables.com.au/news/?tag=storage
The sweet thing about this is that the heat goes directly into storage and is pulled out later for steam when needed.
There’s another thermal site being built in which the heated material is located below ground (easier to insulate) and a large concave mirror is placed over the storage material. A surround of small tracking mirrors feed sunlight to the overhead mirror.
A combination of idea such as these can create lots of power for late afternoon/early evening peak needs.
I’m not sure what water would be needed to operate a graphite/closed loop turbine system. Aside from washing the mirrors once in a while….
The cost of installing a residential PV system is about $10/Watt (peak). The breakdown of the cost is (roughly):
$4/Watt – the PV panels themselves
$4/Watt – installation
$2/Watt – inverter
PV manufacturers are racing to reduce the cost of the PV panels – $1/Watt seems to be the holy grail. That leaves installation and inverter costs.
So why not eliminate those costs completely? Eliminate installation cost via building integration. This is a non-trivial architecture and engineering task, since roofs must be reliable (and repairable) for decades, in all climate zones. Done right, it would actually produce savings by reducing roofing material and labor costs.
Then eliminate the inverter cost with a DC standard: a standard plug and socket and voltage, just like the standard 120 V, 60 cycle AC with the NEMA5-15 plug and socket we use today. Why a DC standard? PV produces DC. All electronic devices use DC, as does LED lighting. A DC standard would eliminate all the costs and losses of converting DC to AC, and then back to DC for your computer, TV, cell phone, etc.
A DC standard would also eliminate all those cell phone plug adapters, and should be universal – in every home and office in every country, (and in every car, like the cigarette lighter, which is indeed a 12V DC standard, but a weak one.)
A DC standard doesn’t require any new technology, but would need the cooperation of manufacturers and standards bodies. The challenge is finding the optimum voltage — not too low, not too high.
So, what is the best voltage for a DC standard? And where do we start?
By the way, did I mention that if BIPV (building integrated photovoltaic) systems cost $1/Watt, it would become the dominant form of power generation?
And with a DC standard (to eliminate inverters), every electronic device would cost less, too (since there would be no need for AC to DC conversion).
Been there, we moved on…
Lots of off the grid people ran 12 vdc systems until inverters got better and affordable. I know several “offers” and none continue to run anything on DC. (Except perhaps one DC light in their battery room. In case their inverter goes out and they need light to work.)
Low voltage requires large wire to move any appreciable power any appreciable distance. Sure, you can run your cell phone off a small wire, but not your refrigerator.
So we’d be looking at moderately high voltage DC? Then we’ve got to step it down for devices and there’s a loss.
Then, what to do with the power you can’t use? You can’t sell it back to the utility company as they want 120/220 vac. (And don’t even think about changing over every 120/220 vac piece of equipment in the US – and elsewhere. Every range, refer, dishwasher, water heater, AC, heat pump, hot tub heater, iron lung, ….)
Inverter price is probably high due to economies of scale. If we made 10x as many there would be more competition and lower prices.
BTW, First Solar reports that they are making thin film for $1.08 per watt and selling it for $2.50. Not selling to us yet, but to large scale purchasers.
Our $4 prices may start falling soon as panel stock seems to be building up in warehouses. Manufacturers are cranking them out faster than installers can use them.
Installation is the place where prices could best be cut. Thin film bonded to roofing panels – all in one, screw down to the roofing deck, snap the leads together.
That said, I’m not convinced that residential rooftops is the way to go. I went out four times yesterday to clean the snow off my panels. How many people are going to do that? How many are going to adjust the angle 4x a year?
–
Oh, yeah. Not all cars are 12vdc….
I would really like to see some reality based cost figures, for example, comparing the cost of solar thermal + transmission upgrades vs. wind, gas, coal, and nuclear. I a similar way, I would like to see the costs of PVs + net-meters vs. utility scale solutions with transmission upgrades, and under what circumstances the decentralized solution is more economical. Thanks for asking our input, Joe.
Bob –
You hit the issues squarely, mainly the choice of voltage.
High voltage is best for the large appliances, which are mainly in the kitchen and utility room (or in the hospital.) 110/220 VAC stays for them.
So is there an optimum DC voltage which would serve the great majority of electronic devices (and thus most rooms of a house)? Maybe it’s a pair of voltages (as Earl Killian suggested in an earlier thread) such as 12/24 or 36/72. We wouldn’t have to rewire every house — power strips could have both AC and DC outlets. Battery backups for computers would be cheaper and more robust, too.
Having DC readily available at a standard voltage would help spur deployment of LED lighting, which is already more efficient than fluorescent. Today LED lamps need a built-in AC to DC converter.
Yes, grid tie-in would still need an inverter. But it suggests that as PV prices come down and appliances become more efficient, the economies tilt more people to off-the-grid (”offers” – what a concept!)
That still leaves installation cost as a huge opportunity.
All this suggests that while a DC standard might not help energy guzzling first world households like mine, it could be huge for what Thomas Friedman calls energy poverty, the 2 billion people without electricity. (The poorest “offers”.)
A DC standard connects a little PV panel, battery, cell phone, LED (and maybe an OLPC) at the lowest cost because everyone is manufacturing to a standard DC voltage.
Anyway, thanks for reading the comment.
Joe,
As someone who just finished up there who has read the other 3 MIT reports (for some reason the Jeff Tester led Geothermal one didn´t get mentioned), I have a few ideas you can kick around with John and the rest of the team.
I absolutely agree with your point about how should the DOE allocate resources to best augment the scale-up. But I look at the US energy problem and see it in two parts: the first being the scale-up of renewables, and the second the integration into the grid. Clearly these two happen in parallel and not series.
But between the two, I see the first as being the easiest. I look to European nations like Germany, Spain, UK, and Denmark and see that “magic number” of renewable electricity hovering around 10-20%. They managed to get there in a relatively short amount of time, and I think the US can replicate their success.
The main hurdle for the EU now is how to integrate into the grid even more. I hope Gore knows something I don´t when he claims we can have 100% renewable electricity within a decade, but most of my fears comes not from the generation side as from the lack of grid infrastructure to hold it. And yes, I am aware (although likely insufficiently so) of the several smart grid propositions in existince.
What I think the MIT team could lend is some very sophisticated modeling and insight to adequately explain what happens when the US hits 20% renewables, and specifically what role does solar play to fill in the other 80%. Clearly this would not be the central focus of the paper (likely only one of the later chapters), but some discussion points could be:
-What transmission upgrades we need to enable all the clever ideas regarding smoothing intermittent solar resources (wind/solar, solar/hydro, solar/ccgt like FPL is building in Florida)
-Where we need it
-What it will cost
-How we should achieve it…maybe even some policy discussion if you can stomach it
I fear that if we don´t start asking more tough questions like this, we will hit the exact same “20% wall” as the EU.
And please provide your own comments & feedback too. I´m a new reader here but have been very impressed with its quality.
I read with great interest Mark and Bob’s discussion of a DC standard. It sounds like the solution for a typical “first world” consumer is a hybrid solution. Where practical they would run a bi-level (e.g. 12/24V) home grid for LED lighting and any other devices (computers, TVs) capable of using that standard. The remaining PV power would go through an inverter to the existing 110/220 AC wiring. Even if only 1/4 of the power is used as DC, that still gains efficiency for that portion, and reduces the size/cost of the inverter.
New construction could put both sets of wiring throughout the home. If I were to retrofit my home (which has no wall or ceiling access without cutting holes…) I could at least run the DC through parts of my attic for LED lighting on the top floor and drop it down a couple of walls for computers on the same floor. It wouldn’t be practical to try to get DC anywhere else, so the rest of the house would take the DC->AC->often back to DC hit, but at least it would be some pure DC.
Bob:
“Installation is the place where prices could best be cut. Thin film bonded to roofing panels – all in one, screw down to the roofing deck, snap the leads together.”
Damn strait. I live in Seattle(!), and I’d strongly consider this – either at some point in the future when the existing roofing material needs to be replaced, or sooner if the new ones could go over the old ones.
Bob:
“That said, I’m not convinced that residential rooftops is the way to go. I went out four times yesterday to clean the snow off my panels. How many people are going to do that? How many are going to adjust the angle 4x a year?”
That’s somewhat contradictory – if it’s thin film bonded to roofing material, no one will be adjusting the angle, ever. When snow is falling, there’s no sunlight. When the sun comes back out, sure most people will just be lazy and wait for it to melt – c’est la vie. That will only make the installation a few % less effective over the course of a year in most climes.
(corrections to my post)
110/220 is of course not the only AC standard in the developed world. So where I said 110/220, substitute the local AC standard.
strait -> straight. It looked wrong when I typed it, but I wasn’t “thinking straight”.
From the announcement:
Harnessing the particular strengths of this institution, [Deutch] says, the report will be “MIT speaking in a way only MIT can — with an interdisciplinary focus, addressing in depth an area of energy.” The report will have “a breadth of focus that encompasses technology, economics and policy, and looks at how these need to work together.”
Two basic questions:
What characteristics of deployment would generate positive feedback, internationally and nationally, for increasingly ambitious climate/energy policy?
What government action is required to initialize deployment trajectories with such characteristics?
Too many platitudes are bandied about in conjunction with these questions,
in the clean development & clean tech context. They deserve a solid interdisciplinary in-depth investigation.
Hopefully the MIT study will at least seek to provide a thorough review. The questions seem pretty important to the future of solar energy.
GFW/Mark – we’ve wandered away from Joe’s initial request, but that’s how stuff goes in a discussion…. ;o)
We really need more sophisticated software for sites like this.
That said, Mark’s initial problem was the cost of inverters to get from the DC produced by a set of residential panels in order to integrate them with the grid.
My points would be that:
1) it would be very, very much too expensive to convert the existing local consumption grid to DC,
2) if we were to use more inverters the price would likely be closer to $1k than $3k, and
3) even if we switched the grid to DC we would still have to have voltage controllers to match specific sets of panels to the grid. There’s some cost there as well.
Remember that a cloud can drift over your panels, dropping them below grid level. Simple panel controllers simply disconnect the panels from the system when panel voltage is less than system voltage (otherwise you get backflow). More sophisticated controllers trade amps for volts and convert voltage to system levels so that power is not wasted.
—
Low voltage DC standards? Absolutely! One power brick per house/office. Lots of us would retrofit a small wattage DC system. Building into new construction would cost little.
BTW, I recently purchased a “netbook” for travel. Wonderful little puppy (Asus PCee). When turned off the power brick pulls ZERO watts. Every piece of electronic gear should adopt this design.
—
Solar roofing – when (and I don’t even say “if” any longer) thin film drops to well below $1 per watt we can stick it on roofs and not worry about adjusting angles. We’ll probably see houses of the future designed with significant portions of their roofs oriented south and pitched for optimal “summer AC” or “winter heat pump” depending on the climate.
And we could install automatic hot water wash downs for snow if it makes economic sense.
http://www.dailymail.co.uk/ sciencetech/ article-1023315/ Machine-clean-greenhouse-gas-breakthrough-war-global-warming-say-scientists.html
What is the likelilhood of implementation and and efficacy of large scale deployments of Co2 scrubbers like the prototype discussed in this article?
Bob/GFW –
Thanks for thinking broadly about pros and cons of a DC standard, and what the topology would look like. I believe it addresses Joe’s request squarely:
“What questions would you like answered of MIT’s Future of Solar Energy Study?”
The point is to make solar cost effective. A DC standard removes some complexity and waste from the system. Yes, our 110/220 VAC system with central power plants and a grid has served well for 100 years, and CSP (Joe’s solar thermal baseload) fits right in with that model.
But now we produce billions of DC devices annually, each with its own custom power supply: cell phones, music players, computers, printers and other peripherals, TVs, cameras, LED lighting, and more. DC devices are the fastest growing, highest value-added electricity users. PV is a fast growing DC source (with fuel cells and thermionic knocking on the door), and the DC -> AC -> DC conversion impedes penetration.
You have helped think through a DC standard for us first-worlders. But it would add huge value to the 2 billion energy poor people who have no access to a grid.
In “Hot, Flat, and Crowded”, Tom Friedman asks of every promising energy technology, “Does it scale?” PV, with a DC standard connecting directly to electronics, batteries, and LED lights scales all the way down to the poorest, struggling to get just their first watts.
Just a thought.
I spend a good portion of most years in less developed places (non-urban India, Bangladesh, Myanmar, Laos, Nepal during the last three years, for example). I think you might be surprised how far 110/220 AC has reached. Even the “towns” in Nepal that are only reached by foot are starting to get the grid. (Often a local micro hydro-supplied grid.)
Low voltage DC and batteries are good technology for places even further from the grid, but those places are disappearing.
I can see great value for standardizing low voltage DC “stuff”. But even if we paid the trillions and trillions of dollars it would take to change the world’s grids to high voltage DC we would still have the expense of stepping it down to lower voltage DC for our cell phones.
I think the really critical question is “How can we bring down the expense of hooking small privately owned arrays to the grid?”.
It is not clear that a single (even pair of) DC voltage will meet all needs.
Rectifiers with voltage stepdown (power supplies) are quite inexpensive.
Looki8ng at the rectifiers/inverters needed for HVDC shows that the losses at each end are about the same, but 1/2 per cent. Note that the same technology is used both for the rectification and the inversion.
I strongly suspect the cost of low voltage inversion will go down.
Bob -
Wow. Thanks for the observations on developing world power and 110/220 VAC. I think though that as PV approaches $1/Watt it will be able to complement the microgrids you’ve seen.
Your question is well put; I would broaden it to: “How can we best exploit private (usually rooftop) intermittent PV?” Less expensive grid tie-in is one answer, batteries (including UPS) and storing hot (and chilled) water add value too.
Commercial users with a steady demand for DC power (computers, POS terminals, lighting) could use all their rooftop PV, and supplement it with the grid and/or CHP.
I have two suggestions for the MIT study, partly covered by comments already made above:
1. It would be very helpful to define a rigid standard metric, or perhaps a small number of metrics, that would allow clear intercomprison of the different technology options. Perhaps establish a standard energy demand scenario, and determine what fraction of demand each technology could meet on its own, and what fraction would require additional investment in other technologies (transmission and storage) to meet.
2. Some analysis of the “learning curves” for solar systems – are we still on track with the projections of a few years ago for price vs volume?
I thought, as Joe wrote on December 14th in his post on the Stanford study Part 1, “Wind, solar baseload easily beat nuclear and they all crush “clean coal”, under a table of figures, he said this:
“tells us all we need to know — the future is inevitably wind, solar, and geothermal.
MIT sounds like they’ve missed the boat. Everything anyone needs to know is already known, eh? Or did I miss something?