Dark days ahead for roof-top solar

Posted on October 28, 2015
Posted By: Davis Swan
Topic: Solar

I am not a big fan of roof-top solar.

Roof-top solar panels are expensive and inefficient to install. These systems also cause major issues for utility grids because of the need to handle the bi-directional flow of electricity to and from the customer. This additional complexity unfairly imposes additional costs on all electricity users who do not have roof-top solar panels. I have also argued that the value of all solar energy, including roof-top solar will decline significantly between 10:00 am and 2:00 pm as supply begins to exceed demand.

Over the past 2 years some of these concerns have become reality in many parts of the world.

The problems associated with integrating large amounts of roof-top solar into the electricity grid on Oahu have led to a steady decline in installations.

On October 13, 2015 it was announced that the Hawaiian Public Utilities Commission was ending the Net metering program whereby customers with solar panels received a credit for any electricity they returned to the grid. The inequity of that approach was that it equated mid-day electricity generation that was increasingly problematic and would exceed demand on some circuits to expensive peak demand evening electricity that was required by all customers including those that had roof-top solar panels. The new system will pay customers something close to market value for their solar power generation – a value that will be much lower than what was received under the net metering system. Customers with solar panels will now also have to pay a minimum monthly bill to help cover the costs of servicing their more complex inter-connections. Most observers have concluded that the impact will be a further reduction in roof-top solar additions.

Hawaii is not the only place where roof-top solar installations are declining significantly.

The world leader in roof-top solar for most of this century has been Germany. I personally have never understood why a country at a latitude of 48 degrees would spend hundreds of billions of dollars subsidizing roof-top solar when there is very little solar power available in the winter – the peak demand season for electricity use in Germany.

I don’t know if this reality has finally become apparent to the Germans but subsidies have been decreased significantly in the last few years and solar panel additions have dropped pretty dramatically.

As in Hawaii, other jurisdictions with high penetration of roof-top solar utilities are requesting and being granted the right to charge customers with solar planels a fixed monthly fee. They are also being allowed to pay customers market based prices for excess solar power rather than a fixed feed-in tariff.

In Spain the government has dramatically cut subsidies and support for solar power development over the past few years most recently targeting residential battery storage systems. In September the Spanish government won a court case that was an attempt to force restoration of these subsidies.

Most recently the Tennessee Valley Authority published the most comprehensive study I have seen yet on the costs and benefits of residential roof-top solar. The study concludes that the amount paid by TVA for rooftop solar is still higher than the true value to the system despite the fact that the TVA has reduced the payments from $.22/kwh to $.12/kwh since 2012.

When you consider these developments it is difficult to see the installation of roof-top solar panels maintaining the pace of the last few years in the developed world (China is a different story where government mandates will ensure that solar installations continue at an accelerated pace). In my opinion that is a good thing. Development of solar power needs a reboot to take a more rational and less subsidized approach.

Believe it or not I am actually a big believer in solar power. At latitudes below 35 degrees N/S I think it is absolutely the best renewable source available. By pairing concentrated solar power installations which include molten salt storage with photo-voltaic solar panels to reduce costs it would be possible to build plants which can supply electricity 7x24x365 (the Gemasolar plant in Spain can already do that). Utility scale solar plants have the added advantages of being easily equipped with sun tracking which significantly increases plant output.

By centralizing solar panel installations it is also much easier to integrate this generation into the regional grid and to supply battery or other short-term voltage stabilization storage technologies. A recent example of how effective this approach can be is the Kauai Island Utility Co-op’s drive to install 30 MW of utility-scale PV solar by the end of 2015 some of which will have battery backup. These facilities will consistently generate more than 50% of the electricity required at mid-day on the island.

For many parts of the world including the Southern United States, Mexico, Southern China, all of India, north Africa, the Middle East and many other areas solar power is the obvious choice when it comes to renewable energy. Innovative projects such as the Khafji Saltwater Desalination plant point to a future where not only energy but abundant fresh water will be available in these areas. Energy storage systems based upon molten salt, batteries, or other technologies need further development and significant R&D funding needs to be directed towards this effort. But there is no doubt that in the long run solar power in equatorial regions has enormous potential.

North and south of 35 degrees latitude I believe that other technologies including wind, geothermal, and hydro-kinetics offer a better value proposition for both subsidies and R&D funding.



Authored By:
Davis has been involved with energy policy development and the exploration of innovation in energy use throughout his career. For more than 20 years he worked in the oil & gas industry where he had broad exposure to the technologies used in the development of natural gas, conventional oil, heavy oil, and tar sands resources. He has also acted as the energy policy advisor for the Official Opposition in the

Other Posts by: Davis Swan

Related Posts



October, 28 2015

Ferdinand E. Banks says

This is the article about solar - and especially solar in Germany - that I have been looking for. I can add too it that despite all the good things we have Heard about solar in Bavaria, there is talk of supplying that part of Germany with wind Power 'harvested' offshore places like Hamburg. I really wonder what people are after in that country: there is a tsunami in Japan, and they start locking down nuclear facilities in Germany.

October, 28 2015

Richard Vesel says

I think you are entirely misreading the situation on Oahu, in the following manner:

Between 2010 and now, over 1.5 MILLION solar panels have been installed on rooftops in Oahu. That's over 1.5 panels per capita (and 1 for every resident of all the islands combined!), making it one of the most dense solar installations on the planet, at a nominal 300W of "capacity" per capita. Compare that to Germany, who has been developing solar installations for two decades, in a much more industrialized society, having an installed capacity of slightly less than 40GW for 80M+ people, or bit less than 500W of solar "capacity" per person.

The rolloff in orders reallyresults from two things. First, the area, mostly in Honolulu, is nearly saturated with rooftop installations. Secondly, with the incentives now greatly reduced, there is reduced financial argument to capture the remaining opportunities.

Solar in Hawaii tho, is far from dead. The state has committed to being 100% renewable by 2045. So the scenario that has already played out for 2/3 of the population of the entire state, located on Oahu, will now play out more gradually for the other islands, in a slower but steadier stream of rooftop and solar park installations. What will also now follow is integration of granular and utility scale storage, power flow and stability management via microgrid, and integration of more wind as well, which blows quite steadily in many areas of the islands, as the wind farms on Maui and the Big Island have already been demonstrating.

RWV (Oahu bound in 2016)

October, 30 2015

Ferdinand E. Banks says


I would like the above discussion to continue...please.

October, 30 2015

Richard Vesel says

Some info to feed to your curiosity, Fred:


800k panels with micro-inverters representing ~60% of installations, so 800k x ( = 1.3M, but the actual figure is 1.5M according to other sources.

Some other background:



For anyone who says solar is not effective during the right times of the day, I can prove otherwise. Sitting on my desk next to me is a system load curve for a day in late June for the island of Oahu. That island's total power demand curve reaches to 95% of daily peak demand by 10AM, and sits in the 95% to 100% zone until 6PM. These are prime solar generation hours. That curve was provided to me by personnel at HECO during an energy efficiency study, so it is undebatable primary source material.


October, 30 2015

Richard Vesel says

that calculation should have read as: 800k x (1.0 / 0.6) = 1.33M

October, 30 2015

Malcolm Rawlingson says

I was in Kona last week - on the Big Island of Hawai'i - now in Brisbane Australia from where I write. I toured around the island and did not see a preponderance of solar panels on roof tops so one should try to understand why that is given the subsidies available. The problem lies in both David Swan's and your perception of Hawai'i as a sunny part of the world where solar should work well. There are some locations where it that description applies however before you make the usual broad brush assumptions you need to understand that Hawai'I (The Big Island that is) has a very wet microclimate and only some parts of it have the sunny climate that the State is famous for. As you likely know in order to have rain one needs clouds. Clouds block sunlight from reaching solar panels and certainly on the Kona Coast rainfall as high as 190 inches are not uncommon. Being a curious fellow I asked some of the locals why they did not use solar panels and the answer is simple....even with high electricity costs it is still uneconomic because it is often quite cloudy. Also add to the fact that there are two volcanoes on the island that constantly emit clouds of volcanic gases and it becomes clear that no amount of subsidy is going to affect the availability of sunlight. Even in Hawai'i the Sun does not shine at night. Also worthy of note is that the amount of sulphur and other gasses release by the volcanoes is 700 times the combined total of all the fossil fuelled power plants in the State which makes the efforts by the State to phase out fossil fuels rather puny and irrelevant. They are doing it for economic reasons rather that "climate change" but that is as good an excuse as any I suppose. So have a great time in Oahu. It is one of my favourite places on earth. Been there many times and please do not feel guilty about the amount of carbon dioxide you will be generating via the jet engines that will carry you there as it is an awfully long swim from Los Angeles airport and I forgive you. Malcolm

October, 30 2015

Malcolm Rawlingson says

David, I liked your article. One of the most objective and realistic articles on solar power I have read. As you know I am a proponent of large scale nuclear power but that does not mean that I am against solar power. In some countries it is by far the best option...but not Germany or regions that do not get sufficient sunlight. The only solar installations I would endorse are those that do not use the grid as a back up power supply. That is, stand alone systems that provide their own storage for use in the overnight hours. Of course when you do that it exposes the inherent weakness of solar installations which is that you need enough panels to produce electricity for the demand during the day and a second set to charge the battery so there is electricity available at night. In addition of course you need a battery as well so the capital cost suddenly becomes much greater. Grid connection and subsidies mask these weaknesses. My only criticism of your article is the assumption that between certain latitudes it is always sunny. Generally that is true but every situation needs to be looked at in detail and this is where such assumptions come unglued. In desert areas for example it certainly is sunny but these areas are also prone to severe sandstorms that not only obliterate the Sun but also present a large clean up job as the panels are now covered in sand. If you base your power production on solar without some sort of back up plan the expectation should be that the lights will go out every time there is a sandstorm. Saudi Arabia is embarking on a sensible plan that includes both solar and nuclear as part of the grid supply and that makes good sense to me. Similarly the UAE is building a new 4 unit plant in Barakah on the Persian Gulf and is also embarking on solar power plants too. This too is a sensible use of solar power. In other words rather than these blanket one size fits all solutions proposed by many each situation needs to be judged on its merits. Solar roof panels may be OK on Oahu....but not in Kona simply because the circumstances are different. I would not be a proponent of a nuclear plant on Kona because the area is volcanic and not suitable for such installations in my opinion. Similarly I am not convinced that nuclear is the best option for Japan. While nuclear plants can easily withstand severe earthquakes I have my doubts as to their ability to withstand volcanic ash and lava flows. Nuclear power is not a one size fits all solution neither is solar or wind. However nuclear is a one size fits most whereas solar and wind is a one size fits a few situations. The choice of power supply should be based on proper evaluation of all options -not on blind dogma. Malcolm

October, 30 2015

Ferdinand E. Banks says

Of course the big question that not only is not answered but posed, concerns why a tsunami in Japan would lead to reactor facilities being locked down in Germany. Not only the big question but the only question Worth fooling around with, because as I informed the audience at an energy Conference, science will favor nuclear, GUARANTEED, and it is a lottery with the others, Even if we dont have the nuclear we need and deserve now, which we do, it will be available by mid-Century.

Nobody has ever Heard me say anything good about the International Energy Agency, but now that they have given their estimate of how much energy wind and solar are providing globally, I might have to congratulate them on their honesty. Come to Think of it, that might be going too far. Instead I'll congratulate them on being able to stay in business despite the mismatch between their salaries and their competence.

November, 02 2015

Richard Vesel says


Kona, on the west side of the Big Island, is NOT under the purview of HECO, which is where Oahu is, and where the incentives were. What is the difficulty in understanding that Oahu has over 2/3 of the total population of the entire state of seven islands, at nearly one million people, out of 1.3M for the seven islands? So your "observation" or "comparison" is useless, like saying that the power systems of Saskatchewan somehow represent those of lower Ontario. Honolulu, on Oahu, is a major metropolis, while Kona is a quiet resort town, where millionaires and billiionaires (such as Michael Dell and Ken Griffin) have their getaway mansions. (I was in Kona May 2014, and return there next year for some more fun in the sun).

Did you by any chance drive down to South Point, about an hour south of Kona, to see the 20 wind turbines there, capable of providing all the power needs of the entire Big Island if running at full capacity? Perhaps not... But I did. Besides spectacular scenery, it is a perfect wind farm location due to constant yet prevailing winds off the ocean, placed so as not to disturb the nearby beautiful lava cliff seacoast and one of the world's few green sand beaches. (do not confuse this new set of turbines with the original windfarm, which has since been disassembled and replaced with this one recently)



Also the Big Island is a reasonable candidate for geothermal power (24/7 power!) due to its proximity to volcanic activity, under circumstances similar to Iceland (which has a nearly 100% renewable electricity production mix)

The Japanese, being on the edge of the Pacific "Ring of Fire" should also be capable of exploiting geothermal power production.


November, 02 2015

Ferdinand E. Banks says

Interesting Richard, but if you are interested in giving a comprehensive lesson on wind Power, I recommend that you book a visit to Tierra del Fuego. I haven't been there of course, but I have Heard great things about the placeI.

Of course, wind is not going to make the cut in Germany, regardless of what Merkel & Co think. Wind and solar for that matter. The Germans are going to do their thing with coal and imported Power, although eventually they will see the light, and challenge France for the most nuclear intensive country in Europé title.

As for all the Power needs of the 'Big Island' being supplied by 20 wind turbines, well...I wouldn't Believe it if I saw it.

November, 03 2015

Reigh Walling says

Richard Vesel states that solar on Oahu is "effective" because the peak solar output lies within the period between 10 am and 6 pm when load is between 95% and 100% of daily peak. One could then draw the implication that solar provides significant contribution to system capacity. This is far from correct, however. The solar is effective in displacing energy production from fossil-fueled generation, but contributes very little to system capacity. At 6 pm, even under clear sky conditions, the solar output will be a small fraction of its rated capacity. Furthermore, the load curve is likely to be close to the 95% point shortly after 6 pm when the solar output drops to zero. Therefore, the maximum possible contribution of solar (without energy storage) to system capacity is on the order of 5%-10%, even if skies are always clear. Skies in the evening, however, are not consistently clear and thus the capacity value is diminished even further.

There appears to be a widespread misunderstanding by solar advocates about the difference between energy and capacity. There is, of course, benefit to the renewable energy provided. It displaces burning of fuel and the resulting emissions. It does not, however, significantly diminish the need for dispatchable resources such as fossil-fuel generation or pondable hydro. The capacity value of any variable renewable generation is greatest when it is aggregated over a large geographic area, due to meteorological diversity. The capacity value is least when considered at an individual distribution feeder level where one cloud can shadow the entire area. This truth does not seem to be well understood by utility regulators, consultants performing "value of solar" assessments, and poliitical policy makers who seem to believe that promotion of distributed solar will replace the need for wire. Energy storage changes this, but rarely does energy storage provide an economically attractive alternative.

November, 03 2015

Richard Vesel says

@Reigh: Why would one then draw a(ny) conclusion about contributions to capacity? The word capacity has not been part of this discussion, nor did it appear once in my comments about solar power generation fitting within the prime time of the demand curve.

There are multiple ways in which to deal with the variability of renewable energy from solar and wind. Short and intermediate duration storage, smart grid stability controls, STATCOM with storage, and a low-cost shadow fleet for a percentage of the total capacity requirements.

"ABB has raised the upper limit for microgrid renewable-energy penetration without storage. Research by the company suggests up to 50 percent intermittent generation could be admitted to microgrids without needing storage, provided that automation systems are in place to keep the grid stable.

Traditionally, the upper limit for renewable energy penetration in microgrids without storage has been around 40 percent of total load.

ABB looked at a range of microgrid scenarios, including low-penetration setups where renewable energy covered up to 30 percent of peak load, medium penetration at 50 percent, and high penetration at 100 percent."


@Fred: You are correct, my mistake. The 20x1.5MW South Point turbines would only supply all of the residential power demand, about 30MW. Total Big Island demand is running somewhere north of 180MW average.


About 1/3 of the current load is covered by sustainable energy sources.


November, 03 2015

Robert Bullard says

The "second wave" of roof top solar will begin when the industry produces a true solar roof, one where the entire roof deck and weathering surface is the solar panel array mounted on a roof framing system with a design to optimize support, maintenance and cleaning of panels without risking the life of the cleaner or damaging/degrading the panel surface. Large, uniformly expansive roofs with minimal and/or clustered roof penetrations will lead the way. Unfortunately, in higher latitudes the low angle of tilt of mono-slope roofs (for safe access) combined with less than ideal azimuth orientation will be the ultimate deterrent to roof top solar of all varieties on an extensive scale.

In a related vein, what amazes me is that the grid utilities a so slow in the take up on intermittent renewables and the need for charge storage to arbitrage the functional and economic value of the charge. For example, by far the most cost effective grid-wide storage method (at least 75 per cent round trip energy efficiency) is pumped hydro, a method that does not require being on a river gorge somewhere or even a river, for that matter, just a modest source of make-up water for the closed loop pair of lagoons and proximal topographic features. Several years ago I took a casual look at the opportunity at wind farm sites of the US West and found that about one in three had the water and topography within the foot prints of wind farm grids to situate lagoons with capacity that would match the 24 hour, 100 per cent capacity factor of the wind machines.

November, 04 2015

Ferdinand E. Banks says

I know precious Little about wind and solar, and I don't want to know more. Amory Lovins has praised the success of solar in Germany, but I am not to claim that he is lying. He is ignorant, like Another well known gentleman in the U.S. And I repeat: COAL AND IMPORTED POWER carry the load in German, not solar and wind. In other words, I am helping to pay for the electricity being used in Germany.

Can I prove it. Well I Think that I have a shot if some Money Changes hands. But while I will put together the economics, I will not talk with or glad hand the persons who give me this cash.

November, 04 2015

Richard Vesel says

ZERO penetration mounting is available, and selling vigorously! (My son-in-law runs the warehouse that ships this product)


This company can just barely keep up with orders, and could do a lot better, in my opinion. But anyway, watch the video.


November, 08 2015

Davis Swan says

This post has generated a lively discussion which is exactly what I hope for when I write articles. I do want to restate a couple of points;

1) I am a huge advocate of solar power between 35 degrees latitude North and South. It is absolutely the most cost effective and reliable source of renewable energy. I just don't think that roof-top solar compares at all well with utility scale solar on any metric.

2) I disagree with Richard Vesel's contention that solar matches demand patterns in Hawaii or in any part of the world for that matter. The http://instituteforenergyresearch.org/solar-energys-duck-curve/ "Duck Curve" clearly demonstrates that there is a need for energy storage in some form to time-shift peak solar by a few hours to better match demand. Once we get the energy storage technology figured out there is no stopping solar. Its only a matter of time.

November, 09 2015

Richard Vesel says

Then Hawaii is in "the zone", would you agree?

Hawaii Lat / Long LATITUDE & LONGITUDE: Latitude/Longitude: (Absolute Locations) Honolulu: (capital city) 21º 18' N, 157º 51' W Hilo: (second largest city, on the Big Island) 19º 43' N, 155º 05' W

Maui and Big Island solar installation permit backlog cleared for 3000 more installations:


"DOE’s National Renewable Energy Laboratory is helping out on this front by hosting tests of smart solar inverter capabilities with the Electric Power Research Institute and big third-party-solar provider SolarCity. That work led directly to HECO’s proposal to double the amount of solar it’s comfortable permitting on already-impacted circuits, from 120 percent to 250 percent of daytime minimum load, without causing grid-destabilizing voltage problems."

Right now, Oahu has oil-fired capacity that is available as spinning reserve, or on warm standby, to compensate for when the solar curve begins to fall off in between 3-4pm. However, as I stated earlier, that island's demand reaches maximum by about 9:30AM, and stays there for about 11 hours, so as solar's contribution maximizes between (early day 50% level to late 50% level, compared to 100% at noon, in southwest Oahu) between approximately 9AM and 4PM



A very effective means of adding storage would be to use "ice bears", as a sizable portion of the Hawai'i's electrical load is due to air conditioning and commercial refrigeration.


Superficial arguments aside, the technology exists, and better technology is coming down the road soon, to make Hawai'i's 100% renewables by 2045 completely realizable, and rooftop solar will be a significant portion of that solution (and I include sizable commercial rooftops as part of that).


November, 17 2015

Wim Bakker says

David: In your analysis of battery technology could you comment on VRB vs Li-ion in utility applications ? Wim Bakker

November, 19 2015

Davis Swan says

Richard - absolutely agree that solar is the best path forward for Hawaii despite a fair amount of cloud cover at times. I just believe that utility solar like KIUC is doing is preferable to roof-top solar. Distributed solar has some advantages in terms of resiliency but it is expansive to install, roofs are never at the optimal azimuth and pitch, and the need for inverters adds cost and complexity. Maybe the contribution towards the total cost by the homeowner makes it worthwhile but I am skeptical.

Love your suggestion about the "ice brears" - there are a few chilled energy storage projects out there - I need to do a blog on that.

Wim - I can't comment on large scale battery longevity. I know that Toyota chose Ni-Cad for the Prius because of concerns about longevity and they have held up exceptionally well. My concerns about Li-Ion may be unfounded because the new battery management systems seem to make a huge difference by controlling charge percentage and temperature. There is however a class action lawsuit over the Leaf batteries which is based upon poor battery performance. I want to see how well Leafs and Model S's are doing in a couple more years.

November, 30 2015

Richard Vesel says

David - One thing the Hawaiians treasure is their land and open spaces. I think there definitely IS room for solar parks in many locations. It is just a matter of convincing folks that some selective disruption of open land areas for a few solar parks would be in their better interests. Meanwhile, the rooftops are already there, ready for use. Also, covering large parking lots with solar canopies would also serve multiple benefits, and create solar parks as you suggest. These canopies can be optimized for efficient construction, optimized angles, and then include charging stations for electric vehicles, which should begin to proliferate there soon.


December, 01 2015

Richard Vesel says

Power, in GWhr, from rooftop solar increasing dramatically in USA:



Add your comments:

Please log in to leave a comment!
back to top

Receive Energy Central eNews & Updates