An Overlooked Danger of Solar Thermal Plants: Fire and Explosion

April 8th, 2013
submit to reddit Share

After doing some reading about solar energy I came across some information on a danger that had, quite frankly, not occurred to me before.  It seems that many forms of solar thermal energy production carry a very high risk of severe fire and even explosion.

Solar thermal power plants use large mirrors to concentrate the suns light on collectors.  In most systems this works through trough mirrors which focus the light onto long pipes.  In others, an array of mirrors focuses the energy on a central receiver, which contains a fluid that is heated by the light.   The fluid is normally pumped continuously through a system of collection tubes in order to transfer as much heat as possible.

Leaks have always plagued solar thermal power systems.   The tubes must be thin to maximize thermal transfer and to keep costs down, and the scale of the systems is necessarily very large.   This, combined with the stress of daily heating and cooling has lead to what might be called a “plumbers nightmare” and explains why so much labor is required to keep solar thermal plants up and running.

The engineering challenge of providing a leak-resistant way of transferring the fluid through miles, under such harsh conditions has resulted in a number of approaches, including the use of ball joints and flexible hoses.   The sheer size of utility-scale installations has also been a problem, since the connections and tubing must be made as economically as possible.


Hot fluid leaks not only reduce system performance by changing the fluid pressure, but they increase the risk of insulation fires or auto-ignition. With risk of fire, high operation costs, and the necessity for large numbers of replacement parts, some pursued other options while improvements to flex hoses were developed.Ball joint assemblies, which have graphite seals and both rotation and angular deflection capabilities, were chosen by those with flex hose issues. Many suppliers and spare parts were available, and seal maintenance after thousands of hours of use resulted in lower maintenance costs.

The greater danger comes from the heat transfer fluid itself, which is generally oil-based.  They may be synthetic or mineral oil based.   The fluids themselves are usually designed to have a relatively high flash point and be less prone to combustion than other oil, but they are still flammable, especially when they are heated to high temperatures.  To maximize the efficiency of solar thermal power plants, the highest temperatures are desired for the transfer fluid, although this increases the danger of fire.    Since the fluid is often pumped at high rates, even a small leak can result in high pressure, high temperature fluid being sprayed.

A greater danger comes from the potential for the fluid to be over-heated within the system.   Either errors in routing the hot fluid or blockages in the system could potentially cause the thermal transfer fluid to absorb too much heat and be pushed beyond the boiling point, potentially to the point of explosion.

YES, this has happened before.    In 1982, Solar One, a concentrated solar tower was built with ten megawatts of capacity.  In August of 1986, a problem with the facility’s heat transfer system resulted in 240,000 gallons of heat transfer oil erupting into flames.   The explosion destroyed much of the plant.

A combined solar and gas fired power plant, SEGS I experienced similar fires at least twice. In 1990, it experienced an explosion resulting in a huge volume of superheated oil, which continued to burn for many hours before being brought under control.

Via the LA Times:

Blasts Rip Desert Solar Power Plant

BARSTOW — A series of explosions and fire shut down electricity generation at the world’s largest solar power plant near here Wednesday.

Thick plumes of black smoke spiraled into the clear desert air when one of four natural gas-fired heaters used to back up the solar heating system exploded.

A short time later, a second natural gas heater caught fire and exploded as the first of 75 firefighters and 25 pieces of equipment were arriving at the site, about 140 miles northeast of Los Angeles.

“We had a series of explosions, more than two,” said Capt. Sharon Sellers of the San Bernardino County Fire Department. “Our first units got on-scene at 9:16 a.m. and a second explosion occurred at that point, then a series of them during the entire incident,” Sellers said.

“There was a mushroom cloud. The heat was real intense and there were explosions,” said an inmate from the Boron Federal Prison Camp who was pressed into service to help fight the fire. He would not identify himself.

Sellers said two workers at the plant suffered minor breathing problems and were treated at Barstow Community Hospital.

Operated by LUZ International Ltd. of Los Angeles, the $280-million Harper Lake solar plant began generating electricity on Dec. 28 and produces 80 megawatts, enough power to serve 115,000 people. The company operates eight such plants in the California desert. Combined, they generate 274 megawatts, which is sold to Southern California Edison Co. An Edison spokesman said there was no interruption of electric service to its customers.

“We had two oil heaters on line and were bringing up the third and fourth oil heaters when this explosion occurred,” LUZ International spokeswoman Kathleen Flanagan said in Los Angeles.

While no flames were visible 1 1/2 hours after the fire began shortly before 9 a.m., San Bernardino County firefighters had difficulty reaching the blaze deep within the generating equipment.

“There is fire up there somewhere still heating that oil,” Sellers said.

The blaze was contained, but continued to burn late Wednesday.

The 1990 fire was not directly the result of the solar power systems on site but of a gas-fired heater at the plant.  None the less, it goes to show how difficult it can be to fight a fire involving such large volumes of heat transfer oil.

A second fire occurred in 1999 at SEGS I.  As a result of “human error” the temperature of the oil exceeded the safe limit, resulting in a massive explosion that shook the area for miles.   Some 900,000 gallons of oil erupted in flames.

Similar fires, involving heat transfer fluid have occurred in other industrial applications.

The danger of fire or explosion may also be increased by certain design features of some solar power stations.  A number of fires have occurred when leaking thermal transfer fluid came into contact with insulting materials, used in some of the pipes at solar thermal plants.    This has lead Dow Chemical, a manufacturer of thermal transfer fluid to issue warnings of the possibility of spontaneous combustion of hot fluids, should they come into contact with insulating materials.

Some newer solar thermal plants have proposed to use large tanks of molten salt as a means of providing extended energy storage.   This has resulted in additional concerns over the possibility of reactions occurring between the salt and heat transfer fluid.  Because the salt will solidify if cooled, and because of its corrosive nature, oil-based heat transfer fluid is still required for the collectors.  DOE Safety tests found that a break in the barrier between the two fluids could be dangerous.

As stated in a DOE document:

Therminol is a synthetic oil; Caloria is a mineral oil. Since the intention of the thermocline is to provide thermal storage for the Solar Trough power plants, we needed to investigate the reactivity between the oils used in the SEGS plants and the molten salt. It is possible that the oils could come into contact with the molten salt at the operating temperature of 400”C due to a failure in the oil-to-salt heat exchanger. The subsequent reaction could cause a dangerous result such as a fire or explosion.

It should, of course, be noted that no energy source is without risk, and there are certainly other industrial facilities that can also explode or produce massive oil fires.  However, the danger has been largely ignored with solar thermal power plants.   Citing such plants near population areas, without appropriate precautions, could be quite dangerous.  Additionally, the sheer volume of oil and the nature of the fires that can occur could easily overwhelm most fire departments.  Solar thermal plants generally have not been required to maintain the assets necessary to fight such a fire, should one occur.

This entry was posted on Monday, April 8th, 2013 at 7:37 pm and is filed under Enviornment, Misc. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.
View blog reactions

162 Responses to “An Overlooked Danger of Solar Thermal Plants: Fire and Explosion”

Pages: « 1 2 3 [4] Show All

  1. 151
    Bas Says:

            George Carty said:

    Feed-in tariffs, which force electricity companies to buy solar-generated electricity at retail prices…

    renevers wrote about subsidy for the investment. I reacted that there is none for that.

    I agree with you, that you can call some part of the feed-in rate ‘subsidy’ paid for by other electricity users (not the tax-payer or citizen as with nuclear).

    How big the subsidy part is, is a little complicated as the feed-in is done at the consumers location.
    So the electricity company usually only have to transport to the neighbor’s, and saves longer distance grid costs.
    Costs that the company has the moment he gets the electricity from real power plant.
    So the electricity company should pay something more for the electricity from the roof than at the power plant. 5 cent more?
    That would imply that ~6 cent of the feed-in rate of 19 cent is subsidized by the other electricity consumers.

    Quote Comment
  2. 152
    Anon Says:

            Bas said:

    I agree with you, that you can call some part of the feed-in rate ‘subsidy’ paid for by other electricity users (not the tax-payer or citizen as with nuclear).

    Never mind that basically every single tax-payer and citizen is also an electricity user.

    Not only that but it’s a very regressive tax which makes it worse than paying for it out of general tax revenue which could be progressive.

            Bas said:

    How big the subsidy part is, is a little complicated as the feed-in is done at the consumers location.

    Not really, take the wholesale price, divide by two to take into account the fact that the power generated is often useless and you’ve got a decent idea what the power is worth.

    Though in truth if the electricity company had a choice in the matter they wouldn’t even take the solar as it is superfluous.

            Bas said:

    So the electricity company usually only have to transport to the neighbor’s, and saves longer distance grid costs.

    They don’t get to save any grid costs because they still need the ability to transmit power the longer distances for when the sun isn’t shinning.

    That’s the thing you don’t seem capable of understanding, that solar power isn’t available when it is needed so you don’t actually get to save on any infrastructure, at best you might reduce fuel consumption a little.

            Bas said:

    So the electricity company should pay something more for the electricity from the roof than at the power plant. 5 cent more?

    Actually they should be able to decide what they will pay and have the right to refuse to buy the solar power from you if you aren’t willing to accept whatever they offer.

    Then we’ll get to find out what that power is actually worth.

    Quote Comment
  3. 153
    renevers Says:

    If this isn’t subsidy ? This is Dutch law.
    “Subsidie zonnepanelen vanaf 1 juli 2012

    U ontvangt 15% van de het aankoopbedrag van uw zonnepanelen. Van uw installatie wordt maximaal 3,5 kWp (kilowattpiek) gesubsidieerd. Zo’n installatie wekt genoeg energie op om een gemiddeld huishouden van stroom te voorzien. Het maximum subsidiebedrag is € 650.

    Alleen de materiaalkosten (zoals de aanschaf van de panelen en ondersteunende apparatuur, zoals omvormers) komen voor subsidie in aanmerking. Dus niet de arbeidskosten voor bijvoorbeeld installatie van het systeem.

    The troll revolted again .. He is clearly releases a lie on subsidy.
    Indeed there is no direct big subsidy on the structure investment for the end user/producer and it is limited. In the Netherlands it is different organized than Belgium or Germany. Every European country has its own method. In fact most of subsidies were canceled and will be phased out .but there were investment subsidies. Now only production related subsidies for end users remain interesting. In Belgium there was subsidy on the investment for the structure for the end user client until last year. There is investment subsidies however from general economic stimulation funds that relate to house improvement and energy use. There is investment subsidies for establishing an industry to make components for the use of solar panels. It is not end-user related as in Belgium but solar component producer oriented. The subsidies are there to have an own national industry on renewables. There is enormous subsidy in institutes for investigation on solar energy and research that can be used by industry in this field. There is a hidden subsidy by large power corporations that take a share in solar investment industry, forced to do so by public opinion and political pressure.
    There were 15% subsidies in the Netherlands

    The lying troll
    The drop in prices for solar is mostly due to massive undersold stocks from bankrupt companies. Why is it cheap? because they are bankrupt . Why they went bankrupt? Because they sold under production cost level.

    Labour cost and normal necessary structural changes to building structures , roof and reinforcements. which are normally necessary are left out in the Dutch proposals to save money for the government. 2 Euro calculates.without structural work .so it is 3 Euro this work included . 3 Euro is a value used every where in the sector for estimating the cost of solar. In fact it is not that important .. It is the dimension..if a factor of 100 becomes 50 in the future is is still disastrous.
    In fact the subsidy is not important to the story of investment cost..The the full cost of 3 Euro /w is normal investment.. That calculates a 100x difference with nuclear power investment.
    Normal calculation is : 1 euro solar panels, 1 euro electrical connections and grid adaptions ,1 euro structural work. Here a company with price of 1.25 Euro /W

    The troll is not answering the ESSENCE .. 100 times difference in investment..(10000%)..(refered to a Chinese nuclear plant investment now. The consequence is staggering..

    In fact from the large investment cost you could derive that solar energy in the Netherlands is energy uneconomic : it cost more energy in investment as it will deliver in its lifespan. There is a back of the envelope relation between investment en energy investment..

    He tries to wriggle out of the discussion.. 100 times is indefensibly .. He is not going to defend that .. He knows he can’t. He is ducking.. He sows doubt to these points and numbers..
    100 times !!!

    He is a troll and doesn’t speak the truth.. Or doesn’t he knows? .. He is a pretender of knowledge..Or he just writes something in the hope it is true, just to undermine someones reasoning..

    He is not a troll ..He is a SUPERTROLL

    Quote Comment
  4. 154
    Bas Says:


    Agree that your remarks regarding the feed-in rate seem logical, but
    in NL even while electricity companies have free choice and consumers/producers have (between ~10 companies) it shows to be not true. Read my explanation here below.

    Regarding the grid. Solar produces during the day, the moment demand is greatest.
    Even if the sun is behind the clouds the panels still produce, substantially less of course.

            renevers said:

    … If this isn’t subsidy ? This is Dutch law…

    Read the date; 2008. There was a limited amount of money for that subsidy.
    So the kitty was empty in a few months. After that nothing at all. Bad management of the minister.
    Happily for us, that minister (Maria van der Hoeven) became executive director of the IEA.

    We now have a somewhat similar ruling as the Germans:
    Feed-in rate same as the rate you pay for (now ~22cent/kWh) provided that:
    * the feed-in volume is less than 5000 kWh/year
    * the feed-in volume is lower then the consumed volume (so you cannot become net-producer).

    The Feed-in rates for the volumes above 5000 kWh or above your consumption, are free to set for the electricity company.
    Those rates differ greatly between different companies. They range from zero to ~19 cent (the German max feed-in rate) and a few even pay more under conditions.
    Friend of mine is net producer (~7.000kWh/year) and could get a ~21 cent rate, but it took time to find the right company, etc. And he lives in the country side.

    I forgot that we now also got a temporary subsidy until the kitty is empty (still money in it).
    The max. you can get is ~€650 or 15% of the net investment in the panels/materials (not for the installation costs, etc). It is somewhat risky to calc with that type of subsidy because the kitty may be empty the moment you come (I had such an experience). So I forgot that in my previous post. Sorry.

    As far as I know the Germans have no investment subsidy, but they get their feed-in rate guaranteed during (I believe) 15 years. That is better for the consumer/investor even while that feed-in is now ~19 cent max (they pay ~25cent). It delivers security regarding profit => a secure business case.

    In NL the situation may change if we get a new government after elections (in 3 years).
    E.g. into a rule that the feed-in rate is the same rate as the rate you pay, minus x cent…

    Personally I believe our ruling is slightly better for the (flexibility of the) economy than the German rules.

            renevers said:

    …Labour cost ..etc…are left out in the Dutch proposals…

    The €1 per watt is for a 4kW PV system on my roof including installation and converter (to connect to 220V~). On my rood it will produce ~3600kWh (not optimal roof for the sun, not best place in NL).

            renevers said:

    … Here a company with price of 1.25 Euro /W …

    Start negotiations and if they refuse (some have to much work), look for another.

    @renevers: Few advices;
    – abusive language makes your case weaker, as that is done by losers.
    – talk with somebody with more math abilities for your calcs.

    Quote Comment
  5. 155
    Bas Says:

    I have to correct regarding the investment costs of my little PV panel project.
    They said it would be all-in but I just read the conditions.

    Those set such unrealistic conditions regarding (ease of) installation, that I assume there will be a significant uplift.
    So total investment may end up somewhere at €1,50 – €2,00 per watt.
    That reduces my profit to more or less ~7%-10% (assuming feed-in rate stays the same, and electricity prices do not go down).
    That is substantially less attractive.
    May be waiting is better.
    If investment next year is >10% lower, than it is better to wait (prices went down more last years).

    Quote Comment
  6. 156
    renevers Says:

    The super troll wrote

    “@renevers: Few advices;
    – abusive language makes your case weaker, as that is done by losers.
    HAHA.. It is not abusive .. it is the truth .. It was untrue .. There was subsidy to get to the investment in the Netherlands there was subsidy in Germany and I notioned the change in policy.

    – talk with somebody with more math abilities for your calcs.”
    the troll should do first some classes in math and some university work before he is at par with renevers in math. In fact this was not math , this is calculus.. Very simple work.

    The troll did until now just avoid the real impact of the comments..Solar a factor 58 more expensive than nuclear power in investment here in Europe.

    As if he is right and i am wrong.. That is how a troll thinks. That is how a troll goes into discussions.
    It is the reverse .. What I wrote is rulings in 2012 not 2008 Of course that ruling is out of money .
    There is not a dent in my arguments..

    The Dutch government can calculate that every dime spent in it is wasted.Minister of economic affairs Verhoeven has seen that quite well. Perhaps she did the back of the envelope calculation that I did and everybody should do that wants to do an opinion whether investment in a very expensive powersource is a good thing for the economy. Even SPAIN has very little solar contribution in the share of the power mix and they have a lot more sun than the Netherlands. they did that simple calculation too….

    Spain is less rosy than I thought in renewables, I thouht Spain would be a solar succes.. nuclear still a lot bigger than solar+wind , solar to small to see in the statistics.. Average production factor for wind in Spain 22.3% at sea expected at 31.% Thet is just a little better than Germany at 19% ( Production factor is the same as average/peak installed)
    THis is a hosanah story by the wind lobby of Spain.. It is not that impressive/
    Jobs lost and immense subsidies to create employment in the wind and solar business did not lead to a economically self sustainable industry of power system.

    The troll sees a calculation of self interest as a big motivation to go to solar energy.. There is no objective reasoning of the larger interest of society. The money the troll receives comes from tax payers and is pumped around.

    The troll thinks the Dutch Slochteren gasfield will last forever? He thinks he can drive a car in 20 Years? No .. If green energy continues than the Netherlands will be bankrupt in a couple of years.
    The nice live the troll got came from a socialistic system supported with money earned by the Dutch state in their big gasfield, that is 80% empty now. If that gas is gone or goes expensive if it needs “fracking” techniques and expensive horizontal drilling, for just a little bit of gas, easy money is over .The luxuries he had in the past were earned with an oldfashioned energy system.. Green power will not generate luxury, it will be survival. The Netherlands is just finishing its last resources.

    Standard and Poor downgraded the Netherlands.

    The Netherlands do very bad now in the economical sense.. Large shortages..No compliance with the Maastricht treaty norms… extremely rapid unemployment rise..Tax income for the state dropping like a stone.. That in a country with former oil sheikh’s richnesses. The Netherlands were an example to Hugo Chavez.. That says it all. Venezuela is a mess now..

    The Netherlands is a TROLLY country.. A former socialistic ” Potemkin facade” on methane..

    Quote Comment
  7. 157
    BMS Says:

            renevers said:

    Spain is less rosy than I thought in renewables, I thouht Spain would be a solar succes..

    But Spain did have the solar panels that worked at night. ;-)

    Quote Comment
  8. 158
    Anon Says:

            BMS said:

    But Spain did have the solar panels that worked at night. ;-)

    Please use the <sarcasm> tag next time.

    Quote Comment
  9. 159
    BMS Says:

            Anon said:

    Please use the <sarcasm> tag next time.

    It wasn’t obvious? At least I gave a wink.

    Quote Comment
  10. 160
    renevers Says:

    @ ANON and BMS
    night “solar” energy
    Can’t we transform that 3 degrees Kelvin Background radiation between the stars into power like this?;-)

    Quote Comment
  11. 161
    BMS Says:

            renevers said:

    Can’t we transform that 3 degrees Kelvin Background radiation between the stars into power like this?;-)

    Of course we can. We just need to find another universe with a background temperature of 2, 1, or 0 Kelvin to use as the heat sink.

    Quote Comment
  12. 162
    Shafe Says:

            BMS said:

    Of course we can. We just need to find another universe with a background temperature of 2, 1, or 0 Kelvin to use as the heat sink.


    Quote Comment

Pages: « 1 2 3 [4] Show All

Leave a Reply

Current month ye@r day *

Please copy the string HOUZJL to the field below:

Protected by WP Anti Spam