It’s no secret that I think the best hope we have for transitioning to a post-petroleum economy is through solar power. I am optimistic that the thin film solar crowd – led by companies like First Solar and Nanosolar – will be able to deliver cost-effective solar power to the masses. I have also lately been looking at the possibility of a solar hot water heater, as I think these will be very good investments if energy prices continue to rise – especially given that there is a tax credit on these systems through 2008.
USA Today just published a new story that suggests that one solar firm (not a thin-film producer) will be able to deliver solar power for 7 cents a kilowatt hour by mid-2009:
Start-up: Affordable solar power possible in a year
I always take these claims with a grain of salt. I am hopeful, but also recognize that the majority of these sorts of promises generally fail to materialize. Nevertheless, it sounds promising:
SUNRGI’s “concentrated photovoltaic” system relies on lenses to magnify sunlight 2,000 times, letting it produce as much electricity as standard panels with a far smaller system. Craig Goodman, head of the National Energy Marketers Association, is expected to announce the breakthrough Tuesday.
Under its plans, which experts call promising but highly ambitious, SUNRGI would initially target utilities and large industrial and commercial customers. The company — founded by veterans of computer, digital design, aerospace and solar industries — would market to homes within three years.
Executives of the year-old company say they’ll start producing solar panels by mid-2009 that will generate electricity for about 7 cents a kilowatt hour, including installation. That’s roughly the price of cheap coal-fired electricity. “We’re bringing the cost of solar electricity down to be competitive with” fossil fuels, says Bob Block, a co-founder of SUNRGI.
Of course there are still barriers to transitioning to a solar economy. We need energy storage solutions, better batteries, and the price needs to continue to come down. But as I argued before, the future still looks to me to be solar.
“On Tuesday, Sen. Kay Bailey Hutchison, R-Texas, announced she will introduce legislation to freeze a biofuel mandate required in a bill passed last year. Pushed by Democrats, the bill required a gradual, five-fold increase in ethanol production by 2022.
And Sen. James Inhofe, R-Okla. — the top Republican on the Senate Environment and Public Works Committee — called on the EPA to institute “an immediate waiver from (biofuels) mandates.”
He wants a review of the effects mandated ethanol production has had on the global food crisis. The waiver power was given to the EPA by Congress in last year’s energy bill.”
http://www.foxnews.com/story/0,2933,353249,00.html
My sentiments are that solar, boosted by wind, geothermal and nukes, can easily supply enough juice for us to migrate to a PHEV transportation fleet.
In fact, we will do so while raising our living standards and getting cleaner air in the bargain. we will save $100s of billion in oil import costs, and that money can be spent on our fellow Americans.
Our grid is adequate – indeed, PHEVs will charge up at night, when demand is lower. And look for buildings to continually use less, not more, electricity in the years ahead. New lights use far less juice than old lights.
This may result in massive oil gluts.
I think we are at a peak price now, and a long, long decline — a generation possibly — of price declines lies shead.
And that may pull the rug out from under alternative energy and transportation. Just like last time, following the 1980 spike.
A conundrum.
Although, if we bombs away on Iran, we may see even higher prices soon.
“We need energy storage solutions, better batteries, and the price needs to continue to come down.”
We need a law that mandates grid-tied solar systems on new construction. No need for storage solutions or batteries with grid-tied systems. And the price would fall with mass adoption. I realise solar wouldn’t be practical everywhere. But,congress should mandate it where practical. It would still take 40 or 50 years to go solar that way. But,we need to start somehwere. At least this way,people could finance their systems for 30 years,along with the mortgage.
This is totally off topic, but it’s an idea I just wanted to explore (although I’m sure others have thought of this).
If we ignore economics for a moment, would it be possible to build a power plant on top of a nearly spent oil field, use the oil for electricity generation (remember, we’re ignoring economics) while capturing the CO2, and then inject that CO2 back into the ground for advanced oil recovery? We would keep doing this until the field is utterly spent, and then we’d pack up and move to the next nearly spent field. That would avoid many of the environmental problems of coal, and eliminate the need for a huge sequestration infrastructure since it is sequestered on site.
Or would that not generate enough CO2 to keep the process going?
CCE, sequestering CO2 is a nice idea,but if you’re going to go through the effort,why not spend a bit more and turn it into gas or jet fuel? We can use nuclear power to capture CO2 from the air and turn it into gasoline. It’s a carbon neutral process,even though the gas would be burned again. Estimated cost of $4.50 per gallon. We’re closing in on that price now.
I would say that First Solar is the leader of thin-film at the moment. They actually sell lots of product, which Nanosolar is only getting going on. Nanosolar clearly still needs to build their marketing force up as well as spool up production.
Sharp’s triple junction silicon thin film is also interesting, but probably too expensive a process to beat the simpler direct band-gap semiconductors.
I’ve never been much of a fan of the concentrating solar systems. They rely on clear skies to work (period, you got clouds, you get bupkus) and require tracking systems. Adding a tracking system to PV means all of a sudden you have moving parts whereas before you didn’t. That’s going to generate maintenance costs in the future.
Our grid is adequate – indeed, PHEVs will charge up at night, when demand is lower.
Ummmm … yeah, but the sun doesn’t shine at night.
Does it make sense to generate electricity during the day using solar, then store it (somehow?) for a few hours, then use that stored energy to charge up millions of PHEVs at night?
I’ve never got a convincing answer to this question.
carbonsink,the oil drum has a story of another company using ammonia as a storage medium. I’d like to see a cost analysis of using gasoline. It’s been demonstrated that solar can be used to make gasoline from CO2. What if companies paid a solar operator to take the CO2 off their hands? What if regular gas got that extra $2.00 tax,but CO2 gas wasn’t taxed at all? What if it were subsidised to boot? WHAT exactly would it take to make the process viable? I really think it’s worth considering. More American jobs,less OPEC oil.
I would say that First Solar is the leader of thin-film at the moment. They actually sell lots of product, which Nanosolar is only getting going on.
Thanks for that. I added it to the article. I have heard lots about Nanosolar, but not so much about First Solar. Are their thin film products similar to what Nanosolar is planning to sell, or are they still pretty different?
Thanks, RR
FirstSolar uses Cadmium Telluride. Except for a couple underfunded startups they’re the only company using CdTe. Nansolar and about a dozen other thin film companies use CIGS (copper-indium-gallium-diselenide). The other major thin-film technology is amorphous-silicon, used by United Solar Ovonic, EPV, etc.
CdTe is cheaper than silicon, especially with the polysilicon shortage of recent years, and FSLR is cleaning up. Revenues are approaching $1 billion and double almost yearly. Whether that justifies a stunning $24 billion market cap is another issue.
SUNRGI is questionable in my mind. 2000x concentration doesn’t have a real advantage over 500x, by the time you get that high your active PV cells are so small they represent less than 10% of the total system cost. It’s all mirrors and motors, and 2000x requires more accurate pointing than 500x so that cost will be higher. The structure of their modules doesn’t strike me as particularly cheap. I’ve seen some innovative ideas like inflatables ballons with Mylar reflective coating which could drive CPV costs into the dirt, SUNRGI’s modules look like gold-plated jewelry boxes in comparison.
Finally, as Robert noted, CPV doesn’t work with clouds and is thus best suited for large-scale installations in deserts.
Maury,
That’s what I meant by “ignoring economics.” Clearly, oil is worth more as liquid fuel than as electricity. But it would be (at least theoretically) a fairly efficient way of getting to oil that is otherwise difficult to pump, while simultaneously taking care of the CO2 without the need for the infrastructure.
First Solar uses a more standard Cadmium Telluride thin-film technology — which involves a more difficult deposition process than the nano-particle “ink” used by Nanosolar, which can be printed onto the pane inkjet-style, as a closely integrated part of a high-speed production system. So Nanosolar has the better technology — faster to make & lower capital costs — but not surprisingly it’s taking them a little longer to get to market.
Does it make sense to generate electricity during the day using solar, then store it (somehow?) for a few hours, then use that stored energy to charge up millions of PHEVs at night?
Not really. If we got all our electricity from PV it’d make more sense to charge PHEVs during the day (at low rates) and let the PHEVs feed electricity back into the grid to handle our minimal nighttime needs.
Of course we’ll never get all our electricity from PV. For a long time the best use of PV will be to meet daytime loads, which peak on sunny, summer days. Even at 20 cents/kWh PV is cheaper than summer peaking power in places like CA. PV also produces 20 cent kWh’s in winter, though, when rates are much cheaper, so the full year economics don’t work. At least not as long as we subsidize fossil with free CO2 permits. Multiple thin film and CPV entrants claim they can make solar cheap enough to beat even subsidized fossil electricityf ($1-2/Wp). Time will tell.
Sungi appears to be using a fresnel lens concentrating solar array. In the USA Today article they say $70/MWhr but on their website they claim $50/MWhr.
In theory these thing should work. Such devices must track the sun in order to keep the lens focused on the PV cell. So it is a mechanical design issue and manufacturing problem. If they can solve those issues, then perhaps they can get costs down. $70/MWhr? I doubt it not by 2009.
Carbonsink:
The point about solar (and wind, geothermal, nukes) is that additional increments to our grid need not use fossil fuels. Indeed, given that buildings will use less and less energy going forward, we may not need any additional increments anyway. However, as we build solar (and wind, nukes etc), we can sub out the fossil plants.
PHEVs, by charging at night, will not overtax the grid. Grid demands plummet at night.
Thus, the solar is not added to handle the PHEV load; the solar is added to sub out fossil fuels, and handle peak daytime needs.
PS I have a friend working on a process to convert natural gas directly to gasoline, some sort of heat and pressure and other fancy processes. So, if solar replaces gas power plants, we can use the natural gas as gasoline. By the way, I read the planet cooled a lot this year….asnd we are seeing radically less sunspots than normal…August is a great time to buy those fur coats….
Robert,
I have a solar hot water heater (colsed pressurized glycerol type). It is pretty great at heating the water for our family of four.
It cost about 7k (installed professionally). That’s two collectors with a single 150 gallon tank (with electrical backup), pump, heat exchanger etc. With tax credits (federal and state)it became more like 3.5-4k.
It probably saves me 10-15 kwh per day, or about $400-500/year.
It does require maintainance, as it occassionally overheats (if I’m out of town and not using hot water) and has to vent coolant (which needs to be replaced). I live in the south, so there is actually quite an excess of hot water in the summer). Other designs (drainback etc) do not require maintainance.
You can put one up yourself for considerably less, if you are up for it. Be careful on the roof!
I have a solar hot water heater
Couple of questions if you don’t mind. Is the coolant readily available? And have you thought about other options for that excess heat? Not sure what you could do with it, but it seems like a home could be found for it (although in the summer, heat is not exactly in short supply).
Cheers, RR
Massive buildout of solar PV is very similar to US Corn ethanol.
It’s a good idea with surplus electronic polysilicon and CdTe that wasn’t being used in other applications. It won’t be so good as they build out coal-fired polysilicon manufacture in China and run out of practically recoverable Tellurium.
On a small scale Solar PV is a great part of the energy picture, but so is ethanol from surplus feed grains. It starts to come apart on the grand scale and has numerous unintended consequences like Dumping of silicon tetrachloride in fields in China (Washington Post article) and that is prior to it being significant enough on the grid to require a storage solution.
Again, Solar PV is a very good solution for remote power and limited applications with surplus semiconductor material.
would it be possible to build a power plant on top of a nearly spent oil field, . . . then inject that CO2 back into the ground for advanced oil recovery?
Yes, we are looking into it. You need a field that has already been waterflooded, and you need a pretty pure CO2 stream. The problem is that some CO2 is produced along with the crude, and you need a fairly pure CO2 stream. You can’t just capture and compress combustion gases, too much nitrogen and not enough CO2.
You would need an oxygen plant to get a CO2 stream. Besides, crude BTUs are worth a lot (currently about $20 per million BTUs). Better to sell the crude for transportation fuel and use some other fuel to produce the CO2. An integrated gasifier combined cycle plant running coal or petroleum coke would do the trick.
CCE,
IMO, a century from now, I believe that history will not be kind to us for this. Burning the thing that produces so many chemicals, and plastics for fuel…well, lets just say that a medical visit would be a whole different experience without cheap plastic. I firmly believe that we need to move towards electrified transportation RIGHT NOW. Oil is far too valuable to simply ‘burn’, regardless of economics.
Bobert,
The coolant is usually propylene glycol, and yes, it is easily available. You certainly think like an engineer! There are, of course, always uses for the extra head, if a person is handy, and creative. I use mine to warm my WVO when I filter it, and to heat my hot tub. If someone would just figure out a way to turn low grade heat into high grade electricity…
King:
Saskpower is probably going forward on a post-combustion coal power plant carbon capture scheme with the carbon stream going to oilfield injection. Estevan (where the Boundary Dam power plant is), is in the upper middle of the Bakken Play.
Winelover:
This thread on Small 2-5kw ORC generator for home solar thermal was by far the most visited link on r2dot.org, so you aren’t the only one interested in home solar thermal electrical generation.
I believe that solar thermal built from common materials is our only realistic energy solution. The “silver bullet” mentality of large scale PV isn’t going to be realistic in the long run.
The idea of a Organic Rankine Cycle steam generator coupled to non-tracking solar thermal panels hasn’t got the same whiz-bang appear as PV and the major complaint is that there are moving parts that will require service. This is stunningly short-sighted because there a many less parts and technology in an ORC generation system than in a car which people have grown accustomed to maintaining.
If someone would just figure out a way to turn low grade heat into high grade electricity…
Yeah, I was thinking about that earlier. You could vaporize a low-boiling point liquid and generate pressure. You could then use that in a turbine. But I couldn’t work out what to do with it after that. You would need to return the outlet back to a liquid state, or have a constant supply of low-boiling liquid. Neither of those options work.
Not sure an approach like that would work, but there should be something. The problem for me is that I am more of a liquid fuels guy. When I start talking electricity, I am always a little out of my element.
RR
Bob,
Thanks for you link, and thoughts.
I have thought of fiddling with a scroll unit myself, and still may. But having lived with my own PV array since 98, plus various wind turbines, non-moving parts beats moving every time, IMO. Here is a company trying to do heat-to-electricity in a solid state manner:
http://blog.businessgreen.com/2006/11/eneco_details_r.html
There have been others, but not a lot of public info has emerged, to my knowledge. For what its worth. I bought 600 watts for $4/watt in 1998, and have moved it with me for the last ten years. It is currently sitting on the roof on my little rental home here in yountville. Being a conservative by nature(an actual conservative, not the political species), I adapted my lifestyle to the amount of power I had. This alone has been an interesting experiment, at odds with how americans, in general, do things. Excuse me, my ‘stimulus’ check just arrived. I am using it to buy a used bicycle…
have a wonderful day!
“Not sure an approach like that would work, but there should be something. The problem for me is that I am more of a liquid fuels guy. When I start talking electricity, I am always a little out of my element.”
You may be Robert, but you, I believe, see that moving away from them might be in our own best interest. Your posts on solar have always been insightful and inspiring.
If we actually made energy a national priority, and got a Manhattan project scale program going, I would think that heat to electricity would be a no-brainer.
In a solid state manner. I wish someone would finally turn a peltier junction into something really useful! Imagine how that could change things…
The Chena Hot Springs ORC system has a pretty good explanation of generating power from 74C hot water.
The PureCycle binary plant is basically a modified commercial chiller that uses a low-boiling point fluid (refrigerant) in a basic Rankine steam cycle.
This is a good simplified diagram.
I have some issues in believing that this type of system can be scaled down to the single homeowner and the idea behind my SHPEGS project is to build out small community sized renewable power and heating systems.
What kind of temperatures can be achieved with a solar water heater?
What kind of temperatures can be achieved with a solar water heater?
I have a this solar water heater (the “Titan” model) which is limited at 55C. However, I live in a frost-free area, so no need for antifreeze/coolant, and its zero maintenance, no pump required.
If we got all our electricity from PV it’d make more sense to charge PHEVs during the day (at low rates) and let the PHEVs feed electricity back into the grid to handle our minimal nighttime needs.
Yes, but cars tend to be out and about during the and parked at home at night, precisely the opposite of optimal. We might be able to convince employers, parking garages, shopping malls etc to provide parking spaces with plugs for PHEVs/EVs, especially if were profitable and encouraged by some ‘carrots’ from government.
If that’s going to happen, I suspect it will happen at Google first.
Indeed, given that buildings will use less and less energy going forward, we may not need any additional increments anyway. However, as we build solar (and wind, nukes etc), we can sub out the fossil plants.
PHEVs, by charging at night, will not overtax the grid. Grid demands plummet at night.
Thus, the solar is not added to handle the PHEV load; the solar is added to sub out fossil fuels, and handle peak daytime needs.
Sorry, but again, I find that a pretty unconvincing argument. Seems to me that the PHEV load will need to be carried by a clean energy source that’s consistent available at night, like geothermal or hydro.
ROBERT–
are yopu aware of commercialized development of CSP by FPL GROUP/FPL ENERGY llc. two installations in MOHAVE now. 2+ billion$ committed over next 6-8 years in CA and FLA
reference fpl group.com
fran
Hey RR,
One comment on the $.07/kwh figure. Have you found what kind of operating life they are assuming for that figure?
With regards to low-grade heat, not matter what scheme you use to try to harvest it, you still run up against the Carnot limit. Thus, it becomes hard to make it worth the added capital expense.
A better direction, I believe, is to try to make use of the IR part of the spectrum directly before it degrades to heat:
http://www.i-sis.org.uk/QDAUESC.php
carbonsink said, “Does it make sense to generate electricity during the day using solar, then store it (somehow?) for a few hours, then use that stored energy to charge up millions of PHEVs at night?“
They above does not make sense.
What does make sense is to store the sun’s energy as heat and generate electricity at night from that heat. Ausra proposes to do this cost effectively. They claim a 16 hour thermal energy storage plan. Whether they can deliver is an important question.
Nighttime generation from daytime heat would probably not be used for charging PHEVs though. You are probably better off delivering the power during the day to charge PHEVs if that’s when you have it. Wind is an energy source that is skewed toward nighttime, and it makes a good PHEV fuel.
If someone would just figure out a way to turn low grade heat into high grade electricity…
Ever heard of the 1 – Tc/Th efficiency limit? If not, consider a course in thermodynamics, or read about it on Wikipedia. You’ll also find out that 1 – sqrt(Tc/Th) is an empirical efficiency estimation, and that is of course even less encouraging.
benny said, “My sentiments are that solar, boosted by wind, geothermal and nukes, can easily supply enough juice for us to migrate to a PHEV transportation fleet.“
In case it helps here is a feasibility calculation:
2050 US population: 420M
2050 Vehicle Miles Traveled (VMT): 420M * 9300 = 3.9 trillion miles
PHEV Watt hours per mile (Wh/mi): 300
PHEV Wh needed at plug = 3.9 trillion * 300 = 1170 TWh
PHEV Wh at power plant: 1170 TWh / 92% = 1274 TWh
Example 1: Stirling dish CSP
Efficiency: 1780 GWh/year on 1800 hectares = 0.989 GWh/yr/ha
Land required: 1274 TWh/yr / 0.989 GWh/yr/ha = 1288006 ha = 4973 sq.mi.
Hmm, 5,000 square miles of the US desert southwest is a lot easier to defend than the 169,234 sq.mi. of Iraq.
Anyone want the land area calculation for wind turbines instead of CSP?
Robert,
The coolant from the solar hot water heater gets very hot. It can easily heat the 150 gal tank to over 140 degrees (there’s a cutoff so that it doesn’t overheat the tank) on a clear day.
The coolant we use is some “green” type of antifreeze solution. I’ve only had to add coolant twice in a year.
As far as excess heat, the collectors will gather up to 40,000 btu each per “day” under optimal conditions.
I figured out how many kwh this would come out to at some point… around 10 I think. So if you could harvest at 40% efficiency then it would probably be competative with PV, but as has been pointed out, the temp difference between the hot and cold sink will limit efficiency.
You could use it to provide space heat in the winter, I suppose, but you’d have to really overbuild the system to get any real benefit in the winter and then you’d have the problem of the excess heat in the summer again.
If you have more questions, feel free to email me at mbmurphy777@yahoo.com.
I too have a solar hot water heater, but nothing fancy — just a 250-liter tank on the roof hooked up to two collector panels. The tank is filled from the tap, and empties by gravity. Our insolation here is not so good because we’re in a deep valley, yet it provides hot water from March through November (during the winter the sun is mostly blocked by trees). I haven’t calculated how much money it’s saved, but it’s considerable. I think we’ve had it five years, and so far maintenance cost is zero. There’s hardly anything to break.
Winelover and kingofkaty,
The oil would be burned to generate electricity, which could power PHEVs and the rest of the grid. It will also be cleaner than burning coal, and I presume the same (or similar) methods used to clean up coal burning and capture CO2 could be used here.
It just seems to me that capturing the CO2 from the same oil you’re pumping and then using that CO2 to extract more would be an efficient way to do things if it was technically feasible (i.e. you probably need more CO2 than burning the oil creates). The Dakota Gasification Company makes synthetic natural gas from coal and then pipes its captured CO2 200 miles to Canada for enhanced oil recovery. I would think we could do something similar on site if the oil was used for electricity generation. In this case, much of the CO2 would stay underground, whereas when you burn the recovered oil in vehicles, the CO2 is released.
Economically, it wouldn’t make sense to do it since liquid fuels are so valuable (unless the price of CO2 was significantly high)
One of the mental blocks engineers have in moving from non-renewable power systems to renewable systems is getting hung up on thermal efficiency. Thermal efficiency in non-renewable electrical power systems relates to fuel, capital investment and ongoing maintenance. In a renewable system efficiency is related to capital investment, the effort involved to move the heat transport media and ongoing maintenance.
This is a fundamental difference from non-renewable systems. If I have a cheap and serviceable solar capture system, who cares how much sunlight it “wastes”.
In the case of medium temperature geothermal, once the energy expended to pump the geothermal water is overcome the only thing to look at in the feasibility is capital investment and maintenance.
If you have a 10% efficient solar thermal system that costs 20% of a 15% efficient solar PV system you have made a good investment. You are going to need more area, but a lot less capital investment. The basic Carnot Efficiency of a hot water supply at 70C and a cold sink of 15C (the shallow ground temperature in Dallas) is ~ 16%. If you have a system that can be serviced by a HVAC tech. or a good plumber and costs 20% of the equivalent output PV system you have made a good investment and haven’t bound yourself to a high tech factory in China for replacement parts.
The major reason Solar Thermal at the residential level hasn’t been pushed is that it’s impossible to control the IP like First Solar has been able to do with their CdTe technology or other Solar PV firms. The fact that it’s difficult for a solar thermal business to tie up IP and be profitable doesn’t mean that it’s a bad idea. Ausra and similar companies are going after the grand scale power plant with solar thermal, but they don’t really hold any IP and are open to competition. I am holding out that a grassroots movement will start on small-scale solar thermal.
We might be able to convince employers, parking garages, shopping malls etc to provide parking spaces with plugs for PHEVs/EVs,
If we really went all-out PV large parking lots would be a popular place to put panels. They offer lots of contiguous land, they’re already an eyesore and covered parking is a valuable commodity.
Anyone want the land area calculation for wind turbines instead of CSP?
Wind farms have two land area metrics: the area they cover and the actual land they use. GE’s 2.5 MW turbine has a 100m rotor diameter. With 5×10 spacing each turbine would cover half a km2 of land. But actual land used by the turbine base and access roads is only a couple percent of that. The rest of the land remains available for cattle, crops, etc. (at least until PETA starts filling lawsuits to protect the viewshed of the poor, defenseless cows).
A well-sited 2.5 MW unit can put out 7500 MWh/year, so you’d need 170,000 turbines to provide your 1274 TWh/year. That’s 85,000 km2 of land coverage but only about 2000 km2 of actual land usage.
At $4m per turbine the cost is $680 billion. That’s about what we spend every years importing oil.
One comment on the $.07/kwh figure. Have you found what kind of operating life they are assuming for that figure?
No, I didn’t know anything about them until I read that story. And as I told someone by e-mail, I am skeptical of all of these claims. Sometimes these things work as expected, but usually they don’t.
RR
Ever heard of the 1 – Tc/Th efficiency limit? If not, consider a course in thermodynamics, or read about it on Wikipedia. You’ll also find out that 1 – sqrt(Tc/Th) is an empirical efficiency estimation, and that is of course even less encouraging.
This limit can be avoided by not using the Carnot conversion.
I’m talking about thermovoltaics and infrared solar cells. 80% efficient infrared nano-antennas or 50% efficient amorphous diamond semiconductors could be aligned in a stack. Hot water (from the solar boiler in this case) flows through the stack. Ambient air or perhaps groundwater can be used to enable emissivity. Parasitics could be big but high effiency should definately be possible.
These technologies are in the laboratory phase though.
Usage of solar energy has increased in the past few years all over the world
Solar powered water heaters are highly subsidised in many countries
hot water systems