#1 cleantech news, reviews, & analysis site in the world. Subscribe today. The future is now.

Energy Storage

Published on February 3rd, 2016 | by Roy L Hales


Phi Suea House’s Better Energy Storage System

February 3rd, 2016 by  

Originally published on the ECOreport

The sun shines more than 300 days a year in Chiang Mai, Thailand, and Hamburg entrepreneur Sebastian-Justus Schmidt says, “it is the perfect location for a solar energy system.” His Phi Suea House’s solar panels produce 6,000 kWh a month, enough to power 20 to 25 average North American households. Though the first three residences in this development went online last March, it officially opened January 29. The most revolutionary aspect of this development is the hydrogen electrolyser in Phi Suea House’s better energy storage system.

unspecified-7“We got the idea to use hydrogen storage from pilot projects in the telecommunication industry. They used a system like this to supply power to communication masts. In many applications, it is more efficient than the battery systems that are available,” he said.

Though Schmidt is aware that his technology has an incredible potential for grid applications, his focus is off the grid.

Phi Suea House incorporates most of the usual elements of a environmentally friendly lifestyle (rain-water collection system, green building designs, home automation, etc), as well as solar panels that produce all of the electricity need for the residences, a water storage reservoir, big gardens, workshops, and a pool.

Schmidt said there is usually 10-20 kW of excess energy by the end of the day.

If he were in Europe or North America, his system would be feeding electricity to the grid.

Unfortunately Thailand is not yet ready for solar projects of this size. The country does incorporate smaller rooftop systems, but does not offer incentives. As few Thais are wealthy enough to wait seven years for their system to pay for itself, the nation’s vast solar potential is largely undeveloped.

Schmidt says that even in the rainy season, Thailand has a lot of sunshine.


“We will have clear skies for half a day and cloudy for the other half,” he said. “We often reach radiation densities of 1,300 watt per square meter and higher. It is much higher than in Germany. We are closer to the equator and have a lot more direct sunshine.”

Schmidt says projects like Phi Suea House are ideal for islands or mountains, where it is difficult to obtain electricity.

“This is a model for offered sites where there is no easy access to good grid power. We are getting a lot of interest from people on islands who are (presently) using very expensive diesel generators,” he said.

Smaller residences at Phi Suea House - all images courtesySebastian-Justus Schmidt

All images Courtesy Sebastian-Justus Schmidt.

Tags: , , ,

About the Author

is the President of Cortes Community Radio , CKTZ 89.5 FM, where he has hosted a half hour program since 2014, and editor of the the ECOreport, a website dedicated to exploring how our lifestyle choices and technologies affect the West Coast of North America. He writes for both writes for both Clean Technica and PlanetSave on Important Media. He is a research junkie who has written over 1,600 since he was first published in 1982. Roy lives on Cortes Island, BC, Canada.

  • Julie Rosenthal

    Thank you for this interesting article.

  • eveee

    50% round trip efficiency from hydrolysis and what compared to a battery? Hydrolysis alone is less than 50%.

  • Ivor O’Connor

    This all seems like an exercise in engineering. Wish there were more people doing similar things.

    One question I have is how long do the electrolysis machines last? I think they may go bad fairly quickly, like batteries, and can be quite expensive.

    • Riely Rumfort

      That’s a surprisingly tough stat to find, the longevity of electrolysis machine. I just figured you electrify platinum and forget about it, maybe clean the surface and channels occasionally.
      Also yes there is a need in far more ‘exercises’ in human expression with technologic unison.
      Refining a sustainable reasonably durable living system is pretty much my ‘primary objective’, but one mist be careful not to make us independent of nature, as such leads to neglection of that beyond immediate ramification. Basically, nature has to benefit by it’s ends, a negative footprint, durable yet not soo internal as to bring about disconnection with nature at large.
      It’s quite the balancing beam and the toolbox is constantly growing.
      Also no route should be under-utilized, multiple storage should be drawn from, multiple renewables tapped, a half dozen sensor types plotted a hand full of ways, customized automation for nearly every facet.
      When you’re talking >30 aqua-cultivations, aglae, phytoplankton, fish(3-5types), vegetative/aeroponics, vermicomposting(worm/fly), nitrate fixing mosses, compost teas. It’s a lot to keep track of.

    • milliamp

      Don’t hold me to this answer but separating water into hydrogen and oxygen basically requires applying a voltage to it. Distilled water doesn’t conduct electricity but some experiments I have seen add salt to normal water to better allow electrical flow.

      This probably means the energy required for the process is the most significant cost. The figures thrown around above is this results in a net efficiency of about 50% vs batteries at 85-95%.

      It sounds like an open and shut case against the idea but the caveat is that solar costs are continuing to tank. In a make believe world where electricity into the process were entirely free hydrogen storage could be a viable option.

      The problem is I agree with Riely in that I have no data on what those costs are. There is a similar project in NJ called Hydrogen House here http://hydrogenhouseproject.org/

      They have some info on their website and they might have a pretty good idea of what the cost is. They seem pretty open to tours, contact, and questions. Maybe they could chime in 🙂

  • milliamp

    Hydrogen storage has high losses through electrolysis to create hydrogen from water but it offers a lot more density than lead acid batteries. Batteries have high efficiency (~94%) but they have limited amount of cycles before they have to be replaced and are high maintenance.

    Now that solar panels are getting cheaper there are legit cases where it might be worth weighing the benefits of using a less efficient (hydrogen) but lower maintenance storage method.

    Lithium ion is better than lead acid too but it does seem that hydrogen isn’t a terrible idea. This was also covered on gizmag with some additional detail: http://www.gizmag.com/phi-suea-house/41033/

    It looks like an interesting project.

    • milliamp

      Jan-Justus Schmidt who is an engineer on the project replied to some of the comments on gizmag (link above), here is his post with his email address removed:

      Hi everyone, My name is Jan-Justus Schmidt, I am the lead engineer at CNX Construction responsible for the energy system. I would like to address a few points brought up in your comments and give some more information about our system.

      As noticed by @Daishi, the energy system at the Phi Suea House is a central solution for community solar power and storage. It is the first system of its kind, and it is ideally suited for residential or other developments in remote locations or where complete independence from the grid is desired. The novelty and value of the data collected at the project is underlined by the interest of leading universities and research institutes. For example, we have signed an agreement to share our data with the Energy Research Institute of Nanyang Technological University (NTU) in Singapore.

      The energy demand of the Phi Suea House project site is about 6000kWh per month – apart from the homes there is additional infrastructure including a centralized water pumping, collection and treatment system with a 1000m3 reserve tank, a 400m3 swimming pool, and a large fish pond with two waterfalls.

      I agree that lithium-ion batteries are now probably superior to lead-acid in the majority of cases of purely battery storage systems. Lead batteries are popular mainly because of their high power-to-weight ratio and low cost, however their lifetime is very strongly affected by the depth of discharge per cycle and the operating temperatures. We use lead-acid batteries in combination with the hydrogen system because we can maximise the advantages and minimize the disadvantages of each. The bulk of the nightly demand is covered by the fuel cells using hydrogen, while the batteries are only used to help cover short peaks in demand exceeding the fuel cell capacity. By only using the batteries to supply short peak loads during the night, we are making the best use of lead-acid batteries’ strengths. They do a great job at instantaneously supplying high power – and do not have to run for extended periods of time, thus minimizing their depth of discharge and keeping down the operating temperature. This way we can use a relatively small battery storage, keep the overall costs at a minimum and extend the expected battery life significantly. Of course it is correct, that both the hydrogen storage and lithium batteries offer better energy densities and require less physical space and this may be a concern for small residential systems, but in our application it was not a problem.

      Regarding hydrogen storage and safety, the only necessary precaution is that the storage should be set up in an open or well-ventilated space to eliminate the possibility of a build-up of gas in a trapped space. While hydrogen is flammable, it poses no serious risk due to it quickly escaping in ventilated or open areas, and any hydrogen fires should be extinguished using standard firefighting methods suitable to the surroundings. On the NFPA fire diamond, Hydrogen is classified “0” for health hazard and instability/reactivity, which means it “poses no health hazard, no precautions necessary and would offer no hazard beyond that of ordinary combustible materials (e.g. wood)” and is “normally stable, even under fire exposure conditions, and is not reactive with water”. It poses less danger to health and safety than LPG, CNG, or gasoline. Contrary to popular belief, hydrogen does not leak through correctly made and installed tanks and piping. In fact, it can be stored for very long periods of time, making it the ideal technology also for seasonal storages.

      If you have any questions or would like more information about the Phi Suea House energy system, please feel free to contact me directly at

      • Riely Rumfort

        Thank you.

      • harisA

        Really nicely written reply by Mr. Schmidt. Good luck with the project.

    • Joe Viocoe

      There are other chemistries besides lead acid and lithium ion. Zinc and Sodium are serious contenders in this space.

    • Bob_Wallace

      Is density important when we’re talking about residential storage?

      Hydrogen is not a dense energy storage. You have to compress hydrogen in order to keep it from taking up a lot of space and compression eats energy. Even compressed hydrogen is not that dense, compare the Mirai and Tesla S which have about the same range.

      Batteries are not high maintenance. I live off the grid with solar and lead acid batteries. Once a month I push a button to start an equalization cycle. Once every three months I open up my batteries and top them up with water. With lithium-ion batteries one doesn’t even do this rather trivial stuff.

      What we don’t know is fuel cell life and maintenance. I read something not long ago about a bus company that was running a fuel cell bus. The keep a spare ‘stack’ on the shelf because every six months they have to swap out the stack and send the used in to be reconditioned. Is this typical? I don’t know. Perhaps not.

      Battery cycle life? If the batteries are not abused then we’re talking thousands of cycles. My current lead acid batteries are rated for 4,000 cycles. Eleven years. There are lithium-ion batteries with higher cycle life.

      What we don’t have is a price per kWh for each approach over time. We can look up the cost of batteries but I haven’t seen the cost of an electrolyser, compressor, storage tanks, and fuel cell. Undoubtedly those would come down over time but how much? (Battery prices will also come down.)

      And don’t forget. Due to the much lower efficiency of hydrogen storage one would have to install 2x as many solar panels in order to get the same amount of power out of storage.

      • Riely Rumfort

        In my opinion hydrogen is more for long term power storage, it can sit a month, with the primary loss being turning it back into usable power. I also think I’d myself devise a system for voltage opimizing solar by splitting load at peak. For example if your minimal voltage is lacking on a cloud day it shift to a lower circuit portion. In a exceeding your levelized voltage you could split a portion to long term.
        I’ve also greatly considered redirection for base low. The constant load of say a fridge or an aeration pump etc, why lose power and cycle your batteries?
        If hydrite were legal to purchase hydrogen would be highly competitive and densely stored, you can legally make it in a particle accelerator, but yeah too many start up costs, it’s a shame.
        So what kind of system do you run, voltage, amps, panel kWh?

        • Bob_Wallace

          1.4 kWh of ground mounted panels.

          I’ve run into problems with shading a few weeks on each side of the winter solstice. My panels are mounted on two racks and the trees south of one rack have grown high enough to block sunlight during part of the day. I’m planning on adding about 1.5 kW on the roof this summer in order to cut down on generator use.

          When I set up my system panels were retailing for $8/watt. I went as small as possible to save money. I’m now going to upsize based on available roof space, not price per watt.

          Voltage, only 120 vac. A large enough inverter to run wood working tools (table saw, 12″ planer, 8″ jointer, ….).

          • eveee

            Have you looked into optimizers. The solar edge ones work on a per module basis. But they are an extra cost. The MPPTs on typical inverters won’t fix the shading problem that well.

          • Bob_Wallace

            What I’m currently looking at is using three MMPT charge controllers. One on each of the two ground mounted racks and one on the roof panels. I’ll rewire the ground mounted panels (want to move one of the racks anyway) and put them in series rather than paired 24 volts. Move from 5% loss with wires to zero loss.

            I’m best staying with DC to the batteries. Not inverting to AC at the panels and back to DC for the batteries.

            I’ll be able to save money on the roof panel cable – high voltage, low amps. And not have to wrestle big copper.

            The roof panels will be at a good angle for winter Sun since the roof is steep (54 degrees). Optimal angle for the winter solstice is 55 degrees. But that will make the roof panels underperform in summer. Using MMPT controllers means that I can get something useful out of the partially shaded rack in the winter and something out of the badly angled panels in the summer.

          • Riely Rumfort

            Hmm, what’s your battery capacity?
            I’m looking more towards >8kWh solar >3kWh wind myself, and maybe >15kWh storage, and when I say > I could more than double any of these values. My end load and even build location is still TBD, lot of work to do design wise before anythings solid. Gotta prioritize outputs of various sorts(pumps, sensors, heat/cooling, water purification, seedling lighting, etc) first. It’s going to be a hundred piece puzzle with piece sizes to blance and optimize. I also have to still decide which overall design to go with, few top candidates largely location dependent. I’ll be ironing all this out over the next 850days..

          • Bob_Wallace

            12 1.35 kWh Trojan RE-105. 16.2 kWh. 20% = 3.24 kWh usable.

            Trojan introduced the RE series a few years back for off grid use. Thicker plates than golf cart batteries. By the time you start installing there will likely be better options.

            I’m actually ‘abusing’ my batteries from time to time. Taking them down closer to 50% in order to avoid running generator. I might shorten their life a bit but 8+ years from now storage options should be much better.
            Actually I might have been better off buying another set of golf cart batteries as I have been getting 5 or more years out of them.

          • Riely Rumfort

            Read the spec sheet, the ‘smart carbon’ series eh, seems they do best between 40-80`F, and yeah, the occasional 50% DoD won’t kill um. I’d expect 7-9years with 1 in 3 discharges being 50%. Still, for the price 2K-4K cycles isn’t bad. As long are you don’t overheat when full, underheat when empty or go to either extreme during your discharges(especially heat) you should be fine.
            I’m sure the golf kart we’re losing capacity after the 5 years though, I think you’ll more than likely get >9 to match those losses on the Smart Carbon(RE) series. As you said it should more than cover you till a far cheaper replacement arises.
            My minds not made up but I’d like to have 3 chemistries at work in my system. 20%LAcid 70%NiMH and 10%LTO. The Lead will be for gradual constant load and overrage, the NiMH for the majority of tasks, and the LTO for quick high demand dumps and removing load from NiMH, such as a bi-daily primary pump and UV water treatment, protecting the NiMH from the damage which goes along with high discharge rates. I’ll have to truly make that call in the months prior to purchase.
            I would go 40%LAcid, 60%NiMH for stable load and higher demand but I’m not sure if 12 years will cross the chasm I think probable, so the 20/70/10 mix could make it 15-20 just to be safe and also to get a bit more overall efficiency. Either course, storage and REnergy is pretty promising these days 😀

      • milliamp

        With lead acid if you cycle them beyond 50% they don’t last as many cycles so you mostly already have to buy 2x your total storage need. Most lead acid batteries are under 1,000 cycles even when you don’t deplete them below 50% so that’s usually ~3 years. Yours are high for lead acid but I’ll bet they weren’t cheap either.

        It really depends on the costs involved and I don’t even know a baseline for long term costs of hydrogen storage unfortunately but if in theory it were half as expensive as batteries (though I doubt it) if solar panels become a cheap enough purchase it may be possible that it would be cheaper to buy several extra panels and use an inefficient but cheaper storage method to batteries.

        I’m not at all saying that’s the case now in fact it’s extremely doubtful but batteries are indeed a pretty expensive component and there is some value for those numbers for which that may be the case.

        To make my hypothetical case slightly more convincing take this link for example: http://www.altenergymag.com/content.php?post_type=1884

        That gives a hypothetical lifetime cost of $0.67/kWh throughput for VRLA batteries in a hot climate.

        Now take the announcement from First Solar that they plan to be at $1/watt installed by 2017 and now we have a LCOE for solar that is significantly below half the cost of the $0.67/kWh storage solution.

        Another point I want to mention that changes the math slightly is that with a 50% efficiency storage method the panel capacity isn’t quite doubled because daytime usage needs are met without storage so if only nighttime energy needs are met with a 50% efficiency storage method the needed solar panel overbuild may only be another 33%.

        With a cheap enough panels and a cheap enough storage method that may be a plausible solution. There is a similar place in New Jersey that uses hydrogen storage with some information buried around their website: http://hydrogenhouseproject.org/

        • Bob_Wallace

          If, if, if….

          It’s possible that solar could become cheap enough that we could afford to use inefficient storage. But at the same time we’re watching the price of batteries fall.

          My current lead acid batteries cost $633/usable kWh. (Usable means cycling no deeper than 80% charged.) GM will be paying LG Chem $145/kWh for lithium-ion batteries that can be cycled down to 20% without problem. Panasonic/Tesla batteries are expected to be less expensive. Battery prices are likely to drop below $100/kWh wholesale.

          I suspect that within a year we’ll see battery storage for under $250/kWh retail and battery life well over ten years with daily cycling.

          Are we likely to see the cost of compressors and pressure tanks fall? Those are mature products.

          Hydrogen is likely better than batteries for long term storage. But for that application hydrogen has to compete with pump-up hydro and flow batteries.

          Hydrogen might have a storage role. I’m very skeptical that it will be used with vehicles. I wouldn’t put any money on it finding a role on grids. Part of the reason I wouldn’t bet on grid use is because PuHS is already in large scale use with more being installed and flow batteries are going online. Hydrogen storage is not a new idea yet I see no grids giving it a try.

          • eveee

            Sounds like you might switch to lithium when your LAs wear out. That will be a while. That equalization and running at less than full DoD is the key to long life. Even LAs improve with battery management.

          • milliamp

            All good points, and yeah if hydrogen was cost competitive with where lithium ion batteries will be in a few years we would have likely already seen common use of it.

            Pumped hydro tends to still dominate that landscape so I am curious what the cost/kWh of pumped hydro is at large scale.

            On a sort of related note with grid solar hitting $1/watt installed it should start to open up the grid storage market a bit. PV seems to be gaining on thermal for the tech of choice for grid solar. The largest PV solar farms are hitting ~500 MW of capacity now.

          • Bob_Wallace

            It’s hard to get a handle on PuHS cost. There are multiple ways to build it. The most expensive is probably building a new reservoir in a valley/canyon with a concrete dam.

            But there’s also the much simpler route of excavating two reservoirs at different heights and boring a hole between them.

            Or, possibly cheaper yet, convert one of probably thousands of usable existing dams by creating a modest sized reservoir in the ‘safety zone’ and installing a pump/turbine.

            Possibly cheapest of all, use an abandoned rock quarry that already has a couple of ponds at different heights.

            Thing is, we don’t need long term storage and may not for 20 years. There’s data that says storage won’t be a big need until wind and solar are much larger than 50% of our electricity supply.

          • milliamp

            Right, the greatest need for storage is to handle a 3 to 4 hour window of peak usage just after solar power tapers off. That way coal plants don’t need to be brought online for just 3-4 hours before being shut down as demand drops again later in the night.

            Right now the cheapest grid energy comes at night so solar is still meeting demand during mostly high demand daytime hours.

            As prices differ based on time of day it will create a market opportunity for people to buy and sell on daily margins which will help flatten out the price and bring in storage revenue.

            Right now the problem is largely unsolved because we have mostly not yet had a need to solve it but it could be very well be pumped hydro. Once there is an abundance of daytime energy from solar anyone with a hill and land is free to pump water up it cheaply to sell it back to the grid at night.

            It’s a problem the market can solve.

      • Frank

        One tiny nit. I think in this application, they can use a much larger tank, so they likely don’t need to use nearly the pressure they use in a car. That would reduce the energy wasted in compression, and maybe lower the cost of some of the components.

  • Riely Rumfort

    Need way more stats and design specs.

  • Marion Meads

    So what’s their overall round trip efficiency for Electricity to hydrogen and hydrogen to electricity? If this super important metric cannot be answered by Phi Suea House, I am 100% certain that they are overhyping their system!

    • harisA

      When the choice is between have a resource and not having it, then efficiency is secondary.
      Cheap automotive battery lead acid battery/inverter based system are extremely in efficient (<50%), but are used by millions of people.

      • Bob_Wallace

        Less than 50% efficient? I think you made a mistake.

        • harisA

          Higher? Lower?

          80% efficient charger, 80% efficient inverter, 10% harmonic loss due to square wave AC, and 80% charge efficiency, gives about 46% efficiency.

          I am not talking about state of the art units.

          I did this calculation a few years ago and do recall the sources.

          If you saw some of these installation (hissing batteries, rotted cables etc.) it would be difficult to believe that these setup do better than 25%.

          • Andre Needham

            Sources for all those numbers? At the very least, it’s easy to see inverters are closer to 96-98% efficient just from a simple Google search.

          • milliamp

            The biggest efficiency loss is likely due the the energy that needs to be put into electrolysis to separate water into hydrogen and oxygen.

          • Bob_Wallace

            Battery efficiency about 85%.


            Inverter efficiency over 95%.


            No battery charger would be needed when using DC inputs as would be the case with this solar house. If one is starting with AC what I’ve seen is a 5% loss in converting AC to DC.

            “If you saw some of these installation (hissing batteries, rotted cables etc.) it would be difficult to believe that these setup do better than 25%”

            Just takes one tiny hole in the storage tank to make hydrogen 0% inefficient.

          • Mike Shurtleff

            80% x 80% x 80% = 51%
            0.80 X 0.80 X 0.80 = 0.51 right?

            10% loss to harmonics would mean 90% efficiency for your inverter. I think you’re making up numbers.

            Typical lithium battery system:
            DC charging directly from solar panels, so very high efficiency there
            >95% round trip efficiency for lithium battery
            >95% efficiency for inverter
            95% x 95% = >90% efficient

            I don’t understand. In the context of this article you are comparing a super duper state of the art 100% solar house with H2 energy storage to junk solar+storage systems?

            Bob has better number down already and I think he’s using 85% for lead-acid batteries, nothing fancy. Lithium batteries are typically over 95% round trip. He also correct that low cost systems in poor areas can benefit from using DC only.

          • Otis11

            Even terrible inverters and chargers are at 92% now… 96% is not hard to find cheaply (Can even get 98% in some cases, but that’s besides the point). Harmonic loss due to square wave AC? No, just no.

            Now, if you’re really using old lead-acid technology, that’s only about 70%. More realistically – new system, new batteries – 92% lithium ion (or better), but if you really wanted to use old batteries I suppose you could make that argument. (Though you’d be oversizing the solar panels to do so, so you’re likely better off buying new batteries… and if you’re comparing it to this system – that argument is bunk.)

        • Mike Shurtleff

          Maybe he has corrected since your comment, but looks like simple typo to me: “in efficient” should be “inefficient”

          • Bob_Wallace

            I read it as inefficient. Less than 50% efficient.

            My last word should have been “efficient”.

      • Mike Shurtleff

        Agree, but I would say this differently. Even without the have or have not choice, efficiency is just one factor in cost. It is an important factor, but cost of usable electricity …in this case 100% of the time …is what’s important.

  • Steve_R

    “it is more efficient (electricity to hydrogen to electricity) than the battery systems that are available,”
    Huh, what, how is that?

    • Marion Meads

      It can be if they were referring to the batteries made during the early 1900’s.

      Today’s round trip efficiency for large grid-scale battery energy storage system is 90%, accounting for all losses. I haven’t seen any electricity to hydrogen to electricity with a round trip efficiency greater than 40%. It can go near 90% if they count heating as part of it, but you certainly don’t use a lot of heating in tropical countries like Thailand.

      Of course, if you use the heat in the battery energy storage system, the efficiency can go as high as 99%.

      • Bob_Wallace

        Nope, not much heating going on in Thailand.

        About a week ago it got down close to 70F and people were wearing coats….

        • Marion Meads

          Taiwan, man, Taiwan. People are dying there when I could have just worn my bikini in the low 30’s temperature.

    • Riely Rumfort

      Possibly system longevity wise or long term? *shrug*

Back to Top ↑