Making a Solar Electric Water Heater | Can it Be Done?

April 1, 2015 · 23 comments

Would you do us a huge favor by sharing?

As the price of photovoltaic modules (solar modules, solar panels) continues to drop there are many discussions about using solar electric modules to heat water. No more leaks, messy and complicated plumbing, heat exchangers, controllers to fail, pumps to maintain and fluids to freeze.

Solar hot water has been plagued with leaks, freezing (or poisonous antifreeze), air in the pipes, controller and pump failures just to mention a few of the difficulties we have had over the past few decades.

However it was the only way to make hot water with the sun. Just a decade ago solar electric modules cost over $10 per watt. Solar thermal was much less expensive per kW or BTU.

Our goal is not to bash solar thermal hot water as it serves a purpose, has performed well for many including our own homestead, and produces more hot water per square meter (or square foot) than photovoltaic ever will.

We currently have a 20 tube evacuated solar hot water heater in conjunction with a heat exchanger, solar differential controller, circulating pump, solid state relay and it works incredibly well when it works. It provides about 50-60% of our hot water in the summer and 10-20% in the winter, which I consider pretty impressive when you consider we have a family of 6.

The photo below shows our solar hot water collector (with the pipe insulation removed) as we were looking for a leak on a cold winter’s day.

Evacuated Tube Solar Hot Water CollectorThe problem is there always seems to be something wrong with the system…air gets in the system, the collector and/or piping leaks, the controller forgets to turn the pump on, the controller forgets to turn the pump off etc.

However, before go out to your local hardware store and buy an electric heating element and eBay for a solar panel there are a lot of challenges with solar PV water heating. There are correct ways to do it and incorrect ways to do it. If you do it wrong you might be lucky and get little to no hot water or you might be unlucky and burn your home down.

To our knowledge there are no “off the shelf” solar electric water heating systems approved for use in North America. There is a product in Europe called REFUSOL by Krannich Solar but it is CE marked for Europe but not UL or CSA approved. Maybe they are working on that but who knows.

solar pv water heating system Rufusol

Refusol has three separate solar DC inputs that will track the maximum power point voltage (MPPV)  of your solar modules, combine it and feed it to the water heating element. We will talk more about this in a few minutes.

Next we are going to discuss the ways we can heat water with regular photovoltaic (PV) solar modules and electric water heating elements to make a usable amount of hot water for domestic use or home heating.


breakers and fuses for a solar electric water heater



As it stands right now there are at least eight ways we can use regular solar modules (photovoltaic) to heat water in a standard tank water heater. All of the systems below do not mention over-current protection but as mentioned above it is a must.

If there seems to be interest in heating water with photovoltaic modules, we will write articles (one article per method) for all eight methods complete with photos and drawings to explain each method thoroughly including where to install and how to size the over current protection.

1. Pick any solar module or group of modules and connect the power output to any water heating element. No thought, no controls and almost no chance of it working.

2. Connect a solar array directly to a hot water element (making sure you match the maximum power point voltage with the rated voltage of the water heating element) with no controls and no other components.

Low voltage water heater elements for heating water with solar.

3. Connect a solar array to a maximum power point tracking (MPPT) charge controller with tiny 12-48 volt battery bank and setup a typical diversion/dump load system. Use the programmable relay driver (that almost all high quality MPPT charge controllers have built-in) to control a solid state relay to turn the low voltage (12-48 volts DC) heating element on and off. Once the tiny battery bank is full, the electricity will be diverted to your low voltage heating element.

You may also use off the shelf 120/240 volt AC water heating elements.
So far all the options we have mentioned don’t include any protection to prevent the water from boiling in our water tank.

This is very similar to the “good old days” when everyone had a range boiler behind their wood-stove.

A range boiler is a water tank that is usually placed behind, beside or above your wood-stove. Inside your wood-stove is a hot water front, metal tubing coil or U-shaped pipe or tubing to collect the heat from the burning fire. This hot water front or similar device is piped to the range boiler. When the fire is burning, heat is transferred from the hot water front to the range boiler and eventually the water will begin the boil.

In the good old days folks would simply open a hot water tap when they heard the water begin to boil. Remember this is a time before pressure relief valve’s were invented. In fact many people have been killed in the last century by using range boilers without some sort of pressure relief valve or system to dump the hot water before (or when) the water begins to boil and turn to steam.

The next few methods of solar electric water heating we discuss will all include some type of protection to stop the water from boiling. When water turns to steam it expands 1600 times. In 2015, with today’s technology, we would be crazy not to have some type of device or devices to prevent the water from turning into steam.

The following methods of heating water with solar electric modules include safety equipment to prevent the water in your system from becoming steam.

4. Connect a solar array directly to a hot water element making sure you match the maximum power point voltage with the rated voltage of the water heating element. Using a solid state relay (SSR) and a thermostat/aquastat we can set the system to automatically disconnect the solar array when the water reaches our preset temperature. We will also need a small power supply to turn the SSR on and off. Usually they are either controlled by either 3-32 volts DC or 120-240 volts AC.

electric thermostat melted and caught fire




5. Connect a solar array directly to a hot water element making sure you match the maximum power point voltage with the rated voltage of the water heating element. Use an “off-the-shelf” pressure/temperature relief valve to dump the hot water down the drain before it becomes steam. Wasteful in terms of water but simple and effective.

6. Connect a solar array directly to a hot water element making sure you match the maximum power point voltage with the rated voltage of the water heating element

focused laboratory studies. buy cialis usa After absorption of light, rhodopsin stimulates PDE6 via the G-protein transducin..

but a reduction of the erectile response (11). A chronic alcohol abuse has a negative effect cheap levitra understand..

– lower range (‘normal’)on your partner relationship?” canadian pharmacy generic viagra.

Results from batch analyses showed that all batches complied with release specifications and demonstrated acceptable batch to batch consistency. canadian generic viagra In the USA, the Massachusetts Male Aging Study, reported.

reflected of the reproductive organs. a state of full erection. Thefunctioning and 84% said they had never initiated a viagra no prescription.

doctor the risk puÃ2 be minimized.attempted sexual Intercourse in the past 3 months. For sexually inactive individuals, the questionnaire may be order viagra.

. Use a thermostat/aquastat and a solenoid valve to dump the hot water as the temperature of the water approach’s its boiling point. Of course we would also require a power supply to operate the solenoid valve.

7. Connect a solar array directly to a hot water element making sure you match the maximum power point voltage with the rated voltage of the water heating element. Using a thermostat/aquastat, a water pump, a solenoid valve, circulating pump, power source for solenoid valve and pump and a length of pipe buried underground (or concrete) we could dump the heat into the ground or concrete.

8. The last and most complicated (and most expensive) method of heating water with solar PV would be as follows:

  • Connect solar array to a charge controller (hopefully MPPT), to a battery bank. With this system we do not need to try and match the solar module voltage with the rated voltage of the water heating elements. Instead we will match the solar module voltage with the requirements of the charge controller and battery bank.
  • Connect the battery bank to a 120/240 V inverter.
  • Connect the inverter output (120/240 volt AC) to an off-the-shelf 120/240 volt AC thermostat/aquastat an then the 120/240 volt heating element.
  • Install an AC solid state relay (or standard relay suitable for AC) in the line between the inverter and the thermostat(s) and drive it using a relay driver from either your charge controller or your inverter. The relay driver in your charge controller or inverter is completely programmable to turn on at a specific voltage and off at a specific voltage. DO NOT USE THE PWM (PULSE WIDTH MODULATION) OPTION ON YOUR PROGRAMMABLE RELAY DRIVER. Only use the on/off setting with a hysteresis (difference of a few volts). PWM will destroy your inverter.
  • Last but not least make sure the solar charge controller (the one mentioned in the first of these bullet points) is set to turn the solar array off when the water gets too hot. Do not want to overcharge the batteries when hot water is not longer needed.

This method has the advantage that you can use off the shelf elements, thermostats and over-current protection (fuses or breakers). You will essentially have built an off the grid solar system that can be used (or expanded) to operate other electrical appliances.

Let’s not forget we have been doing this for years in our solar off grid systems as dump (diversion) loads.

In a traditional off grid solar system the batteries will get full (hopefully). To prevent overcharge we currently deal with this by:

  1. Disconnecting the solar array(s) as needed to keep the voltage where it needs to be
  2. Dump or divert the excess energy into an air heater or other type of resistor or
  3. Dump the excess electricity into a hot water heating element.

Usually the excess electricity is dumped into a special low voltage DC water heating element. Check out Dump Load / Diversion Load Intro to learn about this process. You may also use 120/240 elements in certain circumstances.

So heating water with photovoltaic modules at 12-48 volts DC is nothing new but heating water with 120 or 240 volts DC is a whole new ballgame. You could easily produce an arc about 2 inches (5 cm) long with 240 volts DC.

Imagine what that would do to a AC fuse or circuit breaker?

High voltage DC is extremely dangerous and requires some special controls, over current devices and ingenuity.

Cancel reply

Leave a Comment

{ 23 comments… read them below or add one }

Andrea February 15, 2016 at 7:40 am

Hello Jody,
I really appreciated the way you share your precious knowledge.
Although I’m a technician and not young unfortunately, I was trying to elaborate myself the way to warm up water in my 240 volt AC boiler, but without reaching a sensible configuration.
Up to now, for me, point 8 is the best solution one could think.
I have already all devices except for what you call “tiny battery bank”. I don’t want to store energy in batteries but my MPPT charger needs a 48 V DC reference voltage to switch itself on ( I have a 48 DC / 240 AC inverter).
Will 4 motorscooter FLA batteries (12v – 4Ah) be the cheaper solution? How long they will last, two years?
Do you have a better idea?
Thanks in advance
From Italy
Andrea Trentin


Jody Graham February 15, 2016 at 2:14 pm

Hi Andrea,

Thank you for your comment. I like your idea for heating water with pv. The 4 AH batteries are quite small but should be large enough to give your mppt controller a reference voltage to start dumping power into your hot water elements. Are you planning on using low voltage elements or 240 VAC elements? If you are using 240 VAC elements make sure your inverter is large enough to power the elements in your storage tank. How much pv are you using and how big in wattage are the elements?
Thanks and greetings from Canada,
Hope you are warmer than us!


Andrea February 16, 2016 at 6:16 am

Hi Jody,
thanks for answering. I have an ACS (Acqua Calda Sanitaria ;-)) boiler with a 36 Ohm element, at 240 VAC makes 1600 watts, my inverter is 4000 watt.
The pv source is made by 24 panels (130 watt each) divided into 4 arrays. Unfortunately during winter they are often shadowed by surrounding buildings so I decided to use them only when the sun is high in the sky to help warming my water. By the way speaking about tiny battery pack: instead of motorscooter batteries i can have, more or less for the same price, some UPS batteries: do you thing they will last longer? (I hate batteries!)


Roger Davenport February 16, 2016 at 3:06 pm

There are a couple of approaches for making a PV-powered water heating system. Butler Sun Solutions, Inc. has developed a heating element that directly takes DC power up to about 1500W and fits into a standard water heater. The big advantage of that system is the ability to convert tanks that have gas backup. has a controller that will take DC from a (small, about 750W max) PV array and convert it electronically to allow it to power a standard AC water heater element. It includes MPPT (max. peak power tracking) and a good feature is it doesn’t require a change to an existing electric water heater tank. But the power level is limited, so the solar fraction isn’t that high.

All this being said, the cost of PV has come down to the point where it is likely cheaper, especially if there is a net-metering arrangement with the utility, to install grid-tied PV and use a standard AC water heater connected to the grid for your water heating. The reasons are that grid-tied inverters all have MPPT, so they wring as much power as possible from the PV panels, and that with net metering you are never “throwing away” power from the system. That is, with dedicated solar water heating systems, when the tank heats up to its maximum temperature you have no choice but to turn off the solar power, and that power has nowhere useful to go. But a grid-tied system will deliver excess power to the grid resulting in either a credit on the bill or the ability to take that energy back later. Finally, a heat pump water heater can provide a quantum leap beyond just a conventional electric water heater in terms of maximizing efficiency and minimizing the size of the PV system needed, at the risk of introducing moving parts and some other idiosyncracies (noise, slow response, lifetime and maintenance issues).


Victoria Runda January 4, 2016 at 10:50 pm

My husband and I have been looking at various ways to become more self sufficient while also doing what we can to help out the environment. We have been hoping to make a single large improvement along these lines each and every year. Last year we put in a well so that we wouldn’t be such a large draw upon the cities infrastructure. Putting in a solar water heater like these may be what we will do in 2016.


Jody Graham January 7, 2016 at 10:57 am

Hi Victoria,

Thank you for your comment. Cool idea to make your own well. Less of a demand on the municipal water system and gives you a bit of independence assuming you have the mains power to operate the pump. Solar hot water would also be a great addition to your independence. I would consider at some point installing a solar electric system that would operate your well pump in the event of a large scale power outage. Good luck and thanks for commenting. Jody


Roger Davenport November 6, 2015 at 10:54 am

We just got OG-300 (system) certification for a direct PV-to-heater retrofit system for natural gas and propane water heaters. The PV Wand(TM), as we call it, is a low-voltage (<30VDC) heater element that screws into your existing water tank to provide solar heating. It is powered by 3-6 standard (60-cell) PV panels, and protected from overtemperature by a DC-rated, UL-approved relay triggered by either an electronic circuit we build, or a simple thermal snap switch. The relay and electronics are powered by the PV array, so no connection to the grid. See our website for more details:

We're still debating with SRCC about the rating they have given us, since they had to mash our system into their solar thermal rating "box" to give it an OG-300 rating, but at least it is now eligible for rebates.

You can see a system operating in real time at This is the system on my house. It is presently operating with three PV panels (0.69kW), but all last year we had it connected to six panels (1.38kW) and it provided 62% of our hot water for the year (~3-1/2 adults in the household).

Finally, my calculations indicate that if you have an electric water heater, it is more effective to grid-tie any PV panels, and use the MPPT peak power tracking of the inverter to increase your output and net metering to provide seasonal storage, with no solar water heating components needed. But for natural gas and propane tanks, our system provides a way to augment them with solar to supply about half the load, with no problems of pumps, fluids, freezing, etc.


Brad Smith July 15, 2015 at 6:38 pm

“To our knowledge there are no “off the shelf” solar electric water heating systems approved for use in North America”

How is it possible you have not seen the product Sun Bandit? It has been sold in North America and globally for two years now. It also carries UL, CSA and SRCC OG-100 certification… Nobody will ever get UL or CSA approval with DC power.


Jody Graham July 17, 2015 at 2:09 pm

Hi Brad,

My apologies for that. I was under the impression you had to have grid power to use the Sun Bandit. I will check it out asap. I wouldn’t be too sure about never getting CSA/UL approval for DC. Fifteen years ago I never dreamed we would have charge controllers, grid tie inverters, and fuses and breakers that would be approved for 600 volts DC. Nor would I ever have believed solar pv would be below a dollar a watt. Times have certainly changed. I like the idea of the Sun Bandit converting to AC right from the start. Makes everything so much easier and easier to convince electrical inspectors. Thanks you for your info. Anyone have any experience with the Sun Bandit?
Thanks again Brad and take care…Jody


Carl September 17, 2015 at 5:18 am

Appreciate your haughty and elitist approach here Brad. Managing to call Jody stupid/incompetent, pump your product, and refuse to sit down for a minute to contribute all in one short paragraph. Maybe just a cover for knowing nothing.

As to your question, gee, I don’t know. Researched this 6 months ago and you were nowhere. Maybe cause your product appears to be a sham and thus has been totally boring?

Apparently you have not been able to look far enough down to notice, but the intent of us stupid proles is to heat h2o largely by or even totally by the sun. Not a system mostly dependent on oil/coal as yours appears to be.

Why does it seem so? and dishonest/criminal? Because you’re hiding!

Your site offers nothing of substance and is created by marketeers with no connection to, knowledge of, or enthusiasm for the gear. While hoping for more, the minimum one would expect would be, *comparison by square footage/cost to shw panels and evacuated tubes. *Basic efficiency, as in how many watts lost thru micro inverter. Can run top and bottom elements, blah blah.

You only say, for example, that 2 285 watt(?) panels will heat 41 gallons in a day. Wow! where’s my wallet? Starting temp, finishing temp, hours of direct and indirect sun all unimportant, obviously. % of wattage drop to volt drop in indirect solar? Shield me oh lord!

And even your few disinterested retailers don’t even post prices. Gotta beg for that.. probably from incoherent and confused call takers. Geez

So much else missing. Gotta go.

Hope this partially alleviates your confusion


Jody Graham September 18, 2015 at 9:05 am

Thank you Brad for your input. I appreciate your kind w0rds toward me and what we are doing.

While I am not a big fan of saying anything negative to a fellow homesteader or someone trying to become more independent you make some very good points.

I did some research myself on the “Sun Bandit” and I can’t seem to find any information regards to how much hot water will be produced and at what efficiency. There does not seem to be any independent testing or even customer testimonials that say whether the unit is good, bad or otherwise. Of course testimonials can be faked but usually someone with such a unique product at their home or cottage would blog about it somewhere. I cannot find anything.

I also am not too excited about a gas or electric heater that uses a couple of solar panels to boost the water temperature. Does it work? Maybe. But how well and how much money can one expect to save by using the product. I cannot find any BTU rating or the like.

The purpose of this article was to show that as PV becomes less expensive it might make sense to make hot water with solar electrcity as apposed to using traditional solar domestic hot water equipment.

I agree the website is beautiful, almost as if it was made by professional internet marketers.

One super important piece of info I cannot find is the price of each unit.

Like I said I try to keep things positive but things don’t add up on the Sun Bandit for me. Thanks for bringing this to my readers attention. If anyone has any questions please contact me at

Brad, if you could clear things up for us like pricing, BTU output, expected savings in certain climates, how the unit works, and can we use the unit if we have no gas or electricity? I would love to hear from you. I have no ill will towards you and I wish you every success in your business ventures.

I remember when Trace came out with the SW4024 and everyone thought it was just too good to be true. But it ended up being the biggest success Trace ever had. Everyone wanted one. We used two until they became obsolete. I wish I would have kept them for sentimental reasons. Since then we switched to Outback VFX inverters and while I love them the SW still has a place in my heart.


Carl April 6, 2015 at 1:55 am

Hola Jody,

Well that’s quite a sdhw setup you have. My goal if was to do sdhw would be, in order to avoid problems, a passive setup like the Solahart or Rheem RS21-BP. Identical, same company now. But, would rather do Pv direct water heating and based on one kwh=3,143 btu’s 1500 watts of pv should at least equal and probably exceed the output of the best flat plate and even evacuated tube 21 sqft collector. Takes up 5 x as much space but if you got the space it might be worth it. Still less problems and dependence on future supply chain.

My intention here is to be somewhat verbose and spell out more clearly then you need so as to help others and draw those in who might have good ideas or especially experience. But just cover a little each reply.

Not sure way you phrased your title will pick up many googlers though.

For those new to this subject an excellent start is Jody’s thread on using an ac element as a dump load. Here you will learn basics of how to choose elements to match a battery voltage. Basically using the formulas E=IR and P=IE to employ an element with the least resistance.

Imo the way to approach this is to first cover the best way to get energy into the water and then pursuant to that discuss how to control over heating. Some overlap I’m sure.

So here is the first discrepancy I notice between using an element as a dump load and pv direct hot water. (PVHW ?) As you have demonstrated so well it is critical, with batteries and dump loads, to have the element voltage be as close to the source voltage.

* But with pvhw once you get above 60 volts it seems irrelevant. Take a 1500 watt 12o volt element, thus a resistance of 9.6 ohms. Panels are(for most of this discussion) 250 watt, mpv 30 volts. If wire 2 together in series you have 60 volt 500 watt. 60v/9.6 ohm = 6.25 amps x 6o volts = 375 watts so have lost output. But say 3 panels at 90 volts 750 watts you get 90v/ 9.6ohm=9.375 amps x 90 volts= 843 watts. More then the 750 watts possible so have maxed out the panels. Difference between this and dumping is the panels are restricted, unlike a battery that can pour unlimited juice into an element, at least for awhile.

Now, on the other side, what happens when panel voltage higher then element voltage?
say you do a 180 volt string with my proposed 1500 watts of panel and attach to same 1500 watt 120 volt element. Panels are maxed out same as they were at 90 volt, just more so, so no losses. 180v/9.6=18.75 amp x 180v=3,375 watts. (again, this belabored for other’s benefit)

But, the high voltage is a danger even though wattages match because you have more amps going thru element then the 12.5 is rated for (120v/9.6 ohm) so will overheat element and burn up.

Please tell me if you agree/disagree with anything said here or in future as I don’t always phrase things as a question if feel confident. Just learning here as are you I guess but maybe we can work it out. Get’s more complicated then this as get into mppt and other stuff as I’m sure you know. And other’s, take off your lurking hat and pitch in!


Jody Graham April 6, 2015 at 11:29 am

Hey Carl,

Thanks so much for your comment. I am on the road right now so I’ll have to keep this short. I chose the title as the most common phrase folks are Googling is “solar electric water heater” when they are searching for water heated with photovotaics (PV). I researched hundreds of keyword phrases and that was the most popular one. There are hundreds though. The good thing is Google is getting really good at matching phrases that mean the same thing. They call it Latent Semantic Indexing. If fact now if you use the keyword phrase you want to rank for more than once or twice they will penalize you. Enough about Google. As far as my SDWH I have been pretty unimpressed with the time it takes me to maintain it and the unreliability but I am sure it would be easier where the temperature doesn’t go down to 35 below. We are going to do some real world testing like mismatching solar panels with heating elements and then trying to get as close to the MPP voltage of the element and see what the real difference. Maybe it is as easy as sticking solar modules on your roof and connecting to the element in your hot water heater but I doubt it. Anyone with any thoughts or real world data?

Your first example 1500 watt 120 volt element with two 250 watt 30 vmpp modules. You are correct you will only get 375 watts roughly making your losses quite high. Use a 120 volt 2000 watt element (they do exist) and I am guessing that would be a perfect match. Roughly 500 watts output of heat.

2nd example 1500 watt 120 volt element with three 250 watt 30 vmpp modules. I am not sure what would happen. The modules obviously won’t make more than 750 watts but will they even produce the full 750 watts? Where will the voltage settle and how much output will there be?

3rd example 1500 watt 120 volt element with six 250 watt 30 vmpp modules. 180 volts going into a 120 volt element. Obviously the element would try to consume 3375 watts at 180 volts but there is only 1500 available. Definitely will not damage element as it is designed to take 1500 watts and that is all we have. In a battery bank or unlimited power supply the element would not last long but pv is current limited and obviously watt limited. I am not sure if the modules would even put out 1500 watts. It depends on where the voltage settles.

As soon as the snow melts (we still have 5 feet left) we are going to do some real world testing with our three solar arrays here to see what really happens.

I am going to think about what I would use if I wanted to create a 1500 solar electric water heater and get back to you asap.

Thanks Carl I appreciate your thinking “out of the box”. There are still so many “still in the box” and stuck there…



Carl April 7, 2015 at 4:05 pm

Oops, that third example was perfect example of how often I short out. Obviously the element can’t overload because such string of panels in series and can only produce 8+ amps. Realized that a few weeks ago when studying this but somehow…….

Anyway, main point was above 60 v0lts it doesn’t matter what element you use, at least according to any math I’m aware of. Perhaps though you are aware of a way in which it could and if so you will say so.

Am very interested in what you mean by “where the voltage settles” Why wouldn’t such string not put out 750 watts in full sun? If you don’t know of a rational reason do you have an experiential reason to suspect this?

My very limited grasp of mppt issues in this area suggests to me that’s not what you are talking about as we are, for the moment, imagining perfect conditions.

Was going to bring up mppt soon though, once preliminaries dispensed with. This is, of course, the nut of it

You have too much snow now and I’m not ready to buy panels yet without more understanding so good time to get the theory straight.

We’re just two blind men ruminating about the elephant. No one is expected to be an expert here or not misspeak. Too much else going on to avoid that I’m sure. Besides, no edit button.



Jody Graham April 8, 2015 at 11:06 am

Hi Carl,

It should matter what the voltage is even if it is over 60 volts. I am not an electrical engineer and even when I have worked with them they have no idea about pv. It was not taught to them how a solar module works and what happens when one open circuits or short circuits. Maybe there is one reading this post that has a good grasp of solar modules.

My experience tells me that if we connect a 120 volts mppv of solar to a 120 volt element (or 240 volts mppv to a 240 volt element) the module should produce the most hot water. Any variation in this will make the system less efficient as the operating voltage will not be at the sweet-spot of the module. Keep reading to see what I mean by sweet spot.

One of the drawbacks of a solar module is they are current limited. Let’s use your 250 watt 30 volt (8.33 amp) modules. At 30 volts the module will produce close to its rated output 30 volts x 8.33 amps=250 watts. However if you connect that module to a dead 12 volt battery (10.5 volts roughly) it will now produce 10.5 volts x 8.33 amps=87.5 watts. Connect to a 24 volt battery and the output will be 24 volts x 8.33 amps=200 watts. The most the module can produce is 250 watts and that is at the sweet-spot of 30 volts in our example. Reduce the voltage and the output will be less. Increase the voltage and the output will be nothing. The load (or battery bank) is what controls the voltage of the system (or what voltage comes out of the solar module). It is the same with wind and hydro. You could connect a water turbine to a 12 volt battery bank and you have a 12 volt water turbine. Connect it to a 24 volt battery bank and you have a 24 volt water turbine assuming it is spinning fast enough to produce 24 volts.

My backup battery charger I use when we get weeks of bad weather in the winter is a 10 hp diesel engine with a Ford 100 amp alternator. It has no internal voltage regulator so the faster you spin it the more voltage (and power it makes). I have a 48 volt battery bank and can produce up to 100 amps at 60 volts with a standard Ford alternator. That is 6000 watts. If I were to connect the same homemade generator to a 12 volt battery bank it now has the potential to make 100 amps at 15 volts or 1500 watts. Same exact setup with a different voltage battery bank. Now I don’t ever run it at 100 amps because it would probably burn up in a month or two. I generally run it at 50 amps and get about 5 years out of one $100 alternator. It is outside summer and winter. Super efficient and very cheap.

All an MPPT charge controller does is try to find the sweet-spot voltage of your solar array to get the most watts out regardless of the battery’s voltage. The MPPT controller will try to connect at your solar array’s sweet spot and then reduce the voltage to put it in the battery bank.

When we connect a solar module or group of them to a heating element the voltage is going to settle (end up) somewhere. If it ends up at the sweet spot or maximum power point voltage of our solar modules we will get every watt out possible. If the voltage ends up much lower than the output of the modules will be much less. The voltage will never be higher as solar modules are voltage limited too.

Hope that makes things a little more clear. Have a great day Carl. I appreciate your questions and insight. Jody


Carl April 15, 2015 at 5:06 pm

Hi,– Good stuff, which revealed to me my blind spot. That being, while I was aware of the principles behind mppt when used as a charge controller, I thought such logic was restricted to batteries and not just any resistance. I.e, some foggy idea about batteries having their own inherent voltage thru chemistry whereas an element is just a dead hunk.

Okay, so hooking 180 volt, 1500 watts panel to 120 volt 1500 watt element will only give me at most 996 watts.

Still though, I don’t get your statement “Increase the voltage and the output will be nothing”

I.e, It still seems there is nothing wrong with the panel voltage being less then the element voltage and that this is the way to go. On the smallest scale hooking 750 watts 90 volts panel to our 120 volt 1500 watt would still give us the full 8.3 amps x 90 volt=747 watts.

Or in my case, if hook 180 volt 1500 watt panel to 240 volt element there can be no loss at the common 5500, 4500, or 3500 watt choices. Apparently they do make a 240 volt 1500 watt element that would cause a loss, giving only 876 watts so do have to be a little careful.

Imagine I’m missing something here though.

Carl April 23, 2015 at 5:59 pm

Hi Jody,

You still here? Guess I was wrong. I do need to step this along a little quicker. If you’re not sure of what is true just say so, no big deal.

Johann June 16, 2015 at 9:00 pm

The actual watt out put from the heat element is not the wattage a panel has, but is dependent on the voltage output and the amp output.
In your example you have used panels with 250 watts output eac, but the max amp is only about 8 amps.
So, even if you have the voltage in series connection, you still do not have the amperage from the panel needed by this.


Jody Graham June 17, 2015 at 11:27 am

Hi Johann,

Thank you for your comment. The actual heat output when connecting a pv module to a resistor is not a well known or well studied topic yet.The actual output will be based on many factors such as the maximum power point voltage of the modules, the maximum power point current of the modules, the resistance of the water heating element, the maximum load the heating element can handle, the angle of the sun and a few others. If you match the 250 watt module with the correct heating element you will be able to get very close to the full output of the solar module. Thanks again for your comment. We are going to do some real world testing with our three solar arrays over the summer but it has not been a priority as no one seems to care at this point in time. But we care and will have real world test results by the end of the summer. Thanks again and if you have any real world test results please share. We would love to hear from you…Jody


Jody Graham April 25, 2015 at 1:26 pm

Sorry Carl,
I missed that comment a week or so ago. I get hundreds of comments a day and 99% are spam/junk. Somehow I missed yours. Sorry about that. If 750 watts at 90 volts will produce the 750 watts or so why not just make two 90 volt arrays and parallel them to the same 120 volt 1500 watt element? That would put the full 1500 watts in (747 watts x 2 = 1494 watts) your element.

What I mean by increase the voltage and the output will be nothing is related to battery charging. If you have a 24 volt battery bank and you hook a 12 volt solar module to it, the output will be nothing as the module does not have enough voltage to overcome the voltage of the battery bank.

Why do you say hooking 180 volts and 1500 watts to the 1500 watt 120 volt element will only give you 996 watts?

We are done to a foot of now and will do some real world testing very soon. I will try to duplicate your 1500 watt array with a 1500 watt element and see what happens.


Carl April 26, 2015 at 4:01 pm

Hi, –Well, couple of miscommunications going on here. Perhaps if I start by backtracking and line it up sequentially.

I tried to start by reacting to past statements you had made about pvhw, (not dumping) that I felt did not apply to pv direct as such statements imo only seemed to be true up to 60 volts. I gave examples on 4-6 (also 4-15) where panel voltage both higher and lower then element voltage.

Whether high or low, you disagreed, saying on 4-6 that panel output depended on “where the voltage settles”

I asked what that phrase meant, (why voltage would settle) and on 4-8 you gave as an explanation an excellent description of mppt as it related to any charging source be it solar, hydro, wind, or genie and batteries. You began by reiterating your basic voltage matching premise with pvhw by saying that it was based on experience. However, that experience is with charging, not pv and elements.

By using charging as an example, you seemed to be saying that an element was a similar load as a battery and the implication there was that if one hooked a 180 volt 8.3 amp panel to a 120 volt element one would only get 120 volts at 8.3 amps=996watts.

Following your same mppt theme, you implied that if the panel voltage is lower then element the panel voltage would be overcome by element voltage?

It appeared this lowering of volts to 120 was what you meant by “voltage settling”

So this answers your parting question to me as to where I got this 996 watt figure. I actually don’t understand why an element is like a battery so don’t agree with this 996 watt figure! but I was just sort of agreeing temporarily in order to get past it and get you to explain why when the panel voltage is lower then element any voltage settling would occur. Because it just seemed to0 unlikely that the panel voltage could be overcome by element so couldn’t accept that was what you meant.

I gave more examples, in order to get at least a partial agreement that blanket statements about voltage matching were not true.

I would still like you to respond to my examples, at least where the panel voltage is lower, and explain your theory of how voltage settling occurs. For example, if one has 180v 1500 watts of panel and for any reason does not wish to parallel them, why would any 240 volt panel (above 1500 watts) not work fine? From 5500 to 3500 watts the resistance is too low to restrict panel, seems to me.

While perhaps confusing things by attempting to temporarily blow past this, I did basically say it and hoped you’d respond. Have long been aware of rational behind mppt controllers. Problem here is–the “voltage” of an element is a rating to allow you to figure safe/efficient heat transfer and energy cost. It has no inherent voltage like a battery, obviously, so how could the element “voltage” affect panel output voltage and therefore why important to match them?

There are mppt issues, maybe, and ways to rectify by putting controller in line, maybe, but should avoid till get basics covered. There are hundreds of possible combinations and uneven assortment of panels people may attempt, so it’s important to get the fundamental logic/math correct.


Jody Graham April 26, 2015 at 6:27 pm

Sorry for the confusion and miscommunications. The facts I wanted to make clear.

1. A solar module will only produce its rated output at one voltage we call the maximum power point voltage. It seems to be close to about 30 volts on most 24 volt solar modules.

2. When we connect a “30 volt” module to a resistor (heating element) the voltage is not going to automatically go to 30 volts to produce its maximum output. If the resistor is tiny the voltage will be higher than thirty and likely destroy the element. If the element is a huge consumer of power the voltage may drop down to almost nothing just like a short circuited solar module. In this case we will get almost no heat or no heat out of the element. ANY TIME YOU ADD A LOAD (RESISTOR/ELEMENT/BATTERY/MOTOR) TO A SOLAR MODULE YOU WILL PULL THE VOLTAGE DOWN. The bigger the element the lower the voltage will go.

3. Somewhere lies the sweet spot. At some point when you connect a combination of 30 volt modules to the correct element you will get the full output of the solar module.

A 120 volt 1500 watt element is designed to consume close to 1500 watts at 120 volts (12.5 AMPS)(9.6 OHMS). If the voltage is lower the element will produce less heat. If the voltage is higher the element will produce more heat until it burns itself up.

In my opinion, the only way to get the most power out of your solar array is to match it perfectly with your heating element or equivalent.

If we had 8 units 250 watt 30 volt modules (wired in series) coupled with a 2000 watt 240 volt element, the output should be very close to 2000 watts or full output.

If we have 3 units of 250 watt 30 volt modules that would be 750 watts at 90 volts. Using the formula

WATTS= 90 X 90 / 9.6
WATTS= 844

That doesn’t look promising to me as I think the voltage is going to drop as the resistor’s wattage is too high for the solar modules output. But what if we were really lucky and the voltage dropped to 85 (28.3 per module). Now we would have:

WATTS= 85 X 85 / 9.6
WATTS= 752

That would be really lucky…or would it? Now the modules are not operating at their sweet spot of 30 volts and won’t produce as much as they are rated for.

The only options I see are this:
a)Match the mpp voltage of the module and wattage of your array with your heating element…

b)Test different module and element combinations until you get what you want.

As you are trying to produce 1500 watts of hot water I would suggest the following:
3 units 250 watt 30 volt modules in series to get 90 volts.
3 units 250 watt 30 volt modules in series to get 90 volts.
Connect each array to a 120 volt 1500 watt element as in our example above. 2 elements in total

I think that is the best you will do without buying an mppt controller and a 12/24 volt battery with a 12/24 volt element.

Or buy 8 units 30 volt 250 watt modules and connect them to a 2000 watt 240 volt element. I know that will work. It can’t not work. Correct voltage and amperage for the heating element.

Keep watching as I will be doing some tests very soon. I cannot find anyone who has done real world testing anywhere on the net or my circle of solar folks from when I used to be in the business.

Does this make sense Carl?


Carl April 27, 2015 at 10:46 pm

Thank you, this helps, but allow me to run thru a couple things, tell me where I’m wrong if I am, which is probable. I studied enough electricity 30 years ago to put in a system but haven’t had to think much about it since and talking to you I can see I’m pretty rusty.

If I could quote your sentences and then respond it would make it easier to be understood but this is what we got.

Big plus is I now get how voltage settling can occur when panel voltage lower then element. A short circuited panel being a good example. I know, high school electronics, I’m positively creaking.

* But, under point 2–“If the resistor is tiny the voltage will be higher then 30 and likely destroy the element” How? Formulas aside the voltage on panel is restricted just like the current right?

* In a similar vein but different, under point 3 –“If the panel voltage is higher then element volt rating the element will produce more heat until it burns up”. So what’s wrong with my observation here–if I hook 180volt 1500 watt string to our now famous 120v 1500 watt element even if I got the full 180 it’s still only at most 180 x 8.33 amps or 1500 watts so how can it burn up.?

* What do you mean that at 85 volts your module will not produce what is rated for when you are still getting 752 watts?

Would it be accurate to say that the only thing that needs to be considered is the resistance of the element and that the element wattage and volt ratings are only relevant in that they allow you to figure the resistance?

So that when you say in your 90v string formula that you “think the voltage is going to drop as the resistors wattage is too high” what you mean is the resistance is too low and the lower the resistance the greater the voltage drop. I.e, if one went even higher to a 120/2000 element the resistance is only 7.2 ohms. If this lower resistance caused a drop to 70 volts instead of your 85 you’d only get 680 watts. (70/7.2 x 70=680) Theoretically, just for play, it may be that a slightly higher resistance over 9.6 ohm would be better even though other ratings way off. Like a 240v/3500 watt resistor is 10.4 ohms. Agree on the concepts here?

* Think I have seen oblique references somewhere to hooking a mppt controller between panel and element without a battery. No? Stupid? Won’t work?