Idea for direct solar heat storage

[Spudtoo]

Hello Everyone,

New to this forum so be gentle. I've got a very early concept idea for directly heating a concrete block using second hand PV panels.

The basic idea being a DC heating element embedded in a concrete block (heat storage) warms during the day and emits heat gradually thereby smoothing out the PV output. I'm thinking a max 2kw, the same as my wife uses on a fan heater which cost us a bomb to power last winter.

Any feedback welcome on this idea, particularly any ideas how the PV panels need to be connected to the "heater element" and what sort of "heater element" I could use.

Some requirements:

• I don't want to introduce batteries
• I don't want to use an inverter
• I intend to use second-hand PV panels
• This is a "top-up" heating solution - the house has gas powered central heating

Updated Diagram below:

 

[Maitake]

1600w*3hours=4.8kwh/day = 16KBTU
That's not a lot of heat.....about a pint of fuel oil.

Also without an MPPT into a resistive load you will likely get half or less of that power.

 

[Spudtoo]

1600w*3hours=4.8kwh/day = 16KBTU
That's not a lot of heat.....about a pint of fuel oil.

Also without an MPPT into a resistive load you will likely get half or less of that power.

Thanks for the reply Maitake - how would the MMPT increase the power to the heater element?

Are you talking about something like this:

 

[efficientPV]

I remember seeing this house with a large concrete block in the canter as a heat source. This was common in the pioneer days when fireplaces were massive structures to maintain heat all night. Circuitry wise, it is fairly easy to maintain a panel at power point voltage. Multiple heater strings can be switched in and out to maintain proper resistance. I run four different PV heaters at my house. You want to heat with as high a voltage as possible, forget battery voltage elements. With more watts I would consider concrete. That much power should be dumped when you get it. I have a couple of oil finned heaters which are good for 400W and still touch safe.

 

[Maitake]

Thanks for the reply Maitake - how would the MMPT increase the power to the heater element?

Are you talking about something like this:

View attachment 109032

That is an MPPT, but I think it needs batteries.

 

[OzSolar]

Heating elements are a fixed resistance.

Watts = Volts² / Resistance

Watts = Heat

More voltage = exponentially more heat

Small changes in voltage make big differences in heat.

I'm getting ready to try an experiment with putting modules in series hooked up to standard water heater element. They results may be shocking. (see what I did there?)

 

[efficientPV]

Everyone is accustomed to seeing 30% increases with MPPT in battery charging. This is because the battery limits how low the voltage can go. With pure resistance voltage can drop to almost zero. Here is a little comparison. It is the power of squares.

Panel .......... Increase with power point control over
Current ...... Direct Connect to panel
100% 0% example 10A 60V = 600W (6 ohm resistor heating element)
90% 10%
80% 25%
70% 50% example 7A 60V = 420W 7A X (7A X 6 ohm) = 294W
60% 67%
50% 100% example 5A 60V = 300W 5A X (5A X 6 ohm) = 150W
40% 250%
30% 333% example 3A 60V = 180W 3A X (3A X 6 ohm) = 54W
20% 500%
10% 1100%
It all adds up. If you have perfect skies, you can say it doesn't matter. One study indicated that the ideal resistance wasn't ideal at all for total daily production. A resistance of almost twice the ideal produced higher daily yields by sacrificing some peak production and getting more the rest of the day.

 

[OzSolar]

Everyone is accustomed to seeing 30% increases with MPPT in battery charging

Are you talking about more watts or more daily watt hours?

 

[efficientPV]

Are you talking about more watts or more daily watt hours?

I'm not particularly talking about anything. This is just a number that sticks in people's minds as sounding reasonable, as in the difference between a panel operating at 18V vs 14V. In one study between direct connect and MPPT they quoted about that much of an increase yearly total. However, that ignores practical considerations. When you need heat with less than ideal sun, direct connect will produce next to nothing. While power point can give you useful amounts.

 

[Spudtoo]

Heating elements are a fixed resistance.

Watts = Volts² / Resistance

Watts = Heat

More voltage = exponentially more heat

Small changes in voltage make big differences in heat.

I'm getting ready to try an experiment with putting modules in series hooked up to standard water heater element. They results may be shocking. (see what I did there?)

Hi OzSolar,

Do you have a rough wiring diagram or perhaps a list of the elements of your experiment? Sounds just like the heating solution I want to build.

Thanks.

 

[Maitake]

Direct dc water heating with MPPT

 

[efficientPV]

Those are back on the market again. I don't think you will find engineer777 recommending the techluck these days. He had a lot of failures the guy wouldn't fix and people were stuck with nothing. They have a flaw in the design which makes them very prone to lightning spikes. I was given one to fix and every semiconductor on the board failed. And you couldn't get the programmed proprietary microprocessor. Only the protective devices were still good! The power supply will short and provide a direct path to the micro. Those three energy storage caps are only rated at 0.68A each and there have been reports of those going bad after several years.

This is one I built for myself and that capacitor bank is rated at over 20A. More important it can operate off the same panels as my MPPT charge controller. This 40 gallon is located in the garage and heats water only for the washer. It uses the tanks standard heating elements in parallel. It has lowest priority and the upper 15 gallons heats up to 120F and the lower element preheats water. I do all wash cycles with hot water and that front load washer has no mold build up. I have a 13 gallon tank for the house and just added a BOSCH ES4 because I am too lazy to build extra insulation to the 13 gallon. A new 4 gallon tank only cost $48 with FEDEX shipping included, How could I pass that up. All three tanks work on only the excess energy not used by the house. The HW controller has arc interrupt so existing thermostats can be used, plug and play. I even modified the dishwasher heating element to run off the array DC.

I designed a 4KW microprocessor system for a guy in Canada with six heating elements in an old oil tank. Each element added on progressively as there was more sun. One element had variable power so there were no big jumps in matching the heaters to the array.

 

[Spudtoo]

Those are back on the market again. I don't think you will find engineer777 recommending the techluck these days. He had a lot of failures the guy wouldn't fix and people were stuck with nothing. They have a flaw in the design which makes them very prone to lightning spikes. I was given one to fix and every semiconductor on the board failed. And you couldn't get the programmed proprietary microprocessor. Only the protective devices were still good! The power supply will short and provide a direct path to the micro. Those three energy storage caps are only rated at 0.68A each and there have been reports of those going bad after several years.

This is one I built for myself and that capacitor bank is rated at over 20A. More important it can operate off the same panels as my MPPT charge controller. This 40 gallon is located in the garage and heats water only for the washer. It uses the tanks standard heating elements in parallel. It has lowest priority and the upper 15 gallons heats up to 120F and the lower element preheats water. I do all wash cycles with hot water and that front load washer has no mold build up. I have a 13 gallon tank for the house and just added a BOSCH ES4 because I am too lazy to build extra insulation to the 13 gallon. A new 4 gallon tank only cost $48 with FEDEX shipping included, How could I pass that up. All three tanks work on only the excess energy not used by the house. The HW controller has arc interrupt so existing thermostats can be used, plug and play. I even modified the dishwasher heating element to run off the array DC.

I designed a 4KW microprocessor system for a guy in Canada with six heating elements in an old oil tank. Each element added on progressively as there was more sun. One element had variable power so there were no big jumps in matching the heaters to the array.View attachment 109214

So how much voltage goes into the heating element, assuming there are about 13amps?

 

[OzSolar]

Hi OzSolar,

Do you have a rough wiring diagram or perhaps a list of the elements of your experiment? Sounds just like the heating solution I want to build.

Thanks.

On the surface it is as simple as wiring the modules in series then hooking the positive and negative up to the element.

In practice there's more going on. Switching the DC off when the load is satisfied being a very important one. Also being aware that you are working with voltages above 200v create some unique challenges in that its dangerous. It appears that @efficientPV is several steps ahead of me so I'll will be very keen to hear more from him or her (or them).

I did some experimenting on Sunday but didn't take any notes so more details will follow soon. The main point is that I did use PV to heat water via standard 240v water heater element. I had an old 600V safety switch that I used to turn it off. It might have been sitting in a 5g plastic bucket and I was most certainly wearing Crocs.

 

[efficientPV]

So how much voltage goes into the heating element, assuming there are about 13amps?

When using a capacitor bank to heat water nearly the full voltage and current is always applied to the heating element for short periods. Ideally the heating element resistance is sized for the panels MPP current or slightly lower. Lower voltage of panels when hot also has to be accounted for. Some of that current comes from the solar panel and the rest from the capacitor. Around 50% duty cycle is the highest average current for the capacitor as it sees charging and discharging. Capacitors have internal resistance and welds to the exiting leads are a potential failure point as well as heating over time. A capacitors life at rated temperature is only 4,000 hours. Lowering that temperature dramatically increases life. Most commercial electronics only see a couple hours of use a day. A water heater will see 12+ hours a day of use. One thing that China learned is you can get away with a lot designing things on the edge. In a throw away society this is a valid option. There is always a loss of energy when a capacitor charges and discharges. Keeping that differential voltage as small as possible makes the conversion more efficient. Having higher capacitance can allow a smaller differential voltage. More and better capacitors will add reliability and efficiency. There will also be better low light level performance. To some that is worth the extra cost. Even a design which is only 80% more efficient than direct connect at lower light levels is a dramatic improvement.

 

[MattiFin]

Hello Everyone,

New to this forum so be gentle. I've got a very early concept idea for directly heating a concrete block using second hand PV panels.

The basic idea being a DC heating element embedded in a concrete block (heat storage) warms during the day and emits heat gradually thereby smoothing out the PV output. I'm thinking a max 2kw, the same as my wife uses on a fan heater which cost us a bomb to power last winter.

Any feedback welcome on this idea, particularly any ideas how the PV panels need to be connected to the "heater element" and what sort of "heater element" I could use.

Some requirements:

• I don't want to introduce batteries
• I don't want to use an inverter
• I intend to use second-hand PV panels
• This is a "top-up" heating solution - the house has gas powered central heating

Updated Diagram below:

Water is better storage medium than concrete.
Water has higher specific heat capacity than concrete measured both in volume and weight.
Per weight water is 4.2kJ/kg/K and concrete is only 0.9kJ/kg/K
Per volume water is 4.2kJ/L/K and concrete is about 2.2kJ/L/K
Extracting the thermal energy out of water is also much easier and more efficient. (high thermal conductivity, can use pumps and radiators)
Concrete makes sense only if you can use structural concrete part as thermal mass. Like using concrete floor slab.

I don't know what you pay for electricity or where you are located but 2kW doesn't sound much.
Also depending on how long your wife likes to keep the heater running you'd need lot more panels than 900W to collect the same amount of energy in limited sunshine duration. 900W panels with 3 hours of efective sunshine would be bit over 2kWh or same as your wife keeping the heater on for 1 hour per day.

 

[32 volt boater]

Electric you only catch something like 20% of suns energy.

I would use the greenhouse effect. Use mirrors to shine alot of solar through a window into a dark colored room. Or just sunlight direct into every window.

Heliostats are computer-controlled mirrors which keep the sun reflected on a target as the sun moves across the sky. The LightManufacturing H1™ heliostat delivers up to 1300 watts of heat to a target – or up to 150,000 lumens for lighting applications

Or some mirrors all pointed at a single solar air heater. Old idea Archimedes' mirrors,

Heliostats | LightManufacturing Systems

Unlimited heat and light for factories, offices, homes, and more! Heliostats are computer-controlled mirrors which keep the sun reflected on a target as the sun moves across the sky. The LightManufacturing H1™ heliostat delivers up to 1300 watts of heat to a target - or up to 150,000 lumens for…

lm.solar

 

[32 volt boater]

Thread I made on this here https://diysolarforum.com/threads/r...f-dark-rooms-heliostats-to-aim-mirrors.45567/

 

[32 volt boater]

.

 

[Leeds]

This thread is almost exactly what I've been working on! It seems I wandered upon it a little late as I've already cast the first stone, so to speak but it's only the first test cast and I intend to do more anyway. (That said,

In my scenario I'm trying to heat concrete with an embedded resistive heater driven directly from a single solar panel. The purpose is I need to add some subtle heat to a greenhouse that has yet to be designed and built and I feel direct solar heat storage is the way to go.

I hope to find a way in the daytime to heat the concrete up more than it can dissipate the heat over that day and the following night and perhaps a few hours more. The plastic covering the greenhouse will be the main source of heat of course but I'd like to have this greenhouse above the frost point for a month before the normal last frost. If I can find a way to make a block of concrete do that I can then determine how many pairs of blocks and panels I'd need to achieve my goal. Then I can get onto the task of designing the summertime use of the panels, fans.

I would really like to keep this project simple, I don't want to burden the gardeners with batteries or active electronics that I then have to maintain. I'd also not like to have oil or water to heat as they have maintenance issues.

My first attempt to achieve these mini heaters was to wrap a 1.75 ohm heating wire around a brick and sink it into concrete with two bolts protruding for the wiring.

All that said, I matched the resistance to roughly the mpp of full power for the panels I have (10.2A) Was that a mistake? Should I have targeted a higher power level due to these being used in the winter? Or should I have fixed this resistance to be closer to max power at lower currents I'd see when the lighting is not great? Each experiment is about $8 out of pocket so I can afford some do-overs, I'm just looking for any insight you folks want to add. Thanks in advance!

 

[efficientPV]

Low resistance heating elements are bad because you end up with a lot of resistance in everything before the heating element. Studies have shown for direct connect that and element of twice what you think is ideal gives higher total daily output. Power point heater control is the way to go to maximize output. It takes twice to three times the panels to get the same performance. The science is out there, no need to just waste money experimenting with disproven ideas.

 

[Leeds]

You make a number of good points, it's time to rethink this operation. The benefits of adding some smarts to this system outweigh the downsides of the added complexity. I'll still use concrete as the heaters but I need to rethink their operation and composition. Fixed power modulated over time is better suited than modulated power over fixed time.
Cheers

 

[JoergK.]

Hello all, particularly spudtoo,

I am new to this too, but have done a lot of brain investment into this whole thermal storage for over 20 years.

Here MHO:

1. concrete is not a good conductor, different then we would believe
2. storing energy in a 2/3 full 5 gal pail is good for a test, but nothing to keep your breakfast egg warm
3. using PV panels to create heat to then store it is (again MHO) thinking around the corner

So, the thermal storage has to be of size, retrievable on demand and not at random.

I feel.