PV panels used as heaters.

[Dacian Todea]

Today I managed to test heating using PV panels. I had some 60 cell unused panels (had no time to install them) so got two inside and modified to use as a heater.

By modifying I mean disabling 20 of the 60 cells by shorting them as seen in below photo.

It just so happens that 40 cells will be ideal to get the max power point voltage needed for a 60 cell PV panel. So each group of 40 cells can handle two parallel 60 cell panels.

Notice the yellow wire link I soldered to disable the middle 20 PV cell's on this panel so it is now a 40 cell PV panel.

I then connected this to two parallel panels (60 cell 260W) that are out in the sun and I got the results below

At initial connection panel is cold 

29.3V   @ 16.7A  (489W)   (not sure about +20C ambient)

28.8V   @ 16.8A  (484W)   (a few minutes later +35C panel)

28.06V @ 17.5A  (491W)   (panel at about max temperature around an hour after test started +50C)

27.77V @ 16.5A  (458W)   (a bit latter some transparent clouds sun closer to sunset).

26.8V   @ 12.3A  (329W)   (even later more dense cloud's)

25.5V   @ 7.7A    (196W)   (again thicker cloud cover).

25.9V   @ 7.6A    (197W)   (panel started to cool so slight voltage increase at about the same current as above).

24.2V   @ 2.6A    (63W)     (as I write this close to sunset panel now at room temperature still fairly close to max power point if not right there).

PV cell are nothing else than large diodes so is the same concept as my diode heater based on cheap bridge rectifiers I mentioned before. Those are more compact and installed on a water block can be used to heat some water thermal storage but since PV panels are geting cheaper and you can find used ones for maybe even below $100 it can have some uses as heaters if you do not dislike the look of PV panels inside :)

 

Thermal image above of the DSSR20 at 17.5A while supplying the PV heat panel.

Hottest point on the PV panel about 50C at 17.5A (current may get to 20A in some solar conditions from two parallel panel but temperature will still be OK as panel can be more than that in summer sun).

Full thermal view of the panel to see the 20 cells on top and bottom heated and the 20 cell in the middle deactivated.

Best use case will be with two panels for heating using 3 groups of 40 cells and 6 panels (same 60 cells) to supply them with energy this will provide very simple and reliable maximum power point heating.

But this can not be a solution for full house heating as this will heat the air and the objects in the room but if you have to many it will get to hot during the day and then when is cloudy you need the large thermal mass storage to deal with those periods.

[Jason Striegel]

That's pretty interesting. I am trying to imagine what that feels like. I'm guessing this is sort of like having warm sun shining on you from a window about 1/6th the area of the equivalent wattage panel (assuming ~16% efficiency) , right? So if you're powering that with 400W, and 400W of panel is about 5'x4', that's sort of like having a 20"x24" patch of sunny window that you can place anywhere in the house.

Does this have any negative effect on the cells over time? On cloudy days you could still use this approach with battery storage, though I suppose that's more expensive. What would be the potential benefits of this that would complement the floor heating you have installed?

[Dacian Todea]

Jason,

I'm only using 66.6% of the panel area as a heater as I only use 40 cells out of the total 60 cells and to supply this heater I use two full 60 cells panels connected in parallel and of course out in the Sun.

So the surface of PV panels is 3x larger than the surface of the PV heater inside those the amount of heat provided by that heater panel per unit area is similar to a window as it will be 17% external panel efficiency x 3 (as inside area for heater is 3x smaller) that is 51% about what a typical two or three panel window will allow inside but the feeling is not the same as photons that go trough the window travel in a straight line so is like a laser and will heat the object that it lands on where the PV panel as a heater (sort of an infrared heater) is way more diffuse and you will need to be a few cm from the PV panel to feel the heat on your skin.

No cells are not affected at all as operating temperature is just around +50C in this sort of setup while the outside PV panels in summer get to 70 even 80C in a hot summer day so there is way less stress on the panels used as heaters this way and so they will last "forever" more than the panels that are outside and those should last 30 to 40 years at least.

 

[John Mundy]

So 2 72 cell panels in parallel should work on a full 60 cell? There are no diodes to disable?

The panel being used as a heater is a straight connection not a reverse polarity, correct?

[Dacian Todea]

60 cells will be to much for a 72 cell panel ideal will be 48 cells so you can use a 72 cell panel as heater with one diode / set of 24 cell disabled.

To fully take use of panels you can have 6 panels outside connected to two heater panels inside all panels same type either 60 cells or 72 cells.

Both 60 cells and 72 cell panels are divided in 3 sections of 20 and 24 cells respectively.

So since two panels need two sections you can use 2 panels for first two sections of a heater PV panel then same for another set and the remaining group from each heater panel can be supplied by the other two panels. That will take advantage of the heater panels.

Yes polarity is direct no reversed.  If you connect in reverse the diodes will conduct and heat up so voltage drop will be just 1 or 2V.

Bellow is the diagram of a 60 cell panel. The 72 cell is the same there are just 3 groups of 24 cells instead of 20 cells.

 

[Julien]

Hi,

If I well understood this principle,

2x 60 cell panels in // should be OK for one 36 cell panel without mod. (which can be found cheap also, that's why I mention them.)

[Dacian Todea]

Julien,

Yes it will work with a 36 cell panel if the panel is made with 6" cells same as the 60 cells.  The smaller 36 cell panels made with 5" cells may have to small surface area and get maybe to hot.  The 36 cell made with 6" cells are rated at around 160W and those are maybe not as cheap.

[Dacian Todea]

I got a link to an article from Colin and you may be interested in that  https://www.sciencedirect.com/science/article/pii/S2352728519300508

It is about using a 100W 36 cell PV panel to heat a string of 20 diodes in series that are then epoxy to a water pot to heat water or cook food.

The was almost no difference between a 3Ohm resistive immersion heater and the chain of 20 diodes but the immersion heater had the advantage of being able to transfer more of the heat to the water than the glued (epoxy) down diodes on the outside of the pot even considering the thermal insulation present in all cases. The difference as expected is when is not sunny but then there is also less energy available.

The diodes seems to work way too hot to close to the limit of melting the diode junction so to improve that they should have used two sets of 20 diodes glued to the pot that way each will see just about half of the current and max temperature of the diodes will have been lower also due to a bit larger surface area they will have also improved the heat transfer to water a bit. Also they used iron pot with enamel on both sides since they were preparing food that makes sense and is likely a lower cost more available pot. But just for water heating an aluminium pot will have worked much better.

Seems they used a 160C thermostatic switch but that is not the diode junction temperature and diodes where rated for 150C max junction temperature.

I did not read the article (no time) just took a quick 2 minute look.