Are there such things as heaters for solar panels?

[EverAzure]

Sprinkle some salt on the board

[peakbagger]

I have about 50 acres of predominantly blighted beech that regenerated from root stock after a big ice storm. It chokes out the other more desirable hardwoods. So its either girdle the trunks and let it rot standing until it falls over in 10 years or so or cut it for firewood. The carbon still ends up in the air ether way but in theory the regrowth will be more vigorous and will sequester more carbon then currently. Of course its hardwoods so I will be in the ground before some future owner has to decide the harvest the new wood versus sequester. California is throwing a lot of money at Northeast Woodland owners to buy carbon sequestration rights but from the chatter I have heard from forestry type folks a lot of the calculation used are not consistent with reality.

[SunEagle]

I have wood boiler and "free wood". I have considered running a glycol loop under the lower edge of my difficult to access roof panels. I find that if I can expose just a few inches of bottom of the solar panel with a roof rake that the panel will self clear much faster. I have two rows high and its noticeable that the seam between the two rows is a problem with the top row as I cannot get to it with the rake unless there is 2 story ladder involved. My guess a bit of heat applied to the back side of the panel post storm might do the trick.

Why havent I done it, lots of reasons.
-Potential panel failure due to differential heating of the panel.
- I hate roof penetrations
-Settting up a glycol loop is PITA
- I have a great net metering plan that is grandfathered and produce more power than I need so it would be a acedemic exercise.

If I build a new house and retired I may play around with projects like that.

Sounds like a great plan to use the available wood to burn to make heat. Of course some people (not me) will be mad at you for burning that wood and not being "green"

[peakbagger]

Assuming you are "bitcoin mining", then from what I understand you got lots of heat to deal with. The goal is to get the heat you got to where you want it to go. Glycol is antifreeze. For basics, two types of easy to get glycol, Ethylene and Polypropylene. Ethylene is old school, cheap, toxic and hard to get rid of, Polypropylene is used in many vehicles and campers, food grade can be used in food products and no doubt you probably have drank a fair share of it. It turns into goo at very low temps so if you are in very cold climate, below zero you need a high concentration and pump that can move it due to its viscosity getting thick.

Heat rises so if your CPUs are below the panels it helps. Essentially every closed loop solar hot water system is a glycol loop. I have one running in Northern NH for over 20 years. Not a lot of movign parts and mine uses a DC pump fed from an ancient solar panel so controls are minimal. So if you got heat you need to heat the glycol up with a heat transfer coil and using pex tubing run it up to the panel preferably keeping all the piping vertical or sloping upwards and insulate the supply and return lines. The devil is in the details and how much heat and how much tubing area you are going to need is all experimentation. The trick is the coils should not touch the back of the panel, you need an air gap and possibly some reflectors to direct radiant heat to the back of the panels.

[miner_tom]

I have wood boiler and "free wood". I have considered running a glycol loop under the lower edge of my difficult to access roof panels. I find that if I can expose just a few inches of bottom of the solar panel with a roof rake that the panel will self clear much faster. I have two rows high and its noticeable that the seam between the two rows is a problem with the top row as I cannot get to it with the rake unless there is 2 story ladder involved. My guess a bit of heat applied to the back side of the panel post storm might do the trick.

Why havent I done it, lots of reasons.
-Potential panel failure due to differential heating of the panel.
- I hate roof penetrations
-Settting up a glycol loop is PITA
- I have a great net metering plan that is grandfathered and produce more power than I need so it would be a acedemic exercise.

If I build a new house and retired I may play around with projects like that.

peakbagger,

Thank you for your input. I would not have thought of setting up a glycol loop, since I really know nothing about it but it does sound like circulating a non-freezable heated fluid, somewhere around the panel. As I was saying to another poster, this panel is in an area where there is no grid to attach to, so, it is on its own. If you have any data or sites that might save me a bit of time doing the research for a glycol loop as you put it, I would appreciate the links.

Regards and thanks
Tom

[peakbagger]

I have wood boiler and "free wood". I have considered running a glycol loop under the lower edge of my difficult to access roof panels. I find that if I can expose just a few inches of bottom of the solar panel with a roof rake that the panel will self clear much faster. I have two rows high and its noticeable that the seam between the two rows is a problem with the top row as I cannot get to it with the rake unless there is 2 story ladder involved. My guess a bit of heat applied to the back side of the panel post storm might do the trick.

Why havent I done it, lots of reasons.
-Potential panel failure due to differential heating of the panel.
- I hate roof penetrations
-Settting up a glycol loop is PITA
- I have a great net metering plan that is grandfathered and produce more power than I need so it would be a acedemic exercise.

If I build a new house and retired I may play around with projects like that.

[miner_tom]

Bruce,

Thank you again for your most impressive help. Of course, you are doing practical work. Like a wise man once said: "The difference between theory and practice is that in theory there is no difference between theory and practice, but in practice, there is". I spent two years in the early 80's in "FAB" at Intel, so I have a background in semiconductor physics, which is the only reason why I mentioned the physics in the first place. Yes, you are doing this in a practical sense, that is probably why you get things to work. I did not mean to push any buttons.

I am grateful that you have done the work that you did. It is going to prevent me from wasting a whole lot of time. I am sure that you are correct in that some kind of "manual removal" is in order.

Best Regards
Tom

[bcroe]

Several of us have discussed ways to minimize snow buildup on
panels, in the end some manual removal is in order. My aim was
to see if melting it off here was remotely practical. When the power
applied reached 3 times the panel peak rating, I concluded the
answer was NO, experiment ended. There was some concern of
going so high that the panel was damaged.

No temp measurements were made, possibly thermal insulation on
the back of the panel would have cut the power requirement in half,
again not practical here. Even this power level would have required
over 100KW here, how could that be controled and switched in? My
house system is limited to 48KW. The capacity of grid tie stuff is
generally orders of magnitude greater than battery systems, good luck.

I am not too concerned with the semiconductor physics and the
theoretical junction heat, I know the power is the V x I at the leads
connecting to the panel. Experiments here tend to run toward doing
something useful, not theoretical research. Some have suggested
we can collect energy under a full moon or that wire insulation is a
problem here, others suggested probably not. I took the next step
and ran a test proof. Bruce Roe

[miner_tom]

Tom, The voltage applied needs to be the same polarity as normal
operation, otherwise I would just burn out the panel bypass diodes.
This panel eventually went back in service.

CellMod.jpg
Here left is an accepted PV cell model, the sun generates current,
all can be bled off with a short on the connections, no power. As
the voltage is allowed to rise, some V x I power is taken, but the
diode starts to conduct and steal some of the current. Of course
an open circuit load shows max V, at zero I. An MPPT load will
find the best load point for max power, and track sun drive.

On the right is my melting test setup, done at night min sun, the
VI graph is for one 4A cell, this panel has 60 8A cells in series. It
was connected only to a dedicated variable supply for this test.

Kwsupply1.JPGKWsupply2.JPG

Since nothing on my bench would handle this, I tossed together this
extremely crude KW supply. Serious regular work would demand
a switching regulator (in past have used), lots more development time.
Bruce Roe

Bruce,

Very impressive! Really. Of course, I have a couple of questions that result from your pictures.

Firstly, I was wondering about the increase in the temperature relative to ambient. Was that measurement done? Would the heating actually melt ice? I, personally, am only interested in "heating" a panel because I have a panel (rather, a friend who is part of a non profit has a panel) and during the winter it tends to ice up. I have just become involved, as a favor, so I personally have seen the panel but only during the summer, so I don't know how much it ices up. I am the only EE he knows so, of course, I would know about all things electrical (NOT!).

Secondly, as you say, the backfeed voltage is the same as for normal operation. I expect that for brief periods, the backfeed voltage (I have trouble writing backfeed since I am always dealing with feedback ) is higher than the "normal" voltage, then I would expect that temperature would be the enemy. Correct? I expect that one would have to go back into semiconductor physics to find out what the theoretical junction heat would be and that probably would only be theoretical. Perhaps a thermal sensor would be better? They make those pretty inexpensively. Of course, in the field, if not generated from the battery storage, the higher voltage would have to be generated off of the batteries by either a generator/alternator with a rectifier and that would require power in and of itself.

Thirdly, if, in a 24 hour period, the heating of the panel and all that it involves, left the batteries with less reserve power than they would be left with just because of icing preventing maximum sunlight from reaching the cells, then the whole process would be pointless. What are your thoughts on this.

I thank you so much for the information that you have provided

Tom

[bcroe]

Tom, The voltage applied needs to be the same polarity as normal
operation, otherwise I would just burn out the panel bypass diodes.
This panel eventually went back in service.

CellMod.jpg
Here left is an accepted PV cell model, the sun generates current,
all can be bled off with a short on the connections, no power. As
the voltage is allowed to rise, some V x I power is taken, but the
diode starts to conduct and steal some of the current. Of course
an open circuit load shows max V, at zero I. An MPPT load will
find the best load point for max power, and track sun drive.

On the right is my melting test setup, done at night min sun, the
VI graph is for one 4A cell, this panel has 60 8A cells in series. It
was connected only to a dedicated variable supply for this test.

Kwsupply1.JPGKWsupply2.JPG

Since nothing on my bench would handle this, I tossed together this
extremely crude KW supply. Serious regular work would demand
a switching regulator (in past have used), lots more development time.
Bruce Roe

[miner_tom]

I set up this 250W panel one cold winter day, and gradually back fed it 750W.
After 20 minutes it had zero effect on the snow.

The current MO is get some dark parts of panels cleared and exposed, and let the sun
do the rest. good luck, Bruce Roe

PVmelt.jpg

Bruce, thank you for your reply. Being a EE I have to ask questions that might seem obvious to long time users of Solar Panels but I have not used them before.

For example, when you say "back feeding" power, I am going to assume (possibly incorrectly) that you are turning off the feed FROM the panel and feeding a voltage ACROSS the panel, thereby forward biasing what are essentially PIN diodes. Is that correct? If that is so, is the voltage that is being back fed a higher voltage than the voltage that is generated by the panel? If that is so then it has to be generated somehow by an inverter/converter setup, which seems pretty power hungry or by some other method.

Can you enlighten me as to the technical procedure involved in the "back feeding' as you put it.

Regards and Thanks
Tom

[littleharbor2]

So at 7 deg F below freezing, with a lab supply for 2 X the panel output
rating, it melts in 3 hours. Have seen more than 50 deg F below freezing
here. Bruce Roe

And for these type of scenarios the Super Duper Reverse Current Flow feature never was implemented.

[peakbagger]

I have talked to folks who have done so and observed how well snow reflect sun on vertical panels even under substantial overhangs. On a sunny day with low sun angle and vertical panels snow just does not build up on panels.

[bcroe]

So at 7 deg F below freezing, with a lab supply for 2 X the panel output
rating, it melts in 3 hours. Have seen more than 50 deg F below freezing
here. Bruce Roe

[littleharbor2]

I sure would but there's no link.

What ever happened to.. (midniteftp.com)