Solar panel directly connected to immersion heater?

[SunEagle]

Not scary at all. I solder in converter FETs with the power on. That is an old picture just after the electronics was moved from the basement stairs. That was scary, only three steps were left and those were cracked. Micro board was above the fridge. Software updates were always interrupted by my wife wanting to get in the fridge. All that stuff ran everything, fridge, two solar converters, two hot water tanks and a pump. It had to be up and running in less than an hour or the fridge would overheat. It looks better now, but I have tried to keep that look to show people that electronics can be done by anyone. The shed is really nice with a sun roof, Anderson windows and Hardie plank for siding. I am new to this host software and see the saved icon pop up all the time. However, I have lost posts at least four times as I have been writing them and found no way to recover them.

If your posts had any weblinks attached then they went to an Unapproved area first. One of the Mods would have to review it first before it would show up as a view able post.

What I was trying to covey about it being scary was that open circuitry and wiring is not necessarily a safe way to have a system due to someone being able to touch live contacts or insects getting into the works which may add paths for shorts.

[PNPmacnab]

Not scary at all. I solder in converter FETs with the power on. That is an old picture just after the electronics was moved from the basement stairs. That was scary, only three steps were left and those were cracked. Micro board was above the fridge. Software updates were always interrupted by my wife wanting to get in the fridge. All that stuff ran everything, fridge, two solar converters, two hot water tanks and a pump. It had to be up and running in less than an hour or the fridge would overheat. It looks better now, but I have tried to keep that look to show people that electronics can be done by anyone. The shed is really nice with a sun roof, Anderson windows and Hardie plank for siding. I am new to this host software and see the saved icon pop up all the time. However, I have lost posts at least four times as I have been writing them and found no way to recover them.

[SunEagle]

I currently do this at my camp and an able to put about 3KWH into making hot water consistently. That is from 900W of panels connected as a 36V string and about a 52V power point. Keep in mind that this is only the excess power after everything else is powered. There are two tanks, 10gal with 2000W 125V element and 20gal with 1500W 125V element. These are connected in series with the 10 gallon as the final. That allows fast heating in the morning and can usually get to 45C by 10am. Either heater or both can be on. A couple FET control each heating element by PWM from a UNO. FET are driven at low frequency via a simple opto isolator driver. It operates from a simple program that monitors the fixed power point voltage of the panels. Really no need to track. If voltage is over the setpoint, the PWM counts up. And down if lower. You could do the same with a TL495 switching regulator chip. Controls are powered by wall warts. A good percentage of them work easily at 50V. The whole secret is a large capacitor bank that stores panel power between on pulses. I use an old TURNIGY 130A to record power. Variants of these sell for only $10. This picture is my power shed that holds the two tanks and all my electronics.

I also tried a variant of this at home using cheap ebay boost converters. Boost converter runs at full bore and the input voltage is monitored with uno driving a FET. Even just a 250W panel is quite effective since normal water heater losses are about 150W.

Man. That breadboard system with all the loose wires looks scary to me. I hope that area is dry and bug free or you will find it can get real messy after a while.

[PNPmacnab]

I currently do this at my camp and an able to put about 3KWH into making hot water consistently. That is from 900W of panels connected as a 36V string and about a 52V power point. Keep in mind that this is only the excess power after everything else is powered. There are two tanks, 10gal with 2000W 125V element and 20gal with 1500W 125V element. These are connected in series with the 10 gallon as the final. That allows fast heating in the morning and can usually get to 45C by 10am. Either heater or both can be on. A couple FET control each heating element by PWM from a UNO. FET are driven at low frequency via a simple opto isolator driver. It operates from a simple program that monitors the fixed power point voltage of the panels. Really no need to track. If voltage is over the setpoint, the PWM counts up. And down if lower. You could do the same with a TL495 switching regulator chip. Controls are powered by wall warts. A good percentage of them work easily at 50V. The whole secret is a large capacitor bank that stores panel power between on pulses. I use an old TURNIGY 130A to record power. Variants of these sell for only $10. This picture is my power shed that holds the two tanks and all my electronics.

I also tried a variant of this at home using cheap ebay boost converters. Boost converter runs at full bore and the input voltage is monitored with uno driving a FET. Even just a 250W panel is quite effective since normal water heater losses are about 150W.

[bcroe]

Building and evaluating control schemes is a huge step forward from blindly connecting things together. Maybe the thermostat
could be wired in to control your DC rated relays. Bruce Roe

[Roop27]

Hi

I do exactly what the OP suggested, I have two 175W 35V MPP panels connected directly to my original 3kW immersion heater. The immersion had only ever been used once in 20 years just to see if it worked.

I have a relay that switches parallel to series at about 38V and back to parallel at approx. 35V. It was only done for a bit of fun, not expecting it to pay for itself or even work very well but in the summer it is actually pretty good. In full summer sun the impedance match is very good, it often gets up to about 72V (280W) in series and on bright cloudy days is often sat at around 36V (70W) in parallel.

Output on dull days is virtually nothing but like I said it is just for fun.

Not too worried about the possibility of the heater thermostat arcing, I have never know mine to operate even when the water got really hot. OC voltage is just over 80V so it may be ok anyway.

Feel free to criticise this inefficient use of the panels but I have had a lot of laughs with it, mostly by blowing up control circuits I was experimenting with!

[bcroe]

The other issue is the heater thermostat. It isn't designed to break a DC power source; could arc over and burn up. Bruce Roe

[Sunking]

arent you better off getting a solar water heater since heating is what you wanna achieve

He is educated in the USA. They have no clue what you are talking about. That is how they got stuck with Chump or Billiary as Potus

[VTSE]

Hi experts,
I am planing to buy a 250/500 watt solar PV panel and connect it directly to my 2kw immersion heater attached to hot water cylinder without any convertor/inverter in between. (pure DC to heating element). I believe this should work in principal and should raise temperature of water by 10-15 degrees in one day.

My question is - will this work? Are there any problems connecting this way? Is there a better way of achieving the same result without additional cost?

Your views/comments will be greatly appreciated
Thanks

arent you better off getting a solar water heater since heating is what you wanna achieve

[bcroe]

In addition to the good advice above, here is an old time GRAPHICAL solution to performance of
a PV panel source connected to a resistance heater load. With a 0.3 ohm heater 3V gives 10A
of current, 6V gives 20A, and so on. Plotting these point gives a straight load line from 0,0.

Then plot the power curve of a 12Vmp 20Amp 240W panel. 15Voc, 25Asc. These 3 points give
a rough curve as shown. That gives a max power point at A, 12V X 20A = 240W.

However, conditions hardly ever give the ideal maximum, so lets plot a more achievable PV curve
at 80%, or 192W. Vmp will hardly vary at 12V, Imp drops to 16A. At B power is 192W. The
heater load line intersects at C 5.8V, 19.5A, this is where they would operate if connected
together. But only 113W, or 59% of the available power is delivered.

100% efficiency could be achieved if the load line ran right through the panel MP point B. 12V
divided by 16A (ohms law) gives a load line of 0.75 ohms. If the heater ohms are off by a
factor of 20 (in either direction) the efficiency might plot out around 5%. Not the way to use
resources.

What about when the sun drops to 40%, point D of 96W potential? The curve intersection
at E shows our "ideal" heater drops to 7.4V, 9.7A, or 72W. At 75% efficiency some
adjustment would help. The 0.3 ohm heater efficiency is in the toilet.

So plot out the panels and loads under consideration. If the efficiency isn't above 60%, it
will likely be a big disappointment. A severe mismatch can easily get into single digit
efficiency, a total waste of time.

Electronic means of improving the match are possible. But off the shelf devices pretty
much work with some sort of power reserve (grid, battery), not direct load connection.
Bruce Roe

[Sunking]

To make this work better without a Converter is to use lower voltage Battery panels of 18 volts. Example you could use 125 watt panels, two of them wired in parallel. Each panels has a Vmp = 18 volts and Isc of 7.35 amps. When two panels are parallel you have 14.7 amps of current. Using the same 12 volt 250 watt heater means you know have 14.7 amps x 14.7 amps x .576 Ohms = 124 watts from 250 panels. Only 50% efficient, but a batter match of resistance.

[jflorey2]

I am planing to buy a 250/500 watt solar PV panel and connect it directly to my 2kw immersion heater attached to hot water cylinder without any convertor/inverter in between. (pure DC to heating element). I believe this should work in principal and should raise temperature of water by 10-15 degrees in one day.

Sure, it will work. But you have to match the resistance of the coil to the output impedance of the solar panel array. To do that we need to know what solar panels you have chosen, and then you will have to choose a heating element accordingly.

[Sunking]

No, I was thinking of a Non-grid tied systemsystem. So are you saying that the power from the panel will be too less?. All I really want is to raise the water temperature by a few degrees in one day. Can a 500w panel work with a 500w immersion element? Obviously with some sort of resistance to limit the current to avoid short circuit

A solar panel is a current source. A 250 watt panel has a Isc of roughly 8 amps.

A Resistance Heater is a fixed amount of resistance. A 12 volt 250 watt heater is a resistor with a value of .576 Ohms

So you have a current source of 8 amps flowing through a fixed resistance of .576 Ohms. Ohms Law for Power = Current x Current x Resistance. Now use 8 amps and .576 Ohms.

8 amps x 8 amps x .576 Ohms = 37 watts. @ 4.6 volts

You need a power Converter, a Buck Converter, to change the panel output from 32 volts @ 8 amps to 12 volts @ 21.3 amps = 250 watts.

So your idea will work, just not work worth a damn because you are changing your panel wattage from 250 to 37 watts of heat. In other words only 15% efficient. You loose 85% of your power. Which is perfect for democrats taking 85% of your money. If you are a democrat, that is the way you should do it because it is the right thing to do.

[Nathan251145]

Well if you are talking about using a Grid Tied Panel say a 250 watt panel with a Vmp= 32 volts and Imp = 8 amps and say connect it to a 12 volt 500 watt immersion heater with 8 amps from the panel generates 8 x 8 x .576 = 37 watts from your 250 watt panel. So you tell us, does that work for you?

No, I was thinking of a Non-grid tied systemsystem. So are you saying that the power from the panel will be too less?. All I really want is to raise the water temperature by a few degrees in one day. Can a 500w panel work with a 500w immersion element? Obviously with some sort of resistance to limit the current to avoid short circuit

[Sunking]

Thanks adoublee, so is there a way to avoid the short curcuit? I guess adding some sort of resistance?

No, you need a power converter which none exist to do what you want. Easy enough to build for a designer.