Help with trouble shooting hot water

[lile001]

And now you are adding batteries, a lower operating voltage on the panels an inverter etc to the mix. By the time you factor all of that in you might as well just do a true thermal system. your costs will be about the same.

No reason for batteries and inverters at all. If you have, say a 2 KW heating element, and you just used a 2KW array, then it would only really work very well under perfect conditions. But solar panels are constant voltage devices (as long as the load is small enough..) if the solar panel is oversized, say 4 KW, then it would not deliver too much voltage to the heater - voltage is constant. Enphase uses this very idea in their microinverters, arguing that a panel almost never delivers full power, usually some fraction.

Like the guy sez - panels are CHEAP now! Especially if you can use an off-brand, or a non-UL panel. These things are going for a couple dozen cents a watt these days. As long as the safety systems are in place (disconnect, fuse, wire in conduit, DC rated thermostatic control on the heater, overtemp/overpressure valve on the water side) there isn't any reason this wouldn't work.

[lile001]

Numerous problems not the least of which is the NEC.

Really? Cite chapter and verse that sez you can't! The NEC says that a solar panel has to have a fuse, a disconnect, DC wire in conduit, and a few details about wiring methods. It has no prohibition about using DC power for loads, in fact the next version will have a whole new chapter about modular DC distribution systems in building.

[Naptown]

No reason for batteries and inverters at all. If you have, say a 2 KW heating element, and you just used a 2KW array, then it would only really work very well under perfect conditions. But solar panels are constant voltage devices (as long as the load is small enough..) if the solar panel is oversized, say 4 KW, then it would not deliver too much voltage to the heater - voltage is constant. Enphase uses this very idea in their microinverters, arguing that a panel almost never delivers full power, usually some fraction.

Like the guy sez - panels are CHEAP now! Especially if you can use an off-brand, or a non-UL panel. These things are going for a couple dozen cents a watt these days. As long as the safety systems are in place (disconnect, fuse, wire in conduit, DC rated thermostatic control on the heater, overtemp/overpressure valve on the water side) there isn't any reason this wouldn't work.

You are a PE and condoning and or encouraging that practice? You should know better than that.

Really? Cite chapter and verse that sez you can't! The NEC says that a solar panel has to have a fuse, a disconnect, DC wire in conduit, and a few details about wiring methods. It has no prohibition about using DC power for loads, in fact the next version will have a whole new chapter about modular DC distribution systems in building.

I never said the NEC forbid it. My point was that to comply it would be more work than most all other methods.
You are now into an extremely complex system that few other than EE's and the original owner installer will know much about. Particularly the local AHJ which in my experience generally are not fond of ground breaking innovation that is not addressed in the code.
Good luck with the project.

[Sunking]

But solar panels are constant voltage devices (as long as the load is small enough..)

Wrong! Solar panels are current sources, not a voltage source like a battery.

A solar panel is pure current sources from MPP to Isc. Voltage is unknown and determined by the load resistance. A solar panel equivilent circuit is:



So for example I have a 100 watt solar panel. With

VOC = 22 volts
Vmp = 17 volts
Imp = 5.88 amps.
Isc = 6.1 amps.

Assuming I face the panel into the sun with 1000 watt/M2 and connect a 2.9 Ohm resistor across the panel I will see 17 volts, 5.8 amps delivering 100 watts.

Take the same panel under the same conditions except use a 1 Ohm resistor I will see 5.8 volts @ 5.8 amps or 30 watts.

Same circuit, except this time I place a 10 Ohm resistor across the panel, I will see 21 volts @ 2.1 amps or 44 watts.

[bcroe]

Model

Wrong! Solar panels are current sources, not a voltage source like a battery.

A solar panel is pure current sources from MPP to Isc. Voltage is unknown and determined by the load resistance. A solar panel equivilent circuit is:



So for example I have a 100 watt solar panel. With

VOC = 22 volts
Vmp = 17 volts
Imp = 5.88 amps.
Isc = 6.1 amps.

Assuming I face the panel into the sun with 1000 watt/M2 and connect a 2.9 Ohm resistor across the panel I will see 17 volts, 5.8 amps delivering 100 watts.

Take the same panel under the same conditions except use a 1 Ohm resistor I will see 5.8 volts @ 5.8 amps or 30 watts.

Same circuit, except this time I place a 10 Ohm resistor across the panel, I will see 21 volts @ 2.1 amps or 44 watts.

It would appear, the diode in that model would short out any + voltage? Bruce

[LucMan]

Why not just put in a solar thermal panel and keep it simple, there is a reason why it's not mainstream practice to use PV for heating domestic HW.
Solar thermal is your biggest bang for your buck.

[muskrat]

that was then this is now

Why not just put in a solar thermal panel and keep it simple, there is a reason why it's not mainstream practice to use PV for heating domestic HW.
Solar thermal is your biggest bang for your buck.

Solar thermal was the best deal for the dollar until panels got well under a dollar a watt.I have a friend with solar thermal who has had several problems.Difficulties with overheated glycol loop when the pump jammed or the heat exchange tank reached its maximum safe temperature,pop off valves that stick open and of course with the possible exception of the more expensive vacuum panels they don't work well or at all when the temperature is below zero.Also in my experience plumbing is harder to do right than wiring.I had photovoltaic panels on my roof for 30 years with only occasional cleaning.I've never seen a system with a pump in it ;especially in hot water systems that held up anywhere near that long.I think solar thermal is great but I think the amazingly low panel prices now make for different economics.One thing I still need to research is the effect of less than optimum lighting.Of course lower than rated amperage will drastically lower the heat out put but would that somehow damage the panels?It would be easy enough to add a low voltage cutoff but would it be necessary?The load would be undamaged since unlike an electrical motor it is purely a resistive load.You would still need the usual diodes to prevent a brightly lit string from back feeding a dimly lit one of course.Over sizing the panels somewhat for the heating element would minimize the times when the voltage dropped off.

[Naptown]

Solar thermal was the best deal for the dollar until panels got well under a dollar a watt.I have a friend with solar thermal who has had several problems.Difficulties with overheated glycol loop when the pump jammed or the heat exchange tank reached its maximum safe temperature,pop off valves that stick open and of course with the possible exception of the more expensive vacuum panels they don't work well or at all when the temperature is below zero.Also in my experience plumbing is harder to do right than wiring.I had photovoltaic panels on my roof for 30 years with only occasional cleaning.I've never seen a system with a pump in it ;especially in hot water systems that held up anywhere near that long.I think solar thermal is great but I think the amazingly low panel prices now make for different economics.One thing I still need to research is the effect of less than optimum lighting.Of course lower than rated amperage will drastically lower the heat out put but would that somehow damage the panels?It would be easy enough to add a low voltage cutoff but would it be necessary?The load would be undamaged since unlike an electrical motor it is purely a resistive load.You would still need the usual diodes to prevent a brightly lit string from back feeding a dimly lit one of course.Over sizing the panels somewhat for the heating element would minimize the times when the voltage dropped off.

Are the panels grid tie? If so then adding some panels to increase the system output may make sense in a situation where there is not a large load on the water heater. Say 1 or two people in a house. Where this does not make sense is when the hot water loads increase in say a house with 5 or 6 people living there. There is a break point where the thermal system will be more cost effective and require less roof space than the energy equivalent in PV.
If there is already a grid tie in place then this may make sense to do it this way. In either case you will need a back up source of heat for the water heater and batteries is no way to go on that front.

[muskrat]

I was assuming alternate power but not grid tied panel

Are the panels grid tie? If so then adding some panels to increase the system output may make sense in a situation where there is not a large load on the water heater. Say 1 or two people in a house. Where this does not make sense is when the hot water loads increase in say a house with 5 or 6 people living there. There is a break point where the thermal system will be more cost effective and require less roof space than the energy equivalent in PV.
If there is already a grid tie in place then this may make sense to do it this way. In either case you will need a back up source of heat for the water heater and batteries is no way to go on that front.

I was assuming grid tied electrical for back up heating but not having a grid tie inverter on the water heating electrical panels. The idea there was to keep the cost of the system low by eliminating inverters batteries etc.

[Naptown]

Here's the down side to that logic.
say you have a 50 gallon water heater and you don't use a separate tank for the back up.
It's cloudy for a day or two and you have guests at the house. Since the bottom element is now run from solar the tank from about 6" below the remaining element to the bottom is cold. You have taken the 50 gallon water heater and turned it into a 15 gallon water heater whenever the sun is not out or at night. Plan on a lot of cold showers in the winter.
This might work but would need to be a separate tank feeding the existing tank with tempered water. If you are grid tie and can add a few panels to offset the water heater you will get by with fewer panels as once the thermostat satisfies the system is essentially shut down. To make it work you will need more panels than you would need as a grid tie and you will not get the credit on net metering or SRECs (Low but still available in Mo I believe) The controls needed for this plus the additional wattage and lack of production when at temperature will be equal to or greater over the long run to do it than the cost of an inverter.

[bcroe]

Dual Tank

Here's the down side to that logic.
say you have a 50 gallon water heater and you don't use a separate tank for the back up.
It's cloudy for a day or two and you have guests at the house. Since the bottom element is now run from solar the tank from about 6" below the remaining element to the bottom is cold. You have taken the 50 gallon water heater and turned it into a 15 gallon water heater whenever the sun is not out or at night. Plan on a lot of cold showers in the winter.
This might work but would need to be a separate tank feeding the existing tank with tempered water. If you are grid tie and can add a few panels to offset the water heater you will get by with fewer panels as once the thermostat satisfies the system is essentially shut down. To make it work you will need more panels than you would need as a grid tie and you will not get the credit on net metering or SRECs (Low but still available in Mo I believe) The controls needed for this plus the additional wattage and lack of production when at temperature will be equal to or greater over the long run to do it than the cost of an inverter.

How about a large "Pre Heater" tank, heated by the PV? If the water coming out of this into the regular heater is hot enough, that tank
(propane here) would never run, or much reduced. Problem I see, making sure the pre heater doesn't get too hot. Bruce Roe

[Naptown]

How about a large "Pre Heater" tank, heated by the PV? If the water coming out of this into the regular heater is hot enough, that tank
(propane here) would never run, or much reduced. Problem I see, making sure the pre heater doesn't get too hot. Bruce Roe

A preheat tank is really the only way to make this work assuming you are using standard electric water heater for the purpose. Controlling the temp in the preheat tank is not as much of a problem other than you will need to use a DC relay rated for the voltage tied to a more or less standard thermostat to turn off the power.
A bigger issue I think is keeping the panels on the maximum power point. Not enough watts and the voltage collapses with the resistance of the element.

[muskrat]

100 gallon tank

A preheat tank is really the only way to make this work assuming you are using standard electric water heater for the purpose. Controlling the temp in the preheat tank is not as much of a problem other than you will need to use a DC relay rated for the voltage tied to a more or less standard thermostat to turn off the power.
A bigger issue I think is keeping the panels on the maximum power point. Not enough watts and the voltage collapses with the resistance of the element.

Actually I was thinking of either a hundred gallon tank or a preheat tank.over sized electric tanks are not that much more.Good point about the thermal gradient in the tank.using the bottom element for the solar might work better.I thought the over sized tank would allow the line element to stay caught up by running longer if the solar wasn't cutting it.They do have mppt devices for high voltage strings I'll have to figure out if it would be better to add panels or mppt.most of these problems are also true of a thermal system but I guess the heat exchanger tank acts as a reserve.

[muskrat]

bigger tank

Here's the down side to that logic.
say you have a 50 gallon water heater and you don't use a separate tank for the back up.
It's cloudy for a day or two and you have guests at the house. Since the bottom element is now run from solar the tank from about 6" below the remaining element to the bottom is cold. You have taken the 50 gallon water heater and turned it into a 15 gallon water heater whenever the sun is not out or at night. Plan on a lot of cold showers in the winter.
This might work but would need to be a separate tank feeding the existing tank with tempered water. If you are grid tie and can add a few panels to offset the water heater you will get by with fewer panels as once the thermostat satisfies the system is essentially shut down. To make it work you will need more panels than you would need as a grid tie and you will not get the credit on net metering or SRECs (Low but still available in Mo I believe) The controls needed for this plus the additional wattage and lack of production when at temperature will be equal to or greater over the long run to do it than the cost of an inverter.

40 gallon hot water heater is 329 dollars a 119 gallon heater is 629 dollars.Three times the capacity for less than double the price.thermal stratification might still be a problem though if the solar wasn't kicking in.If you had a couple days usage in the tank one element might be able to keep up.Bigger families would need bigger systems of course but this would also be true for thermal systems.

[inetdog]

... most of these problems are also true of a thermal system but I guess the heat exchanger tank acts as a reserve.

A thermal system benefits in two ways from drawing from the bottom and inserting halfway up (equivalent to being the lower element):
1. It lets the main or backup heater do the last bit of heating to get useful hot water if the solar is not able to keep up.
2. It gives the thermal panel first crack at the incoming cold water, since the panel will be more efficient the colder the incoming water to it.

For resistive heating, the advantage is that it can heat all of the water in the tank if the power is available, while the grid powered upper element will be called for only when the outlet water temp is too low. And it will benefit from the preheating done by the opportunity load heater.
If the solar electric heat is only available occasionally, the right thing go to would be to put it at the bottom of the tank and also have a grid powered element in that same position. A little harder to do with a conventional tank.
In either case, you need to allow it to raise the tank temperature higher than the normal elements, and that then requires a tempering valve on the tank output for safety.