Help with trouble shooting hot water

[solar_realist]

I have a 80 gal Ruud (solar servant) about 3 yrs old. I found a very old 4x8 collector and pump.
With advise from a contractor who sold me new thermal release valve and air purge fitting, I set up the system myself. All seem to be working fine with the diff controller I got from a put together kit.
It was last year when I notice the pump would run a little at night in cool weather. I then put a timer on pump to only allow it to run during the day time. I opened the cover on the check valve and cleaned off the build up with some vinegar. Lately I have not had enough hot water in the evening if I turn the breaker off. I occasionally turn the breaker off in order to see if solar is making enough on its own. The collector is on the roof directly above the tank. There is very little piping between the two. Last evening I checked the temperature on the roof with inferred. On the air bleed fitting that does not have any insulation, I was getting 89° compared to 60° from the metal racking nearby.
I realize this is not the most robust system but it was saving me power. My biggest concern now is if I leave the breaker on all the time, will I be losing more power then I was saving before this started happening?

[Kevin_in_Denver]

Reverse Thermosiphoning is usually the culprit. A spring check valve is better than a swing check. Either style has a surprisingly high rate of leak failure.

Forward thermosiphoning has also been seen. A spring check with a stronger spring can fix that.

Make sure the controller is not set for recirculation freeze protection.

[peakbagger]

Is this an open loops system where you are pumping the residential hot water through the panel? If so you probably have some scaling issues in the piping including the collectors and the fittings. Its going to be hard to keep a check valve happy if there is scale in the water and a non functioning check valve will creat reverse thermosiphoning where the cooler night time air cools the tank off. If you have "hard " water, you may have to descale the system on occasion otherwise the scale can really cut back on the heat transfer.

[solar_realist]

Thanks for the reply
[open loop. The controller is a basic differential. Freeze is not an issue in my regional.]
It looks like I’ll be on the quest for a better check valve.
Forward thermo siphoning could be an issue as well. At the end of a good sunny day, every part of tank inside and out is hot. Then at night it has both pipes to carry the heat away. I was wondering if the copper pipes aid as a heat sink as well.

[muskrat]

direct dc to electric water heater

With photovoltaic panels getting in the sub dollar range I have been thinking of doing a direct dc panel to electric hot water system.By stinging panels together in series until they add up to 240 volts dc and hooking them straight to large tanked electric hot water heater I could avoid the cost of an inverter ,charge controller etc.This would avoid the complication of glycol loops or drain back systems.I would mount the panels vertically.This would of course decrease summer heating but would increase winter heating at a time when water panels are nearly non functional.Most electric hot water heaters have two heating elements and one thermostatic controller.Extra thermostatic controllers are easy to come by as is usually the tank that fails on a electric hot water heater. If one was hooked to a solar array and set to a higher temperature and the other to the power grid at a lower temperature the solar would do the heating most of the time with the grid taking over when the solar didn't cut it.Vertical panels would be very cheap to mount and self cleaning for snow.you would get only two thirds the total power per year but the total per month would be more even.You would probably need a cutoff switch for the array for safety's sake but other expenses would be minimal.Has anyone out there tried a system like this?Can any one point out any problems I may have overlooked?Thanks for your input.

[Naptown]

With photovoltaic panels getting in the sub dollar range I have been thinking of doing a direct dc panel to electric hot water system.By stinging panels together in series until they add up to 240 volts dc and hooking them straight to large tanked electric hot water heater I could avoid the cost of an inverter ,charge controller etc.This would avoid the complication of glycol loops or drain back systems.I would mount the panels vertically.This would of course decrease summer heating but would increase winter heating at a time when water panels are nearly non functional.Most electric hot water heaters have two heating elements and one thermostatic controller.Extra thermostatic controllers are easy to come by as is usually the tank that fails on a electric hot water heater. If one was hooked to a solar array and set to a higher temperature and the other to the power grid at a lower temperature the solar would do the heating most of the time with the grid taking over when the solar didn't cut it.Vertical panels would be very cheap to mount and self cleaning for snow.you would get only two thirds the total power per year but the total per month would be more even.You would probably need a cutoff switch for the array for safety's sake but other expenses would be minimal.Has anyone out there tried a system like this?Can any one point out any problems I may have overlooked?Thanks for your input.

Numerous problems not the least of which is the NEC.

[russ]

Numerous problems not the least of which is the NEC.

As Dave pointed out - not good. The switch you mention would be expensive for 240 volt DC.

All in all this is a good idea to keep at the idea stage.

[muskrat]

what part of the nec code are you talking about?

Numerous problems not the least of which is the NEC.

I consulted with a licensed electrical engineer with extensive solar experience about possible code problems and he said he didn't know of any part of the code that would forbid it.The safety switch I was talking about is standard equipment for almost any solar array and although not cheap isn't ridiculously priced.The 240 volt thermostat switch is part of any standard hot water heater.Do you have specific code in mind or is this just a case of "we ain't done it that way before".

[Naptown]

The problem lies with the thermostat which is not rated for DC. Particularly high voltage DC. That safety switch is expensive because it has to arrest an arc that will become huge with high voltage DC.
If you bring DC into the house then it will have to be in metallic raceway from the point of entry to the water heater with the disconnect placed at the closest accessable point where it enters.

[bcroe]

240v dc

The problem lies with the thermostat which is not rated for DC. Particularly high
voltage DC. That safety switch is expensive because it has to arrest an arc that will become huge with high
voltage DC.
If you bring DC into the house then it will have to be in metallic raceway from the point of entry to the
water heater with the disconnect placed at the closest accessable point where it enters.

In addition to the above, your 240V Maximum Power Point set of panels will only actually deliver that voltage
when there is best sun. At somewhat less optimum sun, available current will drop, so voltage will drop,
power delivered drops as the SQUARE. So the system will be hugely inefficient most of the time. This problem
could be cured by some kind of MPP control reducing heater voltage to optimize panel power output.
Greatly increasing the panel capacity would work, again at poor efficiency. Bruce Roe

[muskrat]

much more helpful

The problem lies with the thermostat which is not rated for DC. Particularly high voltage DC. That safety switch is expensive because it has to arrest an arc that will become huge with high voltage DC.
If you bring DC into the house then it will have to be in metallic raceway from the point of entry to the water heater with the disconnect placed at the closest accessible point where it enters.

this comment was much more help full. The problem might be overcome with a low cost modified sign wave inverter with a low voltage cutoff.The metallic raceway doesn't sound like too much of a problem and the cutoff switch was already mentioned.I had already expected to oversize the panels and am not worried about efficiency but only installed cost compared to water panels.Using the hot water heater thermostat to control a dc rated relay might be an option I'll have to check prices.Thanks for the feedback.

[Naptown]

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.

[muskrat]

relay would work

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.

I just checked mauser a 250 volt 25 amp dc relay is $150 .That's about 6000 watts which is greater than the amount of panel I expect to use.The same relay could be used to shut off the current if the voltage dropped too low.No need for cheap inverters or low voltage.According to pv watts a 6 kw array place vertically will produce an average of 15 kwh per day in the winter. This will raise 75 gallons per day of water 85 degrees F.I can get multi crystal panels in pallet lots for about 48 cents per watt.This brings it into the price range of a similar hot water system without the complications of a glycol loop or drain back controls.It should function just as well in 20 below zero temperatures as in 40 degree temperatures;something a thermal system cannot do.

[russ]

Thermal evacuated tube types work well under those condtions - have fun as you have already decided what you want to do though.

[lile001]

The problem lies with the thermostat which is not rated for DC. Particularly high voltage DC. That safety switch is expensive because it has to arrest an arc that will become huge with high voltage DC.
If you bring DC into the house then it will have to be in metallic raceway from the point of entry to the water heater with the disconnect placed at the closest accessable point where it enters.

None of these are insurmountable problems. The safety switch really isn't that expensive, I have a DC safety switch on my 3 KW solar array, and I seem to remember it was under $200. The NEC requires fuses or circuit breakers - a DC circuit breaker with the proper rating should be about $30. Square D makes a number of odd breakers, including, DC, Switching duty, and remote controlled breakers - it may be possible to use some of these, or a DC contactor, to control the current to the heater, if the thermostatic controls that already come with a water heater can't be pressed into service.