Hot/Cold water tank to supplement home Heating/Cooling

[hamburglar]

Hello everyone.  I've been thinking about ways to use solar energy to supplement home heating and cooling.  I would like for it to be somewhat affordable and worthwhile financially, but that is not a sticking point, because I will have a lot of fun with it as well.


The most basic system for supplemental heating that I can imagine would be a large water tank (50-200 gallons) placed somewhere central in the home.  It could be hidden away in a closet or pantry….whatever.
Plumb in appropriately sized 12/24/48v DC heating elements.  Maybe slap on a cheap thermometer for monitoring.  Run whatever solar panels you want directly to the element terminals.

No pumps, pipes, batteries, controllers, or inverters.  Keep it simple.  I guess you could get these fancy elements that have built in thermostats if you are running high power and worry about boiling the water.
I think it is reasonable to say that every bit of heat you add to the water will eventually and slowly escape into the house.  Slowly is good because it will also release at night when it gets the coldest.


From that starting point you could probably do whatever you want, like adding multiple tanks with circulation pumps, radiant floor heating, radiators with fans….ect

Does anyone have any thoughts on this, like where it can go horribly horribly wrong?

The cooling idea involved the same “large tank” concept, but you use a water chiller to cool the water.  But that would also require plumbing and a drip pan for the tank, charge controller, small battery bank (for compressor start up current), and an inverter.  That is much more expensive and complicated.

[DamonHD]

Schemes like that (though more like tens of thousands of litres per house) do exist.  You can also use the ground under your house as a thermal store if you don't have flowing ground water and if you dig insulattion down deep all round the edges of your house.

Rgds

Damon

[GaryGary]

Hi,
You might want to consider a home made solar thermal panel instead of a PV panel.

PV panels are selling for about $1 per peak watt -- maybe a bit less if you can order a lot of them (plus shipping).

A good homemade solar thermal water heating panel runs about $5 per sqft. 
On a comparable peak watt basis, it delivers about (1 sqft)(1 sqm/10.76 sqft)(1000 watts/sm)(0.6 efic) = 56 peak watts per sqft, or $5/56watts = 9 cents a peak watt -- so, the homemade thermal panel is about 10X more cost effective than the PV panel (roughly).

Since the PV panels are about 15% efficient and the water heating panels are about 60% efficient, it only takes about 1/4 of the roof space for the same heat output panel from a thermal collector.


You could basically use a system like this: http://www.builditsolar.com/Experimental/PEXColDHW/Overview.htm
but, just leave the insulation out of the storage tank so that it can lose its heat more rapidly to the room.

If you wanted to take it a step further, you could also use the same system for some cooling in the summer by using a scheme like this: http://www.builditsolar.com/Experimental/RadiationCool/EvaproRad.htm
It achieves quite good cooling rates in my climate, but may not work as well in cloudy more humid climates.
You would want to be careful to monitor for condensation on the tank if you use it for cooling -- all depends on your climate.

If you use the PV approach, I'm not sure if you need some kind of matching between the heating element and the PV panels to get good efficiency out of your PV panel.  Most PV systems now have inverters that keep the PV panel at the point on its power curve where it generates the most power (MPPT) -- would something like that be needed between the PV panel and the heating element?

How much heat storage would a 200 gallon tank give you?
200 gallons of water cooled from 130F down to 80F gives you (200 gallons)*(8.3 lb/gal)(130F - 80F)(1 BTU/lb-F) = 83000 BTU -- about equivalent to a gallon of propane burned in an efficient furnace.

Would the tank be able to get rid of the heat fast enough between collection periods?
If the tank walls were about R1 and the tank wall area was 80 sqft, then when the tank temperature was 110F and the room temperature 70F, then the tank would deliver about (80 sqft)(110F - 70F)/R1 = 3200 BTU/hr, or about (24)(3200) = 77K BTU/day -- so, its not such a bad match to the amount of heat it stores.  A tank that is made thinner and longer would have more radiator area, and would get the heat out faster.

How much collector do you need to heat up a 200 gallon tank on a sunny day?
The rule of thumb is about 1.5 to 2 gallons of water per 1 sqft of solar thermal collector area, so the 200 gallon tank could handle about 100 sqft of solar thermal collector.

Gary

[Bruce S]

Gary;
Great to see you posting answers !!
As a small hijack, when did you get into cloning dinosaurs? IS that a Brontosaurs I see 
Cheers
Bruce S

[GaryGary]

Hey -- that ain't no ordinary dino -- its solar powered with a very loud airhorn

http://www.builditsolar.com/Projects/Educational/Dino/SolarDino.htm

Gary

[Bruce S]

Pretty Nifty!! That would be pretty cool to look out and see that , sure would get me to slow down and go play if I saw it along the road 

So pray tell what is that small world doing hovering just above the Dino's head?
Balloon or UFO?

[GaryGary]

That (somewhat amazingly) is a ball that was thrown up in the air just before the picture.
I'd like to say that it was good photography, but it was actually just luck.

Gary

[hamburglar]

Gary, thanks for that great information!  I can already tell you that I will use a lot of those calculations and ideas in the future.

[bob g]

you might want to check this out

http://mit.edu/solardecathlon/solar1.html

MIT's solar heated house from 1939, lots have been written about the project.

bob g

[hamburglar]

Hey guys, I have been reading more about this over the past couple of months, and I have another question.
Does anyone know of a “controller” that can be used to keep the panels running at peak output?  I understand that MPPT charge controllers are used when charging batteries.  But can you put a MPPT charge controller between a PV-Array and a pure resistive load, without batteries?
I saw one solution written in the early 90's that involved a custom built controller that used relays to change the Ohms of the heating elements (parallel-series combinations of multiple elements).  This was done so that the Ohms fit into the power curve of the PV Array at different times of the day for better efficiency.  But that sounds too complicated.  I wondered if these newer MTTP charge controllers could be used without burning up.


I am not ignoring the traditional thermal panel response above.  That post as well as the links in it are awesome.  I will play with that for sure (probably as a preheating stage).  I understand that heating water with PV panels will be much less efficient.   But, I plan on having a lot of PV panels to produce backup electricity, and possibly run a mini-split AC in the summer.  This will just be a dump load for them in the winter months.

[GaryGary]

Hi,
There is one system here to optimize the pv to water heating resistor -- http://sel.me.wisc.edu/trnsys/downloads/trnsedapps/demos/pvsdhw.htm

Maybe there is a simpler way?

Gary

[ghurd]

"Does anyone know of a “controller” that can be used to keep the panels running at peak output?"

As a matter of fact, yes. 

Seems like you know most of this, but,

The thing about PVs is they are designed to operate at a fairly fixed voltage.
Less sun means less current, but about the same voltage.  Chart below.

Problem, mainly, is a lower ohm resistor than the PV can feed, and still maintain the correct operating voltage.

Minor problem is "peak power" is a good percentage above standard battery voltages.

Solution?
Ghurd controller set up to switch very fast at the proper voltage for the PVs (probably about 16~16.6V for a 12V PV), and a big fat cap instead of a battery.
It would keep the PVs operating at about their design voltage, and the fast switching speed would effectively be 'a self adjusting variable resistor' (for lack of a better term at this time of the morning).

I agree that solar thermal would be a better and cheaper project on any scale large enough to do any good.
200W of PV is not going to make much heat in mid winter.
G-

[hamburglar]

I would like to talk with you a little more about this controller.   One thing I have failed to mention before is that this project would be in the 1000-2000 watt range.   So im not talking about hooking up a single small panel.

Im about to be living the single life and ill be able to spend my money like I want to again.

[XeonPony]

How fancy do you want to get? The morning star Mppt 60 will handle 3Kw at 48v and has allot of logging built in and a built in lan interface.

[hamburglar]

XeonPony,

Do you know if you can run a controller like that without a battery bank?  And I don't mind getting that fancy. 

[XeonPony]

well I don't see why not, alternitevely just use a small battery to act as a slow capacitore to make it think it is charging a alrge bank