Posted by tim on November 12, 2007, 2:03 am
My water is heated by an instantaneous (continuous flow / non-storage)
gas-fired water heater (http://www.rinnai.com.au/hotwater/home/?
whs=home&pg=0). Using clever electomagic controllers, it burns just
enough gas to maintain the set water outlet temperature.
The gas heater takes water from mains temperature to 50dgC. My plan is
to reduce this temperature step, thus reduce the amount of gas
required, by pre-heating the water. The electromagic side of the gas
heater can manage the fine details of how much gas to burn to maintain
a stable temperature at the tap.
I have a redundant storage gas water heater to donate to the cause.
It's about 20 years old, past its design life, but the tank is in good
condition - it was holding mains pressure up until the day it was
retired, so should have no trouble with the couple of metres water
head I'll be subjecting it to.
I plan to build I build a cheap and cheerful collector panel, and
install it on the roof. A small circulation pump will cycle this water
to the tank. If I use a solar panel to power the pump, it will be
conveniently self-regulating: pump when the panel is collecting heat,
stop when it isn't. Might be more effective to use a photoresistor to
switch a mains powered pump and a solenoid drain-back valve.
So, this setup will give me a tank full of warmer-than-ambient water.
How hot? Depends on how big and how effective my panel is, I guess.
I'm thinking really simple here - a zig-zag of black plastic pipe
between two sheets of polycarbonate.
I'll strip the insulation off the top half of the tank, wind copper
water pipe around it, and re-insulate over the top. So the only part
of the system that will see mains pressure is copper pipe. Unlikely to
Best case, my pre-heater might occasionally raise the water up to (or
slightly above?) the 50dgC my has heater is set to; hopefully the gas
heater will be smart enough to know that it doesn't need to burn gas
in this case. More typically, it will pre-heat the water less than
this, and the gas system will add the final few degrees. Worst case,
my gas system gets incoming water at ambient (mains) temperature, as
it currently does.
For thermal expansion safety, I might install a small un-sealed header
tank at the top of the collector panel. I don't imagine my simple
collector getting hot enough to boil water, but if it does, my top
tank can vent steam. We get occasional winter frost, but rarely get
more than a couple of degrees below freezing. Plastic pipes should be
able to handle that, or if necessary I could dose the low-pressure
side with automotive antifreeze.
Any comments or suggestions?
Posted by Sundug on November 12, 2007, 2:28 pm
I did a similar thing with a tank, here`s mine-
It would transfer heat to the tank better if the coil was embedded in
grout. PV pumped systems are the most efficient, but do you have any
below freezing weather?
Posted by gary on November 12, 2007, 3:29 pm
If you arrange the plumbing correctly, you should not need the
solenoid valve -- the water should drain back to the tank by itself
when the pump turns off. This is more reliable than depending on
solenoid valves to work -- reliability is important if you have
freezing temperatures, and the drain back is your freeze protection.
I like the cheerful part :)
I think that the stagnation temperature of such a collector will be
too high for black plastic pipe.
You may also be disappointed with the performance.
If I am understanding you correctly, the mains water takes one pass
through the copper coil heat exchanger on the way to the tankless
I think that it may be difficult to get good enough heat transfer from
the storage tank to the mains water with just a single pass.
There are some designs in which the coil of copper pipe is immersed in
the storage tank, and even with this good heat transfer situation, it
has to be a large coil of copper to work well.
If you don't plan to pressurize the tank anyway, you could cut the top
off, and put the coil inside the tank. Or, use a small version of a
tank like this one:
If you use a drain back system, this would take care of freeze
protection and expansion (since its vented to atmosphere).
DON'T use automotive antifreeze -- its poison. Use polypropylene
glycol -- its readily available and not poison.
You might just take a look at a batch system:
Even simpler than what you are planning.
Posted by tim on November 12, 2007, 8:23 pm
On Nov 13, 2:29 am, g...@builditsolar.com wrote:
My mistake of incorrect terminology. Perhaps I mean an anti-
I intended to have the collector full of water, so the circulation
pump need only work against pipe losses. Over night (or whenever the
sun isn't shining enough to trigger the photoresistor), I'd close the
valve so my tank full of hot water doesn't thermosyphon to the roof
and radiate to atmosphere.
I'm new to all of this (obviously), so never considered the option of
leaving air space in the system so it will all drain back to the tank
when not pumped. Would need a bigger pump, but it's an elegant
solution to freezing or overheating.
I don't think freezing will be a problem. We have a few -2dgC (28F)
nights each winter, but it rarely gets colder than that. I've never
had frozen water pipes in my 5 years here. I'm at 37S, 450m elevation,
Posted by daestrom on November 13, 2007, 11:07 pm
Well, 'an ounce of prevention...' Collectors facing a night sky can cool
down to ambient rather quickly. Depending on the tubing material, it only
takes one night to do many $$ damage. (I've actually heard of some folks
*wanting* thermal syphon on those rare nights to prevent freezing)
But you're right about the pump size. When using 'drain back' systems, the
pump must be able to fill the collectors in the morning. That can mean a
total head from the tank to the collector. That can be quite a bit more
than a simple 'circulating pump'.