Tank Updates

The EPDM lined tank on our $1K solar water heating system has been in operation for about 10 months.  

I decided to spend a little time on the tank to accomplish the following items:

Bottom line is that everything looks fine, and is holding up well -- I expect it  to live longer than I do :)


How does it look after 10 months?

I took the lid off, and removed the HDPE (black poly) heat exchanger pipe coil.  This gives a good look at the full liner, the edge sill pieces, and everything inside the tank.  It all looked fine to me.  The liner showed no signs of distress or brittleness.  The plastic deck wood pieces that go around the top edge of the tank, and provide a place for the lid to seal down to looked fine.  I was a bit concerned about whether these would hold up to the tank temperature, but they are doing fine.


The EPDM padded concrete blocks that are put in the tank bottom to support the pipe coil heat exchanger were not disintegrating.


click on pictures for full size

The tank water has a very light yellowish
cast.  The picture does not show it well,
but the bottom is very clearly visible with
only a small amount of sediment on the bottom.

The sill pieces that the lid seals down
to are made from plastic deck wood.
They show a little mineral staining from
the water, but not signs of deterioration
or swelling.


Water Condition and pH

As the picture above shows, the water was relatively clear.  There is nothing growing in it that I can see.  There is a slight yellow tinge that might be some dissolved material from the EPDM liner, or who knows what.  There is a little sediment at the bottom, and a few things I managed to drop into the tank and did not recover.  The tank water tastes awful.


I did not add anything to the water when I initially filled the tank.  It is water straight from our cold water tap.  We do have a water softener, so some of the dissolved Mg and Ca minerals are replaced with Na by the softener.  None of the pipes that I could examine showed any sign of mineral deposits.


I checked the water pH, and it is 8.4.  I was prepared to add some baking soda if the pH was acid, but this is slightly alkaline.  As near as I can tell from asking a few people, the 8.4 is OK, and no action is needed.


So, I'd say that this is some tentative evidence that you can just put plain water in the tank, and it will probably be OK.  In any case, I'd stay away from putting anything toxic in the tank, as the heat exchanger is a single wall type.

Changes to reduce heat loss, and provide better lid sealing

I did this test on tank heat loss last month.  It shows somewhat more heat loss than I would like.  So, since I was working on the tank anyway, I did a few things that might help reduce heat loss, including:




The wood strips on top of the top of the lid clamp the lid down tightly to the grey plastic deck lumber sills.
6 inch long lag screws go through both layers of foam and into the sills -- one of the lag screws
is visible in front of the coffee cup.
The part of the lid to the right is fixed to the tank, while the large part to the left comes off for access to the tank.
There is a half inch CPVC pipe that goes through the tank lid (hidden by the coffee cup) -- a dipstick can be inserted
in this pipe to measure the water depth, or test the pH.  The same pipe can be used to add water if needed -- to date the
evaporation rate has been quite low.


The construction of the lid from the bottom to top is:

All of the layers are glued together with Great Stuff polyurethane foam or silicone seal.


I did not add anymore insulation at this point, so that I could test the effect of the steps above. 


I added a dipstick that allows any drop in the tank water level to be measured more precisely.  The pipe the dipstick goes through is sealed up with silicone to the EPDM and insulation board.  I leave a very small part of the pipe that the dipstick goes down through open to vent the tank and prevent any pressure difference between inside and outside, although I would guess that it is not so pressure tight that this is needed.

Replace the HDPE plastic pipe coil heat exchanger with a PEX pipe coil

I've been recommending for some time that people building the system now use a coil of PEX pipe for the heat exchanger rather than the coil of HDPE (black poly) pipe that I used in the initial build.  There are two main reasons for this:

- With the HDPE pipe, I could sometimes get a slight smell of poly pipe when taking a shower.  This was not really that objectionable, but using the PEX pipe coil heat exchanger eliminates this slight smell.


- When used as a heat exchanger, the HDPE pipe is at, or a bit beyond its design specifications.  When I got the HDPE coil, I intentionally got the lowest pressure rating they had (80 psi) to make sure that if the exposure to 140F and house water pressure were going to result in a failure of the HDPE it would show up on the prototype.  After 10 months, the HDPE pipe coil showed no signs at all of distress, and seemed to be holding up fine to our up to 50 psi water pressure. 


But, using the PEX instead of the HDPE completely eliminates the pressure/temperature rating as a potential source of failure or leaks.  The PEX has temperature/pressure ratings well over what it will ever see in this application.  Based on its track record in applications like radiant heating, I expect it to be going strong 50 years from now.

The price difference between a 300 ft roll of 1 inch PEX and the same in HDPE is about $100 -- probably money well spent -- this extra $100 is already included in the total cost estimate for the system, and it still stays inside the $1K target.


Both the PEX and the HDPE coils were installed as a single length with no fittings at all inside the tank.    Whatever pipe coil material you use, it must be NSF approved -- this will be stamped on the outside of the pipe.  I got my 300 ft coil of 1 inch PEX from BlueRidge Hydronics here for $167 with free shipping...  While the this 300 ft coil of PEX heat exchanger has not yet stood the test of time, my feeling is that it will be about as bomb proof and efficient a heat exchanger as you will find. 


In a nutshell, the advantage of using a big coil of plastic pipe like this is that it actually stores about 9 gallons of water right in the pipe coil.  This water warms up to full tank temperatures, so for most water draws there is no temperature loss due to the heat exchanger at all -- its a 100% efficient heat exchanger.  Test results for the pipe coil heat exchanger here...


One sort of odd thing is that the PEX 1 inch pipe is smaller in diamter than the HDPE 1 inch pipe -- the PEX is 0.875 inch ID, while the HDPE is right around 1 inch.  The result is that the PEX coil holds 9 gallons of water, and the HDPE holds around 12 gallons of water.  Not really enough to make much difference.


If you want to think about a cheaper solution, a 200 ft roll of 3/4 inch PEX costs about $95, and holds about 3.8 gallons of water -- this will probably  work  well -- if you go this way, I would take more care to separate the layers in the pipe coil well so that tank water can flow through the pipe coil easily and improve its heat exchange efficiency.  At 3 gpm, 200 feet of 3/4 inch plastic pipe has about 2 psi of pressure drop, so as long as you have descent water pressure, this should be OK.  Another option is to use a heat exchanger similar to the one Kevin used that uses a mix of copper and PEX -- this is more compact and will fit in a smaller tank.


For the connection of the 1 inch PEX coil to the 3/4 inch copper house piping I used SharkBite reducing couplings.  This was basically the most straight forward way to go, and did not require renting or buying a PEX connection crimper.  These fittings are kind of pricy at $12 each, but, they have the plus that they can be taken apart in the future if need be.  The valves that I put in two 3/4 inch copper lines to the coil made it fast and easy to isolate the pipe coil from the house plumbing for the change over.  To drain the HDPE coil, I disconnected one end, drilled a 3/8 inch hole in the other end, and used a compressed air hose inserted into the 3/8 inch hole to force the water out the other end into a couple 5 gallon buckets.  It would have been very awkward to get out of the tank without draining it first.


Note that both the PEX and the HDPE coils when filled with water have essentially neutral buoyancy.  They float up to the top of the tank so that just a tiny bit of the top set of coils is breaking the surface.  They don't sit on top of the nice EPDM padded concrete blocks that I put in the tank to give them a place to rest.  So, one might just leave out the concrete blocks, although they do provide a good landing place in case you wanted to pump the water out of the tank for some reason. 


On the original HDPE coil installation, I separated the coils from each other by pushing some short lengths of half inch CPVC pipe horizontally between the pipe coils.  The idea was to separate the pipe coils enough to allow the tank water to flow around them, and improve the heat exchange.  Most of these CPVC pipes had fallen out, so I only put a few of them back in for the PEX coil.


After thinking about this, I think that the best way would have been to make some T shaped pieces from CPVC such that one leg of the T could be pushed between the coils from the top -- the other legs of the T would rest on the top of the pipe coil and keep it from falling out.  This would make for vertical flow channels through the PEX coil, which is probably better for heat exchange than the horizontal channels.


click on pictures for full size

The 300 ft by 1 inch NSF PEX coil
ready to go into the tank.

The PEX coil in the tank and hooked
up to the copper house plumbing.
Note how it stays just barely
submerged right at the top of
the tank.

The PEX comes out through the sill
board, so that no penetration of the
EPDM liner is needed.
The slot in the sill board is sealed up
with silicone to keep vapor from
leaking out.
Note the SharkBite fitting that connects the
house copper pipe to the PEX.

PEX coil in place.  Between the
temperature loggers and
the differential controller, there
are 4 temperature sensors in the tank --
this is what all the wires are.

The PEX pipe must be secured well where it goes through slots in the sill board.  I used a copper pipe clamp on the inside, and a plywood support on the outside to keep it in position.  Be sure to seal around the penetration and the pipe clamp with silicone.

When you fill the tank, its very important to make sure there is an air gap between the collector return pipe and the water surface.  Air must be able to go up the return pipe to replace the water in he collectors as the collectors drain down.  Not allowing this air space will result in the collector not draining, and freeze damage to the collector piping.

I guess that had I been thinking about this more carefully, I would have done the CPVC pipe coil separators with a T shaped CPVC pipes placed vertically between the coils (as described above), and would have cut the polypropylene ties that I used to keep the coil together when placing it in the tank.  Both of these would result in more tank water circulation around the coil and faster heat exchange.  I'll do this next time I open the tank in 10 or 20 years :)

Gary June 11, 2009