I made some changes to the tank for the 07-08 winter. This included adding more insulation inside the tank, and a new arrangement for the lid and for the way the plumbing and sensors are brought into the tank. These are kind of incremental improvements to the design -- the old design was/is fine, the new one is just a bit better. If you are wondering what in the world a "Solar Shed" is, go here ...
New insulation going in the tank
after removing the lining. The insulation is glued in place using
"Great Stuff" polyurethane foam in a can.
Not much is needed, in that the water pressure will hold it in place. All the edges and seams are also sealed with Great Stuff.
The new layer of tank floor insulation has not yet been installed.
I also made some changes to the lid arrangement.
The wood board that ran around the top edge of the tank to provide a good place for the lid to seal down to was looking like it would have a limited life in the moist environment. I replaced this with an edge board made from plastic deck lumber.
I moved all the plumbing and sensor connections that had been made through the west part of the old lid. These connections now all go through slots cut in the west edge plastic deck board. I cut a slot for each connection, then placed the pipe or wire in the slot, and then silicone caulked the slot full. This way the entire lid can be removed without disturbing these connections.
The connections are:
The 3/4 inch copper return line from the collectors
The tank vent pipe (vents tank to ambient air)
The tank temperature sensor used by the control system inside the house (enclosed in half inch copper pipe)
The 3/4 inch CPVC supply line to the house
The 3/4 inch CPVC return line from the house
My caulking job was a bit messy (well, maybe a lot messy), but the arrangement works well, and does not leak any water vapor out of the tank.
The pump still connects through the tank sidewall near the bottom on the east end of the tank.
The only penetration through the lid of the tank is a quickly removable tube that contains the tank temperature sensor used by the differential controller for the collector pump. This also serves as a place that a dipstick can be inserted to measure the water level in the tank.
I also took this opportunity to run the return line coming from the house to the far corner of the bottom of the tank. So, this separates the supply and return lines better, and returns the cold water from the house to the bottom of the tank.
Left to right: 1) return from collectors, 2) tank vent pipe, 3) tank temperature sensor, 4) hot supply line to house, 5) cold return from house.
The tank lid itself now looks like this layer by layer from the bottom up:
EPDM liner for the lid (bottom layer -- seals against the board running around top edge of tank)
2 inches Atlas R-Board insulation board
Layer of half inch OSB plywood
2 inches of polystyrene blue board (XPS)
A number of old insulation bats piled over the top
The lid is held down by extra long coated deck screws that screw through the OSB, through the 2 inches of R-Board, through the plastic deck board that surrounds the top edge of tank, and into the tank frame. The steel lag bolts that held the old lid down were showing some signs or rust, so I thought the coated deck screws might be better (not that the lag screws would have failed within my lifetime).
I also replaced the tension tie strap that ties the tow long edges of the tank together with a stainless steel strap. The galvanized strap was corroding with the continual exposure to hot water vapor. I think that its worth using a non-corroding material here in that the tension tie is structurally important, and the tank might well fail if the strap corrodes all the way through.
I had planned to actually measure the temperature drop of the tank overnight under cold conditions, but other projects have gotten in the way. This is less straightforward than it seems, as the tank tends to develop a temperature stratification that changes with time.
Gary Feb 7, 2008