The current installation.
The TMO500 limits temperature transfer above 80°C. I envisage "tinkering" with the controller, so having a panel that will do the least amount of damage when water is not being circulated is pretty importatnt.
The TMS20 manifold connects 20 tubes, providing a net absorber area of 2 m2.
The kit is excellent and is securely packed and complete.
Assembly is straightforward, the time consuming bit is preparing the roof and weathering.
I employed a roofer to install the panel - he made up two lead shrouds for the 22mm water pipes where they enter the roof.
Silicone lubricant on each manifold recess helps ease in the tubes during assembly.
There is a cheaper alternative.
Water is incompressible. Unlike air, it's impossible to squeeze any quantity of water into a smaller volume. This is important in a "closed" water system.
When the water gets heated it expands and that extra volume has to go somewhere. Without an expansion tank the pressure could rise to a dangerous level.
A diaphragm type of expansion tank is a pressure vessel that contains a flexible membrane totally separating the water from a captive volume of air. Air compresses and, as the water temperature rises, creates extra pressure; the air in the vessel accommodates that increase.
Instructions state a min working pressure of 0.5 bar. Filling 1bar. Safety valve 3 bar.
I'm using a Flamco Flexcon 4 Litre. Chris Laughton suggests an 8 Litre. I would agree: there's a noticeable pressure change between different temperature conditions in the system. When the system is cold, the pressure drops by 0.2 bar from the warm state. The larger capacity tank would "mop" this up.
UK water bylaws require this isolation. Should the mains pressure drop, this arrangement avoids contaminating drinking water.
Connections to closed circuits.
(I failed to find a web link to corroborate this statement.)
Alternatively, do away with this! Having had to figure out a way of charging the system with antifreeze, you could try this "cheapskate" approach:
Make no connection whatsoever to the water main. Use a 5L house-plant sprayer (Hozelock). Remove spray head, fill container with antifreeze water mix, attach hose to system drain and pump. The pump will easily deliver up to 3 bar. The volume of the my system is 8 Litres, and was filled in a few minutes. Pumping to 1 bar is no problem.
This prevents water from thermo-syphoning. Having the solar panel cool down the hot water at night is not a good idea.
Continuously purges trapped air/ gasses from the water. I'm using a Honeywell EA122. All pipe work from the hot water cylinder coil rises to the vent, that is all "horizontal" pipe runs are set with a sprit level bubble displaced.
Chris Laughton comments: " T'max and myself no longer recommend siting the air vent near the manifold unless it is a 130C. type. I have personally seen a few melted ones in 'no-pump, lots of sun' conditions."
(As it happens, I chose to install the automatic air vent away from the panel for easy access. It's in the roof space sited 3 metres from the "hot" side of the panel.)
This is a purpose designed break in the overflow pipe. The operation of the safety valve can then be immediately seen.
I could do away with the pump if the solar panel was significantly lower than the hot water cylinder coil. Warm water would thermo-syphon around the system, cooled water from the heat exchanger in the cylinder returning to the solar panel.
In my installation, the solar panel is slightly higher than the cylinder, so a pump is required.
I'm using a Brauchwasserpumpen für Trinkwasser-Zirkulationsleitungen (Industrial water pumps for drinking water circulation lines.) These pumps designed for hot water systems where you turning on a hot tap and hot water comes out - no waiting.
Grundfos UP 15-13 B The specification is: head (1.5m H20) / flow of 0.7 m3/hr single phase 25W. Continuous operation. (The head and flow metrics are similar to the definition of load lines in electronics.)
This may not the ideal choice. The pump is probably over specified as we're not worried about potable water.
Playing with the pump, transparent hoses and a bucket of water is educational. Water won't flow if there is air in the pump (cavitation). It must be primed by flushing through first. As soon as flow is established, the pump circulation improves, extracting air from the rotor.
Uses a Synchronous Reluctance Motor - very similar to an old record deck motor.
The solar panel is exposed to frosts: the lowest recorded temperature in England is -26.1°C (10 January 1982).
Fernox Alphi-11 is carried by the local plumbers' merchant - and it "does" -30°C. The liquid is non toxic - just in case of a leak in the heat exchanger coil in the hot water cylinder.
A 4 bar gauge is sited in a visible location - this can be checked regularly with ease. (The expansion tank has one as part of the kit, but as it it mounted in the roof space it's not so easy to check.) System pressure, when cold, is set to 1 bar at the gauge, 8ft beneath the highest point, the vent.
This is present for experimental use. I was interested in slowing down the water flow. This could easily be deleted, or a companion valve could be added to isolate the coil.
The DS1820 probes are wired to the controller. Heat shrink sleeving protects the soldered pin connections. Silicone rubber is used to provide a good coupling between the package and the pipe work. An air gap must be avoided. Insulation tape secures the assembly while the rubber cures.
Cable ties provide strain relief. The manifold probe is mounted in a 10mm pipe immersed in the "hot" manifold T piece.
Water was introduced and flushed out through the drain cocks. (The pressure relief valve was not used as an easy flushing point: there is a risk of wire wool and debris fouling it.) System run for a day with water only. Pressure at 2 bar. Check for leaks / pressure drop.
Water flushed out again. Antifreeze added: 4L pumped in, and water and pressure set to 1 bar.
Encouraging the pump to work involves bypassing the single check valve and coil with a piece of clear hose between the drain cocks. As soon as the pump decides it is a pump - determined by listening and watching the clear pipe, the drains are closed and the coil gate valve opened.
The red cap on the automatic vent was removed. The cap has a "clever" paper filter that seals ups should the vent start to leak. The installation notes suggest this removal during charging - we don't want the thing sealing up just because water sloshed in to its float chamber.
The vent cap is now replaced.
1, I wanted a totally separate system to the existing gas boiler, gravity
feed arrangement. I could have taken pressure from that existing system, but
it would need isolating for maintenance. (I assumed that the solar system
would be intermittently out of commission.)
2, I'm using antifreeze. To have that flowing in the house radiators - I'd need lots of it. (I've seen an American installation that automatically bleeds when there is a risk of freezing - so it doesn't need antifreeze.)
3, I could not site a header tank in a convenient place.
From my notes from the CAT solar heated water course:
Some snaps of the installation.
After the two week's use - yes it works! 3 sunny October days provided hot water for 2 adults for 4 days. That's 3 sunny October days in England.
The installation was commissioned in October 1999. At the time of writing, January 2002, I have to report that I haven't had to tinker. The only maintenance has been a quick "squirt" to get the pressure up to 1 bar - the gauge read 0.9 bar.
The solar panel installation is subject to regulations.
This page was last updated on the 11th March 2006.