We ended up not choosing solar thermal, and just using photovoltaics which have no moving parts and nothing running through them that can freeze or boil. Below are some ideas from the planning phase of our building based on theory and various research.
The ideal solar thermal system stores heat in a single hot water tank which can collect heat from solar panels, add heat from a backup boiler or heater, provide hot water, and distribute heat to the radiators, underfloor heating or the ventilation system.
The problems and challenges facing solar thermal systems include overheating, chattering, hygiene and freezing. People often worry about not getting enough heat from their solar heaters, but apparently overheating is the biggest cause of failure for solar systems, and one reason why systems in Japan are usually under-sized, meeting demand only at maxiumum output. Overheating causes steam and high pressures, which can shorten the lifetime of elements within the system. It should be possible to design a system that can withstand a range of pressures and temperatures. After all, steam heating systems have been around for over a hundred years. Evidently they are not always designed for the pressures and temperatures that sooner or later they well reach, and solar systems often fail within five years.
Another problem is chattering. This happens when the hot water from the solar system is fluctuating around the level at which extra heat must be added. As with photovoltaics, sunny days are are no problem, as there is a large and constant amount of heat. Overcast or rainy days are not a problem, because there is a small and constant amount of heat. Partially cloudy days are a problem. Because solar systems cannot guarantee to supply enough heat all the time, there must be a backup system. If the solar heating is sufficient, the backup heater is not needed; if the solar heat is insufficient, for example on a snowy day, the backup must switch on. In certain weather conditions, the backup system may be switching on and off several times, rapidly wearing out motors and switches. A certain level of sophistication is needed in the control system. This problem should not be insurmountable, for example the backup system could only switch on after the sun has gone down.
Another issue is storage of water within the tank. The tank could most efficiently contain the same fluid used for the panels and for the domestic hot water. This is most efficient, however there may be issues with legionnaires disease, which thrives in the kind of temperatures that you will often get in a solar water system. In addition, tap water is liable to freeze, so any water left in the panels at night time could lead to problems, since freezing water expands and tends to burst pipes. Using a refrigerant within the solar circuit, and a heat exchanger for the domestic hot water also has advantages, for example in allowing higher pressure within the panels. Because of the heat exchanger it is, of course, less efficient.
One choice with solar systems is whether to use a drainback or continuous system. In the continuous system a refrigerant is used, and as soon as the water in the panels heats up, flow starts, transferring the heat from the panels outside into the tank inside. In the drainback system, water is sent to fill the panels in the morning, or at least when the panels start heating up. It is then returned either when the water has reached a certain temperature, or at the end of the day when it's got all the radiation it can get, before it starts cooling down. The advantages with this system are that overheating is avoided, since the hot water is sent into the house before it overheats. Tap water can be used, since there is no danger of the panels freezing as water will not be sent there if it is too cold, and will be sent into the house before the temperature drops.
Another issue is the performance of the hot water supply system. Ideally there should be time and volume settings so that the bath be filled to the desired volume and temperature. There should ideally be a means for re-heating the water within the tub, so that water may be saved as is customary in Japanese houses. Most commercially available hot water systems in Japan have all these functions ready fitted, although the controllers cannot necessarily be acquired separately. Designing your own system would be a challenge.
From the beginning solar thermal seemed very attractive as a way of reducing energy. Solar thermal panels turn over half the sun's energy into heat, and are three or four times more efficient that photovoltaics. In the end a major considerations for not choosing solar thermal were to keep the systems simple and the roof elegant. Thermally it may have made sense, but financially, in terms of initial cost and the relative costs of purchased and re-sold electricity, photovoltaics were the obvious choice. And if the solar thermal system was going to fail within five years, even the superior thermal efficiency of the system becomes doubtful. Another factor was our NEDO grant application, which I'll write more on later.
