I wrote a bit about hybrid photo voltaic thermal panels in April 2011. We looked into this, and even found a manufacture, Solimpeks in Turkey, and an importer, but in the end we didn't use these panels, and did not use solar water heating. The basic reasons were the uncertainty of maintenance costs, impacting the return on investment, and increased complexity leading most seriously to an ugly roof. In the case of the PV/T panels, which deliver both electricity and hot water, we would not have been able to connect them to the grid and sell the electricity, since the panels were not licensed in Japan.
Photovoltaics have been pretty much a no-brainer to install on new builds here, as in many cases they have no moving parts and will still be producing electricity when the house is knocked down. Photovoltaics can literally be plugged in.
Solar thermal, on the other hand, needs to store heat when there is more than you need, supplement it when there is less than you need, and deliver the heat where you need it.
The most obvious worry is what to do when there is not enough sun, but the three big threats for solar thermal are freezing, overheating and hygiene. Overheating seems the most serious problem. When it gets hot nothing magical happens; water turns to steam, volume expands and pressure goes up. Watt solved similar problems for the steam engine in the nineteenth century, so it should be fairly predictable in a solar collector.
Although Watt is usually credited with having invented the steam engine, like many great inventors, he did not create it out of thin air, but innovated existing ideas to make them practical. Before Watt, the steam was thrown away after pushing the piston around in the cylinder. This was fine for a static engine by a big pool of water, and in the case of mining there were often big pools of water to get rid of. In the case of textile mills, irrigation was already in place to get water there. These engines wouldn't go very far without running out of water. Watt great innovation was to turn the steam back into water, and keep it in the system. When you are collecting heat from the sun, if you cannot guarantee that the heat will be taken away from the water, then it will inevitably turn to steam, and it needs a system like the one Watt invented all those years ago. It's not rocket science—that was Stevenson—but steam in your system could be explosive, and if any air bubbles get in there after it condenses, water may stop flowing.
Unless your system can handle overheating, you either need a much smaller system than your needs, or you need to be able to store the heat for a rainy day. Heat storage strategies are a rabbit hole you can get lost in. Adding a supplementary heater will increase the cost, but it shouldn't be so complicated to add an electrical immersion heater, fixed up to a thermostat to switch on when the tank is too cold, and a timer to check that it's night time. Electricity is an expensive way to heat, but if you are getting most of your heat from the sun, and you are rarely using this heater, then it's not a big issue. In Japan systems are typically undersized, so and electrical heater would become very expensive, and effectively the solar panels are supplementing a regular heater system, which you need to spend more on to increase the efficiency.
There are seven different ways of getting heat from the panels into your hot water tank (according to Rob Harlan interviewed on Back Woods Home). Drain back systems are perhaps the most simple. Water is sent up to the panels when the sun comes out. When the water gets hot enough, or at the end of the day, it is all sent into a tank in the house to be used. No water is left out in the cold, so there is no chance of freezing. An alternative is a continuous circuit, either open-loop heating water directly, or closed-loop using a refrigerant that will transfer heat into water in the tank. Running domestic water through a solar panel is probably bad news, since there are risks of freezing, and of water standing at a lukewarm temperature that is ideal for legionnaires disease. The continuous circuit should probably have a closed loop, which will be slightly less efficient. Since you are not using the water directly, you can use a coolant instead, and make sure it has a freezing point below your minimum temperature.
The solar collectors can either be flat panels or vacuum tubes. An interesting twist on the storage problem is to add phase change materials into vacuum tubes, so heat is stored by melting one or two materials with a high melting point. Water can be passed through the panels any time, and it will be heated up from the phase change materials. There used to be commercial models here on Made in China,com, but all Google will find me now are research papers and patents.
References
Papadimitratosa, A., Sobhansarbandib, S., Pozdina, V., Zakhidovc, A. & Hassanipour, F. (2016).
Evacuated tube solar collectors integrated with phase change materials. Solar Energy, 129, 10-19. Available from: https://www.researchgate.net/publication/294111899_Evacuated_tube_solar_collectors_integrated_with_phase_change_materials [accessed Apr 21, 2017].
