Facts

During summer, the solar production is so efficient that a lot of energy cannot even be consumed. It would be very useful to store this surplus of energy for later use, for example in winter when the days are so short. An electricity storage with a capacity of 500 KWh would be perfect. Unfortunately such a battery is currently not affordable for an individual. We have to find something else to use this energy...

Following chart captured on August 19th, 2014 depicts the situation:

The yellow area  in the background is the solar irradiance (available power). The green area is the energy used by the system. The surface of the green area is 13 KWh. The visible yellow area is the energy that has not been used by the system. The surface of the area is obviously more than 15 KWh.

An option would be to use this available energy to heat water for domestic use. Our sanitary hot water is produced by a natural gas boiler. If I insert an electrical hot water tank in series before the tank of the boiler, the gas boiler need not to heat the water anymore (provided solar power was sufficient to heat the water). So our natural gas bill will decrease thanks to solar electricity. 

That seems to be a great idea ! - Let's see if it is feasible and if it's worth the money.

Feasibility of Photovoltaic Hot Water production

A 4 person family uses up to 200 L of hot water per day. The electrical power of 200L tanks is generally 2 KW. That figure is compliant with my Inverter rating.

To heat up 200 L of water, one needs approximately 10 KWh. This is compliant with the surpluses of my PV system.

It will not be possible to manage the hot water tank simply with an ON-OFF switch. The available sun power will often less than 2 KW. So an ON-OFF scheme would lead to eventually drain the battery. That is not acceptable. I will  implement a triac dimmer to manage precisely the power drained by the hot water tank. This triac module will be connected to the PV System CAN bus. The microcontroller managing the hot water dimmer will be able to compute preciseley the available power thanks to the data available on the bus and set the triac firing angle accordingly. 

How much natural gas bill reduction can I expect ?
My calculations predict a 100 € yearly saving on the gas bill

How much does it cost ?
A tank costs  400 €
Pipes and accessories : 100 €
Electronic board + box: 200 €
Total cost: 700 €

Provided the gas price does not increase, the ROI will be of 7 years. That's OK, let's start the project.

Design

This project is fully designed now and is currently being tested with a dummy load. See my HackADay project.