Hot Water

The heating of hot water is another important need for heat energy. Here there is the possibility to store excess electrical energy as low grade heat energy. This is done in insulated water tanks. The storing of this energy acts to stabilise the system, absorbing excess energy for use when required.


The first step is to identify the hot water energy demand which is to be met. The relevant variables for this calculation fare as stated in the table below. It is assumed that 100 people will to consume 50 litres of heated water which is heated up to 60 °C.

Variable definition table for hot water

Hot water storage tanks (6)

To calculate an estimate of the energy required for each month of the year, the following formula was used:



Ereq = [ C · (Ts – Ti) · 30 / 3600 + (ql / 12) ] · Vc · pop






where 'n' denotes the number of days in a month.


This formula gives the energy required to increase the temperature of the water to the set-point defined and average heat losses. For the sake of simplicity the heat loss was considered homogeneous throughout the year. While not necessarily realistic it paints an accurate enough picture of the heat losses and helps the evaluation of the hot water strategy.


Using the suggested water inlet temperature as defined by SAC 2012 [2] the monthly energy requirements were calculated as demonstrated on the “Energy requirements for monthly hot water demand" table below:

Energy requirements for monthly hot water demand table. This table shows the water inlet temperature for the hot water tank and the energy necessary to supply 50 L of water at 60°C for a 100 people.


With these figures it is indicated that a total of 115,258 kWh per year is sufficient to meet Eigg’s hot water energy demand that is required for a 100 people to consume 50 litres of heated water up to 60 °C.

excess energy


Excess energy, or surplus energy, is a term to define a portion of the supplied energy that there is no immediate demand for. In this case, there is "excess energy" in the network after the electrical demand is met. Unlike on a macrogrid system, an autonomous system such as Eigg does not have the ability to feed back the energy to the national grid or to sell it on.


While a potentially destructive phenomenon it can be harnessed to produce positive results.


The graph to the right shows the monthly hot water demand with the excess energy derived from the simulation made in HOMER. The graph shows a large amount of excess energy compared to the hot water demand, and so it is likely that using excess energy to store hot water is feasible.



Average monthly demand and excess energy

frequency controlled switch

A frequency controlled switch is a consumer device which can measure the frequency of the network and, based on this, power an appliance. The Amber Control [7] sold through Wind & Sun [5] will be used as an example to explain its function. This device measures the frequency. It must be installed between the power supply and the load that is to be activated when there is an excess of energy. As shown on the connection diagram.


During installation the frequency set-point must be defined to inform the device when to switch-on (close the circuit). This setting will also determine the switch-off (open the circuit) frequency, or the value at which the device will be powered off. This way, when there is an under-frequency (lack of supply) scenario the device will decouple the load from the grid.


One possible risk of using frequency dependent devices is that the frequency might oscillate, and having the devices being turned on and off destabilising the grid even further. This can be solved by using a time delay that waits some minutes (or seconds) before allowing the device to be powered on or off. The device is only switched on in instances that there is an assurance of excess energy.


This way the controller ensures that the over-frequency scenario is not just a small disturbance. The device is only switched on in instances that there is an assured excess energy.


Another risk arises from the possibility that due to the physical properties of the network there will be one or a group of users that will benefit more frequently than others. This would happen if their devices sensed the disturbance before others. On top of that there is an issue of having power peaks from multiples devices being turned on at the same instant.


The way to solve this is though randomised delays. The delay chart shown  to the right illustrates a possible solution. At the time the controller performs an action (switches on or off) there is a randomised  delay that equalises the probability among the many users and creates a ladder like demand increase. This way both risks of power peaks and privileging specific individuals is greatly minimized if not nullified. This randomisation strategy is similar to what some types of communication protocols do. They randomise some waiting periods to avoid sending data packages at the same time.


There are two important limitations to this controller:


One is that the frequency sensing device's output is limited to 13 A.


The other relates to the load type that can be connected. Due to the potential frequency switching, appliances that are not made to withstand (e.g. pumps) it cannot be used in conjunction with the controller. Therefore, resistive heaters are the ideal load type, such as immersion heaters in hot water tanks.


Frequency Controlled Switch (5)

Frequency Controlled Switch Connection Diagram [5]. Function block (f>) reads the grid’s frequency and closes the contact when it is above the set-point allowing energy to flow from the supply to the load.

Frequency Controlled Switch Delay Chart (5)

excess energy hierarchy


In a system that takes advantage of the excess energy produced, it becomes necessary to define a strategy of how, or in which order the many subsystems will be switched on. Therefore, a hierarchy must be established.


For this reason it is necessary to analyse all frequency dependent devices and prioritise them. In practice this involves either changing the frequency set-point of the device, or changing the delay incurred after a given set-point is reached.


Devices higher in the hierarchy (with higher priority) will be configured with lower frequency set-points and/or lower delays before activation. Consequently, devices lower in the hierarchy will have higher frequency set-points and/or higher delays before activation.


The following hierarchy is suggested for the upgraded system. Take note that the only new addition to the hierarchy (after the suggested additions to the network take place) is the immersion heater for hot water. The batteries are the island’s de facto storage method and must be prioritised.



Excess Energy Hierarchy

On this hierarchy the community buildings (e.g. church) were deemed more important then the residential water heating because they play an important role in Eigg’s community dynamic.


Lastly, there is the “shut-down strategy”. Since allowing excessive over-frequency can damage the devices within the grid it becomes necessary to limit production if the excess energy strategies fails to stabilize the network. Therefore, as a last resort, partial shut-downs must take place. This means decoupling arrays of PV panels or/and decoupling wind turbines. The wind farm has an additional defence line with open air heaters that can also limit output.


Naturally, this last item requires a slightly more complex algorithm, since the outputs will vary according to the weather and different action will be more or less effective in different conditions. Every grid that has frequency dependent devices will have different priorities and must be analysed with its specific socio-political inclination.

Under frequency


Although not relevant for the excess energy, or over-frequency, there is also the issue of under-frequency that happens as a consequence of greater demand than supply. Currently this is remedied with the battery bank and the back-up generators.


Additional strategies might become necessary with the increase in risk offered by a greater potential demand.


© University of Strathclyde | TEC Eigg | Sustainable Engineering 2016