Year end results for 2021/22 water year

It's been a dry year in 2022 and I've been itching to see what my year-end graphs look like. Remember, the year for me ends on September 30th, - a 12 month period I call a 'water year'.

The bold black line in each graph represents the data from the past year:

1. Daily energy and power output

The solid stretch of generation that usually happens from November through to March never happened this year; only in early January and late February / early March was there enough flow to see the turbine generating at its maximum level of 900+ W; with rainfall being so poor, ground water was depleted for the rest of the year, and added to that, generation didn't see the usual pick up from a 'June monsoon', as happens in most years.

2. Cumulative energy (kWh)

With the year being so dry, unsurprisingly the total of kWh generated was a meagre 3599; only one year previously has had such a low rainfall, 2016/17, and in that year 2773 kWh was generated; the fact that more generation resulted in this year when a similar amount of rain fell, is because I've learnt since 2016/17 how to obtain greater efficiency from the system. 

3. Power duration curve

The dire level of generation is nicely revealed in this power duration curve; it shows that maximum power was generated for only 50 days in the year, and for more than half of the days in the year, generation languished at 200 W or less; remarkably however, generation did continue for the full 365 days, and that is a happy consequence of the geology of the hillside which sees groundwater being not too adversely affected by lack of rainfall, - if the lack of rainfall doesn't go on for too long.

4. Rainfall vs energy generated for the past 9 years

Despite groundwater keeping the numbers sweet by maintaining a reserve sufficient to keep the turbine running in drought times, this graph shows there is a clear relationship between rainfall and generation; from 2016 onwards that relationship is noticeably consistent; this consistency is not seen in the early years because in those years I was missing out on how much generation I could capture, - mostly from ignorance of how to operate the system optimally.

Conclusion

Every farmer has a bad year from time to time but always looks to the coming year to be better; that's how I'm viewing these results; already rainfall in September is higher than it usually is so it looks as if the next 'water year' should get off to a good start. With energy costs as they are, a good year would be nice.

Battery back-up for grid outage

 The electricity grid where I live is well maintained; grid outages are rare and when they do happen they are usually brief.  But like everybody we have a freezer and a central heating pump so the consequences of a power failure are not without reckoning: the possibility of food lost or the inability to keep warm in the middle of winter.

A more subtle pressure to consider a particular type of back-up comes from having a Powerspout contributing to house electricity 24 hours a day; though the turbine will remain operational during an outage, it cannot deliver its output to the house unless its inverter is working, and when the grid goes down, the rules are that a grid connected inverter must stop working; how frustrating is that ! - to have hydro power available and yet not be able to make use of it !

So the subtle pressure is to opt for a particular type of back-up which would allow the Powerspout to continue supplying the house.

The differing factors to be considered make the decision a delicate matter; it's delicate because the added expense and complexity are considerable for a system that would see the Powerspout continue to supply the house.

As often happens, the factors which sway a decision are in the detail; not all grid outage back-up systems are the same and household electricity safety requirements rule out some configurations which might otherwise be possibilities. 

So what are the details of the different options, and what operational technicalities have to be thought about?

Back-up by way of a secure outlet.

This is the simplest option because it doesn't require the house distribution system to be islanded (ie: isolated) from the utility grid, - but in not requiring islanding the option doesn't allow the Powerspout to supply the house in an outage; it is also simple in that it is not automated and would require manual switching, both for switching on and off.

The way it works is that power is supplied from the battery, via the inverter, to a single socket which is not a socket like any others in the house distribution system. It stands alone outside from any house ring main and has no connection to the house fuse board (consumer unit). 

Into this socket, either directly or via an extension cable, can be plugged those loads which are deemed essential. The maximum power that can be drawn is limited; for SMA inverters it is 3.68 kW. How long power will be supplied depends on the size and state of charge of the battery, as well as the size of the load connected.

The position where the secure outlet is located is clearly of importance in order that loads deemed to need power can be connected to it.

Manufacturers of battery charger / inverters offering a secure outlet include SMA in their SunnyBoy Storage 3.7/5.0/6.0 range.

Back-up by way of islanding.

This is the option which is the attractive one because it allows the Powerspout to continue to contribute during an outage.

To island a house means to completely separate it from the utility grid; in a typical house supplied by single phase power, this entails interrupting both the live and neutral conductors where they first come into the house; the isolation has to take place between the utility company's meter and the house consumer unit. 

For safety reasons, because the neutral conductor is always made to be at zero potential (or at least near zero) by being earthed outside of the house by an earth connection constructed by the electricity company, interruption of the neutral as it enters the house will mean the neutral conductor loses its earth connection and will need to be re-connected to earth on the house side of the dis-connect; this has to happen at the same moment it is dis-connected from its outside earthing and at the same moment the live conductor is disconnected.

The new earthing of the neutral can be to the earth rod the house will already have, or it can be to a new and separate earth rod. Details about this can be found in this very informative on-line document.

The switching necessary to accomplish everything is achieved easily enough, and automatically, using relay contacts housed in a secure box, and it is best if this is sited near the point where the utility supply first comes to the house. 

Many manufacturers offer this type of hardware, for example Tesla, Sonnen and SMA. 

Whether power from the battery supports the whole house or only some circuits within the house is a question which can only be answered by estimating the load for each of these two options and weighing that against the ability of the battery and inverter to meet it; 

Hybrid inverters are not all equal in how much power they can output and not all batteries are equal in the maximum power they can deliver, so careful consideration needs to be given to ensure that battery and inverter have matching power output ratings sufficient to meet the load of either the whole house, if that is chosen, or just selected circuits such as lights / freezer / central heating pump.

A point of detail: "What would happen if..."

Attention to the detail of how my system would operate if I choose the option of 'islanding' and continuing to have the Powerspout contribute to the islanded grid is the question "what would happen if too much power is being generated into the islanded grid"; it is a question to which as yet I'm not sure of an answer !

The situation could arise (a worst case scenario which might never actually arise but which needs to be catered for) where there is a grid outage, the house is islanded and being powered by the battery, nobody is at home so house consumption is very low, say < 300 W, the state of charge of the battery is 100% and so can receive no further energy input, and yet both solar and hydro 'generators' are feeding into the islanded grid their outputs, of perhaps over 2 kW; a surplus of 1700 W exists in the system.

Where does the surplus power go ? what happens to ac voltage and frequency in a situation like this ? what can be done to manage the situation ?

These are all questions which still need answers and solutions; until I find them, for me back-up by secure outlet rather than islanding may prove to be the safest, maybe even the only way to go.

Any suggestions are welcome !

Figuring battery size

There are a host of considerations to take account of when deciding how many kWh of energy storage are needed for a domestic battery storage system; if I choose too big a battery, the surplus available from self-generation may not be enough to fully charge it, and additionally it will be needlessly expensive; choose too small and it may not have enough storage to meet the functionality I want.

So an early decision in the journey to installing battery storage is to be sure what I want. 

There is a divide based on complexity between two options, with a further divide in the second option; the  technically simplest option is to store energy so it can be used at times of peak demand, thereby reducing drawing from the grid at those times; the more complex alternative is to have that capability plus the capability to back up the house in the event of a grid outage.

The further divide in this latter alternative is whether it is 'whole house back-up' or 'secure outlet back-up'; the distinction between these is important because 'whole house back-up' will allow the Powerspout to continue to feed its output into the house grid when the house is disconnected from the utility grid, whereas a 'secure outlet back-up' only provides a single, switched, power outlet to which emergency loads can be plugged in. In a grid outage, being able to make use of the Powerspout's output is obviously very desirable, so if at all possible, that is what I would like to go for. 

Grid outage back-up of either type requires the battery to be able to store more kWh, and questions immediately flag up - how much energy does my house use, when are the peak times of usage, and how great is the power drawn at times of peak demand. 

With these 'demand-side' unknowns in one pan of the scales and 'supply-side' unknowns in the other, - supply-side questions being: how much surplus energy does my installation produce, and how does this surplus differ between summer and winter months, - a delicate weighing up operation needs to be undertaken to balance the one with the other.

The whole matter becomes rather complicated, and working through to a properly reasoned conclusion can only be done by having data, - the more data the better, and data covering both 'demand-side' and 'supply-side'.

Below are graphs to show my 'supply side' data; they document my hydro and solar generation over the past 6 years: 

The bars in the main graph show energy generated each day of an 'average' year; such a year is a hypothetical one concocted from the averaged data from the 6 years from 2016 to 2021; the individual years are shown in the smaller graphs above and below the main plot, to make the point that real years are never quite the same as an 'average' year.

The graph shows that 'typically' the lowest level of self generation happens in September / October and is around 10 kWh / day. Highest generation peaks at 30 to 35 kWh / day in March / May when solar output is picking up and yet hydro output is still good due to groundwater having been replenished by rain which fell in the previous December to March period.

So much for my 'supply-side' data, which being accumulated from 6 years of record keeping is about as reliable as one can hope for; what about 'demand side' data? 

Below left is a plot of energy consumption from the grid in a month; for our house, all months, winter or summer, look the same; 

note it is not a complete picture of house consumption because it does not include energy consumed which is self generated; to get that full picture, the number of kWh consumed which are self-generated need to be added, and that number comes to, I have established, an additional 10 kWh / day; the graph below right shows this daily addition (in blue), so the graph now represents the complete demand-side picture over one month:

It will be clear from the above graphs that on days (in truth, nights, - so use is made of cheap rate, night-time energy) when my Nissan Leaf is charged, the grid energy taken is in a different league to other days, 20 kWh compared to 3 - 6 kWh; this is important to keep in mind when deciding on battery size; if I rule out any hope of storing sufficient kWh to charge the car from energy stored in the domestic battery, then the battery only needs to be big enough to cover each day's domestic consumption taken from the grid; a battery of 5 - 6 kWh capacity would be plenty large enough for that, whilst to charge the car from stored energy would require a 20 kWh battery.

I have established that the demand each month is much the same in one month to the next, regardless of the time of year, so the pattern of monthly demand can be scaled up to a year pattern simply by repeating it 12 times; it then becomes possible to see how data for 'supply side' and 'demand side' relate to each other if the one is overlaid on the other; as the following graph shows, where blue bars are not overlain by red bars, that represents surplus generation which is available for storing in a battery:

Interpreting the graph further, one can conclude that for the 9 months between December and August, the excess of generation over consumption which is indicated by blue bars not being covered by red bars, amounts to at least 5 kWh/day, and this should make it possible to have enough surplus energy to fully drain and re-charge a 5 kWh battery each day throughout that period; within that 9 month period, there are 7 months from January to July when the margin of surplus is much more generous, at 10 to 15 kWh /day.

Here however, a caveat needs to be remembered: these calculations are founded on 'averaged' data taken over a 6 year period; if the actual data of each of those 6 years, given above, is scrutinised, it will be clear that in any given year, the 'supply side' data is quite variable; the import of this is that in some years things will not work out as anticipated.

...but a battery of 5 kWh seems to be the size for me and it is the size I am contemplating getting; it will be big enough to cover what we usually take from the utility grid each day and it should be enough to cover 'grid outage back-up', whether I go for the option of 'whole house back-up' or 'secure outlet back up'.

Next time, we'll look into those two options.

Towards battery storage

For quite a long time, I have been thinking about adding battery storage to my set up; the recent eye-watering rise in the cost of grid-purchased electricity here in the UK is finally driving me from thinking about it to acting on it.

It is not a simple matter however; choices about battery size, whether to include grid outage back-up capability, where to physically locate the battery and charger/inverter, - all these and much more have to be worked through, and inevitably compromises identified and accepted.

I have persuaded myself that the case for battery storage is unassailable. My reasoning goes like this: my Powerspout is good at delivering energy over long periods at low power levels; batteries are good at delivering energy over short periods at high power levels; by combining the two, advantage can be taken of their respective strengths.

In effect, for a grid connected Powerspout like mine, a battery can be used to address the limited capability of the Powerspout to deliver sufficient power when domestic appliances like a kettle or cooker are turned on; the battery opens the door to saving on the energy which would, without a battery, be drawn from the grid.

In the complicated journey of working out how to configure a battery system, a major 'road block' has been how to integrate the existing diversion device I have, a Solarcache; it harvests surplus energy and diverts it to 3 different heating loads in the house, always maintaining a bare minimum of power to trickle back to the grid; it has been a very valuable component of our system which we have come to rely on as our sole source of domestic hot water in the summer months; I am keen not to have to jettison it with the introduction of battery storage.

The 'road block' was because my Solarcache was an early model which lacked the later firmware which enabled Solarcache to be adjusted so the battery management system (BMS) got first priority of surplus power; 

Happily, yesterday that block was removed; I discovered I had a spare Solarcache with the required firmware update; below is the screen from it which shows how the Solarcache's settings can be adjusted to give the BMS priority.

The settings allow for a threshold level of Watts flowing back to the grid, and a delay time during which that power needs to flow, before Solarcache will start to divert power to the loads connected to it. Discrimination between the settings of the BMS and Solarcache can thus be achieved to ensure the BMS gets first priority.

I was really excited at discovering I could keep Solarcache alongside the battery system; having removed that particular 'block', I now feel free to progress the battery scheme further.

I'll post more in due course.