Background story. This thread of the story starts when we bought a woodburning stove from Mr Roger Kirk of Individual Fires and Stoves, back in late 2009.  It was one of the very first things we did, and it was just as well we did so, because we were then hit by the worst winter in 25 years, in a house with no other heating, apart from a very elderly coal-fired range in the kitchen.  As recorded at the time in this blog entry “Cold“.

Mr Kirk must have been a good salesman, because he also persuaded us to buy a Wamsler 1100 cooking stove with boiler, and an Akvaterm 1400EK accumulator tank.

The general concept of the accumulator tank is explained reasonably well here (accumulation tank basics).  In a normal hot water cylinder, the bulk of the water in the cylinder is the water that comes out of the taps.  Heat from the boiler is transferred to the water in the tank via a coil inside the boiler.  Water heated by the boiler flows through the coil and thus transfers its heat to the main water in the cylinder.  The accumulator tank works the opposite way round.  The boiler heats the bulk of the water in the cylinder, and when you want some hot water, it is created from the normal cold supply by passing the water through a heat exchanger in the tank.  The advantage is that all the water – both hot and cold – throughout the house is driven directly from the mains supply, so that (a) you get good pressure delivery of hot water, and (b) no  cold water tank is required, only an expansion tank about the size of a lavatory cistern.

At the moment neither the Wamsler nor the accumulator are actually working.  The Wamsler can’t be lit until it has water flowing through it, and this means that we have to have the accumulator tank in and working.  But the accumulator tank can only be inserted into the house via a window – it’s too big too go up the stairs.  And the convenient time to stuff it through the window is when we have the windows replaced – an event that is scheduled to take place around the end of July.

The final stage in this story is the immersion heaters.  These were something of an afterthought, but if they work as planned they will kill two birds with one stone and solve an issue with the photovoltaics.

The basic “problem” with being off-grid (for electricity) is that we cannot export our surplus to the grid.  Of course we can put it into our batteries, until they become full, at which point we have to just throw it away.  The way this works is that if the batteries cannot accept the surplus power from the photovoltaics, then this information is communicated to the PV system (I will do a separate blog page at some point explaining how this works) and the PV system turns itself off completely.  It then waits for about 3.5 minutes and tries again.   During the time that the PV is off we are not earning the feed-in tariffs (FITs), so we are “losing” money (compared to people who are on the grid and therefore don’t have this problem).

In fact the problem is a bit more subtle than I have described above.  The amount of charge that the batteries can accept goes down as the batteries get nearer to being full.  For example if the batteries are 80% charged then they can accept power at the rate of 3kW, whereas when they are 90% charged they can only accept power at the rate of 1kW (these are not precise figures, but they illustrate the principles involved).  The only way you could get more than 1kW into my system when it is 90% charged would be by raising the charging voltage above the safe level.  Therefore, if at some point the batteries are at 90% and the supply (from the PV) exceeds the demand (from the house) by more than 1kW, the system has to disconnect the PV completely, notwithstanding the fact that the batteries are not entirely full.  So not only do we have to throw away PV input, we have to do so even though our batteries are not full.  And paradoxically, we can charge the batteries more fully on a cloudy day than on a sunny one.

The solution to this problem is something called the Load Controller, a piece of equipment which is being built by Energy Solutions, who also supplied the PV system.  The Load Controller will switch in extra load when the above situation is detected.  In fact it switches in extra load in six steps.  If this works as planned, then the power being pushed into the batteries should always be a little less than what they can accept.

Where all this fits together is that the surplus loads will in fact be the electric immersion heaters in the accumulator tank.  So basically we will be using the surplus electricity to make lots of lovely hot water, but only when (and to the extent that) we have a surplus after the normal loads and after giving the batteries what they want.

Energy Solutions have delivered and connected up the Load Controller system, but they haven’t yet made it work.  I’m confident that they will, although I think that they may have under-estimated the difficulties a bit (they did say that they planned to visit this week with modifications to a couple of parts of the Load Controller system; it is now Friday, so this obviously isn’t going to happen).

My main residual doubt is the reaction time of the system.  The PV output can vary fast if the sun comes out from behind a cloud, and the load from the house can go up or down suddenly and substantially (if a kettle comes to the boil, for example).  If the surplus power is too much for the batteries, the Inverter/Charger will switch the PV off in 3 seconds.  So (it seems to me) the Load Controller will have to have a reaction time of less than 3 seconds (the 3 second parameter is configurable so I suppose one could increase it, but this might not be very good for the batteries).

Another thing I have been a bit worried about is whether the relays in the Load Controller are big enough.  However, I’ve looked at them closely and I see that they are Omron model number G7L-2A-T and the data sheet here http://www.omron.com/ecb/products/pry/123/g7l.html confirms that they can handle 25A at 250VDC, which would be over 6kW.  I don’t think the immersion heaters will be bigger than this.  I also see that they can handle 1800 switches per hour, which would be one every two seconds.  I just hope they aren’t noisy.  The spec web page has lots of data but is (ironically) silent on the subject of noise.

Nevertheless, I would like to know more about the immersion heaters in the Akvaterm Akvantti.  When we ordered it from Mr Kirk of Individual Fires and Stoves we ordered it, rather vaguely, “with immersion heaters”.  The Akvantti page on the Akvaterm web site says that it can be ordered with 3 “resistor units” so I assume that this is what we’ve got but I’m not quite sure what this means.  Is it one physical heater with three connections for different power levels?  Is it three different heaters one going 1/3 of the way down the tank, one going 2/3 down and one going the whole height?  What is their power rating?  I phoned Mr Kirk two weeks ago (while Energy Solutions were here installing the Load Controller) asking for this info; he said he would get back to me and he hasn’t.

So I chased Mr Kirk with another phone call this morning.  So far no reply.