Solar PV installation
I’ve not got a history of being especially green, yet I’ve just done something (apparently) rather “eco”. I’ve sold the M5, and rather than replacing it I have invested those funds along with some savings into a fairly noticeable set of solar panels for the house roof. There are three good reasons to do this:
1) It is ecologically preferable. Just 3 years ago we were a 3 car family, totalling nearly 10 litres and 700hp of motoring genius. There were all sorts of excursions such as trips to the Nűrburgring etc, but since then we’ve moved to larger house and become more invested in our local jobs and making a nest for our imminent family. It makes sense to try and put something back in. That said though to be honest I’d be quick to buy another large V8 if I thought I’d have the time and money to drag it sideways around circuits all weekend. Hmm.
2) More convincingly, there are substantial financial incentives currently in place for solar PV installations. Install after March 2012 and these incentives start fading annually, but install before April 2012 and there’s a tasty government backed, indexed linked “Feed in Tariff” (FiT) that should provide ROI for 25 years.
3) Most interestingly, this is some geeky apparatus. I’ve got some rather cool kit and can produce lots of stats and graphs. Fruity!
So anyway, all this started at the beginning of the year when the cost of heating water here at home was starting to get to me. At that point I was interested in solar thermal (and I still am), but my research from that point forward suggested that it would make sense to get solar PV first (to get in before the April 2012 deadline), and worry about solar thermal once I’d recovered from that. I’m fortunate that my friend Henry’s (of three peaks fame) father is the project leader for Rushcliffe Solar, and was able to point me in the right direction on many of these matters.
Still it wasn’t until July, when amidst British Gas’s announcement that they were going to significantly increase energy costs from mid-August, that I took a slightly different route to work and noticed a couple of nearby homes had recently had some panels fitted.
The above photo neatly highlights both the different styles of panels available, and the restrictions that roof size can make on an installation. Personally, I don’t like those silver trimmed efforts on the right, and I really like those black panels on the left. Also, the house on the right was clearly restricted to 12 panels by the size of their roof. With a good panel yielding around 245W, that installation on the right can at best produce 2,940W, whereas the 16 panels on the left could produce 3,920W. Systems under 4kW are eligible for the most favourable FiT, so I wondered if the installation on the left was exactly what I was looking for.
That evening on the way home from work I decided to visit the house on the left and see if they wouldn’t mind giving me some details. I was lucky enough to find the owners extremely enthusiastic and more than happy to pass on as much information as I could digest. Those are 245W Solar World PV modules, and they were installed by a company called Eco Fusion. Later that week I called up and soon had a site survey booked in.
Installation size and pre-requisites
I also realised that I’d probably have to do something about my house’s original 1975 fuse box:
So I booked an electrician from Diane’s business group to come in and replace that. As it happens he arrived on the same day as the PV survey, which handily allowed him to confirm their requirements. Once he had finished the fuse cupboard looked rather more satisfactory:
Anyway, back to the PV survey, which is where things got interesting. I’d roughly measured the roof, and had feared that we would only be able to have a 2×7 panel array up there. The survey guy explained that in order to protect against wind and to ensure running water doesn’t surge over the gutter, it’s necessary to ensure that the panels are at least 200mm from any roof edge. Consequently, we could only fit a 2×6 panel array on the main roof. He therefore suggested that we put 4 panels on the extension roof. Let’s put this into context with a photograph:
Yeah, it’s an ugly house. At least there’s no beauty to lose by coating it in solar technology! The house faces a little east of south, so the extension roof would be okay in the morning but would be shaded by the main house roof later in the day – not ideal. Still, the surveyor left and I mulled things over while I waited to hear back from them. I went out for a run and, having memorized the roof and panel dimensions, pondered the situation.
Suddenly I had a Eureka moment – what if the panels could be mounted in landscape rather than portrait? I knew the panels were 1001mm wide and a 20mm gap is required between each panel. So for panels + gaps + run-off the roof would need to be (4×1001) + (3×20) + (2×200) = 4464mm from ridge to gutter. There was only one way to be sure, and that was to get up there and measure it with a tape measure, and I was pleased to note it could be managed with a few mm to spare! So I called the shop and put this suggestion to them, and after and hour or so they came back to me and said it would work out.
I should note here that since then I’ve done some reading around as to what the minimum recommended distance from the edge is, and I’ve seen a variety of responses from 200mm to 600mm. So it would seem that my supplier (perhaps predictably) was willing to (literally!) sail a little closer to the wind than most. Still, there’s plenty of room from the left and right, and the panels can’t be seen over the ridge at all from the rear of the house (the danger point for wind-loading). We’ve had heavy rain and haven’t suffered with water flowing off the panels over the guttering at the front either.
I had been quoted for 16x245W Solar World SW245 Mono panels (3,920W), and a Fronius TL 3.6 inverter (max input 3,800W, max output 3,600W). Obviously I questioned this – there’s the potential for the panels to overdrive this inverter. I was assured this was standard practice and that the inverter could cope with it just fine. I’ve pondered this and concluded that peak output is extremely unlikely to be delivered by the panels for long, and these inverters are designed to sense overload and shut down. If that happens, or if it’s damaged, I’ll hold Eco-Fusion accountable, but it does seem like a sensible approach.
The panels do degrade with time. Their specification sheet (linked to above) states they carry warranty to produce at least 90% of rated power after 10 years and at least 80% of rated power after 25 years. After 8% degradation they will match the inverter.
So, before the installation we had some scaffold arrive:
Now, that evening I decided to make use of this. Note the yellow tinge on the roof – there was a huge amount of moss and general roof garden up there. So I decided to get up there and blast it all off – with some kind assistance from Robin the second and Ben the first. The roof was cleaned by nightfall.
The following morning the house roof looked rather different.
It was suggested as a result of this process being documented on Facebook that we had removed a protective layer on the tiles by pressure washing them. Certainly there was a fairly radical change in colour which suggested some kind of staining had been removed. I inspected them carefully and compared the way they handled water being poured over them with the tiles on the back of the house (that we didn’t clean). There was no difference. I consulted an experienced roofer and trusted friend on the matter, and he seemed to think there was no issue. Still, to be safe I gave the roof a couple of coats of transparent tile seal.
Installation day itself was miserable – overcast with pretty constant drizzle. Did this stop the installation? Hell no!
At about 10am a 3 man team arrived – two for the roof and an electrician. There was work to be done in the electrical cupboard, and the inverter was to be installed in the loft.
On the roof the rails went up first, and then the panels were fixed in place.
In the loft:
In the electrical cupboard:
By 3pm they were finished – very impressive given what they had to achieve and the weather. The panels are connected into two groups of eight and then feed into the black (DC) side of the inverter. From the inverter a cable runs to the roof edge and then subtly down the side of the house and into the electrical cupboard where the distribution board is driven.
The installation came with a Wattson energy meter. This is a fantastic piece of kit, especially if it is installed correctly, which it almost was. It comes with a couple of sensors which clip around live cables, and by measuring the EMF it knows how much current is flowing in those cables. Sadly though, it doesn’t know which way the current is flowing. It is bright enough to understand that one of the sensors may be for generated input power, which makes it ideal for the job.
It was installed with the generator clip in the right place, but the usage clip connected to our main feed from the grid. Under circumstances where we were generating less than we were using it worked a treat, but as it always assumed that the current sensed between the house and the grid was inbound (not export), when we generated more than we used it got things wrong. I fixed this by installing that clip within the distribution board on the feed to the RCDs (post-generator input), but that meant that it knew nothing about our night-time only circuit which powers things like our storage and immersion heaters. Fortunately the Wattson can have additional inputs, so I invested in another clip and now the system works perfectly.
The Wattson logs all data and I can access it via USB using my laptop and the inevitably named software “Holmes”.
I’ve also got an Environ Current Cost meter and internet bridge. I’ve connected this in such a way that it measures only our generated power and it logs this every 15 minutes on the internet. This means that from anywhere in the world, I can at any time see how much energy the house is generating. Which, as a geek, makes me very happy indeed. So for instance, here’s our generation stats so far today:
Isn’t that wonderful! The blue line and scale on the right are the temperature in the office here at home. Here are the two meters next to each other earlier today:
That’s 3.17kW of energy being harvested from the roof on the left, and an export (hence the minus symbol) of 2.7kW on the right. From that we can deduce that the house is using a little under 500W, which is about right (fridge, freezer, server, laptop, computer, 2 screens, modem, router, switches etc). I have even seen generation figures of just over 3.6kW which is nice and proves that everything is certainly working as it should. To drive all this the electrical cupboard has a couple of wireless transmitters and a total of 4 sensor clips:
So as you can see from the generation meter above, in the ten days since this system was brought online we’ve generated 146kWh. At the current FiT rate of 43.3ppkWh, that’s £63.22 I’m owed. Put into perspective against the massive capital outlay required to install this system that’s not much, but it’ll continue to grow with no additional effort. There are also other revenues. Firstly, my supplier pays 3.1ppkWh exported. As they can’t tell exactly what was exported they assume 50% (which probably suits them more than me), but that’s another £2.26 off my next bill so far. If that is averaged over the quarterly cycle that’s a more meaningful £20 saving. More importantly though, our daytime usage from the grid has more than halved since these panels were installed, so especially in the summer months those bills will be drastically reduced.
As I’d expect from such a big financial outlay, this is a fantastic toy. However, compared with say a car (and I spent more on the 330d when I bought it), it should prove to be rather more profitable! As it happened it was installed on my 32nd birthday, and it is my hope that the capital expenditure will have been recouped before I am 40. I will post again some time around late October when I’ve got statistics for the system being active either side of equinox – hopefully from there I’ll be able to extrapolate with enough meaningful data to make a more accurate prediction. Until then, here are some photos of the completed system: