November's solar generation stats

We have a 2kW 8 panel solar array (Panasonic HIT panels), mounted on a south-facing roof at approximately 35-40° angle from the horizontal, which were installed at the end of August 2013.

This is the results for October 2014 (in red) with last year's results (green), and distribution by average sunlight 1971-2010 (met office stats) in blue:

The generation for November 2013 was 59.03kWh. This year the panels have done 59.07kWh, up a tiny fraction on last year.

With 300 hours of daylight in November (sun up to sun down for all 30 days), the average kWh generated per hour of daylight in October 2014 was 0.20kW (rounded up) , exactly the same as last year.

October's solar generation stats

We have a 2kW 8 panel solar array (Panasonic HIT panels), mounted on a south-facing roof at approximately 35-40° angle from the horizontal, which were installed at the end of August 2013.

This is the results for October 2014 (in red) with last year's results (green), and distribution by average sunlight 1971-2010 (met office stats) in blue:

The generation for October 2013 was 92.06kWh. This year the panels have done 103.99kWh, 11.93kWh (13%) more than last year. 2014 is certainly a significantly sunnier autumn than 2013.

With 341 hours of daylight in October (sun up to sun down for all 30 days), the average kWh generated per hour of daylight in October 2014 was 0.30kW, compared with 0.27kW from last October.

September's solar generation stats

We have a 2kW 8 panel solar array (Panasonic HIT panels), mounted on a south-facing roof at approximately 35-40° angle from the horizontal, which were installed at the end of August 2013.

This is the results from the first month of year 2!

To calculate a revised forecast for 2014-15, I took the total actual generated amount for the year 2013-14 (1929kWh), reduced it by 1% (1909kWh) to account for the predicted degradation of generation from the panels (see previous blog entry) and then divided it up proportionally based on the actual monthly generation results from last year.  These values are shown in green as the forecast amount for this year in the chart below, with September's actual total included:

The forecast based on last year's generation is 158.61kWh for September, but we've actually pulled in 205.13kWh for September 2014, 29% higher than expected. The only explanation for this is that September this year was sunnier than September last year.

With 420 hours of daylight in September (sun up to sun down for all 30 days), the average kWh generated per hour of daylight this year was 0.49kW, compared with 0.37kW from last September.

The First Full Year - Statistics, Analysis and 20 Year Forecast

It's a long post this month - I hope you're sitting comfortably and have coffee to hand!

To conclude the year's measurements, Here's the graph for the year's generation from our array of solar panels including completed August. This is for our 2kW 8 panel solar array (Panasonic HIT panels), mounted on our south-facing roof at approximately 35-40° angle from the horizontal.

The panels generated 244.77 kWh for the whole of August, 52.68 kWh higher than the forecast amount of 192.09kWh.

Now we have have an entire year's results in, the change in monthly generation is easily observed. At 27.kWh for the whole month, December is the lowest-generating month, and at 288.60 kWh, July generated the most. That's just over 1000% difference between the maximum and minimum months.

The highest generating day was May 6th, with 13.7kWh. The 2nd, 12th and 13th of December are in joint bottom at 0.1kWh per day. That's a staggering 13,700% difference between the highest and lowest generating days.

Here is the completed annual sun-strength chart:

I've discussed this graph on previous posts so won't go into it again.


One of the major factors when measuring the benefit of having solar panels installed, is the amount of your solar generated electricity you actually use yourself.  The more of your own ‘home grown’ electricity you use, the less you have to buy from the grid, and the more money you save.  Naturally, solar panels generate electricity during daylight hours, and normally, during daylight hours, people are out at work and so the unoccupied house is not using (very much) electricity.

Measuring how much of your own home grown solar energy you are using yourself is not a straightforward task. On a standard solar installation (like ours) you have two electricity meters.  The first is the normal electricity meter that everyone has, which measures the amount of electricity you have bought from the grid.  With a solar installation, you get a second meter, which measures how much solar electricity your panels have generated.

The complicating factor is that your solar panels could be generating 2kW whilst you’re out at work, and the electrical appliances in the house are only using, say 0.5kW (fridge, freezer, electrical items on standby etc.). That means that the 1.5kW that you’re not using is being fed into the grid and supplied to somewhere it is needed.   There is no meter that shows how much of your solar energy you’ve used yourself, only the total amount generated.

From a purely selfish point of view, to get the most financial benefit out of your panels you want to be using as much of your solar generated electricity as possible.  We are fortunate in this regard in that I work from home, and I have an office with computers / printers / air con in use all day. Also, the dishwasher and washing machine are both regularly in use during the day too.

The inverter system is clever, in that it will use up all your solar generated electricity first, before buying more electricity in from the grid to ‘top up’.  One way of getting an idea about how much of the solar energy you’re using yourself is to look at the main electricity meter during the day whilst the sun is shining brightly.  If the main meter is still counting up (even only very slowly) then it means you are using all the solar energy you’re generating, and you are still topping up with electricity from the grid.  If the main meter has stopped counting up, it means you are not (at that precise moment) buying any electricity from the grid. Unless your household is using exactly the same amount of electricity as is being generated by the panels, you are now (probably) exporting to the grid, since you are not using all the solar electricity you’re generating. 

I have, at sporadic intervals during the day, during bright sunshine, had a look at our electricity meter, and it’s always been counting. Sometimes it only goes very *very* slowly, but it has never stopped counting on the occasions I’ve looked at it.   Now I may be a solar sad act geek,  but even I refuse to spend all day sat in front of the meter watching to see if it should ever stop counting.

See below a chart showing the average daily amount of electricity we have bought over the last two years. The early averages are spread over longer periods because I took infrequent meter readings.

Here are the above meter reads chart shown on a graph:

ELECTRICITY BOUGHT FROM THE GRID BEFORE PANELS WERE INSTALLED
1st September 2012 - 22nd August 2013
Total 355 days
8680kWh total bought from grid
Average of 24.45kWh per day.

ELECTRICITY BOUGHT FROM THE GRID AFTER PANELS WERE INSTALLED
22nd August 2013 - 15th August 2014
Total 358 days
5175kWh total bought from grid
Average of 14.45kWh per day.

This means, since installing the solar panels, we have bought, on average, exactly 10kWh less per day (24.45kWh - 14.45kWh) from the grid. I've checked the maths, it really is *exactly* 10kWh - to 2 decimal places! Over the same period, the solar panels have generated, 1929/365 = 5.28kWh per day.

I have made other energy saving measures over the last 12 months:
1) fitted low-energy lights throughout the office/house
2) our fridge/freezer packed in, and it has been replaced with a new A* energy certificated one (it’s turned on 24/7 so despite not using much electricity, it makes a significant saving over the year)
3) I have also been more vigilant (my sons would say dictatorial) in making sure computers/lights/TV are turned off when no-one is in the room.

This means that even if we are using *all* of the solar generated electricity ourselves, the above measures are also saving us 5kWh per day.  I don’t know whether these measures could possibly contribute more than 5kWh in savings per day, in which case we would not be using all the solar energy we generate.

There’s no way to determine retrospectively how much of the saving is due to the above measures, and how much of it is due to using our own solar-generated electricity; but I think I’m on fairly safe ground assuming that we are in fact using all of the solar energy we generate.  This is the best possible situation to be in, since we are saving the maximum amount of money on not having to buy electricity from the grid.  I put this down to the fact that I work from home, and have computers/printers/air con all turned on during the day when the solar energy is being generated.


20 YEAR FORECAST
Based on this year’s results, I have created a forecast to predict the performance of the panels over the 20 year life span of the Feed in Tariff.  Note that the panels will still be working (all being well!) after 20 years, so there will be an ongoing saving on electricity usage from that point, but we will no longer receive any money from the Feed in Tariff scheme, so I have used that as a convenient cut off point for the forecast.

Measurement discrepancies

I have taken all my daily measurements over the past year from the meter that is built into the inverter itself (the big box in the photo, left) because it is convenient to do so. However, this meter is not calibrated, and there will also be some rounding errors on summing the daily measurements to give monthly values. The actual calibrated meter readings supplied to the electricity company for the Feed in Tariff are taken from the tiny white box immediately below the inverter.  The "official" total for the year from this meter is 1929 kWh, compared to the 1898kWh sum of measurements taken from the inverter, about a 1% discrepancy. 

It is the calibrated meter reading for the year ending August 2014 that I have used for the 20 year forecast data, so there will be discrepancies compared to the total of monthly values.


Assumptions on this forecast
Consistency of usage
I’m assuming that I will continue to use as much electricity over the next 19 years as I have been doing for the last year. Also I have assumed that I will continue to use all of the solar electricity that we generate.

Deterioration of generating capability:
As is the case with all domestic panels, the manufacturer’s specifications state that after 20 years, the panels will only be achieving 80% of their brand-new generating capability. For the sake of this forecast, I have assumed a linear yearly deterioration of 1% of the first year’s generation, which translates exactly to 80% after 20 years.  I will be able to compare the actual deterioration against this as time goes on.

Feed-in tariff index linked:
The feed in and export tariff is index linked to the Retail Price Index (RPI), applied yearly every April. See here:  http://www.fitariffs.co.uk/eligible/levels/indexation/ This year our generation Feed in Tariff rose from 14.9p/kWh to 15.177p/kWh, and the export tariff rose from 4.64p/kWh to 4.731p/kWh.  This is an average rise of 1.88% (quite a lot lower than the 2.7% quoted for April 2014 on the website above). To be conservative, for the sake of this forecast I have fixed on the 1.88% rise for the next 20 years.

Increase in cost of electricity bought from the grid:
According to https://www.gov.uk/government/statistical-data-sets/monthly-domestic-energy-price-stastics, electricity costs rose by an average of 7.8% per year from 2006 until 2013, and 6% from 2013-2014. There are many factors that will affect the price of electricity in the UK over the next 20 years. I'm not going to start trying to second guess Hinckley Point C coming online / other nuclear station switch offs etc.  I don’t have a crystal ball, but I think I’m being conservatively realistic by allowing for an average price increase of of 5% per year over the next 20 years.  

Here is all the data plugged into an excel chart:

(There is no significance to the highlighted box, it's just where my cursor happened to be when I did a screencap) 

Here is the financial data from above shown on a graph:

Observations
Over 20 years, our 2kW array will generate in the region of 34-35MWh. Somewhere about seven and a half years is the breakeven point on the original £5200 investment in the panels. After 20 years the panels will achieve a total of approx £21,736 benefit (feed-in tariff + savings on electricity not bought from the grid), which is a return on the investment of 318% over 20 years, or 15.9% per year. 

Note that if we discount the savings from using all our own solar electricity, using the feed-in tariff alone it would take about 13 and a half years to recover the original investment, almost twice as long. In this case, the total 20-year benefit is £8567. This is a return of 58% over 20 years, or 3.24% per year. This is a significant consideration if your household does not use much (or any!) electricity during daylight hours.

The chart shows that the savings made per year by not buying electricity from the grid, breaks even with the money received from the feed in tariff after 6 years. Due to the fact that the cost of buying electricity from the grid rises faster than the Retail Price Index (to which the Feed-in Tariff is linked), from year 6 onwards the biggest yearly financial benefit of the solar panels is from the money saved by not buying from the grid.

Conclusions
Practically - The Panels
Our installation was a bit of a disruption - we had scaffold up on the back of our house for about 8 weeks all told, although that was due to holiday date clashes and repairs we had to make to our roof prior to the panels going up - it only took a couple of days to actually put the panels up. As long as your roof is sound and the installation is done by a reputable company then the disruption from panel installation is minimal.  

The inverter The inverter is a large box of electrical wizardry that turns solar electricity into a form that can be fed into your consumer unit (fuse box) and then on into the grid (see photo under 'measurement discrepancies', further up). You need to have somewhere to fix it in the house (ours is in the under stairs cubby hole, next to the consumer unit). The installation company said they also mount them in lofts, they can go wherever is convenient for your property - with the only proviso being they have to physically route the wiring from the panels on the roof to the inverter, and then from the inverter to your home's consumer unit. Our wire runs through a sealed hole in the roof immediately under the panels (the guy said they've been installing panels for 10 years and have never had a customer ring up with leaking issues), across the inside of the roof in the loft, down to the eaves, then through the wall and down the outside of the house in plastic cable conduit. Our under-stairs cupboard is against an exterior wall so it could go straight through the wall directly to the inverter.

Once up and running, on a day-to-day basis you don't notice them. You don't need to do anything with them, you don't need to turn them on or off, they just sit on your roof, come rain or shine (although preferably mostly shine) and generate electricity. There's no moving parts to wear out. They make no noise. They don't belch smoke. 

There was only one issue we have had with our panels, and that was during installation, one of the snap-fit electrical connectors that join the panels together hadn't been properly snapped on installation. Thermal expansion & contraction caused it to unseat midway through September 2013, meaning we lost a weekend's worth of generation before the company could come out and fix it.

I have been advised that inverters have a life expectancy of 'about 10 years'.  No examples of our particular model (Power One Aurora Uno) have yet come to end of life, but based on previous inverter technology, I've been told to budget to replace it around 2023, at a cost of around £600. This should be factored into the financial benefit of the system as a whole: £600 out of £21,736 takes the total benefit down to £21,136, making for an annual return of 15.3% over 20 years on the £5200 investment.

Financially
Buying solar panels is certainly not a get-rich-quick scheme. They are a long-term investment with respectable returns, as the above figures show. If you use most/all of the solar generated electricity yourself, on a £5200 investment you can net yourself a whopping 15% annual return over 20 years, and you don't have to put your money anywhere dodgy to get it!

Ethically
Regardless of any financial benefit, solar panels are totally and utterly an ethical choice (obvs). Our 8 lil' ol' solar panels will save the UK a not-insignificant 35MWh of generation over the next 20 years. A drop in the ocean compared to national usage maybe, but if every roof in the UK had a solar array on it, we wouldn't need power stations, simple as. The numbers are compelling (see my earlier blog post  http://gillpanels.blogspot.co.uk/2014/07/a-solar-uk.html ). I haven't done any studies on the environmental cost of manufacturing solar panels, which is something I'd like to look into in future.

The democratisation of energy is a good thing, but we would still need a grid to distribute electricity to where it is needed. And we'd also need to replace the power stations with power storage facilities - and this is the big problem - they'd need to be able to store HUGE amounts of kWh (not yet done the maths...) The majority of solar electricity is generated in the summer, and the excess would need to be stored for use during the winter. In this brave new world, electricity still isn't free. In the absence of a revenue stream from generation, the electricity companies will have to shift to a new model for financing a robust energy storage network and upkeep of the distribution grid.


Roll on the next 19 years...

July's solar generation stats

I'm a bit late with this month's post - we've been on our hols. 

Here's the graph for July's generation from our array of solar panels. This is for our 2kW 8 panel solar array (Panasonic HIT panels), mounted on our south-facing roof at approximately 35-40° angle from the horizontal.

The dark blue line is July - the light orange line is December by way of comparison.  At a total of 288.6kWh for the month, that's 80.86kWh higher than the forecast.

Here is July included on the yearly chart (and a bit of August too....)

July is the highest month of generation of the year. August isn't complete yet, but I'm betting it will be slightly lower than July looking at the forecast.

Here's the annual sun-strength chart. The value for August is based on the first 12 days of the month, so isn't completely accurate:

The shape of the curve is more obvious now - a steep climb from January to April, and a more gradual decline in sun strength from July down to December.  The strange blip in May (as discussed on last month's post) is now very clear.

The total forecast for the year that we were given by the installers was 1645kWh.  As of last night (12th August) the total for the year stood at 1754kWh, with 19 more days of August to go to complete the year. I reckon the final total is going to be between 1850 and 1900kWh, depending on how many sunny days we get between now and the end of the month.  

Next month I'm going to do a wrap up of the whole year, with financial predictions for the 20 year life span of the panels.

A Solar UK!

After reading this article today, I pondered about the UK's solar energy market. And I pondered about how many solar panels you'd need to fulfil the entirity of the UK's electricity consumption requirements.

A case of some simple googling and some simple maths, I thought!   And I was right, I worked it out.... here's how I did it...  Brain in gear?  Then let's start....

Begin with the known quantities, which are the solar panels we have on our roof. Our solar panels have an area of 10.08 square metres (measured) , and generate 1800kWh per year (measured) from 1567 hours of sunshine per year (average stats over 30 years from Met office).

So, 1 square metre of our solar panels generates 1800 / 1567 / 10.08 = 0.114 kW per square metre.

I thought I'd go for a best case scenario, and googled which city in the world has the most hours of sunlight per year. It happens to be Yuma, in Arizona, USA, right on the border of Mexico. They have an average of 4015 hours of sunlight per year.

So in one year, our solar panels in Yuma would generate 0.114 x 4015 = 457.54kW per square metre. To get the numbers into more of a sensible format, this converts to 457,540 MW per square kilometre (1,000,000 square metres in a square kilometre, 1,000kW in a MW).

I then googled how much electricity the UK uses in a year. The most recent data is from 2008, and it is 344,700,000 MWh per year.

So, by dividing the total energy requirements by the amount generated per square kilometre, this gives us how many square kilometres we'd need to generate that much.

It comes out at: 344,700,000 / 457,540 = 753.37 square kilometres.

So, to generate enough electricity to power the whole of the UK, you need 753.37 square kilometres in Yuma, Arizona. To get this many square kilometres, you need a square about 27 km along each side (about 17 miles).

That's about the size of the Greater Manchester area.  Here it is on a map:

Here it is shown next to Yuma on the US map:

There it is. The red square. East coast, just above the long thin stretch of water (Gulf of California). A solar farm that big would provide enough electricity for the whole of the UK.

And now for the final crunch - how many solar panels would that be?  Well, each of our panels is 1.26 square metres. I'll do the maths - it's just under 598 million. (597,918,059 panels to be precise).

And at about £200 per panel, (here) that works out at £119,583,611,737 - lets round it up to a nice round £200 billion.

At current consumer electricity prices of about £150 per Megawatt hour, the UK has a total spend on electricity of 344,700,000 MWh x £150 = £51,705,000,000. Lets round it up to £52 billion.

So, just less than four year's worth of electricity would pay for enough solar panels to power the whole country for ever (well, not quite, but you get the idea...!)

Although this is just a thought experiment (1; I know you couldn't transport all that energy from Yuma to the UK and 2; you'd need about two and a half times this area to do it in the UK due to the lower levels of yearly sunshine) it does put solar energy into perspective.

I believe these figures show, far from being pie in the sky, rooftop solar panels could very feasibly generate enough electricity to power the entire country (and yes I know with solar comes some not insignificant energy storage requirements to smooth out the supply/demand discrepancies).

The feed-in tariff incentives to retrofit panels to existing property are an excellent start - what we need now are regulations forcing new builds to fit panels as standard, and we will be going a very long way to easing, if not completely satisfying the forthcoming energy 'crunch' over the next decade as the old nuclear plants are phased out.

Roll on the democratisation of energy!

AND YOU DON'T NEED TO FRACK AN EFFING SQUARE INCH.

Just sayin'

June's solar generation stats

Here's the graph for June's generation from our array of solar panels. This is for our 2kW 8 panel solar array (Panasonic HIT panels), mounted on our south-facing roof at approximately 35-40° angle from the horizontal.

The dark orange line is June - the light orange line is December by way of comparison.  At a total of 270.6kWh for the month, that's 59.29kWh higher than the forecast.

Here is June included on the yearly chart:

And the yearly sun-strength chart, which is taking shape nicely, showing more of the year:

Since this chart takes account of the total number of hours of sunlight in each month sunrise to sunset  (which doesn't change from year to year), there are only two factors that affect the shape of the graph:

1) The increasing strength of the sun as it gets higher in the sky in summer.
2) Cloud cover.

The changing elevation of the sun in the sky throughout the year creates a smoothly varying rise and fall in the shape of the graph. It also seems reasonable to expect that the amount of monthly cloud cover would decrease steadily as Spring moves towards Summer, causing the average kWh to increase steadily. 

I've double checked the code in my spreadsheet, and it's right - May really did dip like this.  I conclude that May's anomaly must be due to cloud cover - this year on the Fylde coast, there was more cloud cover in May than there was in either April or March, which sounds counter-intuitive, but the numbers point to this being be the case.

Whether this is a one-off, or whether this is normal for the Fylde coast, I won't know until next year....

May's solar generation stats and quarterly feed in tariff results

Here's the graph for May's generation from our array of solar panels. This is for our 2kW 8 panel solar array (Panasonic HIT panels), mounted on our south-facing roof at approximately 35-40° angle from the horizontal.

The turquoise line is May - the orange line is December (the lowest solar generating month) by way of comparison.

May has not quite hit the forecast target for the month.  The forecast suggested 228.85kWh would be generated in May, and the panels actually generated 221.8kWh, which is 7.05kWh down on the monthly forecast.  

This works out at an actual average of 7.15kWh generated per day.  I put this down to May being more-than-averagely cloudy this year.  Although we have hit another daily record this month - on the 6th May, the panels generated a total of 13.7kWh.

Here's the yearly forecast chart:

And the yearly sun-strength chart:

The June value is for the first 3 days of June only and so is not accurate.

Feed-in Tariff
One thing I have completely overlooked in my financial projections is the fact that the amount we get paid for our solar generated electricity under the feed in tariff is index-linked.  The pence-per-kWh increases every quarter, and this quarter it's gone up by a monumental 0.277p per kWh generated, and 0.1p per kWh exported. At the end of the first year I'm going to factor this into our 20-year projections and see how much it brings down the time-to-repay figures.

The spring feed-in tariff is in. For the months March - April - May the panels generated a total of 638.8kWh, for a total of £112.26 (100% generation at 15.1577p/kWh + 50% exported at 4.731p/kWh).

We also benefit financially by using our own solar-generated electricity, as this is electricity we're not having to buy from the grid.  Our electricity meter only measures how much electricity we use in total (solar + grid), and the solar meter only measures how much electricity the panels have generated, and trying to estimate how much of it we've used ourselves is getting tricky. The sun is shining more often, and I can't keep ducking into the under-the-stairs cupboard to see if the electricity meter has stopped counting up.  Having my office at home means we are using a lot more electricity during the day than normally would be expected, but I'm going to hedge my bets and say we have used 75% of our solar-generated electricity this quarter.  This equates to an additional saving of £71.43 of electricity we've not bought from the grid.  This brings the total benefit to £183.50 for the quarter.
Here's a graph showing how much electricity we've bought from the grid over the last 12 months:

The panels were commissioned on the 28th August 2013, which was when I started taking regular readings. The first reading (June 2013) is an average of the previous 5 months.  Still, we're approximately 10kWh per day lower compared to this time last year.  Unfortunately, I can't put this all down to the solar panels, because I've also been fitting low-energy lights throughout the house (and insisting the boys turn the TV off when they've all left the room!) which will also have contributed, but by how much I can't tell.

This has also got me thinking again about energy storage - for every kWh I export to the grid, I get only 4.731p. Whereas if we can use all that energy ourselves, we save 15p/kWh not buying it from the grid. Depending on the set-up costs, it sounds sensible to have small short-term storage (up to a maximum of say 5kWh for 8-10 hours), so that excess generated during the day is used in the evening for lighting/TV etc. -  but I'll save that for another blog!

April's generation statistics

Here's the graph for April's generation from our array of solar panels. This is for our 2kW 8 panel solar array (Panasonic HIT panels), mounted on our south-facing roof at approximately 35-40° angle from the horizontal.

The purple line is April, and I've included March (the green line) by way of comparison.  April has again outperformed the forecast.  The forecast amount for April was 177.07kWh, for an average of 5.90kWh per day.

The actual total for April was 223.90kWh, a 46.83kWh excess (26.45% more than predicted), for an actual daily average of 7.46kWh.  Again, another record generation day this month - the 18th April saw a total for the day of 13.2kWh.  (According to http://www.timeanddate.com/worldclock/astronomy.html?n=1331&month=4&year=2014&obj=sun&afl=-11&day=1 there were 14.25 hours of sunlight on this day, giving an average sun strength of 0.926kWh per hour of daylight.)


Here's the yearly forecast chart:

For the year so far (September - April), the panels have exceeded the predicted generation amount by 61.94kWh.

Here is the yearly sun strength graph:

April achieved an average sun strength of 0.57kWh/hr (or 0.57kW as it should be written). Now the shape of the curve from March to April is more apparent - it seems debatable whether May is going to exceed an average 0.6kWh per hour of daylight.

May's current generation prediction is 228.85kWh for the month. There are 465 hours of daylight in May according to the met office, and using April's sun strength value of 0.57kW, my new prediction for May is 465hrs x 0.6kW = 265kWh, an excess of 36.15kWh.  Let's see how close I am next month!

March Generation Stats

Here's the graph for March's generation from our array of solar panels. This is for our 2kW 8 panel solar array (Panasonic HIT panels), mounted on our south-facing roof at approximately 35-40° angle from the horizontal.

The green line is March, and I've included February (dark red) by way of comparison.  March has performed significantly better than the forecast.  The forecast generation for the month was 117.94kWh, an average of 3.8kWh per day.

The actual generation was 179.5kWh, a 61.56kWh excess (52.19% over). This gave an actual daily average of 5.79kWh.  It shows a record generation of 11.2kWh on the 24th - the best for one day since the panels went in last year.

Here's March on the yearly forecast:

As you can see, March has virtually equalled the forecast amount for April.  We have now caught up on the deficit we had on the forecast over winter, and we've generated 15.11kWh more than the forecast for the year to date. 

Here's the yearly sun strength graph:

The average March sun strength for our array is 0.53kW.  What I find surprising is that already you can see that this graph is skewed.  I had assumed that sun strength would rise and fall symmetrically around summer, but the rise during Jan-Mar is far steeper than the drop off shown in Sep-Dec. Note that the sun strength chart takes cloud cover into account, so the lower frequency of cloud cover during spring-summer will effectively increase the sun strength, and this may be what is contributing to the skew.  It's a bit of guessology, but by following the curve, I think there's a good chance that sun strength will top 0.6kW over the summer months.

As an aside, by comparing year-on-year records of this graph, it will be interesting to see if global warming has any beneficial effects on the amount of electricity our panels generate.

Why the divergence from the forecast?
I've had a mull as to why the forecast has over-estimated the winter months, and (so far) has underestimated the Spring months.  Whenever solar panels are installed, the suppliers give an SAP forecast for generation for the year. This is a standardised forecast based on a set of government calculations.  The SAP yearly prediction for our panels was 1645kWh.  To produce my monthly forecast amounts, I divided that up based on the changing amount of bright sunlight hours we receive per month, (from Met office average statistics from 1971-2010).   If you rummage back through the blog, there's a page about it with links.

The forecast therefore assumes that each hour of sunlight in each month has the same generating strength.  This is not in fact the case. Assuming a cloudless sky, at our longitude here in the UK, one hour of low winter sun generates far less than one hour of high mid-summer sun.  This is shown very obviously in the sun strength graph above.  

I am very optimistic about the next few months. I did a rough revised forecast, based on the conservative assumption that the sun strength will remain at 0.53kW between now and the end of summer, and we are looking to top 1800kWh for the year.

Feed-in Tariff - winter payment

The Feed-in Tariff payment for winter arrived today - I was expecting it to be fairly paltry, and I was not surprised.

For a total official generation figure of 172.25kWh for Dec/Jan/Feb, we received:

Generation payment -
172.25 x 14.9p = £25.66

They assume we export 50% of what we generate back to the grid, and pay us 4.64p per kWh for that.

Export payment -
(50% x 172.25) x 4.64p = £4.00

So we received a total of £29.66 paid into our account.

Because I work from home and have computers, printers and heaters/air con etc running from 9 til 5, we didn't actually export any of our electricity back to the grid, we used all of it ourselves.  But we still get paid for the assumed 50% export. I've read through all the small print, and there's nothing anywhere to say what to do if you think the assumed export value is incorrect, so I'm in no hurry to tell them about it.

Consequently, you stand to get the most financial benefit from solar panels if you use every drop of solar generated electricity yourself - particularly as electricity prices increase - because that's all electricity you're not having to buy from the grid.

We're with npower, and our electricity tariff has just gone up to 15p per kWh, so we've saved 172.25 x 15p = £25.68 this winter on electricity we would have otherwise had to buy from them.

Added together, this is a total benefit to us of £55.34 for the three months Dec '13 + Jan '14 + Feb '14.

Here's a running total of the overall benefit so far:

Autumn 2013 £92.50
Winter 2013  £55.34
----------------------
Total            £147.84