by Tom Mertens © dec 2008
What is this article about? It tells you all about sizing and installing a small solar installation for generating electricity. It explains what components are used in a system that generates solar electricity and how they work.
Although the use of solar energy worldwide is increasing, a lot of people have doubts about the idea. Countries like Holland, Germany and the UK, for example, are better known for their wet and rainy climate rather than sun. Even so, even in these countries the sun delivers more than enough energy to cover all our needs. In the Netherlands, for example, the sun delivers about 50 times as much energy as we are using altogether in the form of gas, petrol, electricity, etc. But how to get the optimum profit from all this energy? Energy from the sun reaches us in many forms, but here I'll mention only the ones we are most familiar with, warmth and light.
When we use the heat of the sun, we call it thermal solar energy. When we use the light and convert it into electricity we call this photovoltaic solar energy. PV for short.
In thermal solar energy installations the heat of the sun is collected and (usually) accumulated for later use. Normally this is done by means of so-called solar boilers. These consist of a heat collector and a tank. Most of the time the storage tank is also used as a heat exchanger. In the heat exchanger tank the heat is transferred to the tap water system in our house or used for central heating. A solar heat collector generally consists of a system of black tubes placed in an insulated box with a window facing the sun. Water is used as the transport medium from the collector to the storage tank/heat exchanger.
The efficiency of a solar hot water system can be very high. Even in winter when the weather is freezing, water at 30 to 40 degrees Celsius or more can be obtained. Another advantage of a solar hot water system is its relative simplicity. The design and installation of thermal solar systems is covered elsewhere.
This article is about photovoltaic (PV) systems. These are of two main types: grid-connected and autonomous.
In this article I'll focus on autonomous PV systems. The other type is the grid-connected system. These are becoming more common and are sometimes called PV micro-generation systems.
Depending on local prices and possible grants and promotional measures, installing a grid-connected system can be an attractive option, but it is outside the scope of this article.
The construction of an autonomous PV system can be compared with a thermal system. In this type of PV system you will also find a ‘collector’ and a 'storage tank'. The collector is the solar panel array and the storage tank is called a battery or accumulator. Before describing the various components of such a system, let’s have a look at the different uses of a PV system.
Without going into all the details we can say that electricity from the grid (in Europe) is cheap compared to solar electricity. If we were to take price as our only guidance, most applications of solar electricity would drop away. In most situations there is a wall outlet nearby. Generally, uses for PV will be found in the countryside, for electric fencing and water pumps, on boats, at weekend houses and on campsites.
In remote areas the costs of an autonomous installation can outweigh the costs and/or problems of getting a grid connection.
Very often we hear the argument that solar energy is expensive, but that we are helping the environment by saving on fossil fuels. Of course that is true, but if we also take into account the energy used for production and transport the balance looks less positive. According to various sources the energy payback time (that is, the time taken to repay the energy used in production, transport and operation) is about 6 years. These numbers are for frameless panels in a grid-connected system, a frame, for example, adds up to 1.6 years to the energy payback time. The energy consumed for installation and recycling is not taken in to account either. Furthermore, autonomous systems use batteries that further add to the cost and have negative environmental issues, although theoretically batteries can be almost 100% recycled. Also, it must be remembered that autonomous systems deliver less energy than grid connected systems can with the same panels.
This may sound a bit negative, but in my experience the biggest gain is to be had from using less energy. When you are using an autonomous system that is not oversized, you will be much more aware of what you are consuming and therefore will consume way less energy than you probably would have done in a grid connected situation.
In most small autonomous systems you will find the following components:
In the following paragraphs I'll discuss the function and workings of these various components.
The solar panel is the actual energy source of the system. Here the light of the sun is converted into electricity. Nowadays almost all solar panels are made out of silicon. Silicon is one of the most abundant elements on earth. Sand is composed of mainly silicon oxide. To make the cells for solar panels, very pure silicon is needed. Producing this pure silicon is a complex and energy-consuming process. In former days solar cells were made out of scrap material from chip factories. Alas, nowadays there is not enough scrap material to fulfill the need, since solar cells are becoming more popular.
There are three main types of solar cells in common use. They use different production processes and can be distinguished by appearance. The main practical difference between these types is the efficiency with which they convert sunlight into electricity. None of them is really efficient. Research is leading to different types of solar cells that will reach efficiencies of 30-40% under laboratory conditions.
Poly-crystalline cells have a market share of about 85%. They have a dark blue or anthracite colour and a nice crystalline pattern is visible on the surface. Amorphous cells are plain brown, black or gold in colour.
In Northern Europe we receive approximately 1000 watts of solar energy per square meter, so with the best commercially available cells we can only get 200 watts of electrical energy. In real life efficiency is often even worse; panels may not always face the sun in the optimal direction or part of the light may be reflected. Maybe worst of all, the panels heat up in the sun, which reduces their efficiency! So during the sunniest days, when we could profit the most from the available sunlight, efficiency drops. A good solution would be to cool the panels with water. By doing this we might profit twice, gaining hot water and (more) electricity. Remarkably, we seldom see this solution in practise.
The batteries store the collected energy. In most applications we need batteries since we cannot rely on the sun to deliver energy at all times when we want it. But batteries not only store energy for use during the night or on cloudy days; they are also capable of supplying huge amounts of energy for short periods of time. They can supply amounts of energy way beyond the maximum peak power output of the panels.
In domestic applications we normally choose batteries that are big enough to cover our needs for a couple of days without sun. For proper sizing we need to know how much energy the sun is supplying during he shortest winter days and how much energy we are using on average per day. I'll come back to these calculations later on. The batteries can be seen as the buffer that balances out the periods of surplus and shortage of solar energy. The size of the batteries needs to be balanced with the size of the solar panels as well. Choosing a battery that is too small compared to the capacity of the panels will cause very fast charging, maybe even too fast. Although it is a good thing to have your batteries full, it is a waste of panel capacity if they sit idling for most of the time on a sunny day. On the other hand, choosing a very large battery capacity, so we can last for a long period of time without sun, means we might find our batteries are seldom completely full. For most batteries, this is not very good and could have a negative impact on their lifespan.
Like with the solar panels, there are different types of batteries that can be used for solar installations. I'll mention a few in order of preference and fitness for solar applications (in my personal opinion).
TIP
For small or experimental installations you can use second-hand car batteries. Very often car batteries are replaced because they no longer have the capacity to start the car. If all cells of the battery are in the same condition, we can still use them. Sometimes you can even get these batteries for free at you local garage.
With the aid of a battery charger, a multimeter, an acid density meter and some car lamps we can test if it is worthwhile using the battery.
The charge controller is an important part of an autonomous PV system. Like the name implies this electronic device takes care of the state of charge of your batteries. Batteries are quite fuzzy on how they are treated. If you don't treat them right their lifespan will be diminished manifold.
Actually it is quite easy to do what a charge controller does. If you have a good voltmeter you can take control of your batteries yourself. Here is what you need to do:
You could have a full-time job doing this, especially on cloudy days or if you have a lot of energy consumers in your system; the battery voltage will vary a lot during the day! It is better to employ a charge controller to do the job.
A good charge controller takes care of your batteries and chooses a charging regime that fits with their current state of charge. The best ones even take the current temperature of the batteries into account and have a built-in calendar to plan certain maintenance events like equalization. (Equalization is a special process that keeps your batteries in top condition by making sure that all cells have, or reach, the same state of charge).
A charge controller is very handy, probably better for your batteries and less stressful too.
State-of-the-art charge controllers use a system called Maximum Power Point Tracking, MPPT for short. This type of charge controller adjusts itself to both the batteries as well as the solar panels. The controller regularly measures how more power can be obtained from the solar panels by changing the current drawn (and thereby the voltage supplied by the panels). It then converts this optimum current-voltage combination (the optimum power point) of the solar panels to the current-voltage combination required by the batteries at their current state of charge. A good MPPT controller can boost efficiency by 10% to 40%.
An added advantage these MPPT controllers is that they can work with a very wide range of input voltages from the solar panels and a wide range of battery voltages as well. This can be of great use, as it means we can connect solar panels in series. By connecting the panels in series, we obtain a higher voltage, but less current flows. The result is that thinner (cheaper) cables can be used to connect the panels to the charge controller while avoiding big energy losses in the cables.
But as with all bonuses, these advantages do not come for free. MPPT controllers cost much more then normal charge controllers.
Very often the charge controller also takes care of a few other things. It prevents current flowing backwards to the solar panels at night and it can make sure that your batteries do not discharge too deep by switching off the load when the batteries get close to empty. Discharging your batteries too deeply can be fatal to them and should be avoided at all costs. It is mainly in small PV systems that loads are connected to the charge controller. In bigger systems the charge controller only takes care of the PV panels and the batteries, while an inverter takes care of the loads.
Because inverters play such an important role in most PV systems, it is a good idea to have a closer look at them. The inverter converts the low-voltage direct current supplied by the batteries into 230 volt (in Europe) alternating current. Most appliances and lamps in a normal household work on 230 volts; this is what we are accustomed to when we are connected to the grid. It is a good idea to have an inverter as we can then use normal appliances.
Nowadays there are two different types of inverters on the market. This was not always so. About ten, maybe fifteen, years ago, you could basically buy one type of inverter. This would deliver a crude-shaped alternating voltage at the output that was okay for lamps and old- fashioned drills, but would not satisfy the more demanding needs of modern electronic appliances like computers, video recorders, audio equipment or modern washing machines. Newer inverters deliver an output voltage waveform that is exactly the same as the voltage you get from the grid.
That is, inverters can be classified into the two types by the waveform of the output voltage they generate.
The illustration shows the different waveforms.
In the illustration below you see how the different components are connected together.
In the simplified scheme above I have drawn an inverter connected straight to the battery. You might expect the inverter to be connected to the output of the charge controller, because it can be seen as being just one of the consumers of power, which is absolutely true. But because an inverter is such a big consumer it is normally connected straight to the battery. In fact, most charge controllers can only supply an amount of power/current that is of the same order as supplied by the solar panels connected to it. For example, if you were to use a charge controller rated for 10 amp PV panels, the output current will probably also be between 10 and 15 amps. A big inverter can consume 100 amps or more so cannot be connected to the charge controller.
To prevent the batteries being completely exhausted by the inverter, the inverter should have built-in battery protection that switches it off when the battery voltage drops below a certain limit. Most modern inverters do this.
Where to put what?
A solar system comprises three or four main components that all have their own environmental preferences.
Because of these differences in requirements, let’s look at the components one by one.
Of course we want to put the solar panels in a place where they receive the maximum amount of direct sunlight. Very often this will be on the roof, but that is no necessity. A place in the garden without buildings or vegetation nearby that could block sunlight might be even better. The cables may be shorter and certainly access to the panels will be easier.
Depending on your geographic location, the angle at which the panels face the sky is different. The more towards the earth’s poles you are situated, the steeper the panels need to be angled. At the equator the panels would be parallel with the ground.
In the Netherlands for example, the optimum angle is 37 degrees relative to ground level. This, of course, is an average, because the angle of the sun changes day by day. If we need more energy during the winter months, we could increase the inclination a bit. We will get less power in the summer and the yearly average will be less as well, but the winter orientation will be better, improving the yield of power.
If the panels are mounted on a roof there is not much room for choice. We have to accept the roof angle and orientation, or maybe we can adjust the angle of the panels slightly for better performance.
If we strive for year-round optimum performance, the best choice is to mount the panels on a rotating pole. (see photo below)
In this photo you see a setup in which the panels can be oriented manually. Vertical inclination can be adjusted and the panels can be turned to follow the sun from east to west. Although automatic tracking systems are getting more popular, I think that there are a lot of situations where the extra cost involved can better be used to buy a few extra panels. Theoretically the gain in energy produced can be quite big, but in real life it is not that spectacular. Why? In summer your batteries will probably be already full half way through the day. So there is no use tracking the sun as it makes its way around a path of more than 200 degrees. In winter you'll probably have more problems getting enough energy, but the curve of the sun is much smaller, maybe 120-150 degrees. So in wintertime there is less gain to be had by following the sun.
Being mechanical devices, solar tracking systems can break down, and they do, while the special poles with automatic tracking systems cost a lot of money. A pole that can be turned manually is much more cost effective. When the need arises, I just walk up to the panels and turn them to the optimal position.
How much space do you need for your panels? With modern poly-crystalline panels you can expect between 100-120 watts per square meter. The six panels in the picture above are about one square meter each, and rated at 120 watts peak power. So that would give 720 watts peak output. With this setup I measured a maximum output of 820 watts with an MPPT charge controller on a cool but sunny day! Remember these are optimal conditions, sunny but with low temperatures.
With a setup as discussed above the cabling can be divided into three segments.
Cables in the low-voltage grid
If you plan to use a low-voltage grid as well, the cables must be sized properly. Cable losses play an important role at low voltages; a much bigger role than in a 230 volt grid using an inverter. A voltage drop of 1 volt in a 230 volt system is not a big thing, just a difference of less than 0.5% and a power loss of less than 1%. In a 12-volt system a drop of 1 volt will have a much bigger impact; the difference is more than 8%, which results in a power loss of 16%! Not only are the differences much bigger in a low voltage system, but the higher currents will easily cause these bigger voltage drops to happen.
This is one good reason to have an inverter in a bigger system. It is not a very attractive option to have many consumers on a low-voltage grid that extends for a long distance from the batteries. But once we know what consumers we have and where they are situated, we can calculate the size of the cables we need. Depending on what cable losses are acceptable, we can size the cables appropriately and decide if we want to use a low-voltage grid.
So far I have given an overview of all the components in an autonomous PV system. But there is still one big question to be answered: what capacity should our system have?
To answer this question we have to evaluate many conditions and think about desires and demands.
A sense of reality is an asset when we answer these questions. An average household in the Netherlands uses about 9 kWh/day. If we aim for a system with this capacity we are going to end up with a huge and expensive system. The best way to go is to make a list with all the energy consumers you have, recording how much they use and how often, or how long you use them per day.
Such a list could look like this:
| Appliance description | Power consumption | Hours use per day | Used power per day |
| Lamps | 40 W | 4 hr | 240 Wh |
| CD/amplifier | 30 W | 2 hr | 60 Wh |
| Well pump | 1000 W | 3/4 hr | 750 Wh |
| Fridge | 45 W | 24 hr | 1080 Wh |
| Desktop PC with CRT screen | 200 W | 4 hr | 800 Wh |
| Video recorder/DVD player in Stand-by mode | 8 W | 22 hr | 176 Wh |
| Video recorder/DVD player on/playing | 30 W | 2 hr | 60 Wh |
| Totals | -- | -- | 3166 Wh |
A list like this gives us insight in our electricity consumption. For example, a fridge does not consume much power, but since it is running 24 hour per day it still ends up as the biggest power consumer we have in our household. Another thing we see, is that appliances in stand-by still use a considerable amount of power during the day. The video recorder I put in the table above is used for two hours per day, but the consumed power in stand-by mode is three times as much as the power used when watching a video! Being aware of these 'hidden' consumers can make the difference. Pulling the plug after use is the only way to be shure that an appliance in not consuming any power. Good for the envionment and good for saving money a well, certainly when installing a solar energy system.
To be added:
Security, fuses, protection measures...