The effect of heat on solar power output

The first really hot day of summer we always get several phone calls from customers who are eagerly watching their PV monitors, expecting to see the highest output yet - and are disappointed to find that actually their array is producing a fair bit less than normal.

If that's you, don't worry. It's not a problem with your inverter, or a dodgy panel. It's just that unfortunately panels become quite a lot less efficient as the cells get warmer. On a really hot day, particularly if there is no breeze to cool them down, panels can easily reach temperatures of 60 or 70 degrees - too hot to touch.

The power output of panels is measured under Standard Test Conditions, which are a cell temperature of 25 degrees C, an irradiance of 1,000 W/m2, and an 'Air Mass' of 1.5. The 'Air Mass' is a measure of how much atmosphere the sunlight passes through before it hits the panel. If the sun was directly overhead (and the panel was at sea level), the Air Mass would be 1. By the time the sun is low on the horizon in the evening, the Air Mass can be as high as 4 or 5. But a value of 1.5 isn't unreasonable for the height of summer in the UK - it's what you would get with the sun being about 45 degrees above the horizon. The irradiance value of 1,000 W/m2 is also a reasonable approximation.

The cell temperature however is normally far above the 25 degrees of 'standard test conditions' on a summer's day. With each degree increase in temperature, a panel will lose around 0.4% of its power. So if the cell temperature reaches 65 degrees, the actual power of a 250W panel becomes: 

250 x (1-0.004)^(65-25)

= 213W

This is about a 15% drop in power compared to what you might expect if you go by the standard test condition values.

Take a look at the power output on a cold, clear spring morning however, and you should see a considerably more pleasing figure!

The effect of shading on solar panels

Shading on solar panels is bad news. But many people fail to realise just how drastic an effect even a small amount of shading can have on the output of your array. The reasons why shading is so bad are slightly technical - but we'll do our best to explain them here in not-too-technical jargon, so stick with us.

In sunlight, each solar cell in an array acts as a little electron pump, pushing electrons from one side of the cell to the other, and giving a voltage boost to the system as they do so. A single cell isn't very powerful though, so in order to get a useful voltage, you need to put quite a number of cells in series. The output of one cell becomes the input to the next cell.

When a cell is shaded, the number of electrons it can pump from one side to the other drops. That, in itself, wouldn't be too bad you might think - you would just lose out by the power output of one cell. But unfortunately, because it is not pumping so many electrons up to its neighbour now, it limits the number of electrons that the neighbour can pump too. Same for the next cell in the line - and the next, and so on.

The other cells can manage to force some extra electrons through the badly performing cell, so it's not quite the case that the whole system performs as poorly as the worst-performing cell in the string - but it's not all that far off. You might easily see a 50% loss in power from a string of solar cells if just a single cell is shaded.

Fortunately, we can help to some extent by fitting bypass diodes to solar panels. Bypass diodes are fitted in parallel with a string of PV cells, and they do exactly what they say on the tin - they allow current to bypass a poorly performing set of cells.

There are a couple of problems with this though.

  • It wouldn't be practical to fit bypass diodes to every cell - manufacturers fit at most two or three per panel. So even if a single cell is shaded, you will still lose at least a third of the panel output.

  • It needs a bit of a shove to get the current from the good cells through the diode - it won't just flow round of its own accord. In fact, quite a lot of the power that the good cells are producing will be used up in forcing the current through the bypass diodes.

  • Inverters are designed to work with a specific size of solar array, with a given input voltage (or at least a band of input voltages). By the time you have lost the voltage from the shaded cells (and any unshaded neighbours on the same bypass diode), and then reduced the remaining voltage further by the amount needed to force the current through the bypass diodes, the remaining voltage will be a lot less. Often, it will drop below the start-up voltage for the inverter - which will then shut down. Even if it's not low enough to shut down, it certainly won't be working at its design voltage, and at lower voltages its efficiency will be a lot less.

According to one study, which you can download here, the output of a 1400W string - which was fitted with bypass diodes - dropped by 10% when only 4 cells were shaded. When 12 cells were shaded, the power output dropped by more than 50%!

The options you have if you have partial shading are:

  • Install as normal - and have a poorly performing system. This, unfortunately, is the answer for many installers.

  • Opt for a smaller array - just don't put panels in the regions which get shaded. Cheaper, and the money you do spend will be well spent. But you lose out on output.

  • Break the array into chunks, and put an inverter on each or use an inverter with dual MPPT. You can even go to the length of putting an inverter on every single panel - Enecsys micro inverters and Tigo voltage optimisers, for example, work on this principle. This will increase the price of your system, but it will be more effective.


Types of solar panel

There are three main types of solar panel in commercial production, all with some advantages and disadvantages. All three are based on silicon semiconductors - the difference being the form that the silicon is in. Panels based on other chemistries are under development: Cadmium telluride and copper indium diselenide panels may well appear in production soon; and also research is being conducted on using the photosynthesis effect that plants use to convert sunlight to useful forms of energy. However, you are unlikely to come across technologies other than silicon for the time being, so we will just consider silicon on this page.

Monocrystalline solar cells

These are made from thin wafers of silicon, sliced from large crystals that have been grown under carefully controlled conditions. Typically, the cells are a few inches across, and a number of cells are laid out in a grid to create a panel. Relative to the other types of cells, they have a high efficiency, meaning you will obtain more electricity from a given area of panel. This is useful if you only have a limited area for mounting your panels, or want to keep the installation small for aesthetic reasons. However, growing large crystals of silicon is a difficult and very energy-intensive process, so the production costs for this type of panel have historically been very high. Production methods have improved though, and prices have fallen a great deal over the years, partly driven by competition as other types of panel have been produced.

Polycrystalline solar cells

It is cheaper to produce silicon wafers in polycrystalline form, as the conditions for growth do not need to be as tightly controlled. In this form, a number of interlocking silicon crystals grow together. Panels based on these cells are cheaper per unit area than monocrystalline panels - but they are also less efficient. In terms of pounds-per-watt, there is not a great deal of difference between mono and poly panels.

Amorphous solar panels

The newest type of panel is based on amorphous silicon. Here, the silicon atoms are not ordered in a crystal lattice at all. The production methods are quite different - instead of growing crystals, the silicon is deposited in a very thin layer on to a backing substrate. Sometimes several layers of silicon, doped in slightly different ways to respond to different wavelengths of light, are laid on top of one another to improve the efficiency. The production methods are complex, but less energy intensive than crystalline panels, and prices should come down as panels are mass produced using this process.

One advantage of using very thin layers of silicon is that the panels can be made flexible. Panels are available that can be curved to the bend in a roof for example, or even attached to a flexible backing sheet so that they can be rolled up and put away when they are not needed! The disadvantage of amorphous panels is that they are not as efficient per unit area as monocrystalline panels - typically you would need nearly double the panel area for the same power output. Having said that, for a given power rating, they do perform better at low light levels than crystalline panels - which is worth having on a dismal winter's day.

Hybrid solar panels

At least one manufacturer now produces a hybrid panel, where a layer of amorphous silicon is deposited on top of single crystal wafers. This gives some of the advantages (high power, but still efficient at low light levels) - and some of the disadvantages (not flexible and relatively high price) of the different types of panels.

Top solar myths

It takes more energy to produce a solar panel than the solar panel produces in its lifetime.

Not true. Typical energy payback periods are about 3 or 4 years, and are quickly reducing as more energy efficient production methods are used. For the manufacture of the very earliest solar panels there was some truth in this myth as they used thick slices of pure silicon crystal, which are very energy intensive to grow. Modern panels use exceptionally thin slices of silicon crystal, which use correspondingly less energy in their production. 'Thin-film' solar panels - such as the Unisolar brand - are even less energy-intensive to produce.

Once solar panels have reached the end of their life they can (and should) be recycled. The process involves stripping the module down to its basic components - glass, ferrous and non ferrous metals, plastic and silicon. These products can then be re used - helping to preserve precious resources. As solar PV is relatively new in the UK, there is a limited market. However, countries such as Germany have been installing PV panels in reasonably high volumes since the early 1990s. Some off these panels are nearing the end of their life - so we can expect improvements in the PV recycling market over the coming years.

Amorphous solar panels degrade quickly - they only last about five years.

Not true - or at least, not true for all thin-film panels. Most modern thin-film panels have usable lifetimes which are almost as long as conventional crystalline panels. Unisolar guarantee the power output of their panels to be at least 80% of rated capacity after 20 years - they wouldn't be that daft to guarantee it if the panels had a usable lifetime of only 5 years!

Be aware though that some cheaper, thin-film panels out there are based on different chemistries, and these may have a significantly shorter lifespan than conventional silicon panels.

You would have to cover the entire country in solar panels to generate enough for the UK's electricity demand

UK annual energy consumption is around 350,000,000,000 kWh. An 85W solar panel measuring 1m x 0.65m produces just over 200 kWh per year in the UK climate - so you would need around 1,750,000,000 such solar panels to meet UK demand. They would cover an area of 1137500000 square metres, or 1137.5 square kilometers. That's a square 34 kilometers by 34 kilometers, or just over 20 miles by 20 miles.

If that still boggles your mind, how much do you think you personally need? Each person in the UK averages a little under 6000 kWh per year, which could be provided by 30 of our 85W solar panels. That's 19.5 square metres, or a square 4.4m by 4.4m - probably smaller than your roof.

The cost for the panels, incidentally, would be about £2K.

How solar panels work

Solar cells are made of silicon semiconductors, very similar to those used in transistors and electronic chips. A solar cell has two layers of silicon. The lower layer is silicon doped with boron atoms: boron atoms have one less electron than silicon, and so there is a shortage of electrons in this layer. The upper layer is doped with phosphorus atoms, which have one more electron than silicon.

Layers of silicon doped in this way exhibit what is called the photovoltaic effect. A photon of light hitting the top layer can knock one of the spare phosphorus electrons across the junction into the lower layer. The lower layer has therefore become negatively charged with respect to the upper layer, and so there is a potential difference (or voltage) across the two layers. Attach a couple of wires, and, say, a light bulb, between the two layers, and bingo! The electron can flow through the wires back to the top layer where it came from, and light the light bulb in the process.

The more light hitting the cell, the more electrons get knocked across, and the more power the cell can produce - although in practice even the best cells are barely 20% efficient, so only one in five of the photons is actually doing any work.

A single cell can only create a small voltage difference, and so to get a useful voltage from the panel it is usual to connect a number of cells in series.