Tuesday, 17 June 2014

Power conditioner - doing so much more than just inverting

I've been using the terms inverter and power conditioner loosely and interchangeably for the past couple of years, but realise this may be very misguided. I'd assumed those boxes were simply turning the direct current of the panels into the alternating current running through my house and around the local grid.

I should have stayed awake for more of those lectures on power electricity.

Perhaps I did realise there was more going on inside the box, but since our roof is basically evenly exposed to direct sunlight, I've never worried about shading. I know that people recommend shading should be avoided. Recently I heard a story of a solar array that stopped working for a couple of hours a day as the shadow of a telegraph pole crossed it.

Looking at things in terms of current and voltage, and seeing the solar arrays as several solar cells in parallel and in series, it starts to make a bit more sense.

Taking the metaphor of electricity as water, we can roughly equate voltage with how high the water is, and current with how wide the channel is. So a tall, thin waterfall would have a high voltage and low current, while a low, wide waterfall would have a low voltage and high current. The amount of power in each could be the same, and in electrical terms, power is voltage times current.

So we can think of a solar cell a bit like a bucket, on your roof, filling up with drops of sunlight rather than water. And it has a hole in the bottom, so that the water can come out and be used. The power conditioner changes the size of the hole.

Actually, several cells are connected with each other in  parallel to make panels, and the panels are then connected together into columns of six, in our case, then four columns of six are connected in parallel. These then go into the power conditioner.

The relationship between the voltage and current is not straightforward, as you can maybe imagine with a hole in a bucket. If the hole is very big, then all the water is going to go straight out of the bucket, so you have no voltage, but a high current. If there is no hole at all, the voltage will get very high, until it's overflowing, but there will be no current. In electrical terms, these are the closed-circuit current, and open-circuit voltage. As you can imagine if you were fitting a little water wheel to the hole in your bucket, you may want to play around with the size of the hole to get the most power out of it.

The curve for a solar panel is something like this, where the voltage increases with a slight fall of current until the MPP where the current starts dropping very quickly. That's the Maximum Power Point, since the power is the voltage times the current.

The job of the power conditioner is to find that maximum power point, and keep the voltage and current there to get the most out of the panels. The only way it can find it is by experimenting with the load it applies to the system, tuning in to the ideal current. It would be great if it could know just from the voltage and current whether it is at the MPP, but since the curve changes depending on the angle of sunlight, cloud cover, temperature of the panels, dirt, shading and birds flying past, it can't find out without doing another experiment.

The strength of sunlight increases the current, but does not make a huge difference to the voltage.

A number of problems can happen. A solar panel is made up of several cells. If one cell is in shade, the current for that cell will drop. Since the cells are in series, that means the current for the whole panel will drop. To stop this problem, each cell has a bypass wire over the top.

This is what causes the telegraph pole problem, and the basic rule with solar panels is: don't put them in the shade. They should ideally never have any shade. But if they do, at least try to keep it to early mornings and late afternoons.

If some of the cells or panels in a column are shaded, the effect is the same as if all the cells in the panel, or all the panels in the column are shaded. If a bird flies past there could be be a dip in the current. If a bird landed on the roof and stayed there, this would cause problems, but this is unlikely to happen since the panels get really hot.

Which brings us onto the problem of temperature. The open-circuit voltage drops with temperature by around 1% per degree, and the current goes up a tiny bit.

I haven't actually measured the temperature of the panels on our roof, but I know the air underneath them can get up to 80 degrees. Since the air is cooling the panels down, the panels at the top of the roof are going to be slightly hotter than those at the bottom, so they will have different voltage characteristics. This shouldn't be a huge problem, since the optimum current is not going to be very different between panels, and the voltages will add up in series. But I need another post to look further into this question.

For now, the clear messages I've learnt are that shade should be avoided from all panels, and that the panels in an array should all be the same under the sun. And of course, if it's a partially cloudy day, the power conditioner is going to be very busy making constant re-calibrations of the optimum power.

The VI graphs are from mpoweruk.com. They also have a cool table of the efficiency you get from panels at different angles to the horizontal and pointing in different directions, for 35 degrees latitude, which may be useful for anyone living around that neighbourhood, which in fact I do. The only part of the UK near that latitude is Gibraltar.