solar panels perth australia image


Over Sizing a solar inverter.

Oversizing is putting more panel power on an inverter than its rated maximum.

For example 6.6kW of panels on a 5kW inverter.

The Clean Energy Regulator allows you to claim STCs, which are the Government discount for solar panels as long as the inverter is not oversized by more than 33.3%, hence a 5kW inverter is fine with 6.6kW of panels. The reason is that in real World conditions, panels typically produce 25% less than their rated output, so a 330W panel at STC (rated output) would be expected to produce a peak of 247W in the real World (NOCT).


The exception to this rule is where you install a battery inverter and DC couple a battery directly to it. Let's take the Huawei M0 battery inverter as an example. As you can see from the datasheet (blue link), Huawei recommend no more than 10.25kW of panel power can be connected to their 5kW inverter.


What this means is that if you had the Huawei battery plugged in, then when the solar panels are producing, the inverter can convert 5kW of that power into AC to power loads in your house, and the other 5.25kW can be stored directly into the battery. The really good news is that the afore mentioned Clean Energy Regulator allows the STCs (discount) on these extra panels, so while your neighbours are making do with a 5kW non battery inverter, 6.6kW of panels, and a Tesla Powerwall (which has a second inverter built-in), you can have a 5kW Huawei inverter controlling 10.25kW of panels and a Huawei battery, and pay many thousands of dollars less for a far more powerful and efficient solution.


Now, at risk of sending you to sleep, the reality is that at the moment, you can't get 10.25kW of panels on the Huawei, you can only get 9.62kW if you install LonGi 370W panels, or 9.36kW of JinKo 390W, or 8.38kW of most 330W panels.


The reason for why you can install more of one panel than another comes down to some regulations we have here in Australia about the maximum open circuit voltage of the string of panels. This is determined by the minimum temperature possible (in Perth -3°C) and the electrical specifications of the panel. The higher the rated output and the lower the VOC and Temperature Coefficient of VOC of the panel, the better, so two 13 x LonGi 370W strings can be installed on each of the two Huawei inverter trackers (MPPT) but only 12 x JinKo 370Ws on each string.


For those readers with too much time on their hands, the formula is like this.

600V is the maximum string voltage.

The temperature difference between the panel test at STC (25°C) and Perth's minimum of -3°C is therefore 28°C


The LonGi 370W panel (click blue link above for datasheet) has an Open Circuit voltage at STC (25°C) of 40.9V, and a temperature coefficient of VOC of 0.27%.


So to work out how much extra voltage is created at -3°C for the LonGi 370W panel we take the temperature difference (28) and multiply that by the panels temperature coefficient 0.0027 and then multiply that result by the panels open circuit voltage (VOC)


(28 x 0.0027) * 40.9 = 3.092 V extra voltage at -3°C


Then add that to the original VOC of 40.9 = 44 Volts per panel


Divide 600 V (maximum allowed) by 44 and you get 13.6, rounded down to 13 panels on the string as you can't cut up a panel into 0.6 of one. 13 x 44 = 572V. Tick. 

13 x 370W = 4.81kW per string x 2 = 9.62kW.


The JinKo 370W panel has a VOC of 44.6 V and a temp coefficient of 0.29% so

((28 x 0.0029) * 44.6) + 44.6 = 48.22 Volts per panel

600 / 48.22 = 12.44, or 12 panels per string. 24 x 370W = 8.88kW


Wake up now, it's all over. Most people dislike maths.
Whatever you do, don't read what follows as it's even worse.


The above formula has always left me scratching my head.

Here's why.


Firstly, temperature. The STC test uses the cell temperature inside the panel. Typically operating solar cells are about 20°C hotter than the ambient air temperature. So in the STC the air temperature is likely to be about 5°C. So why do we calculate 25°C with the minimum air temperature of -3°C to get a temperature difference of 28 when it should be 5°C to -3°C giving a difference of 8°C? It's comparing apples with oranges.


Secondly, irradiance. The formula assumes that the solar irradiance will be a whopping 1000Watts per square metre when the air temperature is -3°C. Not a chance in Perth. Perhaps in Winter on a Swiss mountain at mid day, but not at 7am on a frosty Perth morning in July. It would be more realistic to calculate it using the NOCT value of 800W per sqm, and even that is very high solar irradiation for us here in Perth when we hit our minimum temps.


Thirdly, this is measuring VOC, or Open Circuit Voltage. As soon as the inverter has started up, after perhaps a minute, then the circuit is closed, and the panel voltage drops down by about 15%. Why is VOC not Vmp used?


So, if someone cleverer than I can explain First and Third above I'd be very happy to receive your email. The second part, irridiation, I totally get that. A formula has to be made that works all over the World. Many places can simultanously have very strong solar irradiance and very low temperatures, but it is a bit frustrating to measure voltages here in Perth on really cold mornings and see the voltages are far below what the formula allows. People could install more panels !!