One thing I became aware of by commissioning my balcony power plant is the importance of the voltage and current ratings of the panels: the PV panels used in my balcony power plant deliver a very low voltage 24.8 (Vmp) and in return a high current of up to 16.51A (Imp). When comparing the performance data with some inverters, I found that current inverters for the most part cannot handle more than 13A and would limit the maximum peak power. The Fronius Gen24 as an example can handle 25A on one MPP tracker, but only 12.5A on the second, which would lower the maximum possible power when using the panels on the 2 strings.

Attention: The maximum possible number of modules is limited by the temperature coefficient.

For planning purposes, the open circuit voltage of the modules is crucial. Especially at low temperatures the voltage could become a problem, because the voltage of the modules with decreasing temperature is higher than the specified open circuit voltage. As a concrete example, a Udc max of 1000 volts is given for the Fronius Gen24.

Source: https://www.fronius.com/~/downloads/Solar%20Energy/Datasheets/SE_DS_Fronius_Symo_GEN24Plus_EN.pdf.

As an example, the open circuit voltage when using Meyer Burger White 390 PV panels is : 44.5 V.

Source: https://www.meyerburger.com/fileadmin/user_upload/PDFs/Produktdatenblaetter/DE/DS_Meyer_Burger_White_de.pdf

The STC column reflects the standard test conditions: a solar irradiance of 1000W/m², at a cell temperature of 25°C.

For my system, I planned 19 modules per string, givingan open circuit voltage of 44.5Voc * 19 panels = 845.5V. At an ambient temperature of 25°C (according to STC). The inverter can Udc max 1000V according to the datasheet, so everything is in the green zone.

If the temperature drops, the voltage increases by the temperature coefficient Voc ß: -0.234% / K

Converted to an ambient temperature of -10 °C (35°C less than the 25°C STC temperature), the voltage would be higher than specified by 35 °C x 0.234 %/K = 7.02 V: 44.5 Voc + 7.02 V = 51.52Voc / module; With 19 modules: 1001.11V: With this setup, the open circuit voltage of the panels at a temperature of -10 °C exceeds the voltage specified by the inverter of 1000 V Udc max. 

Finally, a quick look at the current values:
At high temperatures, the voltage decreases, but the current increases. The maximum current would be 10.3A + ((70°C-25°C) x 0.033α) for an assumed 70°C = 11.78 A. The maximum current Idc max for the Fronius GEN24 is given as 12.5 A for the second MPPT. While this value can be exceeded without destroying the inverter, it still cannot handle more than the 12.5A, which can lead to a reduction in power for certain panels.

Re-measured with a multimeter and thermometer

I measured the voltage of 19 panels in series with a multimeter: this is 786 V at 36°C under the module. I measured the temperature with an Aquara Zigbee sensor.

According to the datasheet, the voltage drops by 0.234 V per °C. At 11°C difference (36°C panel temperature - 25°C STC temperature) by 11°C x 0.234 %/K =2.574 V/panel.
So the open circuit voltage per panel at 36°C would be 44.5V (open circuit STC) - 2.574V (11°C x 0.234ß Voc) = 41.9 V.

The calculated value for 19 panels at 36°C is 41.9V x 19 panels = 796.6 V. The difference to the actual measured voltage of 786V is quite close at this temperature. The deviation can be explained by measurement inaccuracies of the multimeter and thermometer, as well as its position, as the cell temperature may deviate slightly from the temperature under the panel. 

My measurements in winter were different: At an outside temperature of 4°C in the shade, the open-circuit voltage of the multimeter shows 833V. This value corresponds to the voltage value when the inverter is switched on. The calculated voltage with the STC values would be 895.85V. If I use the NMOT values, which should be closer to the real conditions, the calculated voltage still comes to 872.9V and is therefore still considerably higher than the measured value.  Evenafter the first winter, I found that the previously calculated open-circuit voltage was far from being achieved in practice. The highest recorded voltage of my east-west system was only 838 volts. The south-facing balcony power station was also nowhere near the theoretically calculated voltage. Nevertheless, to be on the safe side, I would design the system to at least 0°C to avoid any problems at this point. To make it easier to calculate the values, I have put together an online calculator:

Online calculator

Conclusion

The selection of the inverter and the appropriate PV panels is essential for optimal operation. Especially for PV modules with a current higher than 13A, the maximum possible peak power could never be reached when using some inverters. On the other hand, high voltages limit the maximum possible number of PV panels in a string. On the voltage side, sufficient buffer should be planned so that the inverter is not overloaded at low temperatures. For general planning of a PV system, see also: PV - Considerations - Planning - Implementation and Fronius Gen24 Commissioning in practice: Step by step