How do you measure the operating conditions in your PV plant?


How to measure the operating conditions in PV plants? What sensors to use? What advantages and disadvantages do they have? What precautions to take? In this article we review the common practices and present how to get the most out of your data in an aspect that still offers possibilities for improvement when developing PV plants.

The evaluation procedures of photovoltaic (PV) plants are based on determining their efficiency or, in other words, on comparing the input and output of the system. The input is given by the effective irradiance that reaches the generator and the operating temperature of its cells, both known as the operating conditions. In turn, the output of the system is given by the energy produced. Therefore, correctly measuring the operating conditions is essential to accurately assess the quality and performance of PV plants.

To measure the operating conditions, it was usual to use pyranometers for the irradiance measurement (both in the horizontal plane and in the plane of array) and thermocouples or PT-100 for the cell temperature measurement. Then, the technology development derived in using also specific PV devices, such as reference modules and cells.

On the other hand, the development of advanced data acquisition systems for monitoring promoted the simultaneous installation of several sensors in the same plant [1]. The positive side of this redundancy is a greater robustness: for example, if communication with a sensor fails, there are more to turn to. However, the downside is that the use of different devices often results in different results, which adds considerable uncertainty when calculating the performance indexes of the installation, which vary according to the selected sensor [2].

Given this situation, how to know which sensors are preferable? What are their main advantages and disadvantages? In short, how to measure the operating conditions in PV plants? In the following sections, we will review the state of the art and see how, despite being a known aspect, the measurement of operating conditions still presents possibilities for improvement that are worth paying attention to when developing PV projects.

Different options to measure

The first consideration to ensure a correct measurement in the field is to pay attention not only to the measurement device but to also consider some additional precautions. On the one hand, not only the measuring devices but also the associated electronics must be properly calibrated. For example, voltage current converters, transducers, dataloggers… must guarantee a proper performance for the wide temperature and humidity ranges reached in the field. On the other hand, the soling deposition must be controlled, especially in arid climates or when there are nearby soiling sources (crops, unpaved roads, quarries...).

Measuring irradiance

The measurement of the irradiance can be done, mainly, with pyranometers and reference cells or modules. The difference between the former and the latter is that pyranometers measure the global irradiance while cells and modules measure the effective irradiance, which is the result of correcting the global irradiance with the angular and spectral responses of PV devices.

What is relevant in this context is that pyranometers are a suitable device for measuring horizontal irradiance, whose objective is to compare it with the irradiation data used in the design of the PV plant, as also these data sources (satellites, meteorological stations ...) measure the global irradiance. However, pyranometers are no longer adequate to measure the effective irradiance that effectively reaches the PV generator, which is the relevant variable when calculating the performance indexes of the installation. In this case, it is necessary to perform angular and spectral corrections, which imply, in general, an uncertainty increase of up to 3%.

Measuring with PV devices

This uncertainty addition can be directly avoided if measuring the effective incident irradiation with PV devices (cells or modules). Both of them should be of the same technology as the PV generator in order to guarantee the same spectral and angular responses and to lead to more accurate and repeatable results. However, there are still some differences between cells and modules that make the latter more recommendable as reference sensors:

  • Robustness against soiling: on the one hand, soiling distribution in the reference modules is expected to be similar to that of any other module, making it more representative of the PV generator soiling distribution patterns. On the other hand, located soiling (sand accumulation, bird droppings, mud...) has a limited effect on the PV module... However, neither of these two statements is assured for a reference cell: it is directly affected by any dirt, localized or not, and its smaller size can make its soiling distribution patterns to be different than that of PV modules.
  • Repeatability: the response of a PV module is more stable and repeatable than that of any other sensor [3].
  • Temperature measurement: many commercial solar cells include the measurement of the operating temperature, usually by means of thermocouples. This temperature measurement is disregarded due to its low representativeness, as the heat dissipation mechanisms of solar cells are different from those of PV modules (e.g., the area/volume ratio is larger in cells, which makes them cooler than modules).
  • Manufacturing and calibration: reference PV modules are manufactured following strict quality standards that guarantee both their calibration and long term durability in the field, which is not the case for reference cells. For example, round-robin tests performed in European laboratories have shown calibration accuracy better than 2% for crystalline silicon modules [4].


Figure 1 shows the comparison between the irradiance measured by a reference PV module and three reference cells in a PV plant. One of the reference cells was a secondary standard recently calibrated at a European round robin of solar radiation sensors. The other two were commercial devices after two years of sun exposure. Table 1 summarizes the irradiation measured: the difference between the secondary cell and the PV module was as low as 0.6% while the other two reference cells registered 4.6% and 5.1% less irradiation than the PV module. This result is can be due to a calibration problem, possibly increased by an early degradation of the reference cells.

Comparison between the irradiance measured by a reference PV module and three reference cells
Figure 1: Comparison between the irradiance measured by a reference PV module and three reference cells: in blue, the secondary standard.

Irradiation (kWh/m2) Difference (%)
Ref. PV module 2,052 -
Secondary ref. cell 2,065 0.6
Ref. cell 1 1,961 -5.1
Ref. cell 2 1,970 -4.6
Table 1: Irradiation measured by a reference PV module and three reference cells.

Analogous calibration experiments carried out in reference modules of 10 Spanish PV plants showed similar results after 5 years of operation[5]. The maximum degradation rate found was 1.2%, which is below the measurement uncertainty. Moreover, in 8 of the cases the difference was smaller than 0.2%, what gives an idea of thelarge stability of PV modules and their calibration along time.

Cell temperature

A common way of measuring the cell operating temperature is using thermocouples or PT 100 attached to the rear part of the PV module. Other option is to take advantage of the dependency of the PV module’s open-circuit voltage with temperature to use a reference module as sensor. This constitutes a preferable alternative due to the following reasons:

  • Representativeness: the value given by a PV module is the equivalent temperature of the whole device, thus representing the average temperature of the series of cells. Even under normal operation and depending on the size of the module, it is common to find temperature differences between cells ranging from 4°C to 10°C. According to the QPV’s experience, this temperature difference can be divided into 4°C due to voltage variations and 6°C due to dissipation differences between cells, and is not related to any defective performance. Therefore, an equivalent temperature is more representative than any measure taken only at a single point (which is the case of thermocouples).
  • Measuring point: thermocouples measure the temperature of the rear surface of the PV module, which is not necessarily the same as the internal working temperature of the cells. The application of correction coefficients, based on specific experimental cases, entails some additional uncertainty. This is avoided when using PV modules, which give directly the effective operating temperature.
  • Stability: Thermocouples measurement stability in the field is doubtful. It is very common to find detached or bad attached devices after some months of installation. Even if they remain correctly attached, the thermal variations to which PV modules are subject to can give rise to contact deterioration in the module-thermocouple interface, which translates into a defective registered temperature. On the other hand, PV modules stability over time is guaranteed by their manufacturing process.
  • Dispersion: some studies have shown that temperature measurement dispersion is much lower when using reference PV modules than when using thermocouples[3],[6].
  • Signal characteristics: the larger signal amplitude of reference PV modules makes them sturdier against the noise effects associated to the signal transmission.


Figure 2 presents the comparison of the cell temperature measured by two reference PV modules and three thermocouples at a PV plant while Table 2 shows the corresponding results in terms of daily equivalent temperature. A large coincidence is observed between modules (0.3% difference) while significant differences arose with the thermocouples (between 4.0% and 8.7%), although they were perfectly attached to the module.

Comparison between the operating cell temperature measured by two reference PV modules and three thermocouples
Figure 2: Comparison between the operating cell temperature measured by two reference PV modules (grey dots) and three thermocouples (red, blue and green dots).

Measurement Ref. PV module 1 Ref. PV module 2 Therm. 1 Therm. 2 Therm. 3
Eq. temperature (°C) 38.7 38.8 35.3 36.8 42.1
Difference (%) - -0.3 -8.7 -4.4 8.7
Table 2: Daily equivalent temperature measured by two reference PV modules and three thermocouples.

Again, calibration test of reference modules carried out in 10 Spanish PV installations after 5 years of operation showed maximum calibration differences of 0.9% and mean differences smaller than 0.2% [5]. This stresses the great stability of these devices

Using calibrated PV modules as sensors

As we have seen, reference modules are the best alternative for measuring the operating conditions, not only for the precision and repeatability of their measurement but also for their durability and long-term stability. They must be modules of the same technology (generally from the same manufacturer and type) as those constituting the PV generator, but previously stabilized (exposed to more than 60 kWh/m2) and carefully calibrated in an independent laboratory or in the field. In addition, when they are supplied as part of the batch of modules of the plant, their availability and guarantee are assured. Last but not least, they are a totally economically competitive alternative.

Figure 3 shows examples of reference PV modules installed in commercial PV plants with static structures, single and double-axis tracking, as well as c-Si, CIGS and CdTe PV module technologies.

Reference PV module example Reference PV module example
Reference PV module example Reference PV module example
Reference PV module example Reference PV module example
Reference PV module example Reference PV module example
Figure 3: Example of reference PV modules installed in commercial PV plants. Static structures, single and double-axis tracking systems are presented, as well as c-Si, CIGS and CdTe PV module technologies.


[1] Fuentes, M. et al., 2014. Design of an accurate, low-cost autonomous data logger for PV system monitoring using ArduinoTM that complies with IEC standards. Solar Energy Materials and Solar Cells, 130, pp.529–543.

[2] García, M. et al., 2014b. Solar Irradiation and PV Module Temperature Dispersion at a Large-scale PV Plant. Progress in Photovoltaics: Research and Applications.

[3] Caamaño-Martín, E., Lorenzo, E. & Lastres, C., 2002. Crystalline silicon photovoltaic modules: characterization in the field of rural electrification. Progress in Photovoltaics: Research and Applications, 10(7), pp.481–493.

[4] Herrmann, W. et al., 2007. Advanced intercomparison testing of PV modules in European test laboratories. In 22nd European Photovoltaic Solar Energy Conference.

[5] Moretón R. (2016). Contributions to uncertainty reduction in the estimation of PV plants performance. PhD thesis, ETSIT, Universidad Politécnica de Madrid.

[6] Alonso-García, M.C., 2000. On the NOCT determination of PV solar modules. In 16th European Photovoltaic Solar Energy Conference.