Back-of-Module Temperature Measurement Methods
Inside this Article
The accurate measurement of PV module temperature is important for tasks ranging from the determination of normal operating cell temperature (as performed by testing laboratories and manufacturers) to the performance analysis of utility-scale generation plants. Module temperature is a key input to performance models and is essential for the translation of I-V curve data to standard test conditions (STC). Unfortunately, this measurement is difficult to complete with accuracy and is highly dependent on the method by which the measurement probe is attached to the module.
In this article, we present results from a series of empirical tests at the National Renewable Energy Laboratory (NREL) that focused on the method used to attach back-of-module temperature measurement devices to a simulated PV module. The results from this evaluation provide suggested best practices for system installers and operators to successfully monitor module temperatures and to understand the implications of various attachment methods.
NEED FOR ACCURATE MEASUREMENTS
Why do module temperature measurements matter, and why should anyone be concerned with the method of attachment? The temperature of the module—specifically, the temperature of the solar cell junction—impacts the energy production of the module. Temperature coefficients are usually stated by module manufacturers in terms of the effect of temperature on short-circuit current (a), open-circuit voltage (ß) and power (γ), and may be listed in either absolute terms (amperes, volts or watts per °C) or relative terms (% per °C). As the parameter most applicable to system performance analysis, typical values of the relative temperature coefficients of power for various module technologies are listed in Table 1.
Translation equations provide you with a working knowledge of how a module behaves in differing thermal environments. Beyond that, they include temperature coefficients that are used to calculate a module’s electrical characteristics at an arbitrary temperature condition using data measured at a different temperature, such as at a standard reference condition. Standard translation equations are as follows:
Using absolute temperature coefficients:
ISCcorr = ISC + a(T2 - T1)
VOCcorr = VOC + ß(T2 - T1)
PMAXcorr = PMAX + γ(T2 - T1)
Using relative temperature coefficients:
ISCcorr = ISC x [1 + a(T2 - T1)]
VOCcorr = VOC x [1 + ß(T2 - T1)]
PMAXcorr = PMAX x [1 + γ(T2 - T1)]
where ISC is the short-circuit current measured at temperature T1 and ISCcorr is the short-circuit current translated to temperature T2 (see References 1 and 2). Similar definitions apply to VOC, VOCcorr, PMAX and PMAXcorr.
For accurate performance monitoring, modeling and assessment of warranty claims, it is important to know the module temperature coefficients and the appropriate translation of measured temperatures. System operators often use module temperature measurements, in addition to electrical and meteorological data, to commission the system and to predict the output of large-scale systems. An inaccurate measurement of module temperature, which is typically low as compared to reality, results in an overprediction of expected power output. This is due to the negative value of γ, the temperature coefficient of power, which indicates that an increase in temperature results in a decrease in power. For instance, as shown in Figure 1, a measurement that is low by 5°C may result in an overprediction of expected dc power by about 2.25%, a significant amount for large systems.
Beyond performance monitoring, module temperature may be used in degradation rate calculations for systems and modules. An erroneously low temperature measurement during the review of module I-V traces for analysis or warranty claims could trigger an unnecessary module replacement or impact other analyses. This is because comparisons are made against module STC parameters (reported at 1,000 W/m2 irradiance, AM1.5 spectrum and 25°C module temperature) that necessitate translation of the I-V curves from the measured field conditions. Temperatures that are underreported lead to errors in calculated power when translating back to STC, potentially yielding a lower power rating.