Method for Determining an Equivalent Circuit for a Solar Cell
The present invention relates to a method according to Claim 1 for measuring the equivalent circuit of a solar cell.
The invention is used in the quality control of solar cells operating as sub-components in solar panels and for controlling the manufacturing process.
According to the state of the art, the electrical equivalent circuit of solar cells manufactured from crystalline silicon consists of five components, according to Figure 1. The first is a constant-current source Isun, the current of which is generated by solar radiation and the magnitude of which is directly proportional to the radiation energy. This component creates power, whereas the remaining four consume power. The components connected in parallel with the constant-current source Isun are two diodes and a shunt resistance, Rshunt, as well as a series resistance, Rser, which cuts the output power. The behaviour of one of the diodes is depicted by the term /^according to the Schokley equation while the recombination current of the depletion layer of the other diode is depicted by the term Irec, equation (1). In the equation, / and V are the current and voltage available from the circuit while q is the elemental electric charge, k the Boltzman constant, and Ethe temperature in Kelvin.
Conventional measuring devices presently on the market, which simulate the operation of solar cells, are typically used to measure characteristic curves according to Figure 2. The radiation power in so-called STC (Standard Test Condition) measurements is 1000 W/m2 and the temperature is 25 °C. In theory, the various components of the equivalent circuit can be found by applying the equation (1) mathematically, for example, to measurement results using the least squares method. In practice, this has proven to be
extremely difficult and, even in the best case, the equivalent circuit components obtained as a result are at most qualitatively mutually comparable. In the worst case, and unfortunately quite frequently, the components can even acquire non-physical values in these applications. These difficulties have led to the use of simpler and more limited methods, in which, for example, the series and shunt resistances are estimated from derivatives calculated from close to the points of intersection of the axes of the characteristic curve according to Figure 2. The result obtained is typically data that is only suitable for a qualitative examination, and which avoids the non-physical values sometimes obtained by the previous method. The quantitatively most reliable way to measure the series resistance is disclosed in, for example, the international standard IEC 891. This method requires the characteristic curve to be measured at three difference radiation intensities, and is therefore difficult in practice. The parallel resistance can be estimated reliably only by measuring the current- voltage curve of the solar cell in total darkness, using reverse direction voltages. On the other hand, there is no practicable method suitable for industrial use for the reliable differentiation of the physically most important diodes and for estimating their saturation currents.
A reliable estimate of the components of the equivalent circuit will provide extremely useful information in the research and development of solar-cell constructions and manufacturing processes. Known physical models of the components can be further used to estimate the ultimate material parameters dependent on the manufacturing process; such as compound contents and the lives of the minority charge carriers. In the actual production of solar cells too, the aforesaid information can be utilized for more thorough monitoring and control of the production processes. This is particularly important, as it allows each cell produced to be measured and analysed. The time available for this is typically only two seconds, which means that there is insufficient time to apply many of the measurement methods.
Finnish patent 106408 discloses a method for measuring a solar panel's characteristic current-voltage curve. The method can also be applied directly to measuring an individual solar cell. According to the method, a flash pulse 12 is directed onto the solar cell and the response caused by the flash pulse is measured by altering the electrical load during the duration of the pulse. According to the invention, the measurement is made using a peaked flash pulse, in such a way that the decreasing component 13 of the
pulse is used in the measurement for such a short time that the intensity of the pulse is unable to change substantially during the measurement. During that time the electrical load is altered to raise the voltage of the solar cell from zero to its maximum, so that the current will correspondingly drop from its maximum to zero and a characteristic curve according to Figure 2 can be formed from the said current and voltage signals. Figure 3 shows the flash pulse used in the aforementioned patent, which can also be used in the method according to the present invention. In the patent in question, the radiation signal itself is not measured, its time dependence being otherwise known.
The present invention is intended to eliminate the drawbacks of the state of the art and to create a new type of method, applicable to industrial use, for measuring the equivalent circuit of a solar cell.
In this invention, the time dependence of the radiation signal is measured. From the said signals, a characteristic curve according to Figure 2 can be produced, with the aid of the calculation method according to the standard IEC 891.
The invention is based on calculating the components, particularly the diodes and parallel resistance, of the equivalent circuit of the cell, from the voltage signal of the open circuit of the cell and from the radiation signal scaled as the power source.
More specifically, the method according to the invention is characterized by what is stated in the characterizing portion of Claim 1.
The method according to the invention has several advantages over existing measurement methods. The most important of these is the ability to estimate, with a single measurement and reliably, all the components of an equivalent circuit of a silicon solar cell, in connection with measuring the actual characteristic current- voltage curve. In turn, these data make it possible to estimate the magnitudes of the actual material parameters, such as the lives and compound contents, of the silicon material itself. In addition, it is now possible to make precise calculations of predictions of the properties of the cell for various temperatures and radiation conditions, on the basis of material- physical models, without having to first define and then apply average conversion coefficients. This is unique, and has previously been possible only with the aid of
special, extremely expensive measuring devices. A central property of the method is also that, as such, it can also be applied and used in full-scale production. The in-depth monitoring and analysis of m-mufacturing processes provides an opportunity for considerable cost savings and improvements in product quality.
In the following, the invention is examined with the aid of examples of embodiments according to the accompanying drawings.
Figure 1 shows the equivalent circuit of a solar cell according to the state of the art.
Figure 2 shows graphically the characteristic curve of a solar cell according to the state of the art.
Figure 3 shows a flash pulse used for measuring the characteristic curve of a solar cell, according to the state of the art, patent 106408.
Figure 4 shows graphically the current, voltage, and radiation signals obtained by the method according to the invention.
Figure 5 shows graphically the open-circuit voltage, obtained by the method according to the invention, scaled from the radiation signal as a function of the power-source current.
Figure 6 shows graphically the recombination-current temperature dependence obtained by the method according to the invention.
Figure 7 shows graphically the diffusion-current temperature dependence obtained by the method according to the invention.
The method according to the invention applies patent 106408 up to the moment in time T2, shown in Figure 3. Correspondingly, in Figure 4, the moment in time Ti is at the point 0 ms and the moment in time T2 at about the point 2.5 ms. The new and inventive feature of the present solution is that the measurement of the voltage and radiation signals is also continued after the moment in time T2, and typically for long enough for
the radiation intensity to have dropped to at least half, typically to about one-tenth of what it was at the moment in time T This is because after the moment in time T2 the cell is in an open-circuit state, so that no current comes from the circuit.
Figure 4 shows the current, voltage, and radiation signals measured using the method according to the invention. Because after the moment T2 current no longer flows out of the circuit of Figure 1 (7= 0), the equation (1) can be written in the following form:
Because the current produced by the current generator is directly proportional to the radiation signal, the radiation signal can be scaled as the current of the current generator, relying on the simultaneous measurement of the current and radiation signals when the cell is in a short-circuited state at the moment T\. Figure 5 shows the results of one measurement, in which the voltage is shown as a function of the current, obtained as a result of the aforesaid scaling. Equation (2) can now be reliably applied to this measurement material, for example, using the method of the least squares, obtaining as a result estimates for the terms Idiff, Irec, and RShunt- The application is made easier and more reliable by the fact that the exponential terms' exponents no longer have parameters, such as the series resistance, that must themselves be estimated. After this, it is easy to estimate the series resistance reliably, by applying equation (1) to the characteristic curve according to Figure 1 and exploiting the already calculated parameters.
The next embodiment example demonstrates the reliability of the method when estimating the exponential coefficients, Idiff and Irec>. The method according to the invention is used to measure the coefficients at seven different temperatures between 20 and 55°C. Their theoretical temperature-dependence functions are applied to the coefficients thus measured. Theoretical information on this can be found, for example, in: "Physics of Semiconductor Devices", S.M. Sze, Wiley-lhterscience 1969, pages 27,
643, and 648. Figure 6 shows the coefficients (Irec) of the recombination current and
Figure 7 the coefficients Idiff ) of the diffusion current as measured points, and the theoretically applied temperature dependences of the coefficients in question as unbroken curves.