1.
Introduction
Photovoltaic (PV) hot spots are a well-known phenomenon that were described as early as 1969[1] and are still present in PV modules[2–4]. The?mechanism?of hot spot generation is shown in Fig. 1, which shows that if the string working current is higher than the short-circuit current of the shaded cell, the terminal voltage of the shaded cell becomes reversely biased[5–8]. The shaded cell works as a load to dissipate power instead of delivering it and, therefore, operating at abnormally high temperatures[9, 10]. This increase in the temperature of shaded cells will gradually degrade the output power generated and accelerate degradation of PV modules[11–14]. Katherine et al.[15] explored the relationship between the number of cells in series and the potential for hot spotting through simulation research. They found that shorter strings can reduce hot spot risk, but none of the cell types were immune to hot spotting and the temperature rise worsens as string length increases[15]. Limet al.[16] found that there is a positive?correlation?between?reverse bias?voltage of the shaded cell and the hot spot severity. It indicates that the reverse bias?voltage of the shaded cell is an important factor for hot spots.
onerror="this.onerror=null;this.src='http://www.jos.ac.cn/fileBDTXB/journal/article/jos/2018/12/PIC/18050011-1.jpg'" class="figure_img" id="Figure1"/>
Download
Larger image
PowerPoint slide
Figure1.
Schematic diagram of hot spot generated mechanism.
Bypass diodes are used to mitigate the hot spot problem currently[17]. The bypass diode works as a parallel?branch to make the string working current flow through it instead of the shaded cell therefore protecting it from hot spots. In present systems, one PV module configures with 3 or 4 bypass diodes (each diode protects about 20 cells) in consideration of the cost and the limitation of packaging technology. However, this configuration of a bypass diode cannot avoid hot spots and wastes lots of power in some shading conditions[18]. Alrawiet al.[19] presented that the reliability of PV modules increases as the number of bypass diodes increases. More configuration modes of bypass diode are presented as the cost of diode decreases. Li et al.[20] presented a configuration mode of 8 bypass diodes that took into account the cost of the diode and the module output power comprehensively. Afterwards, each cell in a PV module was configured with an individual bypass diode, which has been manufactured and it was found that hot spots can be eliminated effectively when a shaded or faulty cell exists in the module[21]. The efficacy of a weakening hot spot varies with different bypass diode configuration modes, and a bypass diode cannot thoroughly eliminate hot spots[22].
The paper originally shows how the configuration of the bypass diode in the PV module impacts on hot spots and concretely uncovers the relationship among reverse bias voltage of shaded cell, hot spot probability, and the severity and number of bypass diodes. Five commercial polysilicon PV modules (60 solar cells connected in series) configured with different numbers of bypass diodes are used in this experimental research. We test the terminal voltage of shaded cells in 11 kinds of shading cases using a multimeter and the temperature of hot spots with a FOTRIC 255 infrared thermalgraph. Electronic load is used to adjust the working conditions of PV modules in this experiment. The conclusion of this paper provides guidance for the studies of hot spot solutions in the future, and it offers engineers of the PV system a well-informed reference for the manufacture of PV modules with high efficiency and reliability .
2.
Experiment
2.1
Experimental modules
In this work, 5 commercial polysilicon PV modules that are composed of 60 cells in series are used for the experiment. The polysilicon cell has a peak efficiency of 16.83 % with 4 main grid lines and size of 243 cm2. The individual PV module outputs 36 V open-circuit?voltage, 8.6 A short-circuit current and peak power of 275 W at standard testing conditions. The 5 PV modules with different numbers of bypass diodes are designated as No. 1–No. 5, and they are shown in Table 1. Fig. 2 shows one of the 5 PV modules where each cell is configured with an individual bypass diode respectively.
| No. | Numbers of bypass diodes | Numbers of cells that each bypass diode protects |
| 1 | 0 | – |
| 2 | 1 | 60 |
| 3 | 3 | 20 |
| 4 | 15 | 4 |
| 5 | 60 | 1 |
Table1.
The No. of PV modules and the detailed configurations of bypass diode.
Table options
-->
Download as CSV
| No. | Numbers of bypass diodes | Numbers of cells that each bypass diode protects |
| 1 | 0 | – |
| 2 | 1 | 60 |
| 3 | 3 | 20 |
| 4 | 15 | 4 |
| 5 | 60 | 1 |
2.2
Experimental procedures
A positive?correlation?between?the reverse bias?voltage of the shaded cell and the hot spot severity was uncovered in Ref. [16], and therefore the reverse bias?voltage of the shaded cell is a sensitive indicator for hot spot generation. The relationship between configurations of bypass diode and reverse bias?voltage of shaded cell is firstly studied by way of testing the terminal voltage of shaded cells in 5 PV modules under different shading cases and working conditions. Then the severity of hot spot is measured using No. 3 module to determine the necessary conditions of hot spot generation by successively shading different areas on the cells according to the formerly tested results. The minimum shading areas are determined based on the necessary conditions of hot spot generation obtained in the second step to deduce the probability of hot spots. The detailed experimental procedures are designed as follows.
(1) Terminal voltage of shaded cells in 11 shading cases
Opaque baffles are used to shade the same position of the five modules in the conditions of open circuit, short circuit, and maximum power point. Shaded areas are 0, 1, 3, 5, 10, 24, 36, 49, 61, 121, and a maximum of 243 cm2, respectively. The terminal voltage of the shaded cell is recorded using a multimeter. The PV modules used in the experiment are free from an aluminum frame and for the sake of measurement convenience, the electrodes of cells located in both sides of the module are pre-welded before encapsulation, as shown in Fig. 2(b).
(2) Hot spot measurement of No. 3 module
Connect the cathode and anode of the No. 3 module to keep it in the short circuit condition. Shading of the module is implemented in the following successive steps to ensure the terminal voltage of the shaded cell and its relevant bypass diode to meet specific conditions: (1) small shading (cell with positive voltage and diode switched-off); (2) medium shading (cell with small negative voltage and diode switched-off); (3) large shading (cell with large negative voltage and diode switched-off); (4) total shading (cell with very large negative voltage and diode switched-on). The working conditions of the bypass diode are confirmed by testing the terminal voltage of the bypass diode, and the forward conduction voltage of the bypass diode is 0.45 V in this experiment. Keep the No. 3 module in the four conditions for 5 min to sufficiently render the module subjected to a hot spot or not, that can be revealed by infrared thermalgraph, then record the temperature of the shaded cell.
(3) Determination of minimum shading areas
Test the minimum shading areas that the bypass diodes can be switched on of No. 2–No. 5 modules in the short circuit condition and the maximum power point output condition, respectively. In addition, the minimum shading areas that make the terminal voltages of shaded cells in No. 1–No. 5 modules negative are also determined in the aforementioned two conditions.
3.
Results and discussions
Three common working states of commercial PV modules installed outdoors are short-circuited, open-circuited, and power outputted in the maximum value by way of MPPT (Maximum Power Point Tracking) controlling[23,24]. Maximum current output of the PV module in the short circuit condition is most likely to suffer from hot spots, and state of maximum power output is the pattern where PV modules operate at most of the time. The condition of open circuit can be a reference case in this measurement.
Five PV modules connected with different numbers of bypass diodes exhibit drastic difference in terms of terminal voltage of the shaded cells. From Fig. 3, one can clearly find that the diode signifies being of no use at all when all the series cells in an individual PV module are protected by only 1 bypass diode, for the 0 diode and 1 diode curves are precisely overlapped in three cases. However, it should be noted that the 1 diode configuration mode can play an important role at PV module string levels, in which condition the bypass diode can provide a parallel?branch for shunting the string current. The terminal voltage of the shaded cell is always positive in the open circuit state even with a high amount of shading areas as shown in Fig. 3(a), and thus, a hot spot never occurs in this case on the basis of the research in Refs. [5–8,16]. However, in states of short circuit and maximum power output as shown in Figs. 3(b) and 3(c), the terminal voltage of the shaded cell varies from about a positive 0.5 V to a maximum reverse bias value as shading areas increase from 0 to 243 cm2. There is a drastic difference in terms of terminal voltage of the shaded cells for five PV modules connected with different numbers of bypass diodes. We consider an N-cell string that is in parallel with one bypass diode. When one cell in this string is individually shaded, the terminal voltage of the shaded cell depends on the other N – 1 cells in this string. At the standpoint of charged carriers, the photo-generation rate of carriers in the shaded cell is reduced, leading to an accumulation of excessive electrons in its anode and holes in its cathode that are generated from other remnant normally worked cells. The terminal voltage of the shaded cell thus becomes reversed and will turn into a larger degree as the number of remnant cells in the same string increases because of more contribution of excessive carriers. Therefore, different numbers of cells protected by one bypass diode in the five PV modules lead to a drastic difference in terms of the terminal voltage of the shaded cells. An obvious characteristic can be found that all the terminal voltages of the shaded cells in 5 PV modules convert to reverse bias when the shading areas are greater than 5 cm2, which means that there is a large probability for hot spot generation in this condition. The terminal voltage of the shaded cell will reach a stable value as shading areas are greater than 10 cm2 for the 15 diodes pattern and 24 cm2 for the 60 diodes pattern, respectively, as clearly shown in Figs. 3(b) and 3(c). Hot spots can thus be avoided as a result of the switched-on bypass diode. In addition, another characteristic can be drawn clearly by comparing 5 curves that the bypass diode are more easily switched on and results in a decreased reverse bias voltage of shaded cell as the number of bypass diodes increases. Assume the forward conduction voltage for the bypass diode is a constant value of UD, the voltage of a normally working cell is a constant value of U0, and the number of cells protected by each bypass diode is N. According to Kirchhoff's voltage law, the terminal voltage of a shaded cell is a variable Us, and can be calculated by the following equation:
${U_ { m{D}}} = {U_ { m{S}}} + left( {N - 1} ight) {U_0},$  | (1) |
where the UD is the forward conduction voltage for the bypass diode, U0 is the voltage of the normally working cell, and N is the number of cells protected by each bypass diode, because UD and U0 are constant values. In general, N decreases as the number of bypass diodes increases. So it needs a bigger Us to make the bypass diode switch on, that means the reverse bias voltage of the shaded cell is smaller when Us is ?3 V. All the results obtained above have reference significance for the next discussions.
The necessary conditions for hot spot generation are found by the elaborately designed experiments described in Section 2.2. When shaded areas of the No. 3 module are chosen in increased sequence as 3, 6, 20, and 243 cm2, the corresponding terminal voltages of the shaded cell are 0.37, ?4.64, ?9.22, and ?11.23 V. At the same time, it is found by measuring the terminal voltage of the bypass diode where the bypass diode is switched off in the first three shading conditions but switched on in the last shading condition where the cell is totally shaded. The PV module was kept shaded in each case for 5 min. Photos were taken of the No. 3 module sequentially by an infrared thermalgraph to determine whether or not hot spots have occurred and the temperature of them. The captured pictures are shown in Fig. 4. There is no hot spot in the shading case of the 3 cm2 area in Fig. 4(a) due to the positive terminal voltage of the shaded cell, according to the discussion in Figs. 3(b) and 3(c). From Figs. 4(b) and 4(c) one can clearly find that a hot spot has generated in the shading areas of 6 and 20 cm2, respectively. This is because the terminal voltage of the shaded cell is reversely biased but the bypass diode is switched off. As shading is further increased to the total area of the cell, the bypass diode can be instantly switched on, thus avoiding hot spot generation, which can be seen from Fig. 4(d). All the experimental results explicitly indicate the fact that the negative terminal voltage of the shaded cell accompanied by a switched-off bypass diode are the necessary conditions for hot spot generation. Furthermore, the temperature measurements of the shaded cell under 6 and 20 cm2 shading area respectively verify a positive relationship between the severity of hot spot and reverse bias voltage of cell, which is consistent with the results obtained in Ref. [16]. An extremely high temperature above 80 °C shown in Fig. 4(c) is indeed of high risk for outdoor modules, for 20 cm2shading just equals to the size of common leaves.
onerror="this.onerror=null;this.src='http://www.jos.ac.cn/fileBDTXB/journal/article/jos/2018/12/PIC/18050011-4.jpg'" class="figure_img" id="Figure4"/>
Download
Larger image
PowerPoint slide
Figure4.
(Color online) The infrared thermograph of the No. 3 PV module in the state of (a) 3 cm2 shading area and the temperature of shaded cell is 37.8 °C. (b) 6 cm2 shading and the temperature is 64.6 °C. (c) 20 cm2 shading and the temperature is 82.1 °C. (d) Total shaded and the temperature is 38.1 °C.
With the aforementioned discussion on two necessary conditions of a hot spot generating, we can define “hot spot voltage” to illustrate the voltage range where the bias voltage of the shaded cell is reversed, meanwhile, the bypass diode is switched off. All the voltage values within the range of “hot spot voltage” can definitely lead to the occurrence of a hot spot. Similarly, the range of shading areas that result in “hot spot voltage” are named as “hot spot occurrence area” and the ratio of “hot spot occurrence area” to the total cell area is defined as “hot spot probability”. To find the relationship between the specific “hot spot probability” and configuration mode of bypass diode, we determined the minimum shading area for switching on the bypass diode and the minimum shading area for reversing the terminal voltage of the shaded cell under conditions of short circuit displayed in Table 2 and maximum power output displayed in Tables 2 and 3. The measured cell is firstly connected to the Multimeter through lead wires and then subjected to physical shading, the area of which increases from 0 cm2. Meanwhile, from the terminal voltage of the shaded cell displayed by the Multimeter, the minimum shading area is recorded as the value enabling its corresponding voltage converting from positive to negative. The minimum shading area for switching on the bypass diode is determined in the same manner.
| Numbers of bypass diode | Conduction voltagea) (V) | Terminal voltageb) ( V) | Minimum switching areac) (cm2) | Minimum reversing aread) (cm2) | Area rangee) (cm2) | Hot spot probability (%) |
| 0/1 | – | – | – | 2 | 2–243 | 99.18 |
| 3 | 0.45 | ?11.232 | 41 | 3 | 3–41 | 15.64 |
| 15 | 0.45 | ?1.827 | 7 | 4 | 4–7 | 1.24 |
| 60 | 0.45 | ?0.446 | 6 | 5 | 5–6 | 0.41 |
Table2.
The test data in the condition of short circuit.
Table options
-->
Download as CSV
| Numbers of bypass diode | Conduction voltagea) (V) | Terminal voltageb) ( V) | Minimum switching areac) (cm2) | Minimum reversing aread) (cm2) | Area rangee) (cm2) | Hot spot probability (%) |
| 0/1 | – | – | – | 2 | 2–243 | 99.18 |
| 3 | 0.45 | ?11.232 | 41 | 3 | 3–41 | 15.64 |
| 15 | 0.45 | ?1.827 | 7 | 4 | 4–7 | 1.24 |
| 60 | 0.45 | ?0.446 | 6 | 5 | 5–6 | 0.41 |
onerror="this.onerror=null;this.src='http://www.jos.ac.cn/fileBDTXB/journal/article/jos/2018/12/PIC/18050011-2.jpg'" class="figure_img" id="Figure2"/>
Download
Larger image
PowerPoint slide
Figure2.
(Color online) (a) Sketch and (b) photograph of the PV module configured with 60 bypass diodes.
onerror="this.onerror=null;this.src='http://www.jos.ac.cn/fileBDTXB/journal/article/jos/2018/12/PIC/18050011-3.jpg'" class="figure_img" id="Figure3"/>
Download
Larger image
PowerPoint slide
Figure3.
(Color online) The dependence of terminal voltage of shaded cell changes on shading areas of five PV modules with different bypass diode configurations in (a) open circuit state, (b) short circuit state, and (c) maximum power output state.
One can learn directly from Tables 2 and 3 that hot spot problems cannot be eliminated utterly in PV modules no matter what configuration the bypass diode is in. The most sufficient protection mode that each cell in the module is individually paralleled by one bypass diode can reduce the probability of hot spots to a minimum of 0.41%. It should be noticed emphatically that a PV module configured with 3 bypass diodes in a state of maximum power point output has a high hot spot probability of 36.63%. Because this 3 diode configuration is the most common type for modules in power stations, it poses a significant threat on safety and efficiency of PV station operation. Furthermore, in view of the data in Tables 2 and 3, the “hot spot probability” decreases as the number of bypass diodes increases, that is to say, there is a negative correlation between “hot spot occurrence area range” and the numbers of bypass diode. The “hot spot occurrence range” is larger in the short circuit case than that in the maximum power output case for PV modules with the same configuration of bypass diode. It means the PV module has a smaller hot spot probability in the case of maximum power output. In addition, there are many objects such as ornithocopros, leaves, or ruderals etc., whose sizes are within the “hot spot occurrence area range” defined in the tables above, so the site selection of PV power station and the configuration choice of bypass diode are of great importance.
| Numbers of bypass diode | Conduction voltagea) (V) | Terminal voltageb) (V) | Minimum switching areac) (cm2) | Minimum reversing aread) (cm2) | Area rangee) (cm2) | Hot spot probability (%) |
| 0/1 | – | – | – | 4 | 4–243 | 98.35 |
| 3 | 0.45 | ?11.301 | 93 | 4 | 4–93 | 36.63 |
| 15 | 0.45 | ?1.834 | 18 | 7 | 7–18 | 4.53 |
| 60 | 0.45 | ?0.445 | 7 | 6 | 6–7 | 0.41 |
| a) Bypass diode forward conduction voltage. b) Terminal voltage of shaded cell with diode switched on. c) Minimum shading area for switching on bypass diode. d) Minimum shading area for reversing the terminal voltage of shaded cell. e) Hot spot occurrence area range. | ||||||
Table3.
The test data in the condition of maximum power output.
Table options
-->
Download as CSV
| Numbers of bypass diode | Conduction voltagea) (V) | Terminal voltageb) (V) | Minimum switching areac) (cm2) | Minimum reversing aread) (cm2) | Area rangee) (cm2) | Hot spot probability (%) |
| 0/1 | – | – | – | 4 | 4–243 | 98.35 |
| 3 | 0.45 | ?11.301 | 93 | 4 | 4–93 | 36.63 |
| 15 | 0.45 | ?1.834 | 18 | 7 | 7–18 | 4.53 |
| 60 | 0.45 | ?0.445 | 7 | 6 | 6–7 | 0.41 |
| a) Bypass diode forward conduction voltage. b) Terminal voltage of shaded cell with diode switched on. c) Minimum shading area for switching on bypass diode. d) Minimum shading area for reversing the terminal voltage of shaded cell. e) Hot spot occurrence area range. | ||||||
4.
Conclusions
This work designs experiments elaborately using five commercial polysilicon PV modules configured with different numbers of bypass diodes and presents that the reverse bias voltage of shaded cells, hot spot probability, and hot spot severity decrease as the number of bypass diodes increases. Two necessary conditions for hot spot generation are negative terminal voltage and simultaneously switched-off bypass diode. PV module configured with bypass diodes is sure to suffer from hot spot when its shading area lies within a specific range. Hot spot can occur for the most common PV modules configured with 3 bypass diodes when their shading areas are 3–93 cm2, and this kind of module has a high hot spot probability of 36.63% in maximum power output. In an extreme case that each cell has an individual bypass diode in a PV module, hazards of hot spot are inevitable under shading areas of 5–7 cm2, but the probability of hot spot occurrence is reduced to a minimum of 0.41%.
Acknowledgement
We thank the Baoding Lightway Green Energy Technology Co. Ltd. for providing experimental PV modules.
