I’m amazed at how far the solar industry has been able to advance without pinning this question down and making it widely known to the public. In fact, when initially asked this question, I thought I knew the answer, until I was only able to explain it when put on the spot as “well, it’s not good…you’ll lose a lot of power.”
“Well, no duh,” was the consensus.
But when asked around the room, just about every other engineer answered the same way. In fact, amongst a team of ten of us, we could amass so little knowledge as to what was going on, we didn’t even know where to start. The next few months of our job focused around taking apart modules, reverse-engineering them, and subjecting them to various forms of shading. Different shapes, different areas, different directions, big spots, small spots, soft shade, and hard shade, we tried all kinds of different shading types in order to learn in what ways each type of shade affected the performance of a solar array. After about 6 months of learning and researching, we finally had an answer, and we understood how the science works.
From our studies, we learned that there are ultimately two different forms of shade, which were classified as “hard” and “soft” shade. Hard shade occurs when a solid object or obstacle is placed in front of the array, blocking the sunlight in a clear and definable shape. Soft shade occurs when the overall intensity of the light is reduced, such as haze or smog in the atmosphere above.
It is important to note the difference in shading types, because each causes a different effect on a solar array on multiple levels.
It starts off simple enough…
Soft shade cast on a single solar cell is relatively straight forward as the current drops proportionally to the reduced irradiance. Regardless of the irradiance amount, as long as there is sufficient light (~50W/m2) the voltage remained the same. The voltage of a PV cell depends more on temperature and the electron band-gap in the materials than on the light itself.
Hard light, on the other hand, is a bit trickier to explain. As long as there is a solid strip or channel of illuminated material between the two cell electrodes, some electric current will flow. The current was indeed proportional to the surface area of the cell that was illuminated and the shape of the shadow does not appear to matter. However, if we created wider and narrower areas, the current would squeeze and “bunch up” in the narrow portions, creating areas of extremely high temperature, known in the industry as “hot spots.” These hot spots in very rare cases have been known to cause burn-outs and small fires within modules, as they may have the current from an entire string being pinched into a very tiny area of solar cell.
If there was not a complete illuminated path between electrodes or the entire cell is shaded, then no current will flow through the cell, and its voltage will collapse. This has the effect of “opening” the circuit, as there is no longer power being delivered. Most modern solar cells have silver wires imprinted across them to allow charge to be carried more easily across the silicon, much the same way that a freeway allows more rapid travel through a densely packed city. As a result, as long as these silvered-cells have any light exposed to them, they will generate power. The only company that does not add silver to their cells to my knowledge is Sunpower, as their solar cells are made with a different process to allow more light to be collected. The tradeoff is more harvesting capability per module in exchange for letting even tiny amounts of shade completely devastate production.
It’s complicated…
At the module level, several solar cells are connected in series, in order to increase their output voltage. Soft shade, cast on a module will still allow voltage to be generated, but less current will flow from the module. Hard shade cast on a portion of the area on a solar module will open the circuit, causing the voltage generated by the module to drop. Modern solar modules come with small components inside them called “bypass diodes.” These diodes allow shaded cells to be bypassed, allowing the current from other modules in the string to continue flowing. Strings of modules are again connected in series, where the current must be the same throughout all components. That means that without the bypass diodes, any shade on any cell in the string would cause the entire string to stop producing power. Such a devastating loss of power had to be avoided, so typically three diodes are placed in along the solar cells. The diodes are placed in such a way they will allow current to flow through them only if the solar cells they bypass are shaded and opened. Since the diodes have a negligible voltage drop, there are very little losses induced by the diodes, so the only real loss from a shaded group of cells is whatever voltage they were providing.
A great way to visualize the way in which different types of shade effect the module is by looking at the effect on the modules IV curve, or mathematical curve representing the module’s output.
The IV curve is a common tool by which solar engineers can convey information regarding the input or output operation ranges of a device. It is a graph of the output current delivered from the module as a function of the output voltage applied to the module (based on the resistive load applied). The curve’s shape is important, because when multiplied again by the voltage, the IV curve becomes a PV curve, displaying the output power as a function of the voltage of the module.
For every module, there is a point on its PV curve that is higher than any other point, at a specific voltage. This is called the “maximum power point” and it is important because the solar inverter is designed to seek out this point as best it can, in order to most effectively deliver the most power to the grid. However, as the graphs above show, the effects of shade cause this point to shift around.
As a result, whenever shade is applied to a solar array, the inverter loses the ability to deliver the optimal amount of power, and must begin shifting its power-tracking point around, trying to find the new “max power point.” All of this behavior on the part of the inverter has the nasty effect of drastically reducing the array’s output power for a few minutes.
More complicated still…
Up until now, we’ve only been discussing the effects of a single module. So how can these two types of shade interact with the performance of an entire array? Since the shadows are almost never evenly spread over every module in the array, mismatching outputs between modules in a string and strings in the array are induced. From the two different types of shade applied, again two different effects occur.
Soft shade applied on some modules in a string and not evenly to others will cause an effect called “current mismatch,” where the current output of each module is varied. Since the laws of electricity dictate that all components connected in series must have the same current, what typically results is the string settles on the output of the lowest-performing module, reducing the output of the entire string to that of the most heavily-shaded cell in the string. This same effect occurs independently for all strings in an array, as strings are connected in parallel. Despite being independent to each string, current imbalance in one string can still negatively affect other strings, through interaction with the inverter (I’ll get to this later).
Hard shade, on the other hand, causes the output voltage of the shaded modules to drop, but thanks to the bypass diodes and the inverter, the current output typically remains the same unless all modules are affected. However, when two or more strings connected in parallel have shade unevenly applied to them, an effect called “voltage mismatch” occurs. Voltage mismatch is the condition in which two parallel strings are outputting different voltages when measured independently. This can have a confusing effect on the inverter, which sees a much more complicated and messy curve as it adjusts its load, ever seeking the most optimal output.
I’d like to take a moment to note that voltage mismatch cannot occur on a single-string solar array, as there are no parallel string connections with which to create an “imbalance.” A solar array consisting of only one string of modules can only have current-mismatch applied to it. Casting hard shade on a single string will drop the voltage of the string, but the Inverter will detect this drop and immediately adjust, making the drop a non-issue. It’s amazing to me the number of people I have had to explain that to over the years. For some reason, everyone I’ve talked to regarding professionally testing shade on a solar array always assumes they can use a single-string array as a means of testing, only to have nonsensical results when testing voltage-mismatch. So just to clear it up …you need at least two strings for voltage mismatch.
As for the effects of these two different mismatches on the inverter…pretty much any number of events can occur. The PV curve of the entire array exists as the series sum of the modules and the parallel sum of the strings. A shadow moving over the surface of several modules over time has the effect of constantly changing the PV curve from one smooth peak to more of a mountain range.
As the peaks of the PV curve in the inverter change from the shade, the electronics that track the maximum power point can become confused or lost, causing the inverter to choose to operate for long periods of time well outside the optimal output range. This can cause significant loss of power output and eventually annual energy yield. There have been notable advancements in the methods and electronics of the inverters, as well as the birth of new technologies in order to deal with the complications brought on by shade.
Micro-inverters are inverters that are installed on a single module, allowing each module to independently generate AC electricity at its own optimal rate, regardless of whether its neighbors are shaded or not. Unfortunately, the downsides to this approach are the extreme expense and the relative inefficiency of the micro-inverters. As this technology develops into a more useful and competitive form, I fully expect inverters to become fully integrated into modules, allowing them to become “smart-panels.”
Power-optimizers have also been created as a niche answer to the problems caused by shade, but their relative complexity and need to be extremely efficient price them out of the range of practicality. They work similarly to the micro-inverters, isolating each module’s performance from the hindrance of the others, but don’t go so far as to convert the electricity into AC. To put it simply: any solar-engineer worth his weight wouldn’t even consider installing their array anywhere near shade. These devices are a great idea for retrofit conditions where shade affects a few modules at the edge of the array or property, but should not be used widely throughout the array, as they are rather expensive.
Many East-West or dual-axis tracking systems employ a technique called “back-tracking” in order to prevent their arrays from shading one another in the early morning and late evening. This technique basically involves slowly angling the modules closer to the sun in the morning and away from the sun in the evening, in order to make their visible profiles obscure their neighboring modules as little as possible.
In terms of providing a simple answer to the shade question, the unfortunate conclusion to our research was that the result equations were too complex for a single person to calculate…we would need the assistance of a computer in order to fully mathematically model what was going on. I was given the task of creating the simulation engine, which took a year…but that’s a story for a different article.
<< Back to Science and Explanation Articles
