Cell Cracking: The Invisible Performance Thief
Industry awareness and mitigation of PV cell microcracks have improved in recent years, but there are signs that the problem may still be grossly underestimated by many.
Cell microcracks are undetectable by the naked eye, and can even be difficult to spot in standard electroluminescence imaging. They eventually grow and get worse over time, and can reduce the performance of the entire string of modules. Some cracks may result in the development of hotspots, which can lead to catastrophic failure – even fire risk.
In this pv magazine Webinar, our very own expert, Claire Kearns-McCoy, Engineering Services Manager at CEA, will take us through the risks posed by undetected microcracks, and what module buyers can do to prevent their development.
Webinar content:
The business impact of undetected microcracks to a PV project
Which types of microcracks are most likely to affect project performance?
Risk of microcrack development during module lifecycle
Techniques to identify and mitigate sources of microcracks and keep cracked modules out of a project
Q&A
Transcript
Introductions
Anne Fisher (00:00:26):
Good afternoon or indeed. Good morning. Good evening. Wherever in the world you might be joining us from today. Welcome to another PV magazine webinar. I'm Anne Fisher of PV magazine USA and joining today from Berlin to co-moderate this webinar is PV magazine editor, Mark Hutchins. Mark, what are we talking about today?
Mark Hutchins: (00:00:50):
Thanks Anne, and welcome folks. Today's show is all about cell cracking, an issue that still poses significant risk to modules in the field. And one that is quite complex to understand given the way it can slowly progress over time. Once a module has even a tiny crack, and these can form in a number of different ways, it can slowly grow to something that cuts off part of a cell and reduces performance in the worst case, even leads to the risk of fire or other really serious failure modes.
Anne Fisher: So, early detections really key here, and here to share more on this topic we have Claire Kearns-McCoy of Clean Energy Associates or CEA, who will take us through the ways that cracks can form and propagate in a cell and the best ways that module buyers and indeed manufacturers can detect and remove this risk from a project.
Mark Hutchins: And of course, once a crack has been spotted the best strategy will likely be to replace that entire module which can quickly get expensive for project owners and asset managers. We'll hear also from Claire today about the importance of early testing of EL imaging in particular in proving the origin of a crack of damage to a cell in order to make a successful warranty claim.
Anne Fisher (00:02:13):
Well, we'll get right into these topics in just a moment, first though, a couple of housekeeping points we're of course, broadcasting live today. So, please do send in your questions for all of today's speakers using the questions function on Big Marker, our webinar platform. We'll be feeding your questions into the discussion after the presentation. So, don't be shy and we'll try to get through as many of your questions as we can.
Mark Hutchins (00:02:41):
And one question I'll quickly answer now is, yes, the recording and the slides from today's webinar will be available afterwards. These are emailed out to all who registered just shortly after we wrap up today. And you can listen back to this webinar at your leisure at pv-magazine.com/webinars. Here you can listen back to all of our past webinars as well and have a look at what we have coming up lots of great webinars on the way this month. So, please do check those out and I will hand over to Claire in just a moment first though, a quick question for you audience that should be just popping up onto your screen momentarily.
Do you currently use EL testing to assess a module condition before, during or after installing a PV system? And four options there: (A) Yes - Before installation, (B) Yes - During or after installation, (C) Yes - before and during after installation or (D) No - not at all. So, while we give people a minute or so, just to get their answers to that question in.
Claire, welcome. Great to have you with us today.
Claire Kearns – McCoy (00:03:53):
Thank you for having me.
Mark Hutchins (00:03:56):
Now what does this testing typically look like? Is equipment brought to the project site or does it have to be done at night? Things like that.
EL Testing overview
Claire Kearns-McCoy (00:04:06):
What EL testing looks like does depend on if the modules have been installed or not. EL testing does have to occur in the dark. So, if the modules are installed, typically the best way to do it is to leave the modules on the racking and test them at night. The other option is to test in a dark room or dark area and that’s custom with modules that are not yet installed typically, and it is generally preferred. You obviously can remove a module for testing but it's preferred due to the risk of damage and everything I'll talk about today that you leave those modules installed whenever possible.
Mark Hutchins (00:04:42):
Okay. Well, our audience has had a little bit of time and actually we’re seeing just over 50% say no, no testing at all. I mean how does that fit with your expectation? Is this something you think there needs to be more of?
Claire Kearns-McCoy (00:05:00):
I think it's something I see. That's definitely increasing. We see a lot of operators and project owners tend to, unfortunately, there's a lot of people out there that have had a really bad experience with microcracks. And once you've had that first really bad experience definitely more likely to do testing or if you've heard of someone else having a bad experience and want to avoid it. So, we have seen that testing has increased as unfortunately more people have had poor experiences with this to learn from.
Mark Hutchins (00:05:29):
Sure. Okay. Well, well we've got it from a 51 to 49 saying no already, maybe it'll go even lower today. Well, so well clear. Yeah, I know you have a presentation for us, so please over to you.
What are microcracks?
Claire Kearns-McCoy (00:05:47):
Thank you. Welcome everyone. And I'm excited to talk to you all about microcracks, the invisible performance thief today. So, before I jump in, just to clarify, I know I did just tell you that microcracks aren't invisible and throughout this presentation, there are going to be a number of images which do clearly show microcracks. These images are all taken using electroluminescence imaging or EL imaging. And so, this is not what this module will look like to the naked eye, but you can think of EL like an x-ray and like an x-ray if for other materials, for modules EL technology has the ability to show damage that is not visible to the naked eye. So, to start off, what is a microcrack? A microcrack is the term used in the industry to refer to any crack that forms in the silicon of a PV cell.
And for the purpose of this presentation, when I say PV cell or PV module, I'm going to be referring to a crystalline module and these modules are made out of very cells that are made of very thin silicon. Each cell is only about the thickness of two human hairs and so, as you can imagine, these cells are quite unfortunately, they're quite fragile and vulnerable to cracking in the PV industry we do use the term microcrack interchangeably to refer to both a small crack in the silicon as well as a fracture that may go all the way across the cell. These fractures would be enough to break the cell into pieces if the cell were not encapsulated in the module or even shatter a cell but because the module provides that structure and support the cells do remain physically, they appear to be physically intact.
Effects of microcrack in the modules and how insurance claim works
And that's why we're able to see these microcracks instead of just having shattered cells. Microcracks are concerned because fundamentally microcracks have the ability to cause power loss. A microcrack in a module is going to affect the performance of that cell which in turn is going to affect either the current or voltage output of that module. That module power loss is then going to cause the project to underperform, and one important consideration is to remember that because of the way modules are connected, the mismatch. So, that losses that occur once you have some modules that are affected by microcracks is going to cause your entire project to potentially underperform in a way that is greater than the sum of the individual module under performance due to microcracks that under performance then has the potential to or will decrease the output of the project and therefore the project profitability. And that's why microcracks are a big concern.
nd a PV cell is designed to absorb the energy that comes in from the sun and convert that energy into an electrical current which then must be collected and transferred outside of the cell. When a cell has a microcrack, the crack in the silicon interrupts the ability of electrons to flow within the cell and therefore it all also creates opportunities for electron hole recombination and losses. In that case, the best-case scenario for a microcrack is for when the two sides of the crack remain electrically connected. In this case, you're only going to see a very small increase in electrical resistance. The cell is going to lose a small amount of power due that increased resistance, but overall, it's going to have a quite small effect on module power. However, when a portion of the cell has the ability to become electrically isolated becomes of the microcrack. This is when you start to see significant power losses. When a portion of the cell is electrically isolated, you've effectively decreased the cell area and the portion of the cell that is able to operate is effectively a much, it can be a smaller portion of the cell because the current that is generated by the cell is proportional to the area of the cell. When you've isolated a portion of the area, you now have a potential generate less current in that cell. That's going to result in mismatch within the cell stream which can result in module power loss and also represents a risk of module hotspots and can result in the module diode becoming activated and losing the power of one third of that module.
The potential of a power loss as well as the risk of hotspots is dependent on the cell area that is isolated by the modules and the more I start outside by this microcracks and the more cell area you have isolated, the more your module power loss is going to increase. Undetected microcracks are unfortunately, a really big problem out in the PV industry. Many projects that are affected by microcracks, those microcracks are not known. And so, they have the ability to lead to underperformance that doesn't have an obvious cause and leaves the asset owner searching for to understand why their project and why their modules are underperforming. Another big risk associated with undetected microcracks is the risk of a failed warranty or insurance claim. There is often the opportunity to file a claim to get to cover the financial loss associated microcracks if you're able to understand when those microcracks originated and what the cause is, this may be a module's manufacturer's warranty claim, a workmanship claim against the installer or an insurance claim in the event of a weather event.
However, if there were undetected microcracks that happened prior to the cause of the last year is filing the claim with, they may make it challenging to prove and file that successful claim. And so, if you have a project with undetected microcracks, you're not going to be able to prove the origin of those cracks and it significantly risks the success rate of any future claim. Finally, the presence of microcracks decreases the performance and also may decrease the project value during a sale. Since when you go to sell the project, if the potential buyer, through their due diligence uncovers those microcracks that is then going to decrease the amount they're going to be willing to pay for the project.
Here's an example of what happens to a warranty claim with undetected microcracks. So, in this case, modules arrive at the project with small microcracks, the small microcracks initially don't affect performance, but as the modules are installed and it operated those cracks grow and the owner starts to observe underperformance at their project. This then leads to, into an investigation, which they eventually figure out, okay, it's the modules that are resulting in my underperformance. And they go to file a module warranty claim during the testing that is required by the module manufacturer to accept this warranty claim, the microcracks are now, this is the first time these microcracks have been detected. The module manufacturer's going to come back to the owner and ask that they prove that these microcracks weren't formed due to causes that are outside of what is covered in the warranty, say poor installation, a storm, some other potential cause of microcracks that is not covered in the manufacturer's warranty. Since the asset owner did not know these microcracks were present until this moment. They're not going to be able to prove that the microcracks were the result of a manufacturing defect and the warranty claim will be rejected.
Instead, if this asset owner had taken, done a post shipment EL inspection and identified the presence of these small microcracks when after these cracks grow and open and they observe the project under performance, they're then going to have an idea of why their project is underperforming. They're aware that microcracks were present. This will lead to a shorter investigation. They'll be able to focus on the modules more early and then to file that warranty claim and crucially, they can prove that those microcracks grew from cracks that were already present when the modules arrived at the project and have a much better chance to get their warranty claim accepted.
How do microcracks form?
How do microcracks form? As I’m sure you're all thinking it's a very important question. Microcracks can form at pretty much any point of a module's lifetime beginning with, during the manufacturing process and stringing process is one potential as also the cell and string handling and the module handling then as the modules are being transported, if there's any rough handling during the loading or unloading, an accident in which pallets are dropped or containers are dropped, or unfortunately you do see issue instances of a truck getting in an accident during transportation, those sort of incidents can also cause microcracks to form and then finally, there's stress inherent in transportation that has the ability to form microcracks. If not, the modules not properly protected. In my experience, installation is one of the most significant causes of microcrack formation, especially when you have poor installation procedures with rough handling, other improper installation techniques, contact with the back of the module in ways that the module shouldn't be contacted, etcetera.
And unfortunately, not all installers are as well trained as they need to be to avoid microcracks formation during installation. And finally, it comes to the 25 to 30 years that this module is in an operating project. And unfortunately, as we are all aware, severe weather events have been increasing. And so, modules that are subjected to really strong winds, hailstorm, snow outside of the design load, etcetera. Those events also have the potential to form severe microcracking. Sometimes, it is possible to determine the origin of cracks based on things like the location of where the crack is within the module patterns that you may see at a project where you see cracks forming in the same point at every module or similar points or the shape of the crack, but this is not always possible. And so, the most reliable way to establish what a crack formed is to have the ability to compare to an EL image or some other indication of the condition before an event that may have formed a microcrack, which brings us to microcrack detection.
So, it's really microcracks cannot be identified during a visual inspection of the module and this is really important to remember since unfortunately there can be modules that are subjected to an incident such as this module that we have here over on the left. And in this case, this module was dropped and the team that was handling it looked at it. They didn't see any damage to the glass, to the frame and so they concluded the module was probably okay and if had it been up to them, they would've gone and installed this module. Fortunately, the owner of this project did have a policy that modules that had been subjected to potential damage needed to be held aside for EL testing. Since you look at the EL image on the right, we do have five cells with severe microcracks that are present on this module, but there was no indication from the outside that this module was damaged.
Different testing methods to detect microcracks
This brings us to some different testing methods that some people try to use to detect microcracks.
IV Curve Tracing is a great tool and is really helpful for showing the module power output and it has the ability to show the power loss caused by severe microcrack. However, IV Curve Tracing does have some weaknesses when it comes to microcracks. In terms of you can't diagnose that the cause of the power loss in the module is due to microcracks with IV Curve Tracing alone and IV Curve Tracing is also not going to tell you anything about microcracks that may be present in your modules but that are not causing a power loss yet.
Similarly, IR Scanning is going to be a useful tool for showing the hotspots that are caused by severe microcracks and helping to identify and remove those modules before they become a safety risk. However, similarly to IV Curve Tracing, IR is not able to diagnose the cause of the problem. So, you're not going to know that those are microcracks, nor is it going to tell you much about your future risk?
EL imaging is the most reliable way to detect microcracks. EL imaging can detect both microcracks that are present but may not currently be causing a significant problem for the module as well as the types of cracks that would cause damage visible. And IR scanning and IV Curve Tracing and high resolution EL imagery will allow for the detection of all microcrack types. Another tool that can be used in some applications is UV fluorescence.
UV Fluorescence is not useful in all applications, but when it is use, can be used in some applications to provide information about the age of microcracks, which can be especially helpful in proving things for such a warranty claim like a severe weather event. If you need to prove the age of the cracks.
When to inspect for microcracks? The first age of module should be inspected for microcracks is during module manufacturing.
This is an EL inspection during module manufacturing is used to make sure that modules with unacceptable quantity or shapes of microcracks never leave the factory that these ones do not make it to the end customer. This is already a standard practice, and each module manufacturer has their own criteria for what they consider unacceptable based on their quality assurance standards, as well as the design of their module.
The next opportunity to look for microcracks comes after shipment and an after shipment inspection is helpful to establish the conditions of the module as they have arrived at the project site. After shipment is also your opportunity to detect any shipment damage that may have occurred, or to give yourself the peace of mind to know that your modules were not damaged during shipping and after shipment inspection also provides an opportunity for the owner or EPC to establish their own record of the condition of the modules at the time they assume responsibility for them.
EL testing after installation is helpful to establish the conditions of the module after installation and look for any installation damage that may have occurred. It's also very crucial that an after installation inspection may provide the opportunity to improve your installation procedures if the testing is conducted at the right time.
And after installation EL inspection also serves as the baseline in the event of a future weather event and helps you to identify which cracks were approve, which cracks were caused as the results of the weather event versus which ones may have already been present, which brings me to after a weather event detection. So, in inspection for microcracks after a weather event is very important to understand what damage the module might be present beyond those modules with broken glass or the ones that we know were damaged but there may be modules that are still look okay but still have damage.
You want to make sure you prove the extent of damage, and you have the data required to file an insurance claim for all of your damage modules and aren't filing a smaller than necessary insurance claim. How you inspect for microcracks also varies depending on the timing. So, EL inspection is included in the manufacturing line. It's part of the factory line and 100% of the modules will be inspected by the manufacturer and many manufacturers will actually inspect each module, multiple points during the production process. After shipment EL inspection is conducted in a dark room and at this point it is no longer practical to inspect 100% of the modules and so, EL inspection after shipment should be done on a statistical sample of modules and those samples should be chosen to make sure that we're representing all of the shipment batches of modules that have arrived at the project.
Similarly, for an after-installation inspection, you want to use a statistical sample that represents all the installation conditions at the project. So, if there are multiple installation crews, etcetera, you want to make sure your air samples are spread out. Timing is also really important on a post installation inspection since you don't want to necessarily wait to the end of a project especially for a very large project since by the time, if you detect an installation damage at the very end, unfortunately it's too late and many modules have been damaged. So, if you conduct that first post installation inspection early in the process that gives you the time to change your installation procedure as needed and reduce damage in the rest of the modules installed.
Finally, there's an inspection after a weather event and after a weather event you can use EL and, in some cases, UVF may also be utilized if the modules are appropriate for UVF and UVF can help you to identify the age of the cracks, isolate which cracks visible in the EL are new versus older cracks, the sampling for a weather event is going to depend on the requirements for the insurance claim. So, it may be sampling of various portions of the project based on suspected damage or the insurance company may require a higher percentage of models to be inspected. That's going to depend on what the requirements are there.
Microcracks growth
Next. So microcrack growth is really important to consider when considering a microcrack, since you, when you're looking at a microcrack you need to consider both what's happening with that module now, and what is going to happen to the module over the rest of its lifetime. Once formed, unfortunately microcracks don't go away. Silicon is not a self-healing material and so those microcracks are only ever has the potential to get worse. All modules are exposed to stress during operation and this stress worsens the microcracks. It can have small microcracks that begin small, such as these ones over here, and the image on the left. You can see it inside the circles there are actually two small microcracks that this module started off with but over time, those small microcracks became cracks that grew to the entire extent of the cell. Microcracks also open over time and so, when I say open, I refer to the distance between the two sides of the crack is growing and getting larger. So here, and then here's an example of a few different shapes of cracks in this one module.
Sorry let me turn on the drawing. Okay. So, if you have these cracks over here in this cell, these ones here, you can see there's still a path for any current generated in this portion of the module over here, it still has a path out to the busbar. And so, these cracks are going to result, even as they open, you will see an increase of loss due to resistance but you're not going to isolate an area. If you look at these cracks over on this side of the screen, these cracks on the right have a higher chance of resulting in a loss of active area because they have these the Y shape or these branch shapes.
And here we have two different examples. The one on the top, this cell here, this one is you can see areas in the EL image, get more dark as there's less opportunity for current to flow out of an area. So, this one it's darker, but not fully dark and so, in this one, you probably still do have a connection with the fingers across the crack an so some current still is able to flow. However, over time, this cell here is going to turn and look more like this cell where there's no longer a path, those fingers have broken. The crack is opened to the point where there's no longer a path for current to flow from the affected area to outside of the area. And this portion has become electrically isolated.
The reason microcracks grow and open for a number of factors that happen as the modules are, this is their lifecycle. Shipment and transportation is the first opportunity microcracks have to grow and open. All transportation methods expose modules to some level of vibration and so that vibration has the ability to open those cracks. There's also, the stress is applied to modules during loading and unloading. Similarly, the handling during installation, even when an installation is done correctly, still does apply some stress the module and has the ability to open those cracks.
Finally, during operation, once a module is installed. Those modules are designed to be out in the field for 25 to 30 years. And so even small stresses add up, as modules go through a daily temperature cycle that applies a small amount of stress to modules each day, and those daily temperature cycles will eventually open the crack or cause crack growth, similarly wind loading and snow loading, even when loading and snow loading those within design parameters of the modules can cause slow microcrack growth and then there's severe weather events such as a very strong wind, extra heavy snow, etcetera, that can cause cracks to grow rapidly over a short period of time due to the severity of the stress applied.
Factors that affect the risk of microcracks
Effect of Microcrack Shape On Risk
There are a few factors that affect the risk of a microcrack beginning with microcrack shape. So, as I've mentioned a few times here, the most important thing to consider the microcrack is does this crack have the potential to isolate cell area? So, if we start with this module at the top, this crack does not, it's a line crack, there's still a path for current to reach the busbar from any point on either side of the crack, it's this one's likely to result in a very small loss in module power. However, once you start to have intersecting cracks or cracks that form between the busbar and the cell edge, there becomes a potential to isolate cell area. The more cell area, a system of the cracks can isolate the higher the risk is to module power. So, if you compare the middle image, this one here is isolating a relatively small portion of the cell compared to the potential that this large branch crack at the bottom has to isolate a much larger portion of the cell. It's really important to consider both the current shape of the crack as well as the potential for growth. One is one considering the effect of the crack shape on risk.
A small crack does not always mean a small risk because these cracks have the ability to grow. One form of crack where this is especially apparent is an edge driven crack. These are small cracks, which form on mostly in half cut cells where the busbar wire crosses the edge of the cell, and these cracks start off as tiny cracks. And unfortunately, there's a misconception among some parts of the industry that tiny crack doesn't matter. And that is true now, these tiny cracks over here that you see in the EL image on the left, those are not currently going to cause any really any power loss to this module, but you have to worry about what's going to happen to those cracks as the modules in the field for 25 to 30 years. And so, here's an example of three sites we looked at that started off with these cracks pre-installation and beginning, most of the cracks were very small, less than 10 millimeters, but after just installation, these cracks grew to be longer line cracks and most concerningly branch cracks as in the image on the left. And so, this was just during installation. These modules still have to live out their operational lifetime. And so, this is a really good illustration of why you can't take a small crack and decide that it doesn't matter because you need your modules to perform for many, many years and not just perform right now.
Microcrack Considerations by Module Type
There are few considerations for microcracks by module type for a glass-backsheet module. There's additional risk to the module from impacts to the rear of the module since the back sheet doesn't provide that as much mechanical protection to protect the cracks from any impacts that may happen to the back of the module. Glass-glass modules are overall more resistant to microcracks than a glass-backsheet module because the module is more rigid and also cells are located in the middle of the material stack which reduces the stress on the cell for given deflection. However, glass-glass modules do have thinner glass which represents a higher risk of module damage from impact the front of the modules. Cut Cells have the risk of microcrack formation during the cutting process. That cutting process unfortunately does have the ability to weaken the cut edge and therefore serve as a propagation center for microcracks.
Another really important consideration for affecting the risk effect with the microcrack is the Cell Interconnection Method. Certain Cell Interconnection Methods have the ability to decrease the potential for area to become isolated or decrease how much area can be isolated. This includes things like as modules have more busbars, there is likely that cracks are going to isolate less area and also methods of interconnection through the back that makes it so much more challenging to isolate areas of the modules. However, some of these interconnection methods also come with an increased risk of microcrack formation. So, it's generally more busbars, is more soldering, and more risks of forming a microcrack through soldering, as well as we've seen, in some cases, the wires to use a multibus for our cells initiate cracks as they cross the cell edge.
Microcrack Mitigation
Module Manufacturing
Fortunately, there are things that can be done to mitigate the risk of microcracks to your projects. This begins with module manufacturing, with the selection of a reputable module manufacturer with good internal quality control processes. Nothing that's really important to consider in the module manufacturing is that EL acceptance criteria, all module manufacturers are going to allow some quantity of cracks in their final modules, but these are typically cracks that have been determined to be a low risk to the final module performance. However, if the EL acceptance criteria does allow cracks that may grow or have a potential to cause higher risk that, that results in a higher risk from cracks that were formed during module manufacturing, third party module quality assurance or QA is another way you can mitigate the risks for module manufacturing
Module Transportation
During module transportation, it is important to ensure that the packaging and procedures used to transport the modules are adequate module manufacturers will have developed their packaging as well as their handling procedures based on testing that they have done on their modules and so it's really important to follow those procedures, also important to validate that containers are properly loaded and that those procedures have been properly adhered to. The choice of ground transportation to the site is another important way to minimize risk to modules by choosing an appropriate ground transportation method and ensuring that pallets are handled well during unloading, unfortunately, handling of pallets during unloading or as modules are moved around a warehouse is a common cause of damage to modules. So, it's important to make sure that every time the modules need to be moved, that they're handled appropriately. Because it's not possible to know what happens during every stage of the module transportation, a post shipment EL inspection is recommended to ensure there aren't modules that were detected in ways that is not visible at the times the modules arrive at the project.
Module Installation
As I mentioned earlier, module installation is a significant cause of microcracks and damage to modules and this can be mitigated by ensuring that the installation crews are well trained and that they're adhering to the installation procedures. Another opportunity to mitigate the risk associated with module installation is conducting EL on the same modules before and after installation. This allows for really clear understanding of what's happening to the module during installation and for the detection of new or worsening damage. Conducting that EL inspection early in the project installation phase is really important, especially as you'd want to make sure that you know that your installation procedures are poor before an entire project's worth of modules are damaged. This is especially important for large projects where you may be damaging hundreds of megawatts of modules with poor installation procedures. For a large project, periodic EL check should also be used to ensure that quality is consistent and the installation procedures haven't drifted or changed over the months that are required to install a very large project.
Here's a case study of installation damage. This is an actual scenario from a project. In this case, the modules arrived at the project and unfortunately the installation procedures used weren't correct, and the modules were damaged during installation. The owner was unaware of this though, until they needed to conduct a baseline EL inspection to obtain insurance coverage. And at this point, many microcracks were detected but it was too late to go back and fix the installation procedure. Since the entire project had already been installed. They also had not done any EL inspection before installation and so, they were unable to definitively prove that the modules were damaged versus in installation and file a claim against the installers workmanship warranty. These microcracks caused the project to be underperforming and to lose project value and eventually the owner was forced to sell the project at a loss due to the condition of the modules.
Here's what this owner could have done in order to prevent the standard area from happening. So once again, the installation procedure started off poor but had they done an EL inspection early, they would've seen the installation damage. This would've given an opportunity to require an improvement in installation practices followed by another EL inspection to validate the improvement installation practices actually or as a resulting in less damage to the modules. Then the rest of the project could be installed with these improved practices and therefore not damaged by installation. Then when it came time to that baseline EL inspection after all modules were installed, the owner would've found that while some modules that had been installed in the beginning had some installation damage. Most of the project was free from installation damage and the project had a much lower risk of underperformance.
Module Operation
It's also the opportunity to mitigate the risk from microcracks. During module operation, first thing to do here is ensure the modules are correctly installed, ensure that the mounting type is appropriate for the modules, the racking and the conditions the project is expected to face and that the tracker select if trackers are used, the tracker selection is appropriate, including the stow position, given the weather conditions prevalent in your region. Finally, a baseline EL testing is important to serve as a baseline in the event of a need for an insurance claim. And unfortunately, more and more projects are being affected by these severe weather events and severe weather events are starting to hit in regions that are less expected. So, having that EL testing is a baseline that can strengthen your warranty claim in the unfortunate event that insurance claim is needed by giving you the opportunity to prove that the cracks present in the modules after the weather event were caused by the weather event and were not present beforehand.
Key Takeaways
Key takeaways that I hope you all take away from this webinar is that microcracks are a concern since they have the ability to result in project underperformance but then not all microcracks are equally concerning. Microcrack shape is really important for considering the associated risk of modular performance.
Microcracks grow and open over time, it’s very important to remember that what a microcrack looks like now and how a module of microcracks is performing now is not necessarily the way this module is going to continue to perform over time. And so, in evaluating the risk from microcrack, you need to consider the potential for microcrack growth. The good news is that prevention of microcracks is possible through appropriate equipment selection and care during transportation handling and insulation and the use of EL testing to verify procedures.
Finally, in the unfortunate event that microcracks still do appear in a project. The timely detection of those microcracks is important to give the highest likelihood of success in a warranty or insurance claim and this includes proving when microcracks are not present before an event that causes damage so that you can identify when those cracks formed and the appropriate party to file a claim against.
Q&A
There's going to be an opportunity for questions here, but if you have additional questions that are not answered in the Q&A session here, I will be at RE+ next week. So, you can visit me at CEA’s booth which is booth 1888, if you have further questions, and I'm always happy to discuss the risk of microcracks and mitigation and testing procedures. Thank you.
Anne Fisher (00:38:40):
Thank you, Claire. Well, we'll head into a Q & A with Claire in just a minute first though, a second poll question for the audience. Let's see if we've changed any minds today. So, we'd like to ask you, do you plan to use EL testing to assess module condition before, during or after installation of your next PV system? A is Yes - before installation, B – Yes, during or after installation, C - yes, before and during, after installation and D is no. And while we give the audience a moment to answer, Claire, what would you expect here? How widespread is buyer testing at the moment?
Claire Kearns – McCoy (00:39:44):
Yeah, currently we are seeing a bit of a gap between the US market and the European market and testing. We see much more EL testing in the European market, it's been in the US market. It's been rapidly growing over the last few years, and we've seen more and more asset owners, even EPCs understanding the value of testing and timely detection of microcracks and adding testing protocols to their procedures and plans.
Anne Fischer (00:40:11):
Okay. Thank you.
Mark Hutchins (00:40:20):
All right. Well, I think the audience has had a little time to answer now. We see, well, yeah, about 50% saying no before Claire spoke has now gone to “yes - before and during/after installation” is now 54% of people listening certainly plans for. We don't have the breakdown now of where everyone listening is. But yeah, there you go. Well, plenty more questions from the audience have come in as well. Thanks all for getting your questions in. I think the first thing our audience is interested in, I mean, you touched on the cell cutting and interconnection methods. Do you see a difference between actual cell technologies? I guess, particularly PERC, TOPCon and heterojunction and how susceptible they are to cracking?
Claire Kearns – McCoy (00:41:21):
Yeah, we haven't had because right now, currently PERC still does remain the dominant technology in the industry. So, as far as the testing that my team and I have done in the field, we haven't had many opportunities to test on other technologies yet. I certainly know that I've seen some discussions about HJT and the potential for risk there as well and potential for HJT to be less if manufactured correctly. But I think it still remains to be proven and certainly that affects the manufacturing but doesn't necessarily affect how the models are going to perform once they're out in the field.
Mark Hutchins (00:42:07):
Sure. And then at the module level, are there other things like I think you mentioned thicker glass maybe is a good protection. Are there any other steps module makers can take to just protect against cracking?
Claire Kearns-McCoy (00:42:26):
Yeah, as modules have more and we've seen more and more glass module, glass-glass modules that we're seeing the cracks grow more slowly on glass-glass modules. And so, glass-glass modules have been one way to improve reliability for modules and protect against microcracks but certainly the thickness of the glass is important, especially for those weather events and understanding the appropriate glass thickness for your environment and the risk of hail, especially at your site.
Anne Fischer (00:42:59):
Okay. I have a question here. How can you calculate the long-term impact of a crack once it's detected?
Claire Kearns-McCoy (00:43:07):
Yeah. So once a crack is detected, what you want to do. You're going to look at the shape and then there is fortunately an expanding body of research from research scientists. Who've taken different crack types and moved them through simulating the stresses of operation and used that to predict how cracks grow. There still is definitely more research needed on this and more field studies, a lot of stuff is still research based but when you look at the shape of the crack and predict the area that can be isolated and then there are research studies out there that look at the performance effect as correlated to the area that can be isolated by a crack.
Mark Hutchins (00:43:59):
Okay. And once you've detected a crack on an EL image, Is it a given that it will grow over time or a particular conditions, you know, needed for this as well?
Claire Kearns-McCoy (00:44:12):
Yeah, so that crack will grow as the module is affected by stresses and the rate of growth does depend a lot on and that is one area where there's certainly opportunity for more research in terms of actual crack growth rate in different field conditions. But the rate of growth is going to vary based on the stresses those modules are affected by. So, the amount of thermal stresses, the amount of wind that's at the project, the frequency of snow loading, etcetera. So, it is a given that it's going to grow, but how fast it's going to grow does depend on the module design as well as the project location and the actual conditions that module is facing.
Anne Fischer (00:44:55):
Okay. What about test sample size? Is there a recommended test sample size in terms of the percentage of a project's modules at the after installation and after weather event stages of inspection?
Claire Kearns-McCoy (00:45:12):
Yeah, so for after installation, the idea is to get a statistically representative sample. We tend to use the ISO 2859 standard to determine what the appropriate sample size is, and you want to make sure you're testing modules that are expected to have had the same installation conditions or same shipment conditions. So, for a shipment, you're going to want to look at modules on a by shipment batch level, assuming that all modules that were shipped at the same time went through similar conditions. And then for an installation, you're going to want to look for any known differences in installation. So, if there was a change in installation procedures, if there was different installation crews, different installation companies, anything that may have caused some modules to be installed differently from others, you're going to want to make sure that you do a statistically significant sample that covers all different scenarios for other event testing. We do see that driven by the insurance companies and what the insurance company will accept. So, some insurance companies are comfortable with a lower rate of sampling, other insurance, we have seen cases where higher sampling rates are needed. And we've seen that kind depends on the insurance company and also the extent of the storm. If there's evidence that certain projects may have been more affected than others, etcetera. So, there's a lot more variation in a post event claim testing.
Mark Hutchins (00:46:40):
Okay. So, insurers today have a fairly sophisticated handle on this in terms of what they require and understanding that risk. Is that fair to say?
Claire Kearns-McCoy (00:46:52):
I think in my experience insurance companies are still, they're still coming to understand this risk, but they do still, they definitely have opinions on what they're willing to accept and what they're looking for.
Mark Hutchins (00:47:07):
Okay. And what's really the worst-case scenario. I mean, I think you mentioned fires in there. How would something like that come about and how likely is something like that to happen?
Claire Kearns-McCoy (00:47:20):
Yeah. So as far as the risk of fire, other hotspot due to a module that is yeah, the worst case scenario. And that would happen if you have a crack that is at a point where it is causing a hotspot. Fortunately modules do have diodes and those diodes are designed to protect the module from hotspots but diodes also, aren't designed to be activated every day for the entire lifetime of a module. A diode does have a limited lifetime. And so, if you have a module that has a severe crack early in its life, that diodes going to need to be activated for a very long time and the diode may eventually fail once the diode fails that can either lead to the potential for overheating, it depends on how that diode fails. So, you can have overheating in the junction box at the location of the diode or the diode can simply stop protecting the module and you have the ability for that hotspot to then start to act unprotected and cause your module to overheat.
Mark Hutchins (00:48:30):
Sorry, thanks. Just getting through plenty of questions coming in. I'm just trying to keep up with the audience. So, I think the next thing we talked a bit about what you can do within the module to mitigate against cracking. Are there also things within the mounting system or elsewhere within the system design that are good practice?
Claire Kearns-McCoy (00:48:59):
Yeah. So how the module should be mounted is going to be both module and racking technology dependent. There's not one universal, this is the thing to do, but it is really important to make sure that you are mounting modules appropriately for what the manufacturer allows. You're considering mounting configurations that match with your site of configurations. So, some module manufacturers may have this module is rated to a certain snow load in this configuration but a different snow load in another configuration or a lower. So, make sure that you're following those guidelines that are appropriate for your module and racking condition.
Mark Hutchins (00:49:39):
Okay. Sure. Sorry, Anne, please go ahead.
Anne Fischer (00:49:44):
So, what about mounting systems? Do they offer additional protection?
Speaker 3 (00:49:51):
Yes, unfortunately, I'm not a racking expert. So, I would have to defer my colleagues who are, but I do know that there are things that can be done with a mounting. In my experience, I have seen that when I encounter modules that are damaged due to mounting configurations it's usually because of installer error. So, when the installer went to take a well-designed system but didn't torque the modules appropriately or something like that, that leaves the modules poorly mounted in having the ability to, worst case scenario, fly off the racking or move too much on the racking during a storm.
Anne Fischer (00:50:28):
Okay. Well, a similar question having to do with modules. With larger modules showing up and gaining popularity, are there any concerns for increased cracking with larger modules?
Claire Kearns-McCoy (00:50:44):
Yeah, so larger modules need to be appropriately designed to be robust given their size. And so, as modules get larger, you do have larger cells, everything is heavier, et cetera. And so, there is the need to make sure that those modules are appropriately designed and appropriately tested to verify that the design is appropriate.
Mark Hutchins (00:51:12):
And when it comes to doing this onsite, maybe in the field testing, what are the costs involved and how would I like, say as an asset owner calculate when the best time to do this is?
Claire Kearns-McCoy (00:51:28):
Yeah. So, the cost involved does depend on the test method that can be used. Certainly, there are cheaper high throughput methods that the more modules you can test at once, obviously the lower the cost is, but there is the downside of those cheaper high throughput methods that they do have generally lower image resolution and therefore lower ability, they can't detect small cracks as well. And so, the appropriate cost does kind of depend on your need. There's not going to be a one size fits all answer for that. But as far as the benefits of testing versus the potential to certainly testing is way less expensive than a failed warranty claim down the line and dealing with the underperformance that comes from this or doing a small amount of testing at the beginning of an installation versus letting your installers damage hundreds of thousands of modules. It comes out as worth it, for sure.
Mark Hutchins (00:52:33):
Okay. Absolutely. And beyond that, is there a sort of recommended regular time interval or would it just be say, after an extreme weather event or something like that?
Claire Kearns-McCoy (00:52:46):
I personally have not yet encountered an insurance company that is requiring testing at a regular interval, but for a project where that initial test for the baseline didn't show any issues unless required by the insurance company, I would generally not consider it worth it to continue testing. It may be worth it to continue testing in a case where, you know you have an issue to monitor how those cracks are growing and understand at what point are you at a point where it's worth replacing those modules.
Mark Hutchins (00:53:23):
Okay, sure. Is the testing on site something that CEA would do or are there sort of specialists that would do this, that would come to the site and do this?
Claire Kearns-McCoy (00:53:46):
Yes, that is that my group at CEA offers is testing at project sites. We offer pre-installation testing and post installation testing, as well as post weather event testing.
Mark Hutchins (00:53:58):
Okay. Sure. And we touched on this earlier, sorry, but just to recap, so the pre-installation testing that would be, you'd have a sort of EL tester at the site, like in a container or something that can do it.
Claire Kearns-McCoy (00:54:15):
Yeah. So, EL testing then pre-installation is done in some sort of dark room that can be a shipping container, an office, a mobile test lab, some other sort of mobile dark room, something that is configured to test modules as a small dark area you can bring the modules into, but it is definitely there's the opportunity to ship modules to a lab but it's generally preferred to test the modules on site. It's going to be cheaper to test the modules on site. And so, when that's possible, it's recommended.
Mark Hutchins (00:54:48):
Okay. And then post installation, would that look the same? You would dismount them and take them apart, or can you, can you test it?
Claire Kearns-McCoy (00:54:55):
No. Yeah. So, post installation testing, we generally try to avoid dismounting the modules whenever possible, since you don't want to add the question of I mean, there's been a common theme here that you want to make sure you can prove when the cracks happen. And so, you don't want to meet a position where you remove the modules, say “Hey, we found installation damage” and then you go to file a claim and your installation company says, “Oh, you damaged those. Or you may have damaged those” and then you then can't prove it. So, we'd always recommend leaving the modules installed unless they need to be uninstalled for some other reason. Mostly, that most commonly happens after a weather event. If say the tracker is damaged, the roof is damaged. There's some reason those modules need to come off.
Mark Hutchins (00:55:40):
Do you see at some point that there could be a way to, you know, to conclusively prove how a crack was formed or was that's always gonna be a question that of what a warranty claim's always going to be a difficult thing, right?
Claire Kearns-McCoy (00:55:54):
Yeah. I think unfortunately, there's some crack signatures that can form in a number of different ways. There's some crack signatures where it is more clear especially cracks formed by impacts, those ones it is even then though you can tell the crack was formed by an impact, but I can't tell you was that impact because the installer hit the module with a hammer or was it because there was wind and the wind blew something into it? All I can tell you is that something hit that module in that location. And so as long as there are people who want to fight claims, you have to be prepared to fight, to have the evidence to prove your case.
Anne Fischer (00:56:36):
Okay. What about the time interval? Is there a recommended time interval for EL testing on a test sample during its life, during its whole lifespan?
Claire Kearns-McCoy (00:56:50):
Yeah, so that depends on the condition of the modules. I don't generally recommend that modules that are in good condition, be tested on a periodic basis unless there's some reason to believe they're no longer in good condition. The modules are in good condition, and you have no reason to suspect damage and you've done that post installation baseline testing. There haven't been any big storms and you're not required to by an insurance company or some other financial source. I generally don't recommend periodic EL testing. The place where we may recommend periodic yield testing is if you have modules that you know have damage and you're trying to understand how that damage is progressing since you don't necessarily need to see microcracks and immediately yank those modules since they're not going to necessarily perform poorly at the start. And so, if you have a source of damage that you don't have the ability to file a claim on for some reason, periodic EL testing every few years may be helpful to track that damage and understand when it is now worth it to pull the modules.
Mark Hutchins (00:57:56):
Sure. okay. We are coming up to the end of our one-hour slot for today. I will squeeze one last question in though for you, Claire. Do we have any sort of of industry data on the size of the problem, the number of modules that are being installed with cracks and whether this is improving already?
Claire Kearns-McCoy (00:58:18):
Yeah. So, I currently, I have not seen any industrywide database. We are currently at CEA in the process of collecting one from the sites we've been to. There is still a challenge that historically EL testing has mostly been done for projects with suspected damage. And so generally the sample size is a little bit skewed towards projects with damage, but as I mentioned, that is changing. And I think there's going to be a really powerful opportunity for the industry going forward to collect that data and better understand the extent of this problem.
Webinar wrap up
Anne Fischer (00:58:54):
Okay. Well, I think that's a great place to wrap up for today. Thanks to Claire for the presentations and for sticking around for the Q&A as well. Thanks also to everyone at CEA and of course, to my colleagues here at PV magazine for working to put together another excellent webinar. And finally, thank you to all our viewers for joining us today and to all who submitted a question, we really do value your contribution to the discussion and my apologies again that we can't get to all the questions asked today.
