Are you against additional protections for fire fighters?
No, we fully support the intent of adding an additional layer of safety for fire fighters. However, this additional safety should make sense. It should not come with new risks, reduced reliability, limited customer choice, and added cost as it is the case today. Current solutions for 690.12(B)(2)(2) unfortunately do not fulfill the original intention of the NEC and while they reduce the voltage inside the array boundary, they add additional safety risks to the PV system. Therefore, we demand for this requirement to be fixed.
Do you think solar is unsafe because of DC connectors?
No, solar systems are a very safe technology and for many years they've been built without any major safety issues. A comprehensive study by the Fraunhofer Institute has shown that solar systems are very safe. In the rare cases, however, where problems occur, the report showed that DC connectors are the major weak point in a system. While this is manageable in string inverter systems, the adoption of Module Level Shutdown with the currently available options has increased the number of connection points by 2-3 times.
High profile cases of solar related fires in the media mention connection points as a major issue and make solar look unsafe in general. We would like to correct that narrative and make sure that solar systems are built with the least amount of connection points possible.
Do you think that fire fighters should be provided with the highest level of safety?
Yes, absolutely! Therefore, we support an additional safety layer – but it has to be done right, so that safety is actually increased and no additional risks are created. Today’s solutions don’t provide that and add new risks. In fact, they give a false sense of safety to first responders.
Are you against microinverters and DC optimizers in general?
No, as with many things in life, everything is about moderation – just like the body needs some sugar, but if you consume it in large quantities in every meal, it becomes detrimental to your your health. PV is a safe technology by all reports. However, like every technology, there are inherent but acceptable risks. In solar systems, studies (see here and here) show that a risk factor in solar systems are connection points and component hazards. They are needed to some degree, but as an industry we should make sure to keep this risk factor as low as possible, so that it’s not damaging to the industry’s health. Since microinverters and DC optimizers increase the number of connection points and components significantly, it is not good for the industry to require this technology in every system. Therefore, a code requirement like Module Level Shutdown should not be pre-maturely adopted, before better alternatives, as outlined under “Our Vision”, are available. Furthermore, it is necessary that codes and standards ensure the same high levels of safety for every technology. With over 90% of all new residential systems featuring DC optimizers or microinverters, there is an urgent need to address this problem.
Would you like to prohibit current solutions for MLSD?
No, we want to see customer choice and thus do not want to prohibit any technology. However, we would like to have better and safer options available for everyone and we want that codes and standards ensure the same high levels of safety for every technology. When needed, installers should be able to choose Module Level Power Electronics and such, but it should not be mandated for the entire industry.
Isn’t it normal for a solar system to have a power electronic component attached to every single module?
While it seems to be very common these days to use DC optimizers or micro inverters for any system (no matter if there is shading, or if there is a real need for module level monitoring), it does not have to be this way. In fact, for many years systems have been installed with string inverters, where only one power electronic component is needed (the inverter), which is mounted in a safe location such as the garage. String inverters are the major technology in any other solar market in the world. Thus, if the US code would not require module level shutdown or if better solutions (such as module integrated chips) would be available, string inverter technology could be applied in almost every system, limiting the issues outlined on this website to systems where the benefit of Module Level Power Electronics outweighs the downsides – as opposed to requiring them across the entire industry.
What is the value of Module Level Power Electronics in general?
There are applications where Module Level Power Electronics can add value and where the additional cost and labor is justified, e.g. in case of heavy shading or when module level monitoring is necessary. But this should not required for the entire industry, as it would scale up the issues explained on this website. We recommend that safety precautions are taken when using MLPE (e.g. not intermating connectors, checking connection points, regular maintenance, and such).
Shouldn't Module Level Shutdown be stopped forever?
No. we would like to pause it only until safe and reliable solutions are available (such as module integrated, chip-based solutions with an open industry standard), and until customer choice is restored. In the meantime, the requirement must be fixed to avoid the unintended consequences of the current situation.
What about solutions based on UL 3741?
We don’t consider UL 3741 as a good solution because as is, it does not completely resolve all the underlying safety issues for shutdown, such as connection points, additional components on the roof, or noise interference with arc fault detection – so the root causes of these safety issues are not necessarily resolved. Furthermore, the standard does not present a real use case and is rather vague in its description. The certification process for a “safe array” involves many different parties (such as standard PV module, shutdown device, and inverter), who usually do not have R&D collaboration or joint certification processes and supply chains. Even when listed, installers might run into trouble in sourcing the exact components of a “safe array” and will have to do Module Level Shutdown as is.
There are also uncertainties on whether or not AHJs are approving solutions based on UL 3741: https://www.sepisolar.com/sepisolar-blogs/ul-3741-compliance/
Doesn’t the connector standard UL 6703 solve the issues with DC connectors?
UL 6703 was an attempt to ensure intermatability between connectors form different manufacturers. To get this certification, both manufacturers are required to submit their products for testing. However, it has not been widely adopted yet and is hard to enforce as AHJ’s typically do not go on the roof to inspect systems. Thus, it is not a good solution for the mismatching issue that MLPE devices exacerbate with a much higher component and connection count. In addition, with so many different products required in a PV system, enforcing such a standard becomes extremely difficult, given the low probability that installers, designers, and purchasing teams are all paying close enough attention to the products being purchased and whether or not they are certified together. Furthermore, UL 6703 does not resolve all the other connector related issues, such as installation error, force on wiring and aging – all of which, when not minimized increase the probability of arcing and PV related fires.
Why would you like to require 690.11 (Arc Fault Circuit Protection) to apply to all solar PV DC circuits?
While PV related fires are rare, additional connection points and components are a potential hazard and can therefore increase the risk of fire (see studies). With the implementation of Module Level Shutdown, current certified devices increase component and connection count by a very large multiple under both the UL1741 and UL3741 standards. Arc Fault Circuit Interruption (AFCI) adds an additional layer of safety in preventing PV related fires by interrupting and extinguishing DC PV arcs. However, the AFCI requirement is currently not applied to PV systems with microinverters, which ironically have a much higher component and connection count than other systems with AFCI detection (such as string inverter systems). Therefore, the code is not applied consistently.
Because AFCI is a good safety feature to prevent fires in PV systems, we think that the code should be corrected. Today’s code does not require any arc fault detection below 80V, where microinverters typically operate. Studies have shown that you only need as low as 25V to create a stable serial arc that can lead to a fire. This is well below the 80V threshold. Thus, we believe that this loophole in the code needs to be closed, to provide this safety feature to all systems, regardless of the applied inverter technology. The concern of nuisance tripping will be mitigated via our request to require low electrical noise levels as a standard, since these noise levels are often the root cause of nuisance tripping, as explained here.
What should better solutions look like?
These technologies will be based on the SunSpec industry standard for Rapid Shutdown and will be fully integrated into the module and tested to highest safety standards (including testing for noise levels, high temperature, and safe switching). Such a solution means that no extra connection points and no extra complex power conversion components are required to achieve safe and reliable shutdown. With such options, solar companies and consumers can choose PV systems that meet their requirements for their own safety and the safety of first responders. See here for more.
We recommend that details should be asked of your module manufacturer regarding these module-integrated solutions.
You talk about module integrated solutions – wouldn’t it be better to get rid of module level shutdown for good?
We support the 690.12(B)(2)(2) requirement of the NEC 2017 (known as Module Level Shutdown) to add an additional layer of safety for fire fighters. However, this additional safety should not come with new risks, reduced reliability, limited customer choice, and added cost. Today's solutions for 690.12(B)(2)(2) unfortunately do not fulfill the original intention of the NEC and while they reduce the voltage inside the array boundary, they add additional safety risks to the PV system. Therefore, we request for this requirement to be fixed.
If a module integrated solution based on an open industry standard is available, installation and maintenance experience for installers, system reliability, and performance would be just like it was before the shutdown requirement, while providing an additional layer of safety for fire fighters.
Who is responsible for re-writing this code requirement?
The National Fire Protection Association (NFPA) and specifically NEC Code Making Panel 4 are responsible for this requirement and have the authority to re-write it, even during the 3-year code cycle.
The video says that MLSD increases the cost of solar systems. Does this even matter when it comes to the safety of our first responders?
The video goes over a range of unintended consequences of the MLSD requirement, one of which is increased cost. However, the major portion of the video talks about the safety issues caused by the requirement, which increase the risk of fire in the first place. It also highlights that current solutions are unreliable because of electrical noise levels, so fire fighters cannot even rely on them, putting them at actually higher risk. Thus, if cost would be the only unintended consequence while actually increasing the safety of systems, then we would not have to fix the code. However, the current situation has too many unintended safety consequences and thus needs to be fixed. This is what the video highlights.
What does Clayton Hanks mean when he talks about noise and how false errors might be thrown?
Clayton is explaining the issue of electrical noise in a system, due to the added components. Electronics can create electrical noise, such as unwanted signals and fluctuations in current or voltage. Such noise also exists on the DC side of a solar system. This is typically not a problem. However, with the advent of DC optimizers, the noise levels increase significantly and start interfering with an important safety measure: the Arc Fault detection and circuit interruption, also known as AFCI. With noise, AFCI is not able to identify real arcs or to distinguish them from the noise. With many Module Level devices, an AFCI algorithm has to deal with a lot more noise. All this noise increases the likelihood of nuisance tripping, which negatively impacts the system’s reliability and operational cost, as well as the ROI. Or even worse, a real arc might not be detected, and cause a fire.
What is Melissa McCarthy referring to when she says that one component failing causing several others to fail, perhaps even the inverter itself?
Melissa describes her experience with DC optimizers, where components on the roof start to fail. She had a system where one faulty optimizer shut down the entire string and then the inverter, while the monitoring platform did not show the issue. It can also happen that multiple optimizers fail until a string or the entire systems does not turn on anymore. Her experience underlines the compromised reliability that module level shutdown causes: a significantly increased number of sensitive power electronic components has to be installed on the roof, where they are exposed to extreme temperatures and harsh weather. Since every component has a certain failure rate, as well as a 100% wear rate over time, the reliability of systems is compromised.
What does Phil Teague mean when he says that there is less competition in the market because of MLSD?
Since the introduction of rapid shutdown in the code, module level power electronics have seen a sharp increase in market share, with over 90% in residential solar as of 2021. There is only two major players who share this huge slice, creating a duopoly with barely any competition.
How can I get involved and support this movement?
Do you like this page?