HUGE breakthrough for rooftop solar (UL 3741 and news from Ironridge!)

by Duncan Miller on March 4, 2024
 I just came across this great video by Derek Mast, @derekthesolarboi on YouTube. Derek shares technical knowledge with the constantly growing solar workforce, via Youtube  through entertaining and educational videos about electrical theory and rapid shutdown devices.

Watch the video here:



Read the full transcript:

The residential rooftop solar industry in the United States is about to completely change. And if you haven't heard about UL3741 yet, you need to hear the rest of this. What was just announced is going to reduce the cost of installing solar, the labor involved to install it, and the service costs related to failure of your PV system for its entire life.

Hey, I'm the Solar Boy. I repair renewable energy systems for living and love to share everything that I've learned with as many people as possible. Thanks for joining me for this deep dive on safety, regulation, and what the future holds for the solar industry. But first, let's talk some history. Firefighters' relationship to solar as an idea and a product has always been a bit fraught. Firefighters have always been nervous around solar, mostly because electricity and water seems scary together. Normal house electricity is much less scary because you can turn it off. If you have a house of blaze that's connected to the grid, a firefighter takes an ax to your meter base outside and cuts off the entire electrical feed to the house. Pretty easy. However, solar is always on when there's sun. DC solar systems are nominally 600 volts, so when those panels are energized, you're going to see anywhere from 300 to 500 volts or more in anything tied directly to those solar panels. At a surface level, it's a big, scary shock potential for a firefighter that needs to spray a ton of water and semi-indiscriminately hack holes into a roof.

Over the years, this fear has resulted in a number of stories, like the Dietz & Watson warehouse fire in 2013, where the headline was, solar panels hampered firefighting officials say. Quoted in the piece was acting fire marshal William Kramer, who said, we may very well not be able to save buildings that have alternative energy. His next quote exemplifies the narrative that has shaped the solar industry for the last decade.

There is a possibility of electric shock because the electricity to the panels can't be shot off. And not having a clear path on the roof to cut a ventilation hole is another challenge. Are these valid concerns? For sure. And to be clear, this was not a narrative that started with the firefighters of New Jersey. This was a common concern among firefighters and still persists today. And you could take that information and say, okay, that's fine, but that's a firefighter's job to figure out. But if you keep seeing news stories and lobbying from emergency responders saying that they will not fight fires on buildings with solar, the next thing that happens is insurance companies start taking notice. If insurance companies start seeing that buildings with solar are consistently not saved in the event of a fire, your savings from a lower electrical bill might start getting offset with higher insurance premiums for buildings with solar. So the industry needed a solution. And the solution at the time was code based.

The National Electrical Code in 2014 introduced a new requirement called Rapid Shutdown of PV systems on buildings. Or, colloquially in the industry, it's just known as Rapid Shutdown. It's required that PV conductors more than 10 feet from a PV array must be limited to 30 volts or less within 10 seconds of shutting down a solar system. Now, this seems like a reasonable response, right? For firefighters worried about randomly live DC wires that could exist anywhere in a building. This is a legitimately good and reasonable compromise. As a firefighter, as long as you avoid the solar panels and hit the Rapid Shutdown button, you don't have to worry about anything.

But, apparently, it was not enough. Because in the next division of the code, three years later, proposals were introduced that created more stringent Rapid Shutdown requirements. There were several different ideas, but the one that won out was possibly the most onerous. This version of Rapid Shutdown that we find in the 2017 NEC has several parts to it. First, like the previous versions, it addresses electricity outside the boundary of a solar array. But this time, any PV wires outside the array must be 30 volts or less within 30 seconds, so a little longer, but within only one foot of the array as opposed to 10 feet. So, similar to 2014, but a tighter boundary with a little longer reaction time. Second, inside the array, must also be able to be de-energized to 80 volts within 30 seconds. This means you shouldn't be able to take a voltage tester up there, poke into any two random wires under any given panel, and get more than 80 volts once the Rapid Shutdown button has been pressed. It's a small block of text, but with absolutely massive implications for the industry.

Classic setups have solar panels work together by plugging into each other in series, daisy chaining until the ideal voltage is reached, usually around 500 volts or more when not producing. This is generally really easy and convenient to install because you can set a panel, plug one end of the panel into the previous one, clamp it down, do a little managing of the wires so they aren't drooping on the roof, and you're onto the next one. Rinse into repeat until you have a solar array. With this change in the 2017, you basically need to have some way to isolate every single solar panel from every other solar panel within a moment's notice.

Whatever you're using needs to have enough communications to respond to a button press at the service of the house and be able to interrupt power very quickly. And what this suddenly meant for the solar industry is that we had to put extra electronics on the roof to fulfill this safety requirement under every single panel. Now I don't know about you, but when I think of reliability and safety, I immediately don't think of adding extra electronics in an extremely thermally challenging environment. It gets hot, it gets cold, there's a huge range throughout the day. And I certainly don't think of adding 30 of these devices in a residential system, for example, or potentially thousands of them in a large commercial installation.

I mean, how is the solar industry even going to come up with a device to do this all of a sudden with this code change? Well, I guess we already have them. So up until it's pointing the industry, we've had a few different methods of installing solar. Microinverters, optimizers, and string inverters. String inverters tended to be the most cost effective, where microinverters and optimizers touted the ability to monitor each and every panel for its output, as well as optimize the panel for its best production possible, depending on the shade or the angle of the panels. And the way optimizers and microinverters work is by having a device under every single panel to do that monitoring and adjusting. But they're also more expensive, and that cost difference can be difficult to sell to a customer. If only there is a way to squeeze out the competition.

Anyway, in a completely and legally unrelated note, three of the companies that submitted this change to the code to involve module level rapid shutdown are Enphase, Solar Edge, and Tigo, who coincidentally at the time were also the leading microinverter and optimizer companies. Their technology was easy to adapt to this new code change, and the more efficient, but now non-compliant string inverter companies were left fumbling for solutions. In fact, even though there was this proven technology in the space with optimizers and microinverters, this change was so significant that the code post-dated this change to only be effective two full years later in 2019.

So now in 2019, we're at the current status quo for the solar industry in the US. All solar installed on a building needs to have these rapid shutdown devices, whether they be microinverters, optimizers, glorified relay devices, what have you? What I've found during this period as a solar technician is that I have far more callbacks to sites with microinverters, optimizers, and other rapid shutdown devices, whether it be device failure or improper installation due to higher complexity. Reliability of these devices has gotten better over the years, but by and large, the most common things to fail in a system given proper installation is the power electronics and inverters. So more of them in a system, the more likely weird stuff can happen. Plus, that's not even considering the sheer number of connectors we have to use on a roof with rapid shutdown, which ends up posing an exponentially higher risk of something failing.

Helio Volta, a third-party auditor and maker of solar grade, the solar-specific on-site inspection and installation software, they have a massive data set of how solar systems fail and in what way, and they've gone on the record in Solar Builder Mag that these rapid shutdown rules are unbalanced much more likely to create thermal events, which is a nice way to say fire, and increase the danger to firefighters because it simply requires them to show up to fires more often. But we simply have no recourse. The code is the code, and even though anyone can submit to get the code changed, getting rules that are couched in safety to be reverted is always a huge and almost impossible fight. So we pretty much just have to deal with it. Right?

Well, let's look at some actual evidence around firefighter safety and PV systems, shall we? In 2017, the IEEE published a study by Sandy and National Laboratories that evaluated what shock conditions actually exist for firefighters, given worst case illumination scenarios and taking into account the equipment they wear. They found that any risk posed to a firefighter from a solar installation is far lower than the limits defined in IEC, T.S. 60479-1, which in short is the standard everyone uses for evaluating electrical risks to humans and livestock. And not only did this study show that result, but it was also studied in Germany where they found the only life threatening risk to a properly outfitted firefighter was being fully wet and exposed to a thousand volts from hand to hand. The hand of foot was fine. It was hand to hand and fully wet. So again, why did we do all of this? And is there any way to find an escape hatch where we can help firefighters feel safe while limiting the unreliability caused by this code rule change?

Well, here's where I finally get to tell you what UL3741 is. In the 2020 IEC, another three years later, we get another significant change to the code, one that introduces an alternative to module level rapid shutdown. And that is PV hazard control systems. That's just a fancy way of saying, use something listed with UL3741. Now, UL3741 is a standard released in December 2020 to define systems for photovoltaic hazard control. The scope of the document is to evaluate systems that can reduce risk to firefighters in a PV installation. It basically requires the manufacturer of the system to put it through a huge battery of tests. And if under all situations in question, there's not lethal risk posed to a firefighter. It's listed. The standard is wild and incredibly in depth and very specific to the parts specified in the system. It's no surprise that it took a full year and a half before we saw the first system get this listing.

A commercial ballasted roof mount system collaboration between Racking Manufacturer Sellega and inverter manufacturer SMA. But what was so interesting about that was this freshly listed Racking System was not materially different from what Sellega was already shipping. When you read the UL documentation, the first impression you'd come away with is that you would test some new system that would be able to pass this battery of tests that protected firefighters where previous Racking Systems clearly wouldn't be able to. But this Racking System is materially the same. More specifically when it came to wire management and things like that. But Sellega had already been shipping this before UL3741 was the thing. Anyway, just as soon as that happened, we got more and more UL listed products on the market for the commercial space for UL3741. At this point now, today it's fairly easy to find a ballasted roof mount system that you like that is listed with an inverter company that you also like.

The future is bright on the commercial rooftop solar front. But what about residential? We're now installing tons of semi-air fully rapid shutdown free solar for commercial, which is a win for reliability, safety, and efficiency. But homeowners are still stuck with these complex solar options that are less cost effective and less reliable over the long term. Hey, guess what this video is about? As soon as I saw this coming for the commercial solar market, I've been bugging solar racking companies about whether they're working on getting residential roof solutions ready for UL3741. And it's honestly, it's been back and forth. There was a time when people weren't even sure if it was possible, let alone practical. Because just to give some context, the ballast mount systems approved for 3741 are usually plastic, which seems like you'd have a better time of getting certified since there's less metal parts to go shaki-shaki.

Since all the racking for the residential roof mounts are by and large metal, I can't think of a plastic one. It seems less plausible that you'd be able to do the same thing that ballast mount companies did where you'd just take an existing racking system, put it through all the tests, and you're good and golden. Plus, it takes an incredibly long time to just get testing setup and get your product through these tests. So even if you go down that route of putting an existing racking system through the ringer, there's a long lead time regardless.

However, a couple of weeks ago, this post popped up on LinkedIn from Chico USA, and someone from Chico reached out to mention to me that they suddenly got UL3741 approval for residential roofs. And I gotta tell you, this gave me whiplash. I hadn't heard much about all of this from my contacts for a while, but I was generally expecting stuff to start showing up this fall and not from some company that I hadn't heard of before. I reached out to try to get some documentation or more info, but they've been anointingly silent. I don't know if they're busy or they just don't have documentation, or the only place I've seen this from is on LinkedIn. You'd think they'd maybe post to another social or their website.

However, IronRidge reached out this last week to mention that they've been putting the final touches on their UL certification, which that's exciting. IronRidge, dating all the way back to the beginning of my career, is the one I have the most experience with. I've used them on tons of systems, and I know the ins and outs and the quality of their build and their consistency as a company. And so this is very exciting. I've also heard from IronRidge about the lengths they go through for their testing, and they're not going to announce something that they don't have info to back up. And as of this week, they've published their documentation that their residential racking system is listed with UL3741. And as excited as I want to get about this right now, let's look at the documentation first, because like I said, UL3741 requires all parts of a system to be very specifically listed and installed in a very particular way.

So each inverter, wire clip, cable tie, and solar panel needs to be listed with this racking system. Also, who knows what restrictions need to apply, given how thorough this listing is. The first thing to note is the rating section. The railing is rated for a thousand volt systems, but you must use these attachments and components, including those specific cable ties and clips. Any of this type of conduit is good to go, and as well as generically listed PV connectors and PV wire. There's four pages of approved modules at the bottom of the document, which probably covers the vast amount of modules currently on the market.

Then as we scroll down past the rating section, we see the first inverter system and string isolator rated for this system, which happens to be Tesla equipment. Now, so far, this all seems pretty standard, and very much like what you'd see for commercial racking. It does stink that only one inverter is listed so far, but that will inevitably change, just like with the first coming of the commercial racking. And looking at limitations, this all looks very, very similar to the commercial ballast mount requirements, too. If you have one single array, you need to shut down power within one foot of the array, as per code. And if you have two arrays that are close to each other, you don't need an isolator unless they're more than two feet from each other. And it does call out wire management very specifically in the instructions, but honestly, that's expected. And the biggest thing is that you have to use their cable ties and clips. And the wire management instructions is nothing special if you've already been doing it competently.

You can't have it hanging on the roof or sticking out past the panels, it's to be done in workman-like way. Their pictures look like what I would expect any solar system to be managed. So then this installation is not really significantly different from when you install under the 2014 guidelines. There's not really any special parts either. This is functionally the same type of railing that IronRidge has shipped for over a decade. Obviously with tweaks to various things to make the product better in general, because over the course of 10 years, you're gonna just do that if you're being a good company.

Has UL3741 just simply proved that most systems installed competently are of no threat to firefighter safety and you don't need to reduce the voltage to 80 volts at all? Does this just mean the industry has been hamstrung by meaningless regulation based on fear and lack of research? I mean, that's what it seems like to me. It's hilarious and sad. The winding paths this decade is taking, but in short, I'm really glad that we're finally here. It'll still take some time for other racking manufacturers to get their listing as well as get listed with a number of inverter manufacturers, but this is good proof as any to say that we'll finally get there and it'll be totally fine.

We can finally leave optimizers and microinverters behind if we want to. And to be clear, I don't care whether they exist or not, but not having the option of using something else has definitely hurt the industry. We're now so much freer to grasp the future of clean and sustainable energy without being hamstrung by requirements made without research.

Thank you for watching. I'm Derek the Solar Boy and my mission is to help people understand long-term benefits and pitfalls of solar, from the perspective of a field technician. This is a very different video from what I normally do, but if you liked it, please share it with someone that you think might find it interesting and please subscribe. I appreciate each and every one of you.

Source: https://www.youtube.com/watch?v=1MTCvYhjHzA
  • Photo of Duncan Miller

    Duncan Miller

    Founder, Software Developer

    Duncan is the founder and lead software developer for Waivolt. He been working in solar training since 2006 and has been developing software for over 20 years. Duncan has an MBA from Babson College and lives with his wife and two children in Portland Oregon on an extinct cinder code volcano. He is passionate about artificial intelligence, climate solutions, public benefit companies and social entrepreneurship.

Subscribe to our newsletter

The latest NABCEP solar pv certification news and articles, sent to your inbox weekly.