Safety of PV systems: Your questions, our answers
The safety of a PV system depends, among other things, on the design of the overall system. Modern string inverters with integrated features enable a slim system design. This avoids potential sources of error, for example through additional cabling effort when installing module electronics. In a webinar “Analyzing PV system safety and the promise of optimizers” in summer 2022, Dr. Heribert Schmidt from the Fraunhofer Institute for Solar Energy Systems ISE and SMA expert Hannes Knopf discussed the influence of different system designs on the safety of PV systems. Afterwards we received so many exciting questions on the topic, which we would like to share with you. Our colleague Hannes Knopf answered the most frequently asked questions here.
String inverter vs. MLPE: Impact on system safety
Are you saying that the biggest risk is the increase in points of failure?
Correct.
What means less complexity in a PV system?
Less complexity means less devices in a PV system, which means less interconnections, which means less DC-connectors, and less potential points of failure. Less is more – less complexity is more safety.
Several Systems need defined DC Connectors to work good – what is the reason for this? For example, SolarEdge
Typically, PV modules are produced by a different manufacturer than MLPE devices (optimizers etc.). It is very likely that the DC connectors which come with these products are also from two different manufacturers. This is called “cross mating” even if some manufacturers claim DC connector’s compatibility. There is no standard for testing compatibility of DC connectors between different manufacturers, resulting in a significant risk for contact failures, arcing and potential fires. IEC Technical Report TR 63225 addresses this issue and states: “Several countries report that problems with DC connectors are a major cause for failure such as fire hazards in PV systems”.
This is very theoretic. What is real fire by optimizers?
Since typically, neither insurance companies nor firefighters specifically report fires involving or caused by PV systems as a separate class from other fires caused by electricity, it is very hard to get those details. Only very elaborate in detail research, such as done by the TÜV/Fraunhofer study, will be able to provide such data. And this research concludes that additional connectors and complexity are counterproductive in regards to the safety of a PV system.
“Were there optimizers in the famous Walmart fires?”
There are sources that report that optimizers were involved in the Walmart fires, see: https://www.businessinsider.com/tesla-project-titan-replace-bad-solar-panel-parts-2019-8
The fact that there is not much data, is it because the PV system technology and installation is fairly recent, that is, it has not been in place for a long time?
As a matter of fact, in many countries, a large number of PV systems have already been operating for many decades. As a pioneer of this technology, we at SMA have been active is this field for more than 40 years by now. This long history of the use PV systems actually shows that they have always have been a very safe and reliable source of energy and that there are only very few reports of PV systems having caused a fire.
Click here for more answers:
The independent institute Fraunhofer ISE determined that to date, less than 0.006% of all PV systems in Germany have caused a fire. But 0.006% from million PV systems on roofs is not small number, how can you state that the rapid shut-down is not require?
0.006% is in fact a relatively low number. Adding rapid shutdown to PV systems would not reduce this number any further. On the contrary, it would increase the potential points of failure and thus also rather increase the number of potential fires. Instead, to effectively reduce this number towards zero, SMA has decided to introduce AFCI (SMA ArcFix) into all our string inverters.
Why would insurance companies require MLPE and rapid shutdown systems if it’s not for adding safety to the PV systems? Same for requirements from norms and standards as the NEC in the US?
It is unknown on which facts and scientific evidence those requirements in the US and by some insurance companies are based on. The results of the very extensive and competent research done by TÜV and Fraunhofer would strongly suggest otherwise.
You presented the risk from PV system but as you presented all of them are safe. What about the risk of fires from other sources and firefighters need to reach the roof to handle it. How they will do it without RSD?
This question was analyzed intensively in chapter 4 of the study done by TÜV/Fraunhofer. They refer to the safe handling procedures defined in DIN VDE 0132 and conclude “The studies on conductivity and possible leakage currents through nozzles have shown that when extinguishing fires near photovoltaic systems rescue workers are exposed to no serious danger from the leakage currents via the quench water exist at 1,000 V DC, if the safety distances as per DIN VDE 0132 are maintained.” Also see next questions/answer for more details.
In case of fire or other issue on the roof, without MPLE system, how can you shut down the high voltage that running in the roof?
There are safe procedures in place and known by trained firefighters on how to deal with electrical systems in case of a fire. One of the outcomes of a study done by TÜV/Fraunhofer was that Rapid Shutdown (RSD) Systems are never 100% reliable and thus a firefighter should never rely on them. Instead, firefighters must always treat any PV system as fully energized or “life “. Adversely, by adding rapid shutdown devices to a PV system, the number of points for potential failure increases dramatically and thus the risk for fires, making firefighter’s interaction necessary in the first place. Therefore, rapid shutdown systems are not beneficial for firefighters.
PV systems very seldom cause fires. However, is there statistics when PV Systems have resulted in that the fire fighters choose not to act due to PV modules? Is there statistics spilt up between string inverters and systems with optimizers?
There were in fact isolated reports in the past, as PV systems were just newly being introduced in certain regions, that there were local firefighters at the scene, who were still unfamiliar with this new technology and on how to deal it, and consequently did not know how to proceed. Today, there are standards in place and extensive information and training materials available for firefighters on how to act in the presence of PV systems.
What about the discussion about the installation requirements in Australia (NSW esp) REMOVING DC Isolators as they were a (signficant) point of failure?
Regarding DC isolators, PV Magazine Australia reported in June 2020, “Often referred to as the weakest link in the rooftop PV safety, DC isolators were mandated in Australia in 2012 through the Australian Standard for PV installations (AS/NZS 5033). This made Australia the only country in the world requiring rooftop DC isolators. Since then, they have been proven by inspection data to be the largest single source of conventional DC solar system failures.”. This confirms our argument, that the installation of additional components in your PV system also adds additional points for potential failures. As a consequence, and lesson learned from the reported Australian experience, the responsible committee for the maintenance of the Australian Standard for PV installations (AS/NZS 5033) has now finally decided on “removing the mandate for DC isolators to be installed on household solar systems if other safety measures are followed.”, as reported by PV Magazine Australia in November 2021.
PV systems are normal e-systems and I can savely put water on it from 5m?
Correct. See also the pdf file linked here.
String inverter vs. MLPE: Impact on energy yields
Did you do a yield analysis of PV Systems with MLPE optimizers and without?
Such a study was conducted by the University of Southern Denmark and published in 2019. It shows that the yield benefits of MLPE or “optimizer” systems are in most cases non-existent and that as a matter of fact, for most PV systems they perform better without “optimizers” under shaded and non-shaded conditions.
How does ShadeFix increase output for an unshaded PV system? Does it add components & complexity to the PV system and, if so, is it worth the additional safety risks?
SMA ShadeFix is integrated into every SMA string inverter as a smart software function. No additional components are necessary. Complexity remains low, and safety remains high.
Since ShadeFix eliminates the need for additional devices such as “optimizers”, it also eliminates the additional power consumption in a PV system by those “optimizers”. Consequently, as shown by the study done at the University of Southern Denmark, a PV system without “optimizers” typically outperforms a system with “optimizers” in non-shaded, and for most cases, even under shaded conditions.
But unshaded systems are also affected by module mismatch that cannot be properly handled by string inverters.
Manufacturing tolerances for the output power of modern PV-modules are extremely low. The effects of PV-module mismatch in PV-systems installed today are likely very much overstated.
Currently of course we are speaking about disadvantages. What about the better performance in MPP-tracking. I have the feeling shade fix is more a marketing idea than a real performance booster?
The study conducted by the University of Southern Denmark actually shows that PV systems with SMA ShadeFix typically outperform systems with “optimizers” under shaded and unshaded conditions.
SMA sold in the past micro inverters SB240 and Optimizers (Tigo) but stopped. I presume there were too many problems or are there also other reasons?
In the past we sold Tigo “optimizers” through our distribution channels, because customers were asking for them. But since we’ve learned so much about the disadvantages of “optimizers” throughout the last years, we honestly do not see any benefit for our customers buying them. That’s why we discontinued offering “optimizers”.
Is ShadeFix an algorithm specific from SMA that is patented or is it a well know algorithm used by many inverter manufacturers?
SMA ShadeFix is a patented software function integrated exclusively in SMA string inverters.
If one MLPE fails, you only loose one module. So how does that compare to an inverter failure when you lose the whole array?
Because of the high number of MLPE devices in a PV system with “optimizers”, the likelihood of failure of a MLPE device in such a system is much higher than the probability of failure of a string inverter. With the MLPE devices mounted behind the PV modules, their accessibility for maintenance and repair is also an issue, while string inverters are typically installed in a much more accessible location. To minimize the number of truck rolls and associated effort and cost for repairs, it is very likely that installers do not attend to failed “optimizers” as timely as they would to a failed string inverter, thus causing the accumulated energy losses by failed “optimizers” outrunning the losses by a potential downtime of a failed string inverter over the expected lifespan of the PV system.
String inverter vs. MLPE: Impact on energy yields
Can SMA ArcFix be considered a replacement for rapid shutdown?
SMA ArcFix does not replace rapid shutdown; it is a totally different function. It detects potential arcs in the PV systems and suppresses the arcs so arcs cannot develop. It is totally different from rapid shutdown. Rapid shutdown has nothing to do with arcs.
Aren’t most fires from arcs caused by the MC4 connectors in wet circumstances?
Most arcs are caused by faulty crimping of connectors and by cross-mating, i.e. the combination of connectors from different manufacturers in the field.
How does SMA ArcFix work?
This Arc Fault Circuit Interrupter (AFCI) feature effectively detects potential electric arcs and interrupts the electric circuit before a fire can develop. The system automatically restarts if a false alarm occurs. The future integration of SMA ArcFix into PV inverters will offer several benefits. For one, you will not need to install additional arc fault protection devices, which saves you time and money as well as ensuring full system compatibility.
Are the Sunny Tripower 3.0 -10.0 inverter equipped with SMA ArcFix?
All our inverter models for the US market are equipped with SMA ArcFix. Outside the USA, our SMA inverters Sunny Boy (SBx.x-1AV-41) and Sunny Tripower CORE1 (STP 50-41) are currently equipped with SMA ArcFix. Other SMA inverters will follow.
Is there a European (IEC) standard for arc detection?
SMA ArcFix is certified to fulfill the latest existing US standard UL 1699b as well as the upcoming new IEC 63027 standard. This IEC standard currently still exists in a draft version but is expected to be officially released soon. SMA already fulfills both of those standards.
Is SMA ArcFix automatically included in the power converters? Is a F/W (firmware) update enough?
SMA ArcFix is not available as a software update. It must be integrated in the inverter directly at the factory. The inverter’s design needs to be modified accordingly. We have already started this process and we will be converting all our string inverters to include this functionality in the future.
ArcFix has been integrated in our U.S. inverters for more than a decade now, since it is a mandatory functionality in this part of the world. We were the first manufacturer to offer an integrated and certified AFCI solution in the United States. As a result, we have more than a decade of experience in engineering and designing arc-fault detection systems and integrating SMA ArcFix into our series products for the U.S. We are now finally introducing SMA ArcFix in our products for the rest of the world as well.
Does the SMA ArcFix feature work similarly to circuit breakers or comparable devices?
A circuit breaker is a mechanical system that works very differently from our SMA ArcFix function. SMA ArcFix is a combination of specialized sensors and software algorithms for arc detection and power electronic devices for interrupting the circuit.
Tip: Find more information about circuit breakers in the Technical Information.
You want to watch the webinar on the safety of PV systems?
We have an on-grid solar inverter Model No. SOLID-Q 50 installed in our factory.We have 92 Nos 330Wp Mono PV module(30.36 KWp) having Voc-46.30V,Isc-9.24A, Vmpp-38.0V, Impp-8.7A ,FF-77.14%,Now we want to add 54 Nos 575Wp Top Con PV Module( 31.05KWp) having Voc- 51.3V,Isc-14.29A, Vmpp-42.6V, Impp-13.51A . Please suggest It possible to do like this.
Hello Sandip,
Please consider that we are a manufacturer of inverters.
For that reason, we cannot give you any advice regarding your technical request.
Sunny regards
Christiane