A Tigo YouTube video interview from 2019 has recently sparked an engaging conversation on regulations and solar system behavior when the grid is down. I thought this would be an excellent teachable moment since the conversation touched on customer expectation management.
Here is the major premise from the customer:
"I think anyone would assume PV would still work during a grid outage as there is no fundamental technical reason why it should not. Inverters could take their internal power from DC, PV or battery as the latest 'hybrid' inverters seem to. I really hope this regulation does not force us back to this unsatisfactory situation but allows a separate PV 'kill switch' to be fitted next to the grid isolator (with prominent labeling, of course)."
What the commenter is referring to is the anti-islanding requirement programmed into the nonvolatile firmware that stops power production when the inverter detects a grid outage. Every listed PV inverter in the world has this requirement. This important safety feature prevents solar backfeed into utility lines to protect line workers trying to restore power. IEEE and IEC are responsible for this requirement.
Anti-islanding is a frustrating requirement to a homeowner who found out about it the first time their grid went down and their solar inverter stopped producing. All those modules became useless. I blame the installer for this lack of customer expectation management.
Currently, the only way around anti-islanding (for an existing solar system) is to install a hybrid inverter and a battery system along with an automatic transfer switch. The hybrid inverter creates a mini-grid and prevents solar backfeed when the grid is down. It essentially tricks the existing PV inverter into thinking the grid has returned.
Now, this is where the self-taught engineers and hillbilly ingenuity practitioners (of which I belong) will chime in.
"Why can't we just mandate an automatic transfer switch (ATS) on every solar installation and then remove the anti-islanding requirement from the inverter?"
It is a valid question and I thought about it a lot when I started with SMA back in the late 2000's. I even brought it up to a German colleague who had experience with alternative solutions, including mandatory ATS installation. His conclusion was that there was no reliable way to accomplish a mechanical isolation solution that would work harmoniously with the PV inverter. Let's look at a scenario.
The automatic transfer switch detects a grid outage and switches to isolate the house from the grid. The PV inverter restarts. But what would happen if the ATS failed to switch over? The inverter would restart and any excess solar would backfeed into the grid. Have you ever been shocked? It's not fun.
The counterargument to this scenario is to connect a signal wire from the ATS to the inverter so that the inverter would "know" if the grid went down the ATS failed to flip over. Sounds reasonable, until you consider system behavior if the wire was severed or connection was lost.
Anti Islanding provides the safest path to least resistance and does not rely on humans or mechanical choke points.
There are some PV inverters with a gimmicky feature that allows a portion of available PV to feed an internal isolated terminal that connects to an electrical outlet. This outlet can provide some level of power during a grid outage. SMA's Sunny Boy with Secure Power Supply is one such inverter. This feature debuted in 2013 and it was a big hit in the industry. It was the first inverter in the world that could keep your phone charged and even keep a fridge going while the grid was down- without batteries! I have the 5kW model on my home and I love it. Other companies have seen the benefits of this feature and now offer a similar solution.
What do you think? Is there a way to reliably retract the anti-islanding requirements but still offer a high level of safety and reliability?
Please sign in to leave a comment.