Using rooftop grid-tie solar during a power outage.

[zacksc]

I have a small system that consists of eight rooftop mounted 250 Watt solar panels with individual enphase micro inverters. Generally, when the power is on, that feeds into a PG&E electrical grid. I am planning to install new iq8 micro inverters, one for each panel. I believe the connection between each panel and its micro inverter is MC4. I am thinking of putting a splitter between the panel and the micro inverter for one of the panels to provide a way for me to access some dc power directly from a single panel during a power outage. During normal operation the 2nd unused end of the splitter would be terminated in a safe way protected from moisture or contact. During a power outage I can connect to that via an XT60 or something like that and use that to charge a Jackery or a LiFePO4 battery (via a dc-dc converter or MPPT interface). What do you think? Any advice or help is most welcome. I would be very interested to hear of perils and pitfalls with what I am thinking about.

 

[harrisonpatm]

I have a small system that consists of eight rooftop mounted 250 Watt solar panels with individual enphase micro inverters. Generally, when the power is on, that feeds into a PG&E electrical grid. I am planning to install new iq8 micro inverters, one for each panel. I believe the connection between each panel and its micro inverter is MC4. I am thinking of putting a splitter between the panel and the micro inverter for one of the panels to provide a way for me to access some dc power directly from a single panel during a power outage. During normal operation the 2nd unused end of the splitter would be terminated in a safe way protected from moisture or contact. During a power outage I can connect to that via an XT60 or something like that and use that to charge a Jackery or a LiFePO4 battery (via a dc-dc converter or MPPT interface). What do you think? Any advice or help is most welcome. I would be very interested to hear of perils and pitfalls with what I am thinking about.

In addition to a splitter between the panel and inverter, you would also need some sort of a switch; you will want current to go to the inverter or your off-grid setup, but not both.

Your problem, should you choose to take it on, is that what you're planning is almost certainly not going to be "up to code." Since you have a grid-tie system, I can assume that it was inspected at one point, and if so, any modifications to make to it will be something your house insurance provider may want to know about. Or not. Up to you.

 

[offGridDownUnder]

Generally, when the power is on, that feeds into a PG&E electrical grid.

So, your system puts out mains-level power?

What is the physical connection at the output, where it is supplied to PG&E? Is there an option to physically disconnect it and then connect it for local use?

I can't speak to the legal, insurance, or practical issues - outside my jurisdiction - but that seems like an interesting avenue to explore.

 

[harrisonpatm]

So, your system puts out mains-level power?

What is the physical connection at the output, where it is supplied to PG&E? Is there an option to physically disconnect it and then connect it for local use?

I can't speak to the legal, insurance, or practical issues - outside my jurisdiction - but that seems like an interesting avenue to explore.

I believe OP is suggesting putting some sort of splitter between the panel and the inverter, before it goes to the grid. Thus, he's after the panel's DC voltage, anywhere between 30-60 volts depending on the panel.

 

[A-DamW]

Aside from insurance and regulatory inspection, you should be able to tap into the DC of one panel just fine.

About the only thing I can think of to watch out for would be when the AC power comes back on, the grid tie inverter and whatever you have hooked up to the panel DC might oscillate a bit trying to find MPPT, but once the power is on you can disconnect from DC anyway.

 

[offGridDownUnder]

Thus, he's after the panel's DC voltage, anywhere between 30-60 volts depending on the panel.

Without knowing too much myself, in case there might be some roadblock that makes direct impractical perhaps this can be an option.

 

[Chalo]

I'm setting up 24V DC solar arrays at my place, so the grid (and city code) don't interact with them at all. Also that way I don't need to put panels in series and make sure they're all unshaded.

 

[zacksc]

I believe OP is suggesting putting some sort of splitter between the panel and the inverter, before it goes to the grid. Thus, he's after the panel's DC voltage, anywhere between 30-60 volts depending on the panel.

Yes, exactly.

 

[zacksc]

Aside from insurance and regulatory inspection, you should be able to tap into the DC of one panel just fine.

About the only thing I can think of to watch out for would be when the AC power comes back on, the grid tie inverter and whatever you have hooked up to the panel DC might oscillate a bit trying to find MPPT, but once the power is on you can disconnect from DC anyway.

Thanks. When the power goes off I can turn off the main breaker at the panel. That will prevent the microinverters from turning back on while I am tapping into dc power. (Is it a good idea to turn off each individual breaker as well when I do that and then turn them back on one by one? Is that the best practice?)

Although if the micro inverters did turn back on while I was drawing dc power from one panel, I believe that the system would just think that one panel is not working well, exhibiting low voltage. Maybe not that different from if it was in the shade or covered with leaves. I'm not sure, but that would be my guess? (Or maybe it could oscillate? I'm not sure how the MPPT aspect (of the micro inverter) would respond to this situation.

PS. I am thinking that for the dc charging I might use, for example, a Victron Phoenix dc-dc converter that accepts 20 to 40 volt input and puts out 14.2 volts during bulk charging... I will have to research that. I'm not sure what they have or what the exact voltage range of my panel is.

 

[harrisonpatm]

Thanks. When the power goes off I can turn off the main breaker at the panel. That will prevent the microinverters from turning back on while I am tapping into dc power. (Is it a good idea to turn off each individual breaker as well when I do that and then turn them back on one by one? Is that the best practice?)

Although if the micro inverters did turn back on while I was drawing dc power from one panel, I believe that the system would just think that one panel is not working well, exhibiting low voltage. Maybe not that different from if it was in the shade or covered with leaves. I'm not sure, but that would be my guess? (Or maybe it could oscillate? I'm not sure how the MPPT aspect (of the micro inverter) would respond to this situation.

PS. I am thinking that for the dc charging I might use, for example, a Victron Phoenix dc-dc converter that accepts 20 to 40 volt input and puts out 14.2 volts during bulk charging... I will have to research that. I'm not sure what they have or what the exact voltage range of my panel is.

This all sounds good and correct. If you end up going this route, please do share how you decided to switch the panels, I'd be interested to see it.

 

[harrisonpatm]

I'm setting up 24V DC solar arrays at my place, so the grid (and city code) don't interact with them at all. Also that way I don't need to put panels in series and make sure they're all unshaded.

This is the way, I've done the same with 5000w of solar at my house.

 

[d8veh]

These are pretty good if you have somewhere to put it. I have one, and it works perfectly. You can charge it from solar, a generator or the grid. It has 15kwh stored in the battery:

https://vi.aliexpress.com/item/1005007602765814.html?spm=a2g0o.order_list.order_list_main.225.63f11802IXga1U&gatewayAdapt=glo2vnm