diy solar

diy solar

Sol-Ark 12K with AC coupling

ArthurEld

Solar Wizard
Joined
Aug 3, 2020
Messages
2,270
Location
Palm Harbor, Florida
I was asking questions in the Enphase thread and Ampster recommended that I start a new one.

I am about to install a new grid tied hybrid system and I want to use some Enphase Iq7 micro inverters because I have shade and different angles.
Since I knew that batteries are DC I was hesitant to use micro inverters on all the panels. I remember hearing somewhere that it is more efficient to charge the batteries with DC as much as you can within reason. I would enjoy hearing what others have to say about this subject. There are obviously smarter people than me around here. The attached picture was in the email.

This is part of the email conversation I had with Sol-Ark tech support-

Thank you for the information Yang.

I actually want to use micro inverters. The manual shows different ratios of AC to DC described as "Best", "Good", "OK" and "Poor"
Since I am designing my system from the start, I want to set mine up "Best" not "Poor"

Please tell me if the way I placed the panels and micro inverters are going to work well. I can add more micro inverters or take some away. I just want to be sure to do it in the best possible way so the system will work efficiently.

Here's a picture of my roof.

The two strings of 9 panels are facing West. Also the one string of 4 panels with micro inverters is facing the same West direction. There is a string of 8 panels on the East side of the roof. And there is one string of 4 panels with micro inverters facing South. I'm sure there are other possibilities. There are some shadows from 2 palm trees starting to show on the 15' x 43' flat roof. They will move over about half of the panels in the afternoon.

Thank you,

And here's the answer-

Thank you so much for the additional information.
Yes, your plan will definitely works well.

Since you are grid tied, you don't need to worry about derating of AC coupled and batteries.

Generalized application
MPPT1: 2*9*330=5940w
MPPT2: 1*8*330= 2640w
AC:. 2640w
Total: 11220W

Above is my recommendation for your layout
Feel free to contact me yang@sol-ark.com
Or support@sol-ark.com

Best regards,
Yang Chen
Sol-ark engineer support

I'm persistent so I asked this too-

Thanks so much for confirming my setup.
Can you please tell me the maximum # of micro inverters I can AC couple while staying in the "Best" efficiency? I plan to use 50kwh of batteries. And I will use 34 330W solar panels.

reply-

I want to specify even though you can put up to 7.6 AC coupled stuff on Gen input, but there is a max number of Power that can be feed into a 12K.
In your case, 3kW is the max number you can put on the AC coupled side.

And we our Voltage-based on our battery input. As long as your battery is 48V nominal, our system will work with them
We do have some battery partners (Communication to the 12K)
Blue Ion, Discover, Pylontech, Dyness, and fortress eflex (In progress) are some of the common ones that we communicate directly with (no battery setup needed on our system)
 

Attachments

  • panel layout.png
    panel layout.png
    426.6 KB · Views: 33
This sounds like a nice setup.

When I looked into the Sol-Ark, I was concerned about the AC coupling limit. All of my solar is currently 16 panels with the Enphase iQ7 inverters. They are 300 watt panels for a max DC power of 4,800 watts, but the inverters clip it to a max of 3,860 watts. I saw my production flat line there several days in spring when it was cool and sunny.

As much as I liked the Sol-Ark, I did not feel comfortable using it for an all AC coupled system.

For your setup, I would suggest you find all the panels that will have partial shading, and put those on the micro inverters. Then all the panels that are in the clear and facing the same direction and tilt angle go onto the common MPPT input. A string on a different angle then goes on the other MPPT. The micro inverters are by far the bast at dealing with shading issues. A string to an MPPT charge controller is the most efficient at making DC, but will suffer a bit even if just one panel in a series string is shaded. 2 parallel strings are a little better. A shaded panel will lose current, but the parallel panel will still make all of it's current. If you look at panel output graphs, you will see the voltage is fairly constant, it is the current that changes most with light intensity. The Enphase iQ7+ will output up to 290 watts so 10 would be the limit for 3,000 watts. My smaller iQ7-60 clip out at 240 watts, but that would be wasting a fair bit of your 330 watt panels, and if they are more than 60 cells, they are not able to take the higher voltage like the + can. So I would shoot for the 10 most likely to get shade.

11,220 watts of panel DC is 2.33 times what I have. That would certainly completely power my home, but I don't have near that much roof area. I can maybe add 6 to 8 more panels and still be legal on the roof walkway clearance.
 
Now you have your very own thread. The biggest question I have is regarding shade. It looks like your front roof gets some shade and that is where I would put the microinverters. Take a look at Google Sunroof for your address and see if their view shows any indication of shadows. Initially I thought the 8kW SolArk can take up to 7kW of AC coupled solar but I was surprised to learn that the SolArk is limited in terms of AC coupling to 3kW despite what the brochures says. Maybe you lose some efficiency or the inverter clips. Did he say much more about that difference? Designing for shadows first with micro inverters is going to be tricky with that limitation. Was the guy at SolArk aware of the shadows?
There are some devices you can rent to give you a more complete analysis of shade if you don't get the information from Google Sunroof.
@GXMnow made many of the same comments I did and pushed the enter button before I got a chance.
 
For some reason I didn't get a notice that you guys replied to this. I sure get lots of other notifications from this site.
The guy from Sol-Ark was aware of the shadows. I mention them in my email above. Cutting and pasting my emails didn't work as well as I hoped.
I agree that shade is a big consideration. I am fairly sure that all of my panels will get at least 4 hours of clear sun. And I'm putting up a lot of panels.
But I am going to keep an eye on the effect of those shadows on the panels without micro inverters. I may decide that it is best to put more micro inverters up.

I have a lot of roof space and panels aren't that expensive so it makes sense to me to over panel. I plan to charge 2 electric cars in the future. Until then I think the power company is going to like me for pumping so much electricity into the grid. Also, I am not done hammering out all the details with my installer. He very well may sway me one direction or another. Like if he is more comfortable using a different hybrid inverter like Outback or Schneider, I will probably use it as long as it gives me everything I want. There are many possibilities as far as panel placement too. But I do plan to document my system here when I have more firm details.
 
Actually, the electric utilities do not like us for pushing power to the grid. They would much rather charge us for power and have full control of the supply. At first when there was just a few people with solar panels, it was no big deal, but now they actually have issues is some areas with having too much power on the grid during solar noon, and they have to throttle back the power plants. True, they burn less fuel, but it is not easy to adjust the output quickly. Imagine an area with a large amount of solar. The power company rolls back 30% of the power as solar noon comes up. But then a bank of clouds roll in. The power dips, and now they need to make up the difference. Then there is the evening as the sun goes down, just as people are cranking up the A/C, cooking with an electric range, and watching a movie on their big home theatre. The "Duck Curve" is a real issue. This is why adding the "Time of Use" battery storage is a good thing, but the power companies do not want to admit it yet. One of the things that the utilities have been trying is putting their own huge battery banks by the sub stations. If the grid starts to over produce, they kick it into charge mode, and when load spikes, it can quickly switch to discharge to cover it. This is one reason I went with the Schneider, I think it will be capable of them using my system as a small local peaker, and hopefully they will eventually start paying a higher rate to pull battery power when they need it. This is why I need to be sure my system is totally legal.
 
I am going to keep an eye on the effect of those shadows on the panels without micro inverters.
That will be harder to do with panels in a string since you won't have individual panel data. One panel may drop the entire string output which may be a clue. That is why shade is important in panel design because one panel in shade will drop the entire string so it is a multiplier.
Good luck on whatever you and your installer decide.
If you are using an installer he may have tools to tell you where shade will have biggest impact.
I am a big fan of overpanelling. One advantage of hybrid inverters is that you can control how much is sold to the utility. Do you or your installer know much about the Net Metering policies in your jurisdiction.
 
Last edited:
Actually, the electric utilities do not like us for pushing power to the grid. They would much rather charge us for power and have full control of the supply.
Ironically they come up with pricing that drives more people like you, me and others to invest in batteries and inverters so we can do more self consumption. In economics that is known as the death spiral.
Also their whole command and control system was designed for one way flow. The actual transformers can handle power flowing in both directions but the systems that monitor the substations cannot and the infrastructure needs expensive upgrades. I would not mind paying for some of that if they would implement some kind of communication system that would allow me to send power to the grid during the neck of the duck curve.
 
I should see a drop in the whole string as the shadow goes across. Those shadows also move south as the sun moves west so they hit less panels than it looks like in the picture. And when they hit the panels, it is already past peak sun hours.
I will look into the implications of pushing too much energy into the grid. Our net metering policy is that you can pull back whatever you push in for no cost accept a monthly connection charge. It is settled up once a year. And the price they pay for unused is not good.
 
This is one reason I went with the Schneider, I think it will be capable of them using my system as a small local peaker, and hopefully they will eventually start paying a higher rate to pull battery power when they need it. This is why I need to be sure my system is totally legal.
My apologies to @ ArthurEld, I forgot we are on his SolArk thread talking about California issues. Tesla has made an application in California to do something similar to what they are doing in Vermont. This may become more common throughout the country and stimulate the installation of more hybrid inverters which can take advantage of those kindv of opportunities.
Your point about having a legal system is important.
 
I plan to get permits and do everything legal. I know this is a subject for another thread but what is up with the batteries? How come they don't need to be UL listed? I imagine that a home inspector would have a problem with these battery packs if I was selling my house.
 
I plan to get permits and do everything legal. I know this is a subject for another thread but what is up with the batteries? How come they don't need to be UL listed? I imagine that a home inspector would have a problem with these battery packs if I was selling my house.
Good question. I think integrated battery systems like Tesla Powerwall, LG Chem Resu, and some others may have UL approval. There is another regulation that is starting to trickle down to jurisdictions in California about how much Lithium battery storage you can have in one location. I think the limit is 20 kWh. Others may have more details.
 
Since the battery bank is just 48 volts, it may fall under some other guidelines. From what I have seen, only batteries with a built in inverter that puts out 120 vlts or more are UL listed. People have been using large lead acid battery banks for many years without any kind of national safety testing. My LG cells are still in the same package that holds them in a Chevy Bolt. My brother and his wife park 2 of them in their garage. That is 120 KwH's of the same packs. And yes, mine will be in a steel enclosure once the system is all tested at full power so I know my connections and the BMS are all happy. If anyone tries to safe my battery is not safe, I'll ask them about having a Chevy Bolt in my garage. I am going to add Class T fuses at both ends of the battery cables as well, so even if someone takes a machete to the cables, it should pop relatively safely. Everything bolted to my garage wall is UL listed and installed to code. The battery bank will be in an enclosed metal box on the garage floor. I over spec'd all of the cables and have extra current limiting devices. The last thing any of us want is a system that is not safe.

I am at roughly 17 KwH's of storage, but I could add more in the future, it it looks like I need it. I have not found anything about a limit on solar storage battery size, or if there is any special restriction on Lithium batteries. Maybe that will change if an EV ends up burning down a house. But let's all do what we can to be sure the first bad lithium battery house fire is not a home built power wall. I have to admit, seeing entire walls covered in 18650 cells is a little creepy.
 
I looked at someone else's permit application and it didn't have anything about the batteries. It had pictures and descriptions of everything else but not the batteries.
 
Since the battery bank is just 48 volts, it may fall under some other guidelines. From what I have seen, only batteries with a built in inverter that puts out 120 vlts or more are UL listed. People have been using large lead acid battery banks for many years without any kind of national safety testing. My LG cells are still in the same package that holds them in a Chevy Bolt. My brother and his wife park 2 of them in their garage. That is 120 KwH's of the same packs. And yes, mine will be in a steel enclosure once the system is all tested at full power so I know my connections and the BMS are all happy. If anyone tries to safe my battery is not safe, I'll ask them about having a Chevy Bolt in my garage. I am going to add Class T fuses at both ends of the battery cables as well, so even if someone takes a machete to the cables, it should pop relatively safely. Everything bolted to my garage wall is UL listed and installed to code. The battery bank will be in an enclosed metal box on the garage floor. I over spec'd all of the cables and have extra current limiting devices. The last thing any of us want is a system that is not safe.

I am at roughly 17 KwH's of storage, but I could add more in the future, it it looks like I need it. I have not found anything about a limit on solar storage battery size, or if there is any special restriction on Lithium batteries. Maybe that will change if an EV ends up burning down a house. But let's all do what we can to be sure the first bad lithium battery house fire is not a home built power wall. I have to admit, seeing entire walls covered in 18650 cells is a little creepy.
I don't have details but plan on inquiring at my Sonoma County building department. My understanding is that it is a new fire code. Those things roll out to cities and counties at different rates. It may just be a California thing.
 
No worries, it is your thread.

Well, to top it off, I'm thinking optimizers are probably a better idea for me since I'm starting fresh. Then I don't need to worry about AC coupling.
You mentioned optimizers in another thread about micro inverters and it got me thinking. :unsure: I've got more investigating to do.
 
Micro inverters aren't that much more in cost then optimizers when I checked a year ago. My opinion is that if you have to use optimizers you might as well use microinverters.

I have 24 panels and Enphase inverters to net meter. I then added a Outback GS4048 to ac couple through a critical loads panel and have 6 panels and APsystems inverters in the critical loads panel. With a 19.6k battery.

When in grid disconnect I can run continuously and the 2.1k APsystems solar keeps the batteries pretty charged during the day. I can also by throwing 2 circuit breakers, power the whole house off both the Enphase and Apsystems inverters if needed during power outage, though I haven't done that yet. Might try on a cloudy day first.

I do like the sol-ark, just can't beat the price of the GS4048, along with the functionality of it being able to do everything I need....

Good Luck.
 
I was sold on micro inverters.
But I don't want to convert from DC to AC then back to DC to charge the batteries then back to AC.
In my mind, it doesn't make sense to pay extra for micro inverters then AC couple them. I am still researching and I am not very experienced.
If I already had micro inverters on my roof, I would leave them up there and AC couple.
Since I plan to use 50K or more of battery I think it makes sense to dig deeper into this.
I am fairly sure optimizers will help with my shading issues and deliver the energy in the most efficient way for my purpose.
I will use micro inverters if that is what system design engineers recommend. I made the mistake of saying I wanted to use micro inverters when I was researching. I should have let the ones who know tell me how to do it.
I will finish designing my system soon and I will document the reasons for my choices.
 
I don't convert.... My micros convert to AC and I use the power as AC. There is a contact/relay closure that sends the inverter power directly straight through the GS4048 to power the house when grid is up. So the only efficiency loss is at the Enphase inverters which are peak 97.5% efficient. When power is down I convert from battery to AC in the GS4048 and the GS4048 uses AC to charge the batteries back up.

I am guessing by the Sol-Ark spec it does something similar. Sol-Ark lists 96.5% efficient going from Solar to AC to power house but you need to add optimizers if you have shading. I love the Sol-Ark just can't justify the cost. If you are starting from scratch then it makes more sense because you don't have to buy a inverter. I already had the Enphase inverters installed.

I think you are on the right course from what you have said.

Some systems use solar charge controllers to convert DC to DC to charge the battery and then from DC to AC to use the power. Not efficient at all....
 
Back
Top