diy solar

diy solar

Semi trailer complete off grid with micro inverters

Jasondrumguy

New Member
Joined
Mar 1, 2021
Messages
21
I’m building a tiny house from a refridge semi trailer. Getting ready for plumbing/electrical. I got a great deal on 22 used solar panels total of 6500 watts. But some 350, most 300, and some 250. I also will have shade issues. I’m interested in micro inverters but mostly see them in set ups feeding the grid. I have a generator in case I need additional power. I have planned for a lot of DC 12volt items from lights, fans, and water heater DC. Dryer and electric car charge 240 volt. I’m interested in the EG4 stuff mainly because ease of adding batteries. Having a difficult time finding the best way to go from the AC micro inverters to the DC panel. Also consider the isolators. The 25 year warranty for micro inverters is a big plus. Will, a video discussing this and options would be greatly appreciated. Also any recommendations for a dual mini split? I’m planning on a 48v system with step downs. Thank you
 
I have Enphase iQ7 microinverters, and they do work great, but for off grid, I don't think they are the best choice. My system is a hybrid with a Schneider XW-Pro battery inverter. When it goes off grid during a power failure, I have the Enphase micros AC coupled to it, but it is not completely stable. For the few times our grid is down, it works, but I would hate to depend on that for full time off grid.

The sun is down more than half the time, so I would have at least half of the solar array DC coupled to charge the battery bank. I added 2,000 watts of solar panels with a DC charge controller and I am very happy with the results. That DC power is basically what runs my house all night now. The Enphase microinverters provide my AC power during the day while the sun is up and the DC panels charge the battery for use at night. But like I said, off grid, the micros are not the most stable. I have 16 of them, and 5 of them basically refuse to stay producing while off grid. And several more tend to cycle off and on and the battery does not stay charging. I ended up running at a slight loss the whole time. So for a fully off grid system, I would probably use all DC coupled.

If you have a known shading pattern, you could plan the array around that. Series panels that will be shaded at the same time, and parallel with panels that shade at different times. And you can do panel optimizers as well. When I added my DC system, I thought about the Tigo add on optimizers. I may still get them, but my shading is not a big problem and the array is producing well without them. Another good option is using 3 or 4 separate MPPT charge controllers. That way each group is independently optimized, and if something does go wrong and fail, you still have the others to keep making power.

48 to 12 volt step down DC to DC converters work well, but they are likely only 95% efficient, so you will lose a little power. For things like room lighting, see if you can get 48 volt LED lights, or, if you use 12 volt, put the light switch on the 48 volt side before the DC to DC converter so that you are not wasting power running the converter when the lights are off.

For the 120/240 split phase AC power, there are a lot of inverters out there. I would definitely go with a large low frequency type for several reasons. The large transformer really helps keep the power clean and stable under difficult loads. As much as I like to complain about Schneider tech support and software, the hardware in the XW-Pro inverter is an absolute beast. It will drive any load you can throw at it. The 6,800 watt rating is all day, 24/7 without ever a complaint. And you can pull over 5,000 watts on just one leg without issues. It does 8,500 watts for 30 minutes and surge to 12,000 watts for over 30 seconds. And even with the giant transformer, the idle current is quite low. Easily half the wasted power compared to a Sigineer or Ames of the same power capability. My second choice for an off grid inverter would probably be Victron.

While it is nice to use the matching Schneider charge controllers for the data, it's not required. I am using a cheap BougeRV DC charge controller for my 2,000 watt DC array, and it works just fine. But if I was depending on the DC system for all my power, I would probably move up to something like Midnight Solar, Victron, or Outback.

The EG4 batteries look good. Don't cut it close. However much battery you think you need, add 50% to 100% more and don't cycle them as deep. Yes, it will cost more up front, but if you only cycle 40% instead of 80%, they will last more than twice as long, so it does end up cheaper in the long run. I started with 18 KWHs of battery. It was "Enough" to just cover the high rate evening time, but on my first real power failure, I had to run it down to just 20% remaining. I doubled my battery bank to 36 KWHs and now it is normally staying over 55% and after a few days of clouds and storms, it only goes down to 40% running the house all night. And also only charging up to 85 to 90% is easier on the cells as well, and the C rate is also reduced for the same charge power. Since this is a new build, it may be difficult to get a good estimate of how much power you are really going to need. 6,500 watts of solar panels could easily make over 30 KWHs in a good sunny day. It would be nice to be able to store it all instead of having the charge controllers throw it away. If the car will be home during the day, you may be able to have the EV charging to use up the power without having to store it in the house battery first.

I am sure you will have more questions and others on here should be tossing in their ideas as well. I am not trying to tell you what to do, just giving my thoughts.
 
I have Enphase iQ7 microinverters, and they do work great, but for off grid, I don't think they are the best choice. My system is a hybrid with a Schneider XW-Pro battery inverter. When it goes off grid during a power failure, I have the Enphase micros AC coupled to it, but it is not completely stable. For the few times our grid is down, it works, but I would hate to depend on that for full time off grid.

The sun is down more than half the time, so I would have at least half of the solar array DC coupled to charge the battery bank. I added 2,000 watts of solar panels with a DC charge controller and I am very happy with the results. That DC power is basically what runs my house all night now. The Enphase microinverters provide my AC power during the day while the sun is up and the DC panels charge the battery for use at night. But like I said, off grid, the micros are not the most stable. I have 16 of them, and 5 of them basically refuse to stay producing while off grid. And several more tend to cycle off and on and the battery does not stay charging. I ended up running at a slight loss the whole time. So for a fully off grid system, I would probably use all DC coupled.

If you have a known shading pattern, you could plan the array around that. Series panels that will be shaded at the same time, and parallel with panels that shade at different times. And you can do panel optimizers as well. When I added my DC system, I thought about the Tigo add on optimizers. I may still get them, but my shading is not a big problem and the array is producing well without them. Another good option is using 3 or 4 separate MPPT charge controllers. That way each group is independently optimized, and if something does go wrong and fail, you still have the others to keep making power.

48 to 12 volt step down DC to DC converters work well, but they are likely only 95% efficient, so you will lose a little power. For things like room lighting, see if you can get 48 volt LED lights, or, if you use 12 volt, put the light switch on the 48 volt side before the DC to DC converter so that you are not wasting power running the converter when the lights are off.

For the 120/240 split phase AC power, there are a lot of inverters out there. I would definitely go with a large low frequency type for several reasons. The large transformer really helps keep the power clean and stable under difficult loads. As much as I like to complain about Schneider tech support and software, the hardware in the XW-Pro inverter is an absolute beast. It will drive any load you can throw at it. The 6,800 watt rating is all day, 24/7 without ever a complaint. And you can pull over 5,000 watts on just one leg without issues. It does 8,500 watts for 30 minutes and surge to 12,000 watts for over 30 seconds. And even with the giant transformer, the idle current is quite low. Easily half the wasted power compared to a Sigineer or Ames of the same power capability. My second choice for an off grid inverter would probably be Victron.

While it is nice to use the matching Schneider charge controllers for the data, it's not required. I am using a cheap BougeRV DC charge controller for my 2,000 watt DC array, and it works just fine. But if I was depending on the DC system for all my power, I would probably move up to something like Midnight Solar, Victron, or Outback.

The EG4 batteries look good. Don't cut it close. However much battery you think you need, add 50% to 100% more and don't cycle them as deep. Yes, it will cost more up front, but if you only cycle 40% instead of 80%, they will last more than twice as long, so it does end up cheaper in the long run. I started with 18 KWHs of battery. It was "Enough" to just cover the high rate evening time, but on my first real power failure, I had to run it down to just 20% remaining. I doubled my battery bank to 36 KWHs and now it is normally staying over 55% and after a few days of clouds and storms, it only goes down to 40% running the house all night. And also only charging up to 85 to 90% is easier on the cells as well, and the C rate is also reduced for the same charge power. Since this is a new build, it may be difficult to get a good estimate of how much power you are really going to need. 6,500 watts of solar panels could easily make over 30 KWHs in a good sunny day. It would be nice to be able to store it all instead of having the charge controllers throw it away. If the car will be home during the day, you may be able to have the EV charging to use up the power without having to store it in the house battery first.

I am sure you will have more questions and others on here should be tossing in their ideas as well. I am not trying to tell you what to do, just giving my thoughts.
Thank you for all that information. Very informative.
 
Micro inverters, off grid?
"These are not the droids you're looking for "
 
Isn't he basically describing AC coupled system?
Well yes, but...

For a system that will be off grid all the time, microinverters AC coupled to a battery inverter is a big waste. Even if my microinverters stayed operating properly all the time while AC coupled off grid (they don't) it would still be far less efficient than using a DC charge controller. The only time they would be any better is when you are using most of the power they are making while the sun is shining on the panels. So if you do plan to use a lot of power from 11 am to 2 pm, then sure, it might work out. But any extra PV solar production that is being converted to AC and then going backwards into the battery inverter to then be converted back to DC to charge the battery is suffering extra losses.

Even while I am on grid and everything is perfectly stable, I lose an extra 10% of my energy as it charges into the battery and is later inverted back out. Using the DC charge controller into the same battery inverter, the loss is only about 3% getting out to AC again, compared to the energy straight from the microinverters.

DC solar panel energy out to microinverter is 97% efficient.
AC back feed to DC charging the battery is 95% efficient.
DC inverted back to AC is 96% efficient.

1,000 watts from solar panel = 970 watts of AC = 921.5 watts into the battery = 884.64 watts of AC output

DC charge controller is 98% efficient into the battery.
DC inverted back to AC is 96% efficient.

1,000 watts from solar panels = 980 watts into the battery = 940.8 watts of AC output

For a "Grid interactive" system, the microinverters do work very well. All of the power I do use while the sun is shining is more efficient at 97%. But the extra going to and back from the battery is a loss.

So my system is now a split setup. I have 4,800 watts of panels going through the microinverters. They supply all my daytime power and about 40% of my evening power. The battery does charge from the excess AC power, but once the battery is over 80% charged, the extra AC production exports to the grid for credit. Then I have 2,000 watts of solar panels that are DC coupled through an MPPT charge controller directly to the battery. Each morning, I see the DC system starting to charge the battery at least 30 minutes before the microinverters are making enough power to run the house base load. And to make sure I don't consume grid power, I have to stay running off the battery until the AC coupled solar is making at least 400 watts more than the house is using. That is because the lowest charge current in the XW-Pro inverter is about 400 watts. On a decent sunny morning, the DC panels stop the battery from discharging and start putting some energy back in while the XW-Pro inverter is still powering the house base load. Once the DC panels do reach the power level where it is adding charge, the Enphase connected panels are starting to make good power as well. I am getting 25 KWHs a day from the Enphase system, most of which is used while the sun is up, with maybe 8 KWHs of it going to the batteries. And then I get another 8 KWHs from the DC system that is all used at night. They should make over 10 KWHs, but the battery is becoming full by 1 pm and it is going into float mode. I am not using all the power I am making right now. But in 2 more months, that will change when I start having to run the Central A/C system to keep the house cool when it is 110F (43C) outside. Then I will probably be seeing the full 10 KWHs or more coming in from the DC panels. The heat will also drop the output from all of the panels a bit.

With what I have seen, running off grid, I would certainly only use DC charging and skip the microinverters. But for grid interactive, I am getting the best of both worlds. But if I did this from new, I would probably have gone with only 3,600 watts of AC coupled panels and 3,600 watts of DC panels. I am at 6,800 watts of panels now, and 400 more would probably supply every watt hour I ever use. At 6,800 watts, I only every buy grid power when it is cloudy out for 2 days. Even with the storms the last 2 days, I still ended up exporting over 10 KWHs.
 
Coupled to what, though?
You would need another inverter to form the grid to couple to.
I though the big deal with AC coupled systems was that when the inverter started running from the battery that it provides the "Grid" for the micro's to fire up.

Although it was also explained to me that with Micro's the panels are pumping full wattage all the time much like Hydro does and you'd need a way to either shut off panels or have some sort of dump load involved.

Granted, I'm trying to figure this whole micro-grid thing myself as a noodle scratcher. ;)
 
I though the big deal with AC coupled systems was that when the inverter started running from the battery that it provides the "Grid" for the micro's to fire up.

Although it was also explained to me that with Micro's the panels are pumping full wattage all the time much like Hydro does and you'd need a way to either shut off panels or have some sort of dump load involved.

Granted, I'm trying to figure this whole micro-grid thing myself as a noodle scratcher. ;)
Micro inverters are a grid-tied solution. And yes, they don't care what the load is. They push everything available by raising the voltage. But they require a grid connection. Whether locally available or created.
So, you would need another piece of equipment capable of AC coupling to supply the grid source, and control the output of the Micro inverters.
This is almost always a hybrid AIO. Which is already capable of receiving power directly from the solar panels.
Micro inverters would be an addition to an off grid system. Not the primary system.
There are good use cases for this type of setup. But a mobile application wouldn't provide the space for that many panels.
 
Thanks for all this great information. Definitely explains why I couldn’t find many talking about micro inverters for off grid systems. Looks like I’ll need isolators. I’m stuck with location and we love the trees.
 
Look at Tigo optimizers. They will work with any MPPT charge controller.
 
So looks into these. I see TS4 A F. And some TS4 R F.
Another thing to figure out. What’s life of these? I
Found some open box R F
 
I have not tried them myself yet. I was looking into them for my DC system as I will need to add "Rapid Shut Down" anyways, and adding the per panel reporting and optimization is not a whole lot more. The TS4-A-O module is listed as having a 25 year warranty, so they expect it to last. Each one is rated to handle up to a 700 watt PV panel. Since I am using lower voltage 100 wat and 200 watt panels, I was thinking I could group them into section of abut 500 watts and get away with just 4 of them on my system. For panels over 300 watts, you really do need a module for each panel due to the current and voltage limits. And with paired panels, it can only optimize the pair, a shadow on one will still be a problem. If you have shadows moving across the array, you will also want one for each panel.
 
TIGO optimizers won’t theoretically be as good as microinverters unless you parallel all the inputs (because they can only match current or voltage by 30%), with each plane into its own separate MPPT (since TIGO doesn’t do MPPT)

But still better and less janky than using grid tie microinverters off-grid just to get the per panel MPPT
 
Back
Top