diy solar

diy solar

Gridtied AC coupled batt inverter/charger, and contingency operations

can someone escribe this NEM/net/zero/export thing? is it simply trying to not export power that for whatever reason you would end up not getting paid for?
In my case, it is the easiest way to add additional solar production.

I already have enough grid-tied solar to cover my annual electrical consumption but want more for when I have a new EV.

By adding more solar used purely to supply self-consumption (zero export), I reserve the existing grid-tied solar for credit to be used when I charge an EV.
 
One of my early ideas before I went with the Schneider was to use a microcontroller, like an Arduino to monitor the output current of the Enphase Microinverters. I know my house base load is just under 1,000 watts. So any time the solar was making less than say 1,200 watts, it would do nothing and let the solar feed the house, and maybe export a little. Above 1,200 watts, have it turn on a solid state relay that powers up a 600 watt charger on the battery bank. If solar goes over 2,000 watts, turn on a second 600 watt charger. And if solar tops 3,000 watts, turn on a third charger. Build in a bit of hysteresis, so as the production falls, if solar goes below 2,800 watts, one shuts off etc. I already have one 600 watt (58.8 volt 10 amp) charger that I have used to charge my battery bank. It is a proper CC CV so it will taper off current when the batter bank becomes full. When you get to the peak rate time, have all 3 chargers shut off, and let the system export. This time of year, my 16 300 watt panels are still cranking out 1,400 watts, so it runs my house and still exports 500 watts or so as I go into the peak rate 4 pm to 9 pm block. And this is also when I trigger the Schneider to go into "Grid Sell". In your case, this is when you turn on the grid tie battery inverters.
I’ve been thinking more about this idea and am interested enough to start looking into it, so if you made any progress in terms of identifying platforms like Arduino and or clamp sensors any relays, I’m interested in any suggestions.

What I’m realizing is that a single panel on a Microinverter will supply enough power to cover most of my self-consumption during the day. So instead of relying on my GTIL inverter to cover self-consumption at 75% efficiency, it’d be much smarter to keep the GTIL off until the sun goes down and cover self-consumption during daylight hours from a single panel at 95%+ efficiency.

So what I’m thinking about doing is using a clamp meter to monitor export on each leg of my power main and anytime it exceeds my 3KW cap (meaning more than 1.5kW on either leg), turn on a 150W charger until export drops under 1.4kW on that leg (the hysteresis you’re talking about.

This will only ever happen during the brightest ~2 hours of the day and when none of the fridges are running.

Your idea of cascading more chargers would be needed if I was thinking about adding more AC power, but I’m thinking about using this as only a single-panel extension of my existing Microinverter-based system just to cover daytime self-consumption.

MPPT charge controllers get my LiFePO4 charged at efficiencies of over 98%, so using AC chargers at ~80% efficiency as a dump-load to avoid export makes sense, but using them to get the battery filled does not (for me).

NEP, the supplier of my existing microinverters, just realeased a new 500W ‘Macroinverter’ for panels up to 600W so that coupled with a 450W panel makes for an easy ~10% expansion of my AC-coupled system (to be fully consumed through self-consumption).
 
I did not start programming, but I do have a few Arduino controllers. I like the MEGA2560 boards as they have more I/O including 3 hardware serial ports. I also bought a user interface board with a small LCD screen and a few buttons. And then I found this power meter on Amazon.
This meter has RS-485, so you need a small interface converter for the Arduino serial port to read the data. This is bidirectional so it can also be used on the gird side to measure if you are consuming or exporting. It does read only one hot lead though. My microinverters connect only across the two hot legs at 240 volt, so it would work fine to just read the solar production. There are some split phase versions in a single unit, but they are quite a bit more expensive.

The charger I have is a basic E-Bike charger from Amazon. You can order them in several different voltage and current ratings to match your battery bank. The one I have wants to charge to the full 58.8 volts of my 14S Li NMC pack. If I order another for a project like this, I would go a little less on the voltage. The 16S LFP one is 58.4 volts. Mine has a small power switch on it, I was going to wire a relay to that instead of trying to turn on and off the actual power. It does not seem to have much draw when switched off. Since you have no issue exporting up to 3000 watts, you don't have to be super critical. If you measure your main breaker power, it would be easy to switch on a 600 watt charger when you are exporting more than 1,000 watts. Then the export should drop to 400 watts. Then turn on a second charger if export hits 800 watts. How much solar do you have? How much do you want to add?

I am still a bit torn myself. Over 99.9% of the time, I have a solid grid, so it is most efficient to just have the microinverters grid tied. But I want to use my battery bank to power the house during the peak rate time. But to do this now, my battery bank is being charged with AC power from my system. It could be grid or solar. I have it just set to run on time from 9 am to 3:55 pm. That is the cheapest rate time, and when my Solar is normally making plenty of power, so I am still exporting some at that time. But since I am using battery power, it would be more efficient if I charged the battery from DC coupled solar. So if I add 4 more panels, I can't decide if I want to just add to my microinverters, or use a charge controller, and have those 4 panels directly charge the battery. I am using about 6 KWHs during my 5 hour peak rate time of use window. So the 4 x 300 watt panels = 1,200 watts, so I just need a bit over 5 sun hours to top up the battery each day, and not need any grid or AC charging. Rather than have to change the mode on the Schneider to charge, I could have a microcontroller watch the solar charging, and if it is falling short, turn on my 600 watt charge to ensure I have a full battery. If I am only getting 1/3 sun due to clouds, the AC powered charger would still technically be running on solar as my existing system would still be producing 1,000 watts, instead of it's peak 3,840 watts.

My next car is likely to be a plug in hybrid or full electric. In either case, I will also add some charging time. On days I am home, it can charge while the sun is up, but on days I work in the field, I would then want to charge at night. So Cal Edison does offer EV owners a lower 17 cent per KWH overnight rate. That is cheaper than cycling my battery bank to supply the charging power. And as a side benefit, it will also make my solar worth more to me as the little power I do use will be cheaper, and the peak rate time I do run off battery becomes more expensive.
 
I did not start programming, but I do have a few Arduino controllers. I like the MEGA2560 boards as they have more I/O including 3 hardware serial ports. I also bought a user interface board with a small LCD screen and a few buttons. And then I found this power meter on Amazon.
This meter has RS-485, so you need a small interface converter for the Arduino serial port to read the data. This is bidirectional so it can also be used on the gird side to measure if you are consuming or exporting. It does read only one hot lead though. My microinverters connect only across the two hot legs at 240 volt, so it would work fine to just read the solar production. There are some split phase versions in a single unit, but they are quite a bit more expensive.
Thanks, I’ll look into these. I’ll want one per leg since I am controlling for zero export on each leg separately (~300 to 400W of 240V AC translates to ~150-200W per leg).
The charger I have is a basic E-Bike charger from Amazon. You can order them in several different voltage and current ratings to match your battery bank. The one I have wants to charge to the full 58.8 volts of my 14S Li NMC pack. If I order another for a project like this, I would go a little less on the voltage. The 16S LFP one is 58.4 volts.

I’m currently at 24V (8S LiFePO4) and debating going up to 48V (16S LiFePO4).

At 24V, a 5A charger will consume ~160W, slightly more than m
Mine has a small power switch on it, I was going to wire a relay to that instead of trying to turn on and off the actual power. It does not seem to have much draw when switched off.

I was thinking I’d use a relay, but you are right, that may be less efficient than just controlling a low-current control signal. If I decide I’m going to 48V and need new chargers, I’ll keep that in mind.
Since you have no issue exporting up to 3000 watts, you don't have to be super critical. If you measure your main breaker power, it would be easy to switch on a 600 watt charger when you are exporting more than 1,000 watts. Then the export should drop to 400 watts. Then turn on a second charger if export hits 800 watts. How much solar do you have? How much do you want to add?

I’ve got 4kW of panels powered through 3kW of Microinverters. I’m only consuming an average of 350W throughout the day to power my 5 fridges/freezers so a single 450W panel powered by one of NEP’s new 500W ‘Mactoinverters’ should be enough to directly power all daytime use rather than relying on the much-less-efficient GTIL inverters...

It’s hard to know what the detection limits are for overproduction - 105% or 3150kW seems like it should be safely under the radar, but being anywhere near 110% for more than very short bursts would make me nervous.

So I’m thinking of a very simple dump-load where I may cascade my fridges to assure only one is on at a time (per leg) and use the chargers as the dump load of last resort (the true dump load). When I get to ~3075 or 3100W, turn on the charger and keep it on until total export drops under 2925 or 2950W.

My largest 3 fridges all consume over 150W when running, so as long as I spread them out, one of them should pretty much always be running. The two smaller ones are only 100W each but depending on duty cycle, I should be able to absorb most all of the extra generated AC power by just controlling my fridges and the battery chargers will just be there to full any duty-cycle gaps.

I’m only thinking of adding enough new AC-coupled solar to cover daytime average consumption (~10% of my 3kW grid-tied array). And adding more DC coupled to charge the LiFePO4 battery (more below).

The main motivation for the one AC panel is that it is so much more efficient to power with a single panel through a Microinverter @ 95+% efficiency that to put all that energy into a battery and power those loads through the GTIL inverters @ 75% efficiency (ouch!).

Using one 450W panel with a Microinverter will get as much power out of it as if it were a 570W panel generating DC and covering most daytime consumption while the solar is producing.
I am still a bit torn myself. Over 99.9% of the time, I have a solid grid, so it is most efficient to just have the microinverters grid tied. But I want to use my battery bank to power the house during the peak rate time. But to do this now, my battery bank is being charged with AC power from my system. It could be grid or solar. I have it just set to run on time from 9 am to 3:55 pm. That is the cheapest rate time, and when my Solar is normally making plenty of power, so I am still exporting some at that time. But since I am using battery power, it would be more efficient if I charged the battery from DC coupled solar.

Yes, that’s what I’ve concluded. MPPT charge controllers are much more efficient at getting Amp-Hours into a battery compared to Microinverters + AC battery chargers. What is the efficiency of your AC charger?
So if I add 4 more panels, I can't decide if I want to just add to my microinverters, or use a charge controller, and have those 4 panels directly charge the battery.

I’ll have ~4kW of DC-coupled panels charging a 560Ah 24V battery (or a 280Ah 48V battery). That will allow me to self power through 6pm - 11pm and all the way ‘til the next morning using the 2 GTILs, covering all of my peak and overnight consumption.

The lone AC-coupled 450W panel should cover all of my daytime usage without relying on the inefficient GTILs, which should reserve most/all of my 3kW (4kW of panels) grid-tied power for credit towards charging an EV.
I am using about 6 KWHs during my 5 hour peak rate time of use window. So the 4 x 300 watt panels = 1,200 watts, so I just need a bit over 5 sun hours to top up the battery each day, and not need any grid or AC charging. Rather than have to change the mode on the Schneider to charge, I could have a microcontroller watch the solar charging, and if it is falling short, turn on my 600 watt charge to ensure I have a full battery. If I am only getting 1/3 sun due to clouds, the AC powered charger would still technically be running on solar as my existing system would still be producing 1,000 watts, instead of it's peak 3,840 watts.
I’ve done the math and trying to self-power through battery energy which had been put there charging at 80% efficiency is not worthwhile (assuming 80% charging efficiency and with my very inefficient GTIL inverters).

1kWh of AC worth ~$0.20 results in 800Wh getting stored into the battery and then 600Wh of power generated through the GTILs @ 75% efficiency and worth $0.24 @ peak rates of $0.40/kWh.

You could argue a 20% profit is not bad, but then you need to figure in the battery cost. I figure my battery costing me over $1500 should deliver at least 3000 cycles, meaning each cycle is worth $0.50. Even if I’m optimistic and assume I get twice that number of cycles, it’s worth as much as the electricity being generated and I’m better off just conserving my cycles.

With DC solar charging, a whole kWh goes in and is converted to 750Wh going out worth $0.30. Still swamped out by the value of each battery cycle, but if I get 6000 cycles, at least slightly worthwhile.

So on cloudy days or smoky days, I don’t plan to top-off my battery with grid power and may just limit discharge to peak hours until it’s drained...
My next car is likely to be a plug in hybrid or full electric. In either case, I will also add some charging time. On days I am home, it can charge while the sun is up, but on days I work in the field, I would then want to charge at night. So Cal Edison does offer EV owners a lower 17 cent per KWH overnight rate. That is cheaper than cycling my battery bank to supply the charging power. And as a side benefit, it will also make my solar worth more to me as the little power I do use will be cheaper, and the peak rate time I do run off battery becomes more expensive.
Check the fine print. Yes, nighttime rates are cheaper but peak rates are higher and over more days. And I’m pretty sure my utility told me that changing to an EV rate would mean losing my NEM1 grandfathering and being forced on NEM2.

So my plan is to charge at nighttime rates and just build up the credits over summer to cover our expected annual driving.

The poor inverter efficiency of the current-generation of GTIL inverters is the biggest weakness of my current plan, but I assume that will improve over time as new-generation products emerge...
 
Last edited:
I put in solar too late, so I was put on NEM 2.0 from the start. No going back for me. I am also stuck on Time of Use rates. But check this out.
SCEusage03-05-21.PNG
Except for March 3rd, when we had heavy clouds and rain all day. The rest of the time, solar production just about matched our usage. This is half way through this billing cycle, and So Cal Edison owes me $3.00, but that will likely zero out and I may owe a few buck, but it is basically nil. And this is achieved with my Schneider XW-Pro inverter/charger charging the batteries from the AC coming from the 16 Enphase iQ7 inverters, and then using the charge in the battery to run the house each evening peak rate period. I then run on grid power from about 9:30 pm until the sun comes up at 6:30 am. I have the XW-Pro wait until 9 am before it starts charging at 28 amps (about 1,400 watts) which tops up the battery bank by 2:15 pm.

Here is the wild part. My round trip efficiency of AC going in, converting to DC to charge the batteries, and then inverting back to AC and pushing into the grid, it is measuring out to be nearly 90% efficient. So it pulls 1,400 watts for 5 hours to charge 7 KWHs into the batteries, and then pushes back out nearly 6,500 KWHs back to run the house. The charging, battery storage, and inverting are very efficient in this system. The XW-Pro and battery bank run stone cold. No lost energy means no heat production.

In another month or two we will start getting our "normal" hot weather. Then I will need to fire up the Air Conditioning. That beast sucks 14 amps at 240 (3,360 VA) for 7 hours a day from late spring to early fall. That alone can hit 23.6 KWHs a day. And there is not much I can do about it. I have to buy that power, because even if I put up another 3,000 watts of panels, which won't fit on my roof, I will still fall short of running that load. But I want to at least cut that load in half during the peak rate time of use window. Here is my month of usage from last August to September.
SCE-Aug2020.PNG
3 days topped 50 KWHs, and the "average days" were still using 30+ KWHs, on top of the 25+ KWH's my solar was producing.

I do live right next to the Mojave Desert, it gets HOT here. We have had entire weeks stay over 100F. And we have peaks to near 120F. The heat on the solar panels does hurt their output a bit, and just keeping the house under 80F inside is a chore. This is why I am torn. I can easily add 4 more panels. The panels would be aimed for ideal summer evening sun. Turned about 30 degrees west, and only tilted up 15 degrees from flat. Adding 4 microinverters would put the most AC power into my system while the sun is up, and help run the A/C, but I still end up needing it running 3 hours after sunset. If I DC couple them, it won't help run while the sun is up, but it will reduce the loss when I pull the power from the batteries to help run it at night. But now that I have taken some loss measurements, the difference in efficiency will only be about 4-5% The Enphase micros are 97.5% efficient and the XW-Pro is 95% efficient while charging, so I charge at a net 92.6% efficiency. A solid MPPT charge controller, could up that to about 97% charging efficiency. No where near as big of a difference as I was expecting, but when you are talking about over 8 KWHs a day (on just the 4 new panels), it does add up. DC Charge controller losing 240 watt hours a day vs Microinverters to XW-Pro charging losing 592 watt hours a day.

The other idea I was given was to just parallel the new panels with the ones I have that get shaded in the evening. This reduces cost a bit as I don't have to buy any inverters or charge controllers. I would just need a few "Y" connectors and wire. As the existing panel facing 25 degrees East starts to lose power, the new 30 degree west panel will pick up power. The iQ7 inverters will probably clip from the 2 panels for about 2 hours even with very hot panels, but they will carry more power out 2 to 3 hours later in the day, well into my peak rate time, to help run the air conditioner. If it is really clipping bad, I can always add the extra inverters or get a charge controller later. SanTanSolar has some very nice 310 watt panels that look very similar to my existing panels for just $120 each. At that price, I may find a way to squeeze 6 more on my roof. Tie 3 to existing micros, and 3 to a cheap charge controller.
 
"Then I will need to fire up the Air Conditioning. That beast sucks 14 amps at 240 (3,360 VA) for 7 hours a day from late spring to early fall. That alone can hit 23.6 KWHs a day."

Somebody said spraying a mist of water on the condenser significantly improved efficiency. Although I imagine minerals could be a problem. (solar still? Softener?)
 
lets say things really go downhill, power is out for an extended period of time. is it possible to put any of the GT ac-coupled batt inverter/chargers into some kind if "offgrid" mode that will just signal the microinverters to just stay on. of course you'd disco the house main grid connection in this mode.. for just in case purposes, if someone needed that capability, which item would be best suited?

Hypothetically, you might only need an 150-300w (off-grid) inverter plugged in to your grid (with mains isolated) to supply a sine wave signal for the micro inverters to "sense" and "synchronize" with.

I hypothesized using micro inverters to complement my off grid system during daylight hours.
 
Hypothetically, you might only need an 150-300w (off-grid) inverter plugged in to your grid (with mains isolated) to supply a sine wave signal for the micro inverters to "sense" and "synchronize" with.
The micros have more smarts than just "sensing" a sine wave signal. I believe they actually do some tests to confirm the presence of the grid or a hybrid inverter with a buffer. GT inverters are designed to pull optimum power from the panels and push it to the grid so my hypothesis is that if it does work, it will be brief and you will release the smoke on your 150W inverter. Be sure to report back to confirm my hypothesis. ;)
 
Last edited:
I agree with Ampster here. The grid battery inverter that is forming the grid needs to be as powerful as the grid tie inverters you are trying to power up. When the grid tie inverters try to power up, they will push current into the output of the battery inverter. It needs to be able to handle the current, coming out or going in. And if the grid tie inverters happen to produce more power than your loads are using, then the battery inverter needs to be able to run "backwards" and take the current in and push it to the battery.

Using a battery inverter that can't take extra AC might sort of work, "IF" the loads always need more power than the solar grid tie inverters are producing. But you still need a battery inverter that can run all of the load. You power up the battery inverter and turn on the loads. Then as the solar comes up, it will reduce the power the battery inverter needs to supply, but as long as it is making some power going out, it "MIGHT" be stable. But turn off one load too many, and the battery inverter goes POP!!!
 
I agree with Ampster here. The grid battery inverter that is forming the grid needs to be as powerful as the grid tie inverters you are trying to power up. When the grid tie inverters try to power up, they will push current into the output of the battery inverter. It needs to be able to handle the current, coming out or going in. And if the grid tie inverters happen to produce more power than your loads are using, then the battery inverter needs to be able to run "backwards" and take the current in and push it to the battery.

Using a battery inverter that can't take extra AC might sort of work, "IF" the loads always need more power than the solar grid tie inverters are producing. But you still need a battery inverter that can run all of the load. You power up the battery inverter and turn on the loads. Then as the solar comes up, it will reduce the power the battery inverter needs to supply, but as long as it is making some power going out, it "MIGHT" be stable. But turn off one load too many, and the battery inverter goes POP!!!
right... so what to do if I have a decently-sized microinverter array. say 26 microinverters, where each one can sustain 350watts? Thats 9100watts. At that point are you just stacking together multiple sunnyislands or multiple AC-coupled Radians? I guess theres not yet a way for these AC-coupled inverters to not only control the on/off state of microinverters with a clean 60hz signal, but they cannot yet tell microinverters to decrease/increase output?
 
If your microinverters perform frequency-watts, then something like Sunny Island should be able to control them.
If not, they will cycle on and off grid every 5 minutes. Someone's description of Sol Ark indicated it would cycle down to 90% SoC before bringing them back on grid.

A single Sunny Island (5.75kW continuous output at 25 degrees C) can control up to 12kW of AC coupled inverters. If 240V, you'll need a 120/240V transformer. However, internal relay can only let 6.7kW max feed through to grid if you have a grid-tie system with net metering. If using a generator or grid as generator without backfeed, relay limit doesn't restrict AC coupled PV.

Sunny Boy Storage (400V battery inverter) I think is meant to work with (tolerate?) various AC coupled inverters. I don't know what control method it uses.

Enphase has a system for using some of their microinverters in battery backup system.

Maybe what you're hoping for is a batteryless system, using microinverters to create AC so long as there's enough PV. A system with small battery will serve better, because it can supply starting surge to motors.
 
so what to do if I have a decently-sized microinverter array. say 26 microinverters, where each one can sustain 350watts? Thats 9100watts.
Every hybrid has its own unique limit based on the size of the battery connected to it. Some are better than others at leveraging AC coupling. Some micros are also able to be more efficiently coupled. You would have to do some research to find the ones with the limits that work for you.

My guess is that more recently designed hybrids have more capacity. For. Example, the Skybox from Outback was specifically designed to be grid interactive and has a more effective AC coupling alqorithm than the Outback Radian. Also to make AC coupling work the micros need to be wired to the hybrid inverter port through the critical loads panel and you could choose only the number of micros that meet that particular hybrid and battery capacity. The micro combiner box where the branch circuits combine is where they might be conveniently separated. I can go into more details if that is a feasible alternative?
 
Last edited:
...Maybe what you're hoping for is a batteryless system, using microinverters to create AC so long as there's enough PV. A system with small battery will serve better, because it can supply starting surge to motors.
no, im planning on a battery as well.. Ampster's comment revealed to me where my gap was, I didnt realize you tied the GT microinverters to the critical loads panel so the AC-coupling could do its magic with them and also provide the isolation from the mains/POCO grid. This way I can engineer the appropriate number of GT microinverters to critical loads, sized so as to not overload the AC coupling inverter, and the others stay on the mains side so they stay off during an outage. is that about right?
 
is that about right?
Yes, that is the concept. It might be tricky getting data on all of the micros because the powerline cimmunications can be finicky when having to communicate through different breaker panels. Are your micros Enphase?
Also the particular UL spec that the micros were manufacturered to makes a difference. The latest is UL 1741SA ("SA" is the important suffix) is able to modulate incrementally and if I recall correctly, my Skybox, can leverage more capacity of those newer micros. @Hedges is well informed about the ability of SMA inverters to AC couple.
 
Im still torn. I may not do any microinverters and do solark-12k with 4 strings: 6+6 to MPPT1 and 7+7 to MPPT2. Id consider using tigo shutoff modules with this option

Or may get IQ7As and use an outback radian AC coupling solution.

It will come down to which option has better fire/hitemp auto shutdown (if either even support that at all) detection if things on my roof ever start sparking
 
For rooftop, arc-fault and Rapid Shutdown are the protections to have.

SolArk 12k supports 12kW total from PV and 9kW inverting to AC (therefore backfeed to grid) according to datasheet. But some installers on the forum report it can't sustain that, won't deliver over 7.2kW AC continuous, so look into that.

I use Sunny Boy and Sunny Island. Probably any string inverter that performs the optional frequency-watts of Rule 21 would support battery backup inverters. The recent Sunny Boy also have "Secure Power", a 120V AC outlet for up to 2kW (if sufficient PV) batteryless backup. I think I read that Secure Power is now compatible with Rapid Shutdown, but confirm if you want that.

Solar Edge has its own optimizers with Rapid Shutdown that support panels of multiple orientations. They have their own battery backup system.
 
yes Ive read up on the rapid stuff. but it all seems to have to be manually triggered. just wondering if a MI or the tigo rapid shutdown modules have any automatic ability to disconnect everything
 
You can certainly wire up the output of a smoke or fire alarm to trigger a relay to activate the rapid disconnect system. Code does not yet require anything like that, but it is not a bad idea. My home has 7 smoke detectors that are tied together, so all head alarm if any detect. I would bet it is fairly easy to use the tie signal and have it trigger a power cut off.

I am currently running AC coupled. I have 16 Enphase iQ7 inverters feeding into the sub panel on a Schneider XW-Pro. It is working well, but Enphase had a minor software issue that last time I did a power fail test, and had a real power failure, some of my micro inverters went offline and needed a reset to come back on. They just updated the software, and it should fix the issue, I hope to do another power fail test soon. That being said, the iQ7's that did stay online worked perfectly and it was clearly charging the battery bank with the extra power beyond what I was using out of the backup loads panel. I did not let it run long enough for the batteries to top out. If the solar output exceeds the charging current limit set in the XW, or the batteries reach full charge, the XW-Pro will frequency shift up to trigger frequency watt power reduction. The iQ7's can do that and it "should" work to regulate the battery at your desired charge rate or full state of charge. If the power does not drop off enough, or fast enough, the frequency will shift to the point where the micros will shut off and the XW will then power the loads from battery as it ramps the frequency back to 60 Hz. At which point, the micros should go back online after a 5 minute power qualifying delay.

When the system is on grid, there is no way to shift the frequency, so the iQ7's will just put out all they can unless you have the Envoy with CT's on the main set to do zero export. For the Envoy to be able to "talk" to the micros, you may need to add a power line communication phase coupler. They sell these to bridge the high frequency data signals across the legs, and it could bridge o a different panel and should work fine. I have not needed it since my entire Envoy and all my panels are small enough to be on the output on my XW inverter.

I do have to admit, the XW-Pro does have a few small issues in a pure AC coupled setup like mine. The biggest issue is that it will not time shift power without another unit or person commanding it to charge each morning. I have been trying to work with Schneider about this issue, and they don't seem to care. It seems to work perfect for backup power and off grid operation in this mode, but while on grid, it hangs. For this reason, I will likely be adding some DC coupled solar to do the battery charging while the sun is up. It cures a lot of problems. I am looking at just 6 panels of DC to compliment my 16 panels of AC. during the summer, this setup should make over 45 KWHs a day.
 
Or may get IQ7As and use an outback radian AC coupling solution.
I am a big fan of Outback and had a Radian years ago. It did not have the robust AC coupling that the Skybox did but it does have better surge capability. As mentioned below read the specs very carefully and check your assumptions. I know the Radian and the Skybox can be paralleled but the Skybox loses its AC coupling capability when paralleled. I don't know about the Radian AC coupling when paralleled.

SolArk 12k supports 12kW total from PV and 9kW inverting to AC (therefore backfeed to grid) according to datasheet. But some installers on the forum report it can't sustain that, won't deliver over 7.2kW AC continuous, so look into that.
That is an important point.

In order to understand the the difference between what is called a 12kW inverter and 7.5kW AC continuous it is important to distinguish between the sources of power coming into the box and understand the total energy that can go through the box (throughput).
For example, if an AC breaker is 60 Amps that translates to 14.4kW. If the internal Inverter can output 8kW that theoretically might leave 6kW available for an AC coupling solution. In other words the daytime throughput is one number and the evening output is lower because that can only come through the Inverter. My point is that in system design is important and understanding the details of the specs is very important. Test any assumptions that are made. Also some degree of flexibility unique to a site can work to your advantage. So far I have only used kW instantaneous capacity but do not forget about daily kWh production. Using some east and west orientations can avoid mid day peaks and spread out the solar collection window to harvest more kWhs. Similarly do not dismiss higher DC to AC ratios to accomplish the same thing. I know that IQ7s can be overdriven according to the manufacturers specs. Similarly the Skybox and SolArk can be overdriven on the DC side with solar.
 
...snip...
In order to understand the the difference between what is called a 12kW inverter and 7.5kW AC continuous it is important to distinguish between the sources of power coming into the box and understand the total energy that can go through the box (throughput).
For example, if an AC breaker is 60 Amps that translates to 14.4kW. If the internal Inverter can output 8kW that theoretically might leave 6kW available for an AC coupling solution. In other words the daytime throughput is one number and the evening output is lower because that can only come through the Inverter. My point is that in system design is important and understanding the details of the specs is very important. Test any assumptions that are made. Also some degree of flexibility unique to a site can work to your advantage. So far I have only used kW instantaneous capacity but do not forget about daily kWh production. Using some east and west orientations can avoid mid day peaks and spread out the solar collection window to harvest more kWhs. Similarly do not dismiss higher DC to AC ratios to accomplish the same thing. I know that IQ7s can be overdriven according to the manufacturers specs. Similarly the Skybox and SolArk can be overdriven on the DC side with solar.
thanks all, very much for this awesome discussion! I'll have to review all this last one here closely, as i just pulled the order trigger on the solark12k & tigo optimizers option with 24 seraphim 425's. Im hoping to be able to get a nice bank of 16 of the 310AH lifepo4's for it also, and critical loads will be master bed/bath/kitchen/livingroom. will prob use a DC well pump directly to batts also.
 
Back
Top