• Have you tried out dark mode?! Scroll to the bottom of any page to find a sun or moon icon to turn dark mode on or off!

diy solar

diy solar

EG4 18kPV Parallel with mismatched PV arrays

dielectric

New Member
Joined
Dec 5, 2022
Messages
108
@EG4TechSolutionsTeam, @ksmithaz1
How will two EG4 18kPV units share the load when running in parallel?
With each inverter having its own PV array and the two PV arrays producing different power.
For example, having one unit with north facing PV and the other facing south. Or just different sizes.
Is parallel load sharing logic going to make the two inverters produce same amount of power?
Will each use as much solar as they have?

This system is allowed to export to grid and I am considering adding a second unit, the 1st one has worked great for a year now. The 2 units will share a 43kwh battery.
But I suspect the general understanding will benefit everyone. I have not found anything in the manual or this forum.

Tagging EG4 team because they are the EG4 team :) and @ksmithaz1 because I saw mentions of parallel EG4 18kPV setup (and attention to nuances).

Thank you
 
Last edited:
The 2 units will share a 43kwh battery.
In this case, the inverters will share the battery bank, each inverter will push charge into the battery depending on its PV input and the AC output load will be shared equally.
It is completely transparent to the end user.
 
In this case, the inverters will share the battery bank, each inverter will push charge into the battery depending on its PV input and the AC output load will be shared equally.
It is completely transparent to the end user.
by "transparent", do you mean invisible (I'd call it "opaque" :D )?
also, I assume this behavior you've observed 1st hand with your dual inverter system?
 
Last edited:
if the 2 inverters in parallel are effectively as DC-coupled system, I wonder if it's not more efficient to use AC-coupled instead.


EDIT:
Clarification: "coupling" here used to mean distributing energy between devices. AC Coupled configuration is where one inverter feeds into another, or an inverter feeds into grid or vice versa. Here I meant that 48V DC will carry energy between inverters making the system DC-coupled. DC coupled systems exist outside of this question but they are uncommon. In a sense, it would be a DC grid / distribution system.
 
Last edited:
I'm looking to do the same with two 18kpv's.

I'm thinking the batteries stay on the master and I move the PV across the two units.

I have 'on grid' solar with a Net meter, and I'm thinking when that agreement ends.....I move that AC power (micro inverters) to the 18kpv and I'll easily have enough power to be off grid completely. Even in Las Vegas.

The other aspect is starting the 5 ton beast. Although SS says the single unit 'should' start it, I'm skeptical. It has been starting the 3 ton unit I have, but the lights sometime flicker when it kicks on. Both AC's have a mid start from the factory. So the 3 ton basically starts at 1.5 tons, then goes to 3.

I think the light flickering is more a problem with my supply line size. I put two 6 gauge wires for L1 and L2. The plan is to change those to 3 gauge at least, but I'm leaning towards 2 gauge for a one and done setup. Any thoughts ideas that I'm missing?
 
@EG4TechSolutionsTeam, @ksmithaz1
How will two EG4 18kPV units share the load when running in parallel?
With each inverter having its own PV array and the two PV arrays producing different power.
For example, having one unit with north facing PV and the other facing south. Or just different sizes.
Is parallel load sharing logic going to make the two inverters produce same amount of power?
Will each use as much solar as they have?

This system is allowed to export to grid and I am considering adding a second unit, the 1st one has worked great for a year now. The 2 units will share a 43kwh battery.
But I suspect the general understanding will benefit everyone. I have not found anything in the manual or this forum.

Tagging EG4 team because they are the EG4 team :) and @ksmithaz1 because I saw mentions of parallel EG4 18kPV setup (and attention to nuances).

Thank you
Interesting question, so I guess some of it kinda depends on how you set it up. As far as the PV goes, the produced power is ambivalent, it will go wherever it can. You will probably want to have a shared battery bus, I've run it both ways, currently I tell my inverters the bus is not shared to see the reporting from two different sets of packs, but I have a jumper between the busses to keep things from getting lopsided. (This has it's own issues, but I shant digress). My batteries generally always hit 100%, and during the day my PV almost always exceeds demand by a very large margin. If you are running on the edge of capacity you might have a consideration, but your average PV output should be dramatically higher than your average use during the day allowing you to dump all excess power to the batteries. Assuming they are on a shared bus, if say PV drops below on the secondary, it will simply use the batteries if need be. I've always had fairly balanced PV, but I've never seen the inverters get widely lopsided output, say in the late afternoon, or early in the morning when PV is ramping/falling, they seem to always stay +-10% in output, I *think*. I have looked live, but frankly to know for sure you would need to have both a high demand and lopsided PV at the same time. I produced just under 160KWH a couple weeks ago, I ran out of places to put it, charged two cars, mined bitcoin and I still hit 100% before 1600MST.

Having spoken off the cuff I look at the moment, I see quite different PV input but . . .:
1744381786413.png
I now realize all my graphs, and charts simply combine the inverter outputs, I'm not tracking them independently (Though I have the numbers in the tables). You can see the disparate charge numbers, but not the output. I will re-vamp some of my graphs/charts to track the output delta's. 160/PV1 is a fully west facing 12x410 array, the others are generally pretty flat, slightly south, 231/PV1 is a 2x10x410 array, and the rest are 8x455. I'm pretty sure the output numbers are close to the same, it simply tries to keep the output fairly balanced, independent of PV production.

YMMV.
 
I'm looking to do the same with two 18kpv's.

I'm thinking the batteries stay on the master and I move the PV across the two units.

I have 'on grid' solar with a Net meter, and I'm thinking when that agreement ends.....I move that AC power (micro inverters) to the 18kpv and I'll easily have enough power to be off grid completely. Even in Las Vegas.

The other aspect is starting the 5 ton beast. Although SS says the single unit 'should' start it, I'm skeptical. It has been starting the 3 ton unit I have, but the lights sometime flicker when it kicks on. Both AC's have a mid start from the factory. So the 3 ton basically starts at 1.5 tons, then goes to 3.

I think the light flickering is more a problem with my supply line size. I put two 6 gauge wires for L1 and L2. The plan is to change those to 3 gauge at least, but I'm leaning towards 2 gauge for a one and done setup. Any thoughts ideas that I'm missing?

Just get a "Micro Air Easy Start" and be done with it. The light flicker will be drastically reduced, but LED's are LED's. I don't notice the 4T anymore, probably will this summer a bit.
 
if the 2 inverters in parallel are effectively as DC-coupled system, I wonder if it's not more efficient to use AC-coupled instead.
I think you should re-think this statement. The two inverters will be 'coupled' both on the 240VAC output and the 48VDC inputs. There is no way to couple the high-voltage DC outputs from the MPPT's and share them across the inverters. Any of the HVDC you can't use in a given box for direct AC conversion/output will get stepped down to the 54-odd volts needed to charge the batteries which are 'coupled'. Realistically, I think you are hung up on the balancing of the 240V output based on PV, while interesting, not overly relevant in the grand scheme, as an out-of-balance condition on PV will generally only affect the charge rates on your batteries, unless your PV inputs are drastically different under high load conditions. Most output scenarios are very spiky.
 
I think you should re-think this statement. The two inverters will be 'coupled' both on the 240VAC output and the 48VDC inputs. There is no way to couple the high-voltage DC outputs from the MPPT's and share them across the inverters. Any of the HVDC you can't use in a given box for direct AC conversion/output will get stepped down to the 54-odd volts needed to charge the batteries which are 'coupled'. Realistically, I think you are hung up on the balancing of the 240V output based on PV, while interesting, not overly relevant in the grand scheme, as an out-of-balance condition on PV will generally only affect the charge rates on your batteries, unless your PV inputs are drastically different under high load conditions. Most output scenarios are very spiky.
nope, I disagree.
"DC coupled" was about one inverter supplying energy to another via the battery. Yes PV is a DC source and had nothing to do with that statement. AC is not coupled - no energy transfer from INV A to INV B, they are synchronized. You can choose to call that coupling but, "AC Coupling" is a specific term for one AC source feeding into another.
 
Interesting question, so I guess some of it kinda depends on how you set it up. As far as the PV goes, the produced power is ambivalent, it will go wherever it can. You will probably want to have a shared battery bus, I've run it both ways, currently I tell my inverters the bus is not shared to see the reporting from two different sets of packs, but I have a jumper between the busses to keep things from getting lopsided. (This has it's own issues, but I shant digress). My batteries generally always hit 100%, and during the day my PV almost always exceeds demand by a very large margin. If you are running on the edge of capacity you might have a consideration, but your average PV output should be dramatically higher than your average use during the day allowing you to dump all excess power to the batteries. Assuming they are on a shared bus, if say PV drops below on the secondary, it will simply use the batteries if need be. I've always had fairly balanced PV, but I've never seen the inverters get widely lopsided output, say in the late afternoon, or early in the morning when PV is ramping/falling, they seem to always stay +-10% in output, I *think*. I have looked live, but frankly to know for sure you would need to have both a high demand and lopsided PV at the same time. I produced just under 160KWH a couple weeks ago, I ran out of places to put it, charged two cars, mined bitcoin and I still hit 100% before 1600MST.

Having spoken off the cuff I look at the moment, I see quite different PV input but . . .:
View attachment 291478
I now realize all my graphs, and charts simply combine the inverter outputs, I'm not tracking them independently (Though I have the numbers in the tables). You can see the disparate charge numbers, but not the output. I will re-vamp some of my graphs/charts to track the output delta's. 160/PV1 is a fully west facing 12x410 array, the others are generally pretty flat, slightly south, 231/PV1 is a 2x10x410 array, and the rest are 8x455. I'm pretty sure the output numbers are close to the same, it simply tries to keep the output fairly balanced, independent of PV production.

YMMV.
Yes this is exactly the nuances I was hoping you could shed some light on. I have 1 north-ish 1 south-ish facing arrays. By 11am battery is full and by 2pm I start dumping the KW into resistive heater. At 4pm north-east arrays is 1kW and south-west is 4kW. So the specific question in my mnd is this: will there be 2kW of 48v DC flowing from INV B to INV A. It would be ideal if they just output amounts of AC KW as they were able.
 
Been watching the AC power out/split. It looks like the AC stays roughly balanced no matter the PV input, and excess goes to the batteries. On the "DC coupling" front, I have run my batteries coupled and uncoupled between the inverters, they really need to share a bus for things to work like you will want them to. If you don't share your power storage it gets uneven very easily. I just tied the busses.

1745265785716.png
 
Been watching the AC power out/split. It looks like the AC stays roughly balanced no matter the PV input, and excess goes to the batteries. On the "DC coupling" front, I have run my batteries coupled and uncoupled between the inverters, they really need to share a bus for things to work like you will want them to. If you don't share your power storage it gets uneven very easily. I just tied the busses.

View attachment 293720
thank you for confirming,
yeah I have basically 1 battery bus with 3 packs (BMSs) in one "battery" - it's all DIY
so that is all coupled in that sense.
 
The 18Kpv has a 'shared battery' setting that will assume your batteries are all under a single control. I have two different brands with daisy-chained control, one for each inverter, closed loop, mostly just to get a better idea what they consider 0-100% SOC. They tend to vary quite a bit in the middle, at least for a time, but generally very close at the extremes.
 

diy solar

diy solar
Back
Top