Enphase IQ8+ AC coupled to Dual Schneider Conext XW Pro - No PV when off grid

Enphase Gen 3:
Enphase Gen3 not backward compatible with Gen2.
You cannot upgrade your existing system controller/batteries with the new ones and you cannot reuse existing batteries with a Gen3 system controller!

So let's get the Schneider XW AC coupling to work.

AC Calibration:
AC inverter voltage can be changed in:
select Inverter -> Configuration -> Advanced -> AC Calibration
Same password as for setting grid profiles.

Changed my AC voltage from 238V to 248V.
Be vary careful and disconnect all AC loads, do not type in a value that may be off too much, use the spinners to up/down voltage.
Won't change much with increments less than 100 points.

DER Solar said:

Enphase Gen 3:
Enphase Gen3 not backward compatible with Gen2.
You cannot upgrade your existing system controller/batteries with the new ones and you cannot reuse existing batteries with a Gen3 system controller!

So let's get the Schneider XW AC coupling to work.

AC Calibration:
AC inverter voltage can be changed in:
select Inverter -> Configuration -> Advanced -> AC Calibration
Same password as for setting grid profiles.

Changed my AC voltage from 238V to 248V.
Be vary careful and disconnect all AC loads, do not type in a value that may be off too much, use the spinners to up/down voltage.
Won't change much with increments less than 100 points.

Click to expand...

Gen 3 - compatibility, no surprise. They have done it before on the IQ6 IQ7 & IQ8 promised vs delivered compatibility - long story there - off topic for this.

YES = XW to the rescue, hopefully.

AC Calibration - thanks so much! Will definitely check into it. I've been there before and disappointingly this is all I get I need to figure out what this arbitrary number represents.

Gubman said:

Looking to prolong runtime. I can try to go off AC1 in place of AC.Out. (no grid on this project, hence missing on the pic)

Click to expand...

Given the iQ8 AC coupling challenges with XW as the grid forming inverter, I suggest you try moving the Enphase batteries and system controller to the XW input port and enable load shaving above some desired threshold. So, the Enphase batteries will always be the grid forming inverter to better control AC coupling with the iQ8's and the XW output will follow the input from the Enphase batteries to share the load or charge the 48V batteries as needed.

solar8484 said:

Given the iQ8 AC coupling challenges with XW as the grid forming inverter, I suggest you try moving the Enphase batteries and system controller to the XW input port and enable load shaving above some desired threshold. So, the Enphase batteries will always be the grid forming inverter to better control AC coupling with the iQ8's and the XW output will follow the input from the Enphase batteries to share the load or charge the 48V batteries as needed.

Click to expand...

So the Enphase would then just pass through XW's AC1 -> AC.Out? Wouldn't that limit me to the 60A of the XW?

Gubman said:

So the Enphase would then just pass through XW's AC1 -> AC.Out? Wouldn't that limit me to the 60A of the XW?
View attachment 151692

Click to expand...

It looks that way. So, you have 4 x 10T? That would be 64A @240v max continuous output resulting in 4A max power reduction (~6%) going through the XW. Do you need to run >94% power from the 10T's continuously? If it was me I would do it for the small trade off to have more stable and robust AC coupling with the iQ8's since the XW can supplement power as needed. Another potential possibility would be to put one 10T on the XW output.

You are missing the disconnect relay in the AC1 line. There a 4 + 1 relays: one for each leg in the AC1 (grid) line, one for each leg in the AC2 (gen) line and a NFT neutral disconnect relay on the AC (out) N line for preventing imbalance current via the toroidal transformer when grid connected.
AC(out) is always connected to the output of the toroidal transformer and always powered by the inverter. The AC1 lines are not powered and they are disconnected because the AC1 sensing transformers do not detect a grid voltage. When you use AC1 for the AC coupled micro-grid connect it will work because the sensing transformers detect the Enphase reference voltage and the XW Pro connects the relay in the AC1 line after phase sync as in a normal grid operation. Using the AC1 has advantages because in an EPO scenario you can tie in the XW Pro into the Enphase EPO and all DER resources will disconnect. If you stay on the AC(out) line, then the backup panel will be energized in an EPO scenario. You can try this, switch the Enphase off and check the power supplied by the XW Pro.

You are always limited to 28A per XW Pro on the inverter output, except for the surge to 56 A. You can stack 2 or more if you need more power.

There is more to this regarding who provides the micro-grid reference voltage, depending how the subcomponents are switched on. Either the XW Pro if switched on first and Enphase follows, or the Enphase IQ controller if switched on first and the XW Pro follows. Assuming you want to connect the grid to the IQ system controller, the 2nd case is better.

solar8484 said:

It looks that way. So, you have 4 x 10T? That would be 64A @240v max continuous output resulting in 4A max power reduction (~6%) going through the XW. Do you need to run >94% power from the 10T's continuously? If it was me I would do it for the small trade off to have more stable and robust AC coupling with the iQ8's since the XW can supplement power as needed. Another potential possibility would be to put one 10T on the XW output.

Click to expand...

64A total out, yes per spec, but don't forget about the micros. I have 31 IQ8+. The entire system is additive and the micros take priority to reduce battery wear (IMHO). Enphase tries to "simplify" the installation and in turn removes the ability to tweak the specifics of one's setup, specifically output control of the batts - I know, I know what the hell did I buy?

94% continuously, not often, but sometimes I'll go over 64A during the day. (I play nice after sunset.). I'm entirely off grid. As a general rule, one doesn't design systems to run near limits - a wide margin everywhere is your friend, a 125% minimum is nice. Similar reason to up the wire gauge when possible, etc.

DER Solar said:

You are missing the disconnect relay in the AC1 line. There a 4 + 1 relays: one for each leg in the AC1 (grid) line, one for each leg in the AC2 (gen) line and a NFT neutral disconnect relay on the AC (out) N line for preventing imbalance current via the toroidal transformer when grid connected.
AC(out) is always connected to the output of the toroidal transformer and always powered by the inverter. The AC1 lines are not powered and they are disconnected because the AC1 sensing transformers do not detect a grid voltage. When you use AC1 for the AC coupled micro-grid connect it will work because the sensing transformers detect the Enphase reference voltage and the XW Pro connects the relay in the AC1 line after phase sync as in a normal grid operation. Using the AC1 has advantages because in an EPO scenario you can tie in the XW Pro into the Enphase EPO and all DER resources will disconnect. If you stay on the AC(out) line, then the backup panel will be energized in an EPO scenario. You can try this, switch the Enphase off and check the power supplied by the XW Pro.

You are always limited to 28A per XW Pro on the inverter output, except for the surge to 56 A. You can stack 2 or more if you need more power.

There is more to this regarding who provides the micro-grid reference voltage, depending how the subcomponents are switched on. Either the XW Pro if switched on first and Enphase follows, or the Enphase IQ controller if switched on first and the XW Pro follows. Assuming you want to connect the grid to the IQ system controller, the 2nd case is better.

Click to expand...

That is precisely my next general test, on the AC1 and what you describe is exactly what I expect to happen [XW Pro AC Coupling Solutions Guide pg. 13 document 990-6421]. I'm quite excited to monkey around on take 2. Still have to get a bit more comfy on the general operation of the XW by itself. Coming from a buzzing-less environment (Enphase) to a buzzy one (Schneider) is unnerving as I constantly think I'm pushing something to the limit. (Its a reflexive reaction that I need to get over.)

Running the Enphase rig 24x7 for well over a year now, I plan to keep it powered as primary and use the XW to take the edge off, reduce Enphase batt cycling and a host of other nuances. Once this one gets integrated, I'll very likely get another since the Enphase is maxed out on storage.

Some XW Pro's are fairly quiet except for the 60 Hz hum under full load, others are very noisy with annoying high-pitch frequencies. You can check for loose sheet metal parts that separate the different component sections. Have one XW Pro I'm trying to calm down, have 2 XW's that are fairly quiet, will update what works.

XW Pro's are great because you can choose battery chemistry (e.g. Carbon AGM) and extend battery storage to whatever kWh you need. If you get a 2nd XW Pro you can stack them for double power, but then the master takes all the primary load/charge below 6kW and produces the noise while the 2nd one is on standby and quiet. I don't think you can balance 2 stacked XW pro's for 50/50 load/charge aka smart AC coupling, at least not yet with the current firmware. But you could leave them unstacked and connect both to the same battery bank or separate battery banks, gives you a fault tolerance of 2. If one fails you get half the kW power at same kWh storage capacity, much better than Enphase.

Very intriguing, you are building and using the XW Pro as an AC coupled battery addition to Enphase. 6.8kW power, 13.6 kW overload for 1 minute and basically open-ended storage capability.

Gubman said:

64A total out, yes per spec, but don't forget about the micros. I have 31 IQ8+.

Click to expand...

For clarification, my suggested experiment would leave the iQ8's on the XW AC output side along with the loads. So, there would be no limit on iQ8 power going to loads. The 60A limit would apply to iQ8 power backfeed through the XW to the 10T's but with 31 iQ8's the max backfeed current should be well under the 60A limit. Also, loads and XW charging current would further limit that.

solar8484 said:

For clarification, my suggested experiment would leave the iQ8's on the XW AC output side along with the loads. So, there would be no limit on iQ8 power going to loads. The 60A limit would apply to iQ8 power backfeed through the XW to the 10T's but with 31 iQ8's the max backfeed current should be well under the 60A limit. Also, loads and XW charging current would further limit that.

Click to expand...

That's a fantastic detail, and I sincerely thank you for the idea snippet. From the XW's perspective this would fly given that AC.Out is bidirectional so I do not have to worry about any charge imbalance. Unfortunately, like many things in life, the're a catch when one throws in the batts and the need to assemble a complete working system for the 240V split via the NFT. The System Controller has an iron grip of the IQs by way of sense circuitry so I'd have to hack the setup to create a bit of a Frankenstein. [For more detail, follow the Combiner output to the Smart switch at https://diysolarforum.com/attachments/3master-wiring-schneider-off-grid-gif.123909/]. Need to sleep on this as the resulting hack feels like driving with loose wheel nuts...

A knee jerk reaction would be to get 2 XWs to address the 60A, but that also appears to have a catch in that a pair does not go 50-50, but after the master maxes out, the slave joins the party. Work goes on...

DER Solar said:

Some XW Pro's are fairly quiet except for the 60 Hz hum under full load, others are very noisy with annoying high-pitch frequencies. You can check for loose sheet metal parts that separate the different component sections. Have one XW Pro I'm trying to calm down, have 2 XW's that are fairly quiet, will update what works.

XW Pro's are great because you can choose battery chemistry (e.g. Carbon AGM) and extend battery storage to whatever kWh you need. If you get a 2nd XW Pro you can stack them for double power, but then the master takes all the primary load/charge below 6kW and produces the noise while the 2nd one is on standby and quiet. I don't think you can balance 2 stacked XW pro's for 50/50 load/charge aka smart AC coupling, at least not yet with the current firmware. But you could leave them unstacked and connect both to the same battery bank or separate battery banks, gives you a fault tolerance of 2. If one fails you get half the kW power at same kWh storage capacity, much better than Enphase.

Very intriguing, you are building and using the XW Pro as an AC coupled battery addition to Enphase. 6.8kW power, 13.6 kW overload for 1 minute and basically open-ended storage capability.

Click to expand...

So I ran it by itself and at low load (1.2kW) the majority of the buzz appears to come from the relay area. Didn't have my 500V Ansell gloves during the test, else I'd get touchy, but I got what I needed at this time. Its quite amusing how the audible frequency is proportional to varying loads. Given enough time I should be able to estimate the amount of draw given the variety of the audible tones Funny thing is that all of my 3KVA APCs prefer to get buzzed as well, although their DC conversion efficiency is much lower, so that is somewhat expected.

I do like the batt chem choice (I have somewhere about 40kWh of salt water batteries for the 2nd XW) & its demonstrated robustness as the XW takes on a more of a traditional approach to system design, closer to what engineering would do. Enphase is a far more "out of the box" approach as its setup is really a snap. They make you go through the certification, but for me it was more of a review of the basics + some details of what goes where.

I'm taking this on as a personal challenge with some diminishing ignorance, as a shadow companion. If I wanted to take the simple route, I'd hire someone and not worry about it. For me that would remove the fun out of the challenge. I chose Enphase because as of y2021, the report availability of the individual panel (detail), the relative stability of the micros (roof BBQ is no fun) and ease of implementation had no equal. I'm not a fan of DC arcs, the tighter design margins and no matter what I do it never exceeds 240V up there. There are a few more attractive nuances as well.

XW Pro 100A?
Is this the line diagram for the setup of your system?
Is it a full off-grid system or will there be a grid connect?

If it is fully off-grid, there is no need for the MID disconnect, all DER power can be connected to the IQ controller internal 200A bus bar, up to 200A * 1.2 = 240A. Enphase does not officially support off-grid configurations, although they do work.
The XW Pro AC1 and AC load lines must be protected by 60A breakers max, not more. The wiring, terminals, relays, CT primary lines and PCB traces are not able to handle more. In fact, the normal range operating current is 30A, with up to 1 min overload at 60A. The Deltron relays are rated for 30A FLA (aka RLA or carry) with up to 75A LRA overload. Running 100A will overload the AC relay board, or worse, set it on fire.
3 breakers in the IC combiner need to be 15A instead of 10A to match the rated IQ8+ power such that (N * 1.21A) is less or equal to 0.8 PB branch breaker rating. And the main PV breaker needs to be 50A for 31 IQ8's, like 31 * 1.21 = 37.51 which is more than 80% of 40A V breaker.

DER Solar said:

XW Pro 100A?
Is this the line diagram for the setup of your system?
Is it a full off-grid system or will there be a grid connect?

If it is fully off-grid, there is no need for the MID disconnect, all DER power can be connected to the IQ controller internal 200A bus bar, up to 200A * 1.2 = 240A. Enphase does not officially support off-grid configurations, although they do work.
The XW Pro AC1 and AC load lines must be protected by 60A breakers max, not more. The wiring, terminals, relays, CT primary lines and PCB traces are not able to handle more. In fact, the normal range operating current is 30A, with up to 1 min overload at 60A. The Deltron relays are rated for 30A FLA (aka RLA or carry) with up to 75A LRA overload. Running 100A will overload the AC relay board, or worse, set it on fire.
3 breakers in the IC combiner need to be 15A instead of 10A to match the rated IQ8+ power such that (N * 1.21A) is less or equal to 0.8 PB branch breaker rating. And the main PV breaker needs to be 50A for 31 IQ8's, like 31 * 1.21 = 37.51 which is more than 80% of 40A V breaker.

Click to expand...

I put the 100A as a kill switch - its a bolt-on and integrates nicely on the Smart Controller, since the XW will blow its 60A first if it needs to.
The setup is 100% off grid.

Enphase doesn't like "off-grid". They remind me each time I call It will run though and I do so continuously. Its very stable. I'm grounded with an 8' Cu stick in earth and a neutral bonded in one spot. All the usual.
Initially I wanted to place the XW to "fool" Enphase into thinking its on-grid. This would eliminate the whining with each call.
No worries, I would never push the single XW beyond 60A. I prefer the house to remain in its natural non-burnt state. I'm glad you looked!
Yes, I grew the breakers as I added more IQs. Rev3s is dated for its time. I didn't have 31 IQs live then. (Yes the pic totals 31, but got lazy updating the breaker graphics.)
I appreciate the concern. All is well.

Additionally, going into the Smart Switch with AC.Out kind of kills the teamwork out of the 2 systems, and power is what I want as well, so I'll test AC1 in parallel in the load panel and possibly continue Enphase "off-grid" for now - pending successful testing results. If that fails then I'll stick to going in as Rev3s shows. There are several possibilities here.

I'm also going to convert the IQ8+s to IQ8As since on good days I'm clipping and I don't like the IQ8+ being pushed around like that. Afterwards I'll re-run my math to figure out which breakers need to be swapped for bigger ones. The combiner panel maxes out at 80A, so there is still some room left. I'm going to add 14 more panels when all is said and done.

This is finally working after the latest IQ8 firmware upgrade.

wheisenburg said:

This is finally working after the latest IQ8 firmware upgrade.

Click to expand...

Great news. Did you have to update Schneider firmware?

solar8484 said:

Great news. Did you have to update Schneider firmware?

Click to expand...

I did actually update the Schneider XW Pro to the latest version. I also updated the Insight to version 18. I had some troubles with 18. It seemed to keep locking up. I had been on version 16. I backed it off to version 17 and I will see I am still getting lock ups. On version 18, I also needed to rename the file. It wouldn't upload due to something, maybe spaces in the file name.

My understanding based on another person that got this working is that the firmware upgrade was not required. It was purely an Enphase issue. Apparently their anti-islanding algorithm was so sensitive that the inverters wouldn't start up behind most hybrid inverters. They fixed the issue in their new firmware. This will probably allow the IQ8s to also work behind other hybrid inverters such as Sol Arc.

It is also my understanding that other people have gotten IQ7s to work when AC coupled prior to this firmware update. Without the update other people were experiencing similar results to me. Apparently, if you limited yourself to one of two IQ8s you might get them to turn on. There were some manufactures of AC coupled battery systems that apparently got the IQ8s to work by opening and closing a relay rather than frequency shifting. This may have also required installing a special off grid profile on the IQ8s.

wheisenburg said:

My understanding based on another person that got this working is that the firmware upgrade was not required. It was purely an Enphase issue. Apparently their anti-islanding algorithm was so sensitive that the inverters wouldn't start up behind most hybrid inverters. They fixed the issue in their new firmware.

Click to expand...

Good thing Enphase came to their senses. It's likely that telling affected customers to buy their batteries just pissed off the customers and didn't increase battery sales much.