Outback Skybox with AC Coupled Enphase Inverters

[wtfg]

I have a 6.4kw with Enphase Micro inverters with my new Skybox AC coupled to sub panel
In the last 3 weeks since I installed the Skybox I have had 5 SC-Non Recoverable Faults Detected ,witch shut down the Skybox
3 have been between 7:00 and 8:30 AM and 2 around Noon
Outbox is blaming the Enphase micro inverters when they start up.
Strange because it was dark and cloudy this morning, and 2 around noon when AC coupling was already happening.
Any body with enphase have similar problems?
What Grid profile are you using?
I am worried that this skybox is junk!!
Thanks

 

[Ampster]

I have been using Net Metering with Back up for the past week or two. previously I had been using Self Consumption but the weather and short days did not give me enough solar generation to charge my batteries each day to above 80 percent. That left me with little reserve in the mornings.
As a test, turn off the breaker to the micros and leave it off until 9 or 10 in the morning and see if the problem persists.

 

[RCinFLA]

Your best argument would be if you were running Enphase inverters without Skybox, with no issues, for sometime before installing Skybox.

I believe the Skybox is a high freq hybrid grid interactive inverter. HF hybrid inverters with AC coupling is tricky because HF hybrid inverters are vunerable to backfeed surge currents from PV system. Sol-Ark, for example, which is a HF inverter, recommends using AC input port for generator to connect to GT inverters so it can quickly open gen input relay to immediately disconnect PV GT inverters if a sudden backsurge occurs.

As long as there is fast protection from back surge the Skybox should recover with a reconnect processes. It sounds like a Skybox firmware issue.

You did not specifically say you were on active utility grid when this happened. Other important factor is if Skybox was charging batteries as time of issue. The back feed surge is most likely to happen when PV is producing high output when a large house load is switched off. If you had a momentary utility grid glitch it could also be a factor in the issue. A utility grid glitch may cause Skybox to release from grid leaving it to deal with overproduction of PV power with immediate push to batteries. Battery size must be large enough to take the back surge PV power in this case. Skybox cannot handle a large PV power back surge if battery size is insufficient, when not connected to grid, and its only choice it to dump PV GT connection. It should recover from this however. Also a big factor is if you have power push to grid disabled.

If battery size is too small or too fully charged, absorbing a large PV back surge power can cause battery voltage to rise above max Skybox DC input voltage limit. It must shutdown immediately in this case. It can take 10 secs to a minute to gracefully cause GT inverter to shutdown. The battery must be able to absorb the excess PV power while the GT inverter(s) are recognizing shutdown request.

If your battery size is too small for your max PV generation power, that would be your issue not Outback's. With 6.4 kW of PV power you should have about 600-800 AH battery. You could get by with smaller if LFP and leaving the battery no higher then 80% state of charge so it can absorb about 130 amps of momentary PV power induced battery charge current from a PV back surge.

With a 5kW hybrid inverter and 6.5 kW of PV capability, I would move 2 to 3 kW's of PV power off of AC coupling, connecting it to grid side directly. You will lose it when grid goes down but avoid overstressing the hybrid inverter.

 

[wtfg]

Your best argument would be if you were running Enphase inverters without Skybox, with no issues, for sometime before installing Skybox.

I believe the Skybox is a high freq hybrid grid interactive inverter. HF hybrid inverters with AC coupling is tricky because HF hybrid inverters are vunerable to backfeed surge currents from PV system. Sol-Ark, for example, which is a HF inverter, recommends using AC input port for generator so it can quickly open gen input relay to immediately disconnect PV GT inverters if a sudden backsurge occurs.

As long as there is fast protection from back surge the Skybox should recover with a reconnect processes. It sounds like a Skybox firmware issue.

You did not specifically say you were on active utility grid when this happened. Other important factor is if Skybox was charging batteries as time of issue. The back feed surge is most likely to happen when PV is producing high output when a large house load is switched off. If you had a momentary utility grid glitch it could also be a factor in the issue. A utility grid glitch may cause Skybox to release from grid leaving it to deal with overproduction of PV power with immediate push to batteries. Battery size must be large enough to take the back surge PV power in this case. Skybox cannot handle a large PV power back surge if battery size is insufficient, when not connected to grid, and its only choice it to dump PV GT connection. It should recover from this however.

If your battery size is too small, that would be your issue not Outback's.

It has been connected to the grid every time this happens
I have 16 Eve 280Ah batteries, 14.3 kw liFepo4 with a overkill solar 100amp BMS which Outback doesn't like
The solar with the enphase inverters have been in for a year

 

[wtfg]

I will try disconnecting the breaker for a few days
Amspter could you tell me what Grid Profile you have on the micro inverters?
And what loads do you run from your Skybox?
Thanks

 

[Ampster]

My IQ7s are CA Rule 21 if that is what you mean? Since I have been using Backup mode I am only running about 15kWh during the 2PM to 11PM Peak and Midpeak periods. I have 6.6kW of AC coupled micros and about 2kW of DC coupled solar panels.

What do you mean when you say Outback doesn't like the BMS?

 

[wtfg]

Thanks Ampster
Outback doesn't approve of my homemade Batteries and my cheaper CTs
I had 2 refrigerators and my furnace and some lighting on the sub panel averaging 1200 watts.
And I know when the frig and furnace start they draw more amps, I have tested all circuits
They said it couldn't handle that, even though it can pass through 10k
So I took the old refrigerator off and put it back on my main panel again.
Now that there is hardly any load on the sub panel they blame the Enphase micro inverters
And they say Enphase is not there product so they don't support them!!!
They say they tested Enphase Microinveters 2 years ago and they worked using Cal Rule21 profile.
But now there is around 12 Cal Rule 21 profiles.
He said to try each one till it works, why am I the Beta tester for Outback?
They claim it will AC couple 5k with grid profile IEEE1547 2003 and around 7k with Cal Rule 21
But they don't know which one will work, shouldn't they all work?
In the state I live in they have there own grid profile, I tried it and still had issues, its probably against my contract with the utility to use Cal rule 21
But what choice do I have?
I am stuck with what I have and trying to make it work, and I am afraid it will without help from this forum.
Since Outback blames everything but their system.

 

[wtfg]

I checked today their are 20 Cal rule 21 profiles
You just have to look in Enlighten Manager under report, then click grid profile settings report.
It will show what Cal rule 21 profile you're running.

 

[Ampster]

I am running the latest CA Rule 21 which is the same or very similar to UL 1741SA. As far as I know all that means is the way the Micros modulate production. As @RCinFLA mentioned, the issue you are having may be related to too much AC coupled solar, too little batteries. Earlier I suggested turning off the breaker on one string of micros to see if that is the issue. Have you tried that. Each suggestion you receive should be done alone as a science experiment to try to see if you can identify the issue that is causing the problem. I have a lot of respect for Outback service. You can view their comments as blaming other parts of your system and you may never solve your issue. Or you can decide to take their advice and eliminate some equipment scientifically to test the hypothesis. That is what I love about this hobby. It is a great big science experiment.

 

[wtfg]

Its failing at 7:30 in the morning, when its still dark here.
The outback expert said its the Micro inverters when they wake up pushing and pulling the grid causing it to lock out
I shut the main breaker today and didn't start them till 11:00AM and it worked fine, but some times it will be fine for 3 days
I could see shutting off one of the breakers in the combiner box if it was messing up during the day when the sun was shining
I am AC coupling 6.4 your coupling 6.6 I have 14.3kw battery pack the expert at outback said my batteries weren't the problem
I eliminated loads I am willing to try anything that makes sense
I am trying to find people with a Skybox and enphase inverters to see what grid profile they are using, I tried the Cl rule 21 201902 VV VW FW 1.2.8 and that failed
Or ill have to try every Grid profile till it works and if it isn't because of the grid profiles ill keep trying to find answers

 

[Ampster]

I eliminated loads I am willing to try anything that makes sense

If @RCinFLA is correct, eliminating loads is the wrong thing to do. Possibly surge loads but moving that frig did not make the issue go away so if it were me I would put that load back on the sub panel. Since the issue does not happen every day it may take a longer term approach. That is why I suggested doing only one thing at a time and give it a week to see if that one thing eliminated the problem. If I understand your post you are changing the micro configuration without seeing if that is the issue. I can't tell from your post if you have tried dropping a string of micros. The size of my solar system may not be relevant since I have 42kWhs of batteries to act as a buffer. I also don't understand why you are turning off the main breaker? Does this happen when grid connected or only when disconnected from the grid?
So far there are two many unanswered questions and too moving targets for me to give you any better advice until you slow down and try one thing at a time. Have you tried the Skybox thread at the Outback Forum?.

 

[GXMnow]

I don't know what an "SC non recoverable fault" is. Is SC "Short Circuit" or "Self Consumption"? Without knowing why it is shutting down, I am not sure I can help much.

I do know that the Enphase iQ7's on my system do "pump" a bit when online and producing power. I think it is the MPPT system hunting to find the maximum power. My cheap LED light bulbs have a visible pulsing when I am in backup mode on the XW-Pro inverter and the iQ7's are making power as well. When I am on grid, the power seems rock stable. Even while running off grid in backup mode, the output from the system looks perfect on my scope, even with the visible flicker in the cheap bulbs. All my better LED bulbs look fine and all of my equipment runs perfect off the inverter. I do not see any odd current surging with a clamp on amp probe. I do only have 16 iQ7's which max out at just 3,840 watts on clear cool days near solar noon. And the Schneider XW-pro is a 6,800 watt inverter, so it has much more authority to control my local grid. But in your case, you said the faults occurred while you were still on grid. That seems very odd indeed.

Can you verify how it is all wired?
Do you have any pictures of the setup?

The main panel should have a breaker that feeds the grid input to the SkyBox. The inverter output of the SkyBox should go to a backup loads panel. The Enphase gear should connect to a breaker in the backup panel only. In my case, I have the Enphase iQ combiner connected to a 20 amp breaker in my backup loads panel through the safety disconnect on the outside of the house. The input to my battery inverter is coming from my original solar back feed breaker in the main panel. That is also just 20 amps, due to my 100 amp panel with a 100 amp main breaker. Both the Enphase and the XW can back feed to my main panel, so it has to be limited to just 20 amps to meet code. But I ran #8 wire incase I upgrade later. When I put my XW into standby mode, it keeps the contactor pulled in so the backup loads panel is just a sub panel off of my main panel, and the Enphase operates just as it did before I added the battery inverter. Most of the base house load is all in the backup loads panel, so I end up with only a small amount of power that has to back feed.

I really doubt the Enphase gear is causing a problem. Did you have a contractor install the Skybox, or is this a DIY addition? Are you confident in all of the connections? Have you checked for voltage drops? Can you monitor the battery current? Did it show any current or voltage surges when the fault occurred?

 

[wtfg]

Thanks for your replies
I did the install on the solar panels and Skybox
The Enphase combiner box feeds the sub panel to upper left 40 amp breaker
The feed from the Skybox comes into the sub panel lower right 50 amp breaker
6 gauge wire
Bypass box so inverter fails I can still run load sub panel
I shut the 40 amp breaker for solar, my mistake I should have gone outside and shut off one or two of the 20 amp breaker's in combiner box
No I haven't measured the battery current, only have a AC DC clamp on meter
Would would you recommend to use?
This only happens when its grid connected, I have only ran it on batteries for a few half hour times and that seemed to work fine.
Want to get the bugs out before I do time of use.
Thanks again for your help

 

[RCinFLA]

Skybox has some historical performance data storage.

In Charger Commands, there is highest and lowest battery voltage data. Check if they are inbounds. You can reset these values so you can then check min/max battery voltage after next shutdown event.

Another thing you might try is lowering the maximum battery state of charge setting to something like 70% to ensure battery can take max back surge without its voltage rising too much.

Not sure what 'SC-Non Recoverable Faults Detected' means. Is there any other info in System Notification log? Check for seeming unrelated messages near the time of shutdown fault. On the Status Screen with performance graphs what was the history just before fault occurred. Any grid glitches would be important to know, although it might get mixed, mis-interpreted, and lost in inverter shutdown.

Definitely worth trying switching off one of the two outside 20 amp branches to reduce max PV push and run it for while that way. You might have an intermittant defect in one of the microinverters so try running on one of the two branches then try again running on just other branch.

GT inverters attempt to 'wiggle' their output phasing to push against the grid to check for valid (low impedance) grid. The inverter/battery must be strong enough to resist this and if there is too much AC wiring loss between grid and inverter input it can put more of the grid checking loading on Skybox and its battery.

Also too much battery cable line loss can raise the maximum DC input voltage at inverter when PV back surge push happen. You will have to handle a max peak of about 130 amps back surge current into battery, when grid is down, so cable needs to be sized accordingly. I would use 2/0 gauge.

 

[wtfg]

I jut reset the high and lowest battery voltage, they went to zero but they didn't stay at zero.

Before the SC- non recoverable fault shutdown I had
6- PCS-NPC Gate faults
2- PCS DAB FPGA trips
3- PCS-DAB UVP Battery faults
I was averaging 8 cleared PCS faults and Can Communication failures a day , they seemed to stop after latest firmware update
I will try shutting off one breaker on the combiner box and leave it for 3 days than switch to other breaker.
Strange that it happens early in the morning when there is no sun
Today there was 2" of snow on the panels they didn't start producing till I cleared the snow off at 10AM
So far no failure today panels are currently put out 3kw,sun is shinning

I will order some 2/0 cable instead of my 1/0
Also waiting for a bigger BMS
When I started building the battery I had ordered a 100amp 16s from Overkill solar then I figured out it wasn't enough
I thought it would work ok till the 200amp one comes thinking that its grid tied and the grid could take the surges
I am probably wrong!

 

[GXMnow]

3- PCS-DAB UVP Battery faults

UVP is most likely "Under Voltage Protection"

My guess is that it is detecting the battery voltage dipping too low. This could indicate too much resistance in the system. I know my Schneider is very picky about the battery having very low resistance. These large pure sine inverters do pull a lot of current from the batteries, but it is not a steady DC current. It is a series of short high current pulses that build up and roll down at double the line frequency rate. I see it as 120 Hz humps, with both halves of the sine wave going from zero up to 50 amps or so when I am running a moderate load. At full load the peaks would be something like 250 amps.

I looked over the pictures. I understand the two interlocked breakers in the sub panel to choose the inverter output or a direct feed to bypass the inverter. But what is the large transfer switch changing over then? Is that for switching the solar back to the main panel?

Is the 4th image your PV combiner? It looks like the correct equipment to do that, but it does not look like the Enphase combiner box. Where is the Envoy and the production CT?

I don't currently have a manual inverter bypass transfer switch. If my inverter failed to where the grid pass through contactors would not turn on, I would have to moves some wires to get my backup loads panel back on the grid main panel. I bought a 60 amp rated cam switch, but I am a bit skeptical of the quality to be running all my backup loads through it.

 

[wtfg]

The large transfer switch is from the main panel to the input on the skybox and the bypass breaker on the sub panel
The normal breaker is coming from the Skybox load output
The 4th picture is the Skybox BAS for wiring inputs
I have a normal Enphase combiner box outside.

 

[RCinFLA]

UVP is most likely "Under Voltage Protection"

These large pure sine inverters do pull a lot of current from the batteries, but it is not a steady DC current. It is a series of short high current pulses that build up and roll down at double the line frequency rate. I see it as 120 Hz humps, with both halves of the sine wave going from zero up to 50 amps or so when I am running a moderate load. At full load the peaks would be something like 250 amps.

Question on UVP is whether it pertains to battery voltage or AC grid voltage. Battery would likely be a recoverable fault, low AC voltage could be non-recoverable, at least until grid voltage gets back into acceptable limits. Low AC grid voltage should just mean inverter disconnects from grid and pickups up house loads running from battery.

For a sinewave inverters, the ripple current on battery cable will be at twice AC freq sinewave (120 Hz) with a peak current of a little higher than twice the average current seen by a shunt monitor which averages (low pass filters) the 120 Hz ripple current out to get average readout DC current. Peak current is higher if load power factor less than 1. GT inverter's output current follows AC voltage waveform.

DC input current to inverter almost goes to zero current at minimum but the inverter overhead (30-60 watts) will be a near constant DC current that sets the minimum floor with the large 120 Hz ripple current riding on top of it. Some BMS's use this ripple current measurment to calculate the battery Rs resistance by measuring the corresponding ripple voltage across cell terminals caused by cell impedance.

Interesting thing, if power factor on hybrid inverter from an inductive motor load is less than 1.0, the instantaneous current can actual reverse into battery for a small portion of the AC cycle.

Two examples of 3 kW of real AC power consumption with two different power factor loads. This shows why load power factor is important when running on a battery powered inverter. Note difference in inverter loss and battery consumption.

 

[wtfg]

So my BMS and cables can't take that type of peaks, ill have to try and fix that
Still have to figure out what event is causing it to demand that much from grid or batteries.
I am not seeing high loads or high or low voltage from the grid at the event times, looking at Optics
Maybe it to fast for Optics to count
Will wait and see what tomorrow brings

 

[wtfg]

This morning at 8:08 AM I had a Non Recoverable Fault
Preceding it was a DAB-UVP Battery fault again than a Gate Fault

As RCinFLA asked I reset lowest battery voltage, it was 40.65V
So does this prove its the cables or BMS ?
Thanks