Zappi2 Self-Test Failure not reported

[scotflieger]

If Zappi2 EV charger fails its self-test, this is not reported through the MyEnergi integration.

Before starting a charging session the Zappi2 carries out a self-test of the built-in breaker. This is reported in the MyEnergi app and the Zappi2 has to be reset should this occur.

[LeiChat]

Perhaps it's achieved with a push notification in the myenergi app, rather than a property available through the API and as-such, not available for the Home Assistant integration.

[YanisKyr]

I came here for the same. It would be good to have this in HA - if it's being published via the API that is.

[jwillemsen]

Attach the integration diagnostics when this happens, that should show the data provided by myenergi

[YanisKyr]

Attach the integration diagnostics when this happens, that should show the data provided by myenergi

How do I get the diagnostics?

[jwillemsen]

See settings/devices&services/myenergi, press the three dots and select download diagnostics, attach the file here

[YanisKyr]

I think this captures only current state/data.. would we have to pull the diagnostics while the (myenergi) app still shows the red triangle?

[file removed]

[jwillemsen]

Yes, pull it when the app shows the error

[YanisKyr]

Sorry for that, I cleared it straight away as I wanted to get the car charging again. I'll keep it in mind for next time.

[G6EJD]

I’ve been through the API, so what makes you (naively) think that Self-Test failure is reported through the API - it isn’t!

this needs to be closed.

[jwillemsen]

Could you point me to the API description @G6EJD? Always good to read that and see what is all possible

[G6EJD]

The API is on GitHub, but remember it’s been reverse engineered so not documented.

Don’t assume that every capability can be controlled or reported through the API, it was never designed like that from the outset.

[lvo12]

Looking for this as well! Is the assumption incorrect that when the app reports about the Zappi error, that the info must be available within the api (somewhere) as well?

[jwillemsen]

I would expect that it is part of the API/data somewhere, assuming that the myenergi app uses the same API/data

[lvo12]

Just checked the output in HA, because the app showed the "Pect selftest failed".

"data": {
"XXXXXXXXX": {
"deviceClass": "ZAPPI",
"sno": XXXXXXXX,
"dat": "11-07-2024",
"tim": "12:00:22",
"ectp1": 0,
"ectp2": 0,
"ectp3": 0,
"ectt1": "Internal Load",
"ectt2": "Internal Load",
"ectt3": "Internal Load",
"bsm": 0,
"bst": 0,
"dst": 1,
"div": 0,
"frq": 50.01,
"gen": 0,
"grd": -1063,
"pha": 3,
"pri": 1,
"sta": 1,
"tz": 3,
"vol": 2334,
"che": 3.23,
"isVHubEnabled": false,
"bss": 0,
"lck": 27,
"pst": "C2",
"zmo": 3,
"pwm": 5200,
"zs": 6912,
"rac": 4,
"rrac": -4,
"zsh": 27,
"ectp4": -109,
"ectp5": -266,
"ectp6": -696,
"ectt4": "Grid",
"ectt5": "Grid",
"ectt6": "Grid",
"beingTamperedWith": false,
"batteryDischargeEnabled": false,
"g100LockoutState": "NONE",
"phaseSetting": "AUTO",
"wifiLink": false,
"ethernetLink": false,
"ocppEnabled": false,
"mgl": 100,
"sbh": 17,
"sbk": 5,
"sbm": 0,
"fwv": "3560S5.417",
"cmt": 254,
"newAppAvailable": false,
"newBootloaderAvailable": false,
"productCode": "3560"
}
Could it be that the information is in the api, but the call to retrieve it is unknown?

[G6EJD]

I believe they have NO built-in test capability (BIT) from which they can diagnose a Protective Earth Circuit Test failure. It means the Zappi checked the Earth impedance value to see if it is too high (open circuit) or above safe values.

You should be getting your installation checked for earthing integrity. Your Zappi has done you a favour.

[lvo12]

Earthing has been checked twice, no issues found. When this happens, the solar output is around 1500 kwh and MGL set to 100%. My car can start charging at a 1500 kwh minimal charge. It only happens when the solar output is flapping between 1200 and 1600 kwh, MGL set to 100% and my car is connected/charging. It never happens on a sunny day with full solar output or on a rainy day when the solar output does not exceed 1000 kwh.

[G6EJD]

Then it’s highly likely your Solar inverter is creating the impedance earthing issue, or as I believe you have a 3-phase system that there may be an issue of phase alignment between incoming grid and solar phases. What happens during charging is the vector summation of phases can give odd neutral currents, that only exist when charging, ideally the neutral current should be zero so that it doesn’t create a N-E voltage, noting in most installations apart from the cable lengths N and PE are the same point at the transformer. So an out of balance solar generator can generate an Neutral current which is returned by the E to make a circuit which in electrical terms means there is an impedance which can then be measured (correctly) by the Zappi as an impedance error. So I would be analysing your wiring of both grid to the Zappi and solar generation especially if the latter is not a 3-phase system, but I believe it is. Getting this right cdn be difficult but in my experience the issue is solar and Zappi phase alignment.

[lvo12]

The Zappi was installed in 2020 on 3 phases. There was already a PV system (installed 2016) on 3 phases. Recently (2024) a small PV system was added which is connected to 1 phase. The PECT selftest error has been there since the installation of the Zappi. Things have been double checked including earthing. I'm not sure if the average chargingpole installer understands what you are saying. I'm affraid they'll do some basic checks, putting their thumbs up and leave me with the bill. Getting the right specialst understanding the problem will be hard, but even harder to reproduce the issue when such a technician is onsite. Hence my idea to create an automation, reporting me via Telegram on the issue causing me to walk downstars to the Zappi, reset it, and get on with my life. ;-)

[G6EJD]

Yes difficult because getting the right conditions on the day of diagnostics eg sun, etc is near impossible.
If I were investigating I’d check the labelling of the phases installed on the Zappi they are the ME references for all measurements, ideally they are Zappi-T1 wired to your first nominated/labelled phase, the physical phase used is irrelevant, then repeat for the other two phases so you known exactly what’s what. With the Zappi measuring grid voltage it will need to be grid aligned for phase and rotation direction, and it works OK. Then solar production takes over and now the voltage source becomes your new (in effect) Grid supply, now the correctly setup Zappi can have phase connection errors and an incorrect rotation but unlikely for the latter, so this is when the solar connection to your incoming grid may need re-alignment.
The simple test is to check your Zappi Power Factor values on the status page for P123 when there’s no solar generation (important) and all 3 should be positive and in the range ~0.7 to 1, it’s never an issue fact science.
Next do the same when your Solar system is generating and the PF values should be negative in the range -0.7 to -1, the negative value denotes its exporting. However if there are any incorrect values when the Zappi does a vector summation of the 3 voltages that’s when the mathematical process can give unusual results and at the electrical level odd neutral currents can flow creating the PECT error.
Then to further compound your problem the single phase solar generator can skew the neutral point voltage (which should be zero), and it could be its that system that is creating an imbalance, so I would start diagnostics by disconnecting the output of that single phase system to see if it clears the PECT fault detection, if it does then you’ll know where to concentrate on. If it does the solution could be to move the phase it’s connected to, it won’t know or care, then try again, if needed then the next phase.
If they doesn’t solve it (alternative phases) then you may need to add two extra loads to rebalance the neutral point under solar charging.
The key bit is when solar starts working the Zappi is no-longer seeing the correctly oriented grid voltages, the voltage source is new albeit temporary one.

[lvo12]

Should I check the Zappi PF values when it is charging the car or when it's idle? The single phase PV system was added a month ago, I've been having this issue for years already so I don't think that that is causing it. A mismatch between the (older) 3 phase PV system and Zappi seems more likely. Your help/knowlege in this matter is highly appreciated!

[G6EJD]

It’s a pleasure to help.
The normal house load is usually sufficient to get a Zappi PF reading when the solar isn’t generating. So no need for an EV charge.
I forget you mentioned the issue predates the single phase extra.

[lvo12]

So I just checked the Zappi readings (not sure if I have the right values):

CT1: 4.1A | 1034W | -1.00
CT2: 5.0A | 1325W | -1.00
CT3: 13.1A | 3230W | -1.00

I guess it's about the last value per CT? There is no unit mentioned. You can guess where the additional PV system is connected on. :-)

Now I wait until the sun goes down to check these values again and the house is running on grid right?

[G6EJD]

Well they look OK they are saying all solar outputs are currently exporting, so a useful check would be to see if your solar systems are saying the same values for the same phases

[G6EJD]

I’ve been thinking this through. I suspect there is an interaction between your Solar inverter earthing and the MyEnergi connections.

It is highly likely your Zappi has been set to TT / TN (Terra or Earth to Terra or Terra to Neutral) earthing, the Earth testing only works in this setting and when overly sensitive you can desensitise the Zappi by switching it to IT mode in the setup menu, I am confident that will fix your problem.

Dave

[lvo12]

I’ve checked the CT values when PV production stopped.

CT1: 1.1A | 22W | 0.08
CT2: 1.0A | 195W | 0.84
CT3: 0.5A | 61W | 0.46

I made a few snapshots and the last value was always between 0 and 1.

The earthing setting was indeed at TT / TN. I’ve set it to IT and will have to wait for the specific conditions. I’ve set a reminder for over 3 months. If the issue has not occurred since, I consider it solved. If the issue acts up again, I will post it.

Thanks again for your expertise in this case, very much appreciated!

[G6EJD]

Your PF values for CT1 and 3 look far too low if you’re no longer generating any solar.

But I’m confident the problem is fixed.

[G6EJD]

This issue needs to be closed.

[lvo12]

Switching the earthing to IT did indeed fix my problem. Never had it after changing that setting. Thank you very much @G6EJD !!