5kva MPII ground relay test failure during LS, not when syncing

Inverters
61

Three LS later, 3 UPS’es installed at a distance from the inverter, breaker to the inverter is still being switched off.

image

I’m holding my thumbs …

Yeah, should not have said anything …

image
image495×828 91.4 KB

Switched all the breakers on the main DB off, as in every single breaker. :man_facepalming:

The UPS’es definitely have an effect, but then when Eskom went off this morning, after all that time of perfection, this is something else I suspect. Walked around but don’t hear any generators in the street.

UPDATE:
Switched the panels off … UPS’es on … lets see where this goes.

62

Update

So, it seems the 8 and 10 am slots causes the drama, not the 4 pm or evening slots.

image

Have ruled out the UPS’es from the equation. With them connected, and disconnected, the system had errors yesterday morning. As a matter of interest, it seems the UPS’es were getting errors when the system got errors. Will re-test that again when the system is sorted.

With good advice, have done a small change. Now waiting for an 8 / 10 am slot to test if it solved the problem.

63

I’m still inclined to think there is more going on. It should not be this finicky to what is going on on the grid. Once LS lifts up a bit, I’d certainly look into having it repaired. In the mean time, I suppose you can use the time to figure out what might be contributing.

The thing with a problem like this is someone is going to put it on a bench, run it for an hour, and declare that they could find nothing wrong with it.

64

Exactly my thinking.

It works 100% as designed i.e. manually switch Eskom off:
Then back on, it syncs.
Or leave Eskom off, it works as designed.

To trap this error, one needs to test it onsite, 8 or 10am LS slots, it seems.

So you are right, there is more going on, and I can assure you, it is of no hand of mine.

Then again, maybe it is the unit, but why only during certain conditions and time slots?

My cousin lost a relay in one of his 3 x 3kva MPII’s setup a while back due to Eskom LS. Apparently there was a batch with a challenge, which Victron is sorting.

If that was the case here, one would have expected a Victron dealer to say that the serial no is from that time, lets replace it quick. But no-one is suggesting that, on top of which, everyone is in a serious backlog situation. So yeah, not easy to eliminate the inverter being the cause. Can take weeks.

Future: I’m looking actively for a MLT inverter, the older workhorses @Gman so loves. If I find one I will make it “plug and play” backup unit. It is a k_k feeling to be in this position. Should have bought the one unit he suggested a few months back

65

I will keep a extra eye open for you. @BruceHill bought that one and I see he was on the forum 3Aug last so I think his one is running perfectly.

66

check the dark side

67

OOOooooo a 12kw one 15H. I found a 4 kw oasis one for him. the new shape with the touch screen. But that 12kw sound just nice!!! can do 15kw for 30min.

68

I’m blocked on that site for some “dark” reason. :thinking: :smile:

Can anyone ask if it is still available?

69

probably academic but it would be interesting to know whether the “PE2 open” error indicates that:
a) I checked and when I try to close this relay it remains open (or is slow to close)
b) there is a condition that does not allow me to close this relay and I need to close it.

for me the first one suggests hardware problem, and the second one the conditions like swapped neutrals, bridged earths etc. (cursory look through victron community suggests that PE Open is less common than PE closed).

While running tests it is probably useful to keep track if the ground relay fault also remains the same type (open vs close). My thinking is that if some changing condition on the grid is causing the reported error, then the relay check should not [EDIT] necessarily always fail at the exact same point.

Last thought, if nothing on your system changed since the last bout of stage 6 load shedding and it worked without any reported errors, and assuming (a.k.a. hoping) it is not a hardware fault on your multi, then possibly the muni fixed the street box without you knowing or possibly if really limited to daylight hours, the neighbour with the new unregistered solar system installed by his cousin is powering something it shouldn’t.

70

Merely being a member here is sufficient reason I think…

Sadly South Africa does not have a right to be “forgotten” in law (POPIA only allows requesting deleting of personal information). Because I would certainly have liked to be forgotten over there…

1 Like
71

The street box … is open. Has always been open. :slight_smile:
Really limited to daytime hours.
And we are on PAYG, during LS, it is dead.
I switched the entire DB off, every single breaker in it, in case of feedback.
Also walked around during LS, listening for gennies, and solar panels. None.
Contacted my neighbour next door asking if they have a backup of any sort. None.

72

So would I.

Groetnis

73

Mmmm, you are getting closer …

Professional company 1 comes and installs my 1st array, a very long time ago. Groundwork is put in place for expansion too.
A long while later, they come back adding some more panels for me.
Another long while and a 2nd professional company comes and adds even more panels for me.

Then, after the last bout of bad LS, professional company no 3 comes past and adds even more panels for me. Now I have 2 x MPPTs and 15 x 350w panels.

Since the last addition, the MPPTs do their thing, batts are recharged as expected and better, the house is powered without any reported errors, just bliss with the 5.25kW array.

Now, take note:
*without any reported errors - from ±10h30 yesterday, after LS, right through 2 x LS events since, one at 16h00 and one at 00h00, today the entire day (no LS), nary one issue. 100% faultless.

Can anyone explain that?

If it is a relay issue, it should happen during all LS events, consistently, not so?
When I manually switch Eskom off, repeatedly, it should give an error eventually, not so?
Why, 98% of the time, only the 8 and 10 am slots of LS?

Does anyone want to hazard a guess?

74

Something (likely a daytime running device or some such) is doing a back feed on N likely when LS occurs and messes up the N P voltage…

Groetnis

75

You are getting closer too …

Thought of that.

So I switched everything off yesterday morning:
The main DB. Each and every breaker too.
The flats main DB, and their Critical loads DB, breakers too.
Unplugged everything that was plugged into the inverter circuits, fridges etc.
Even unplugged the UPSs.

All that was running was what was powered by UPS batteries.

EDIT1: To also mention, the loads powered at 8/10 am slots, are the same loads powered at 16h00 onwards LS slots.

EDIT2: Every time the inverter gave an error yesterday, I walked around fast asking anyone if they switched anything on/off, till I got tired of all the stairs, and switched the DB’s off. :wink:

76

Right, 16h08, LS started. Let’s see what happens now, rd one so far, the previous ones have no errors.

Let’s see what happens this round.

image

77

Battery grounded at all? On my BYD’s, there is a grounding lug… Battery cable routing and induced voltages due to sudden current demand are other possibilities…

Groetnis

78

No grounding on the batteries.

79

A perfect operational system, not a error or a peep from a UPS, just relays happily clickity clacking when reconnecting with Eskom.

Leaving the entire DB board on, not switching off any breakers.

To sum up: Have now gone through 5 consecutive LS events, with no errors.
EDIT: The 8 am one, i.e LS event no 6, the one I want to test, has started at 8:05.

And then, minutes later, the grid was off again. Back on a few minutes later. That is a first.

Each time, happily clickity clacking to sync.

This is a first for us:

image
image1376×312 20.3 KB

80

And sure as damn, after 5 LS events with no drama, this one at 8:05 am, it’s back.

And loosely the same pattern as the other morning:
Power goes off, then after ±15min, the first error.
±1 Min later, the next one.
… I’m not going to go through them all again.

My temp earthing/bonding of the panels to the earth of a plug supplied by the inverter did not work.

I have now switched the panels off.
During the 5 events with no errors, the panels were on.

image