WOW, I had an overload …
Wanted some coffee, so I made a plan right at the same time the people upstairs did their thing, probably cooking food or also making tea.
The inverter did not even switch off … smart man, smart.
New discovery.
LS is from 06h00-08h30.
Last night there was quite a bit of lightning, so I switched the panels off, No-Ark breakers, just feels better.
This morning I only remembered at 7h50 the panels were off, LS still happening… and …
TADAA!!! I did this…
Steps: (for me to remember)
250/85 String 1 - when switched On = Error
250/85 String 2 - when switched On = Error
250/85 String 3 - Off - (ran out of time.)
250/85 String 4 - Off (spare string)
150/35 String 5 - when switched On = No Error
Note: I switched them individually on.
So to get it working again:
150/35 String 5 - when switched On, No Error
250/85 String 3 - when switched On, No Error
… and then LS was over … once synced, I could switch on 1-2, no problem.
Will try the same again when we have LS at 6-8am again … and later today too.
Plan is simple:
If I can make it happen on demand, then swap inverters.
If the new inverter has no error, then it was the inverter.
If the new inverter has the same error, then it is a problem with the array.
1 Like
When looking at that, why would String 3 not work in the top table and in the bottom table it does?
Is this not just the MPPT giving issues?
Brand new MPPT’, both are.
Good question, fixed the typo …
Right, early days, but I THINK I may have found a potential cause.
I have a 5th set of wires, a spare string, running from the roof, wires connected to the breakers, although the breaker is off, connect to the busbar, connected to the surge arrestor.
So I thought let’s disconnect the wires and put a MM on it the wires just because they are there and I have a MM.
Measure a peak volt of over 9v, hovering below 1v.
EDIT: At 3.21 pm, measuring a constant 0.01v.
So 8 am there is a higher “whatever you call it”.
With the wires totally disconnected, no error this morning with the 8 am LS schedule.
If all goes according to the “schedule”, I can check tomorrow again …
How that is even possible with the breaker off … the tinfoil hat is still being folded.
Or maybe the wire/s touched other parts of the array … I am grabbing at straws here.
PS. I found another problem with the network freezing. Someone, no not me, installed an AC cable right next to a Ubiquity WiFi points UTP wire. Promptly removed it and have not had any, the remaining infrequent, router freezes.
The 2 spare wires, were on the same 250/85’s combiner box.
You’re measuring radio noise… literally, a low impedance meter on a really long wire… you will get a ghost reading.
Cool.
Interesting that 8am has more “radio noise” than later in the day.
The wires have been disconnected completely downstairs with one 8am LS so far with nary an error.
As I accidentally found that switching off/on the 250/85 at 8am LS has a direct bearing on the error, the later the day, the error goes away. The spare string was wired into the 250/85 combiner box, even if the NoArk breaker as off.
It’s picking up the radio noise of the other MPPTs. MPPTs make more noise when they are tracking a power point. Once they start voltage/current limiting (because the battery is full) the noise dies down.
So the bundle of wires coming off the roof has that measurable effect on the two wires hanging loose about 30-40cm from the MPPTs when the batts are at 30% SOC? Then later in the day, the MPPTs still run flat out, batts are full or close to full, and it gets less noisy on those wires. Right?
That would make then 100% sense if I got the picture right, the 8am being the worst.
But I do not see how that radio noise can cause a relay error on the inverter via the MPPTs.
Or maybe the wires are touching something on the roof.
IF, and it is a big IF, the problem is indeed gone now with the wires far away from the combiner box, then leaving a spare string on the roof connected to a combiner box is not such a good idea.
Steps to test:
Wait for 8am LS.
Or 6-8h30, switch off the panels the previous night, and switch on ±8am, the 250/85. That is how I can trigger it. The 150/35 worked fine 3 times before as it is on a separate string, nowhere connected to the spare wires, as the 250.85 was.
Ok, to recap.
Had a spare string from the roof, as Plonkster explained, the voltages I read were due to radio noise, higher at 8am than later in the day.
The breaker was off but the wiring from the breaker was still connected to the surge arrestor for the 4 strings.
So I disconnect it all. Just thought why not.
Also had to switch the NoArk breakers around. Where they were from Load (panels) in at the top, it is now Load (panels) in at the bottom.
And I moved the 250/85, being connected to the inverter direct, to the batteries.
Yeah, three changes, dammit, just realized that now.
This is now the 2nd 8am LS, and there was none planned for the week (yeah, still the pea under my mattress).
The SOC was around about 25%.
And I switched the panels off and on … no problem.
And I boiled a kettle, once before that caused a trip in the 8am slot.
Out of 5 x 8am LS slots, 2 down as “fixed”, 3 to go.
Ps. I did wear my tinfoil hat this morning.
Man, if you want a 8am LS slot, or run a small business, dammit, but they change their minds ALL THE TIME!
I’m leaning towards an unscientific “gut feel” that the 2 spare wires from the roof, the radio noise measured highest at ±8am:
- that the 4 strings, one not connected, all in the same combiner box for the 250/85 on same surge arrestor
- with the 250/85 connected directly to the inverter’s battery connections
- that when the panels are off on the 250/85, no error
… that I may have found the problem.
My arbitrary 5 “tests” at 6-8am LS slots: 3 down as “fixed”, 2 to go.
My arbitrary 5 tests at 8am LS slots, 6 down, and not one VE.Bus Error 8: Ground relay test failed when there is load shedding.
The two changes:
The radio noise from the spare wires, wires are now disconnected.
MPPT, was originally connected directly onto the inverter’s battery connections, now connected to the batteries.
How that can impact the inverters AC relay, and that only during 8am LS slots, I have absolutely no “koeking clue”.
EDIT: I do have a total mess to fix now, as everything was pulled apart during this troubleshooting. 
The error on my inverter was resolved, as above.
So much so I never fitted the replacement unit that was sent to replace my unit. As a matter of fact, could sell it on having arranged that with my supplier, the unit never being used, as a brand new unit.
Just a huge thank you here for Victron SA and my supplier, for having arranged this for me at the time.
They really stepped out far and beyond what most do, can do.
Respect.
3 Likes
I must have missed this… I have mine all to DC busbar’s - never thought of the inverter DC side.
Glad it looks to be sorted TTT!! 
Yes, that is the normal connection.
Way back I was told that connecting the MPPT directly to the inverter was the most efficient way. Also in my case at the time ito the cable run from the MPPT to the then DC busbar.
So it has been like that for many years.
Titbit:
So the one day the sparkie and I was prepping for him to do something.
The process was:
AC off - tick
Inverter off - tick
Disconnect the batteries - tick.
Then because of the moerse caps, switch on the inverter one last time - tick.
Inverter starts right up again with no batteries. We looked at each other in shock … ghost in the machine much!?
After a minute or so … switch off the combiner box TTT - ticked.
PS. It is sorted. Have had many many 8am slots over the last 2 months.
Bugger … AGAIN around about 8 … this time 8 PM just before LS was over.
Caveat:
- It was pouring with rain during that time.
- The wife got a sparkie in to move a light switch and extend a plug a day or so before.
DB redone.
The El.Tech, similar to the Tripconnect device, is also installed.
I set El.Tech to disconnect at 251v, the Victron disconnects from the DB at 255v, I know, I have experienced that.
These last 2 weeks I have had 4 times where the volts exceeded 251v, and the house is disconnected from the grid until the volts drop below 246v
Also disconnects from the grid when the volts drop below 215v, as the general volts here are 238-244v most of the time. The LS switchover “feels” smoother.
Set it to 65 seconds before it re-connects, the Victron then another 65s on top of that before it syncs.
Titbit: The moment I switch on any of the two geysers on the main DB and the IC (Malignant Load Protection value Pure resistive Load Limit) 0.5A - 5A is on, the El.Tech disconnects. Sparkie needs to come and test for why.
The EL will also be re-tested, as will the arrays earth.
Main DB neatened up:
- Main 63a breaker
- Onesto Surge Protection
- El.Tech is also a Surge Protection, but I use it mainly for disconnecting the DB from the grid based on Eskom volts. The Onesto is the “sacrificial” one if you want.
The new Critical Loads DB:
Surge protection is there to stop any surges back from like i.e. a satellite dish and a lightning bolt in the area. Inverter basically sits between 2 surge arrestors, as suggested by my preferred Victron support/dealer/repair shop.
And the new changeover is also in. YES! … and the little lights.
PS. The breakers are on purpose … I have a 4kw inverter, “thou shall not overload it” that it switches off. Nope, “thou shall not”.
Like in the buggers here connected the dryer whilst the washing machine was spinning, the circuit breaker tripped, being 10a. Gotcha, I said! Their faces … priceless.
1 Like
Update on this matter.
With all the expenses incurred as per the above pictures, the system was moved, the earthing of the panels where momentarily not connected (wires from the roof were also moved) … and the inverter had the same error when LS took place. Near instant as a matter of fact.
The moment the sparky connected the panels earthing again … ok, truth be told, they missed it. I connected the damn earth wire to the main DB. Yeah, I was that desperate, I worked on the main DB myself.
The problem was solved instantly.
For the record.
Before the re-connecting of the panel earth, after having re-done the main and critical loads DB’s, at moerse cost I will add (again), I also had the very old CBI Earth Leakage replaced with a brand new CBI Earth Leagake unit.
Eliminating ALL potential causes, “they” said.
So in any and all cases, to eliminate the extreme case of panel frames as a potential cause, even if it sounds silly and even if no relation to anything at all, switch off the panels. Easy enough test.
Maybe my extreme case holds an answer for someone one day when all was tried and nothing worked.
Earthing and tripping issues are not always that simple.
Other times it is extremely simple.
Sometimes it is just plain weird.
This error caused major dramas … and resultant dramas.
What happened after the inverter was taken off the wall, when I decided to move the whole shebang from one wall to another …
BEFORE - after it was moved from the one wall, me, just not having the “cleverness”, or experience, to mount it so that it looks cool …
It was safe … just don’t touch shiite ok!
AND AFTER - all AC is left, and all DC is right. The idea came to me one night late … to KISS.
… and it is not done yet … need to get some paint.
Coolest of this redo, I can now move the system ANYWHERE … AND the inverter can be taken off the wall with ease. Everything is easy now, the batts are still on wheels.
I’ll send you a picture of my upgrade when it is done. Also on wheels. Except the battery, because the battery weighs 150kg.