Just wondering if anyone has noticed any issues with NodeRed flows affecting Miltiplus-ii inverter output. I have not yet tried to recreate the issue, I was away from the setup when my wife discovered the issue and I’m editing the config remotely, just wondered if there is a known issue or if others have had similar problems.
Symptoms:
Inverter AC voltage drops below low voltage cut off when under load. AC voltage, typically 230V drops to <180V. DC voltage was always above 26.5V with reported SoC above 90.
This only seems to happen when I have a simple Node Red flow activated. When I disabled the flow the inverter was stable.
The Flow monitors SOC of the battery bank and turns the Multiplus switch state to ‘ON (3)’ when above 95% and to ‘Charge (1)’ when below 85% (inverter off but still reports temp to GX device). I use this to keep; Engine, Thruster, and, Generator battery banks charged via separate AC chargers with excess power while away from the boat.
I usually deactivate the flow when aboard, however my wife called in when passing and raised an issue of the inverter cutting out when using the coffee machine / microwave only one at once (~1.5kW). On a subsequent visit after deactivating Node Red the system worked as usual with no inverter cut off.
Is it possible you have SOC drift or did the voltages look ok?
Is it possible a weak cell switch off the battery and it recovered? (As in it dropped below the inverter internally programmed voltage)
Thanks for the reply. The SOC of charge of the system was in the low 90s without any significant deviations. The battery was fine, there was no interruption of DC and no significant drop in voltage. - lights on the DC system where on throughout so an interruption would have been noticed.
I could understand if it was predicting the SoC in say 5 mins and this forecast value triggered the flow but I would have assumed it would have switched off the inverter each time. there seam to be instances where the voltage was reduced but stayed above the AC Low voltage cut off.
Screen shots of relevant parameters for the time on the problems:
That is a very weird ac out voltage.
It is happening a lot there. Is that inverting only or with ac ignore or connected to grid?
The ve bus state widget what does that show?
We have only noticed it on this one occasion when leaving the Node Red flow active while using the inverter. I usually disable the flow before arriving to use the boat. On this occasion we where disconnected from external AC power.
I have used the system before and after this incident without issue, both connected and off external AC power.
The VE bus state is shown below. The 3 periods shown as off do not align with the times we where noticing issues with the AC low voltage though. (plot below)
A separate but possibly connected issue: I have recently noticed that the date and time set on the OctoGX had at some point been reset / changed and did not align to real world time out by days / month. This seamed to cause a conflict disabling “real time” control through VRM, only providing data from a few seconds / mins ago on VRM. I could still access and the Remote Console (Classic UI) to amend settings while the clocks where out of sync. After now resetting the date time to correct world time, the “real time” data / control is now available in VRM.
While I assume NodeRed on the OctoGX would work locally only, I’m wondering if the date time issue could have resulted in some unknown cycle elsewhere in the system overloading the CPU… I’ll try to recreate the issue next time I’m at the boat and see if I can get the same symptoms.
The AC Voltage drop on Phase 1 is still occurring but isn’t linked to periods of high draw. With just a third party charger float charging the Engine bank, ~0.4A draw on the AC side, I am still getting output voltage fluctuation and inverter drop out. This is also shown as short OFF periods on the VE.Bus State.
Updating the time setting on the OctoGX did not solve the issue.
The Gateway D-Bus round trip time typically varies between 4ms and 12ms without the NodeRed flow activated. When activated a couple of spikes where recorded at ~17ms though this did not coincide with the periods of AC Voltage drop
Disabling NodeRed flow does stop the AC Voltage variation. (There is only one flow defined, but the service is still active.
Next time I visit the boat I’ll unplug all the electronics on the AC side and see if that changes any of the symptoms.
All plots from the previous 7 days - NodeRed Flow active for first 3 days only. off for last 4 days.
