I’ve had a Raspberry Pi 3B+ running Venus OS for over a year, and it’s worked perfectly. At the weekend, I expanded my PV system and switched off the Raspberry Pi for this, then switched it back on again after some time, without any problems. Since then, the MPPT has been showing me a different generated value from the one shown in the History the following day. For example, on Tuesday evening (the MPPT was switched off), it showed around 5 kWh, but the next day Tuesday’s figure was 7 kWh. This has happened every day since the modification.
By chance, I noticed in Venus OS that the time is exactly two hours behind. No matter what I set, the time doesn’t change. I can only set it manually via SSH; after a restart, it’s two hours behind again. The time zone is correct, and the NTP servers are also being used—I can see this via my Pi-hole, so that isn’t the cause. Internet access is of course also available on Venus OS, and the VRM Portal and app work perfectly when accessed remotely.
Venus OS is 3.79 Large.
Is anyone else having this problem? Thanks!
EDIT: Strangely, the post isn’t displaying any paragraphs, although they appear in the editor!
The MPPT doesn’t have a “clock” at all; it determines day/night from the voltage curve. So if it was switched off briefly, it sees a “new day” and is confused for 2–3 days.
Okay, but why is the MPPT showing today’s yield as 6790 Wh even though it has only been generating for just under two hours? The 6790 Wh was already shown in the history yesterday evening, when the MPPT was switched off at 8 p.m. That doesn’t make any sense at all; this wasn’t happening before either.
Until now, it started at 0 every day, and the yield was counted as soon as the sun came out.
EDIT: The values in the VRM Portal are much more plausible.
The PV system was expanded. It’s very likely that the PV now generates enough voltage at night that the MPPT no longer recognises it as night and therefore doesn’t reset the daily counter. My MPPT in the garage also constantly sees 9–10 V on the PV side at night due to the street lighting.
VRM maintains its own statistics based on the reported power outputs.
That completely distorts the whole count and makes no sense at all in my view. Even if the MPPT sees just under 10 V, no current is flowing and no power is being generated either.
Surely Venus OS or the MPPT should be intelligent enough either to detect that no current is flowing any more or to know from the time of day that there can no longer be any sunlight…
The MPPT’s internal clock is determined by sunrise and sunset. These two moments are detected using the PV voltage. If the PV voltage does not fall below the sunset threshold, sunset is not detected. It’s a simple, pragmatic solution. Whether it’s the best one is, of course, debatable.
Midnight is assumed to be halfway between sunset and sunrise. That’s why the MPPT needs a day or two to calculate this reasonably accurately.
The GX only reads the devices and displays the data. The system should not alter this data either, even if it is incorrect or implausible.
