@Zz1049
I guess I'll count myself lucky that Solar Assistant has been working well for me for quite a while now. (Maybe because I'm reading the battery SOC from my Victron SmartShunt?)
Ian had worked with MPP support to get me a firmware update. But, I never applied it since SolarAssistant was working for me.
However, I am glad that I'm not the only one seeing wonky voltage readings coming from the unit.
JAS
It's a bit of a ramble but I read through your posts and was following up really quick before a meeting at work, the full deets:
I have a SOK 100AH and 206AH (Both Heated/BT) versions, in parallel, yeah I know not a great idea mixing but they're pretty close in resistance to the point where wear is not significant. I use the BT function to see the BMS SOC and voltage of the 4 cells in pack + total voltage, I can confirm that with this as well as a DC charger that the voltage on discharge/load of 500W~ voltage will drop by as far as .3V on the batteries per MPP on SA logs. Meanwhile live watching the batteries, the drop is nearly nonexistent outside of fractions of a tenth of a volt. I am using 2AWG cabling for batteries/MPP. On charging, I will notice up to .2~+/-1 voltage fluctuation where the battery reports a higher voltage than is actually present (verified with BT BMS). Both batteries will pretty much uniformly follow each other and remain consistent from full SOC to 0 (or near 0 at least, since BMS disconnects at 10V).
SA(SolarAssistant) will claim some communication error (I troubleshot this like a year ago with Ian/SA) and then cease to connect to WiFi. It appeared as though the failure to communicate with the MPP unit caused SA to lose WiFi (Super weird) support from Ian was to swap the USB comms board out for a RS232...which had...placebo fix effect lol, it happened as often as it did before and the comms difference was negligible so I just left it vs. switching back. Ian thought that there could be a voltage feed coming in through the USB from the MPP causing the error but that was not the case. (I never heard of such a thing but went with it). Nonetheless the solution to the SA bugging out and needing to be reset was hooking it up to a WiFi plug and just having it reset every day at like 4AM which seems to keep the issue away (it usually has the problem between days and freezes). Personally, I think I have a bad Raspberry Pi.
Is your SA hooked up to the batteries directly via the smartshunt? I think having a direct connection to the batt will fix the issue but the SOK's do not have a comms port on the 12V variety. I didn't know I'd experience the voltage fluctuations but bought the BT version to have more accurate reading of SOC which paid off since the MPP doesn't record voltage accurately. I think that the individual who stated the traces were not robust enough on the board is on point because if they didn't cheap out on that connection we would see less of a drop, your situation with the board swap confirms that suspicion.
Ultimately I was curious as to why I've been noticing this because I'm using the unit to offset peak pricing hours with the 300AH batteries and act as backup power for my server. Since we have rooftop solar and battery for the house this is more of a "subsystem" and as I'm trying to dial it in over this last year (rooftop solar was installed 6 mo ago) I've been trying to get things dialed in so it naturally offsets peak hours while still maintaining backup power in case of outage.
Pardon the lengthy response, I appreciate you bringing up the concern/observation and posting on the forum. I found this thread very helpful.
Best wishes!