This morning just after 6 last shedding kicked in and my system shut down.
This is after I had several shutdowns during the midnight to 2am LS.
The only way to get it to stop tripping… I need to put a decent load on the system. Then after it drops a couple of % on the SoC it’s fine.
But on restarting it this morning I noticed the battery wasn’t full…
Looking at my logs… When the power came on (eventually) at 2:56 I got a low battery alarm and Internal error alarm and then we proceeded to not charge at all. All the indicators on the gx console show normal for the battery though and the battery was only down to 56%
Going to hook up battery view just now to see if it provides better info
Edit…
Battery view shows … Normal temperatures, cell voltages all within 0.01V and “SystemErr”
Edit2…
Power came back after LS now and battery is charging my normally… Colour me baffled.
I’m reading. I don’t have an answer. Sometimes it does that, you reset it, and it never happens again. I vaguely remember older firmware would also do that if the battery was idle for too long, so updating firmware could resolve it. I don’t know… PN support is your best bet.
Please don’t see that as a poke at anyone… reality is I’m sure I’m not the first one with this issue and I won’t be the last. What I was rather attempting to convey there was that the proper path with these things is to actually use the support they provide - you did after all pay for it.
As an interim update… Pylontech did respond asking for logs which I do not have with me a the moment so I’ll send the logs later today.
First impression… for an international company to respond to any kind of support request that quick is quite impressive.
Looking at the logs though… Normal balanced cells, temps all normal… Even SoC logged correctly… But SystemErr and a shut event… Smells like firmware… Or this battery is totally trashed… What do they sell for as spares?!?
I’ve encountered system errors on my stack of Pylontech batteries in the past, sometimes strange combinations like low temperature and cell imbalance errors at the same time. The errors went away when I reseated the communications cables between the batteries and also the CAN cable to the inverter. Those network plugs don’t fit very snugly in the sockets and it seems to me that if the connection is not great that strange things tend to happen.
Perhaps worthwhile swapping the network cables around to see if it has anything to do with cabling, just to exclude it as a possibility if nothing else.
But here’s the interesting bit… The battery has been tripping like clockwork at 71%. So tonight I was proactive… At 73% I went around the house and switched on ALL the lights. This gives me a 700~ watt load (don’t judge, the outside lights are all old-school and the garage light is some fluorescent tubes). We powered straight through “trip-zone”!!
So 300~watt is trip central, 700 is happiness and joy… Now my concern… If I add a battery then we always have low loads running because then 700 would be the new 300!
I do recall an explanation by @plonkster on the complexities of the voltage discharge curve of lithium cells and how the SoC is basically guesswork… So I guess the firmware is doing some weird maths, not carrying the i and then ends up shutting down…
That or I have weird wiring in the BMS that occasionally (and consistently) gets the voltage wrong (doesn’t make sense as the logs show normal cell voltages at shutdown… Also note, these are the pylon battery view logs, not the victron logs)…
I emailed support again this morning as I hadn’t heard from them yet. Just got a message from @Pylontech-Jeff (email from wetransfer) with new firmware… I can only assume this means they are also suspecting the firmware.
Interesting… I wonder if this is then another update of the previous firmware that was supposed to solve the issue. Please let us know what is the version thereof. Good luck, I really hope it will solve your problem.
So the provided firmware is 2.8 (I was on 1.8 before) - The versions suggest that this would be the older version of the US3000C (pre chip shortage). I think 2.8 has been dropped on the firmware thread before… But considering the implications/issues etc of a bad flash… emailing @Pylontech-Jeff and pylontech support and waiting a couple of days for the correct firmware seems worth the effort/wait.
Flash was successful (but based on the number of failed flashes reported on this forum and several other platforms… super stressful.). We’ve now been running a couple of days again but haven’t had a 4hr bout again.
My scenario is an odd one in that sense I guess, I don’t have a big backup load running on this system - averaging between 250 and 350watts normally. Tripping was happening around 71% - that would imply roughly 3 hours into a 4 hour load shedding stint. To complicate it a little more (and as mentioned in previous posts) this only happens when running the small loads, at 700w and above it does not trip - in fact I’m pretty sure anything above 500w would manage this.
New house means I don’t have my wired network to the inverter at the moment so monitoring is a little limited. I will however let it run over the weekend and see if we can get past our tripping points.
It’s been a while… but wanted to give it some time. So I can happily report that I haven’t had the battery error out on me again since the firmware update.
For those looking to do the update simply because of the additional info it feeds to a victron system… the information it adds is a little underwhelming (considering what is available when you use batteryview)
But hey, short version… If your battery goes into fault mode and you can’t see a valid reason for it to do so… contact pylontech support for some fresh firmware… provided you give them the correct serial number for the battery, this would probably be the safest way to flash the battery as you should have a much better chance of it being the correct firmware for the battery.
