Pylontech x Sofar - battery in standby mode far too early

[sharpener]

Pylontechs seem to work without drama. I added a third module 2 weeks ago, and guessing that on delivery the new one would be at about 70% SoC like the original two I ran them down again to 70%, put the new module on top and voila!

They get fully charged most days either from the PV or off-peak grid. Maximum delta V has fallen from an initial 110 mV to about 30 now which strikes me as OK.

My only gripe is that the Pylontech BMS can taper the charge current as the batteries approach full, but does not do the same at the other end, so it will allow the full discharge current until it hits the low voltage protection limit and then cuts off all of a sudden. I have however managed to write my own taper algorithm in NodeRed to prevent the battery shutting down completely under heavy loading, the inverter makes up the difference from the mains.

[dangermouse]

I made myself a Pylontech console cable so I thought I'd check out the balancing on my batteries' cells.

Charging with a variable PSU, one battery is at 52.6V and the difference between the highest and lowest cell is about 30mV. How close do the cells have to be to be considered properly balanced?

Basically I used my inverter to charge the batteries to near 100%, then connected one battery to the PSU and slowly ramped up the voltage from about 50V to 53V, keeping the charge current around 200-300mA. Any lower and the battery goes into standby. I noticed that the individual cell temperatures don't change but the "ambient temperature" reported by batteryview rises sharply.

I'm assuming that this "ambient temperature" is actually the temp of some electronics in the battery, and the increase might indicate that the balancer circuit is operating.

[Tinbum]

The cell balancing kicks in at anything over 30mv. You can read the values using the console cable. Yes, the ambient temp sensor seems to be near the balancing circuit !!

[dangermouse]

The cell balancing kicks in at anything over 30mv. You can read the values using the console cable. Yes, the ambient temp sensor seems to be near the balancing circuit !!

Yep, using the console cable and batteryview software. Just wondering when I should stop the process - based on what you say, it won't balance any more if the cell differences are <30mV, so this is considered balanced?

Current battery is at 52.2V, cell difference 20mV, all cells at 15C but "ambient" is 29C. So I think this one's done?

[Tinbum]

I watch the ambient temp and when that starts to reduce the balancing has stopped.

Re reading values- i was referring to the battery internal BMS settings.

[dangermouse]

Yeah I can see the pattern now ... the voltage gap increases, then the ambient starts to rise, and the voltage gap decreases again.

It seems to go in cycles, mine stopped balancing, then after 5 minutes more charging @ 240mA, started again.

Thanks for your help

[Tinbum]

[dangermouse]

Re reading values- i was referring to the battery internal BMS settings.

I see what you mean now, there are some values there that batteryview doesn't show you.

Also, it wasn't not immediately obvious (to me anyway), but the cells will balance even if the battery is not charging, as long as the cell voltages are high enough (over 3.36V going by the image you posted) and the difference is >30mV.

[Angriness8601]

I've noticed (using solarman app) that my inverter (HYD3600ES) moves to standby mode when the batteries' SOC is way over the limit. I've updated software (Sofar), so it now allows me to set DOD to 90%. Depending on a day it moves to standby mode between 22% and 31% (31 is actually today, first time so high).

Hi all,

New account here, found this forum by searching for a issue to what ducabi is facing.

What's probably happening here is the batteries are dropping to below a magic voltage of 47 volts. At this point, the sofar firmware promptly freaks out and shoves itself into standby mode forever and a day until the sun comes out again or a charge from the grid is initiated.

You can see this behavior from my Home Assistant installation where it goes from standby, into discharge (forced by me rebooting the inverter) and then back into standby the instant we go below 47.0 volts:

You can verify this without home assistant by checking the display on the inverter itself. If you see a history of ID85 errors but you know you have plenty of SOC remaining, that's the inverter deciding you don't have enough volts to continue.

Even with Pylontech selected, and even with the correct battery selected you can see a "minimum voltage" setting which you can set between 47 and 52. You cannot go below 47. With certain versions of the firmware you can go up to 90% DOD, but the inverter will go into standby if the 47 volts number is reached first, regardless of the SOC of the batteries. Even if it was 99% SOC, it would go into standby until the voltage was at or above 47.

Pylontech's own manual says that below 47 volts isn't a problem. If facebook is to be believed than Pylontech themselves have confirmed 46 volts is just fine. The batteries will naturally do this anyway when under load and when working at a lower SOC. When the weather is colder, this will also happen and will be more pronounced. All perfectly normal and nothing to worry about. My manual says 43.5v is the minimum but if you're sitting there for any length of time you're probably causing damage.

So. Solutions?

• Sofar release a software update to allow a minimum voltage below 47 to be set by the user

• Buying more batteries might stop you hitting the lower SOCs as much, but that's hardly a solution

• Get over your first world problems and just live with it

• Somehow hack the firmware to change the lower limit from 47 to 46. This might void warranties if anyone found out about it.

• Perform a "Man in the middle attack" against the CAN connection between the battery and the inverter; telling the truth about the SOC but lying about the voltage until maybe you hit a limit you're not happy with and then start telling the truth again which would then force the inverter to back off. You might be able to drain down to 95% SOC as well if you lied about that (newer batteries support 95%). This might void warranties if anyone found out about it.

• Change the sofar inverter to a more generic battery type which ignores the CAN bus and hope for the best but forever live with an inaccurate SOC being calculated and everything generally being sub-optimal

• Move the setup to somewhere warmer, this will limit the voltage drop but won't fully fix the issue

First option is probably the best option but I'm not sure how responsive sofar are with doing such things.

[Tinbum]

It may be that sofar are not happy to run the inverter down to that voltage.