FW 0206A5 Bug?

[Steve777]

Not sure if this is the best place to ask this, but here goes:

Recently (4-24) had MN upgrade my FW to version 0206A5. It seems to have fixed the issue I was seeing with forced charging, which is great. But after looking at it's operation for a few days, it seems to have introduced a bug around time slot operations:

Have been running with two time slots set to forced charge, 00:00-12:55 and 19:10-23:55. These are for the "cheap time" on our TOU plan. The forced charge keeps the AIO from drawing power from the batt then. The rest of the time, 12:56-19:09, is the high cost time and want to use the batt not the grid. Use mode is self consume with charge from grid off.

With the old FW, this worked pretty well during the active time slot periods the battery was not used as a source of power, and was recharged as PV was available. Since the upgrade to the new FW, the inverter seems to be ignoring the end of the 00:00-12:55 time slot (and maybe the other one too but hard to tell when there is no PV). After 12:55, whenever PV is insufficient to power the loads it now draws from the grid even though batt SOC is at 100% and we are in self consume mode. Min batt SOC is 40%, so that should not be a factor.

Another odd thing is that with the batt at 100% SOC for many hours (18 or more), it still is sending 170W to the batt. It seems hard to believe that it needs that long to balance the cells when at 100%. Is this right?

I have double checked all of the settings and do not see anything which could be causing this behavior.

Is this a know issue with this new FW? Any settings I should check?

TIA.

[Barry Fields]

Tell me about your batteries and the system voltage.

My FLA banks (48v 215AH) draw about 80watts in stabilized float.

[Steve777]

My batt is a Midnite Powerflo16. A LFP type with BMS. Should not be going into "float" as L_A batts do.

[ClassicCrazy]

My batt is a Midnite Powerflo16. A LFP type with BMS. Should not be going into "float" as L_A batts do.

Are you using closed loop ?
Larry

[Steve777]

My batt is a Midnite Powerflo16. A LFP type with BMS. Should not be going into "float" as L_A batts do.

Are you using closed loop ?
Larry

Yes

[ClassicCrazy]

My batt is a Midnite Powerflo16. A LFP type with BMS. Should not be going into "float" as L_A batts do.

Are you using closed loop ?
Larry

Yes

What voltage are you seeing on the battery when  you see it charging for a long time ?

[Steve777]

@ClassicCrazy

Don't know as I did not look or measure the batt voltage when this was happening.

[Steve777]

So just to update: Have fixed or worked around the problem.

Previously there was an unused time slot in the middle of the list, and the used time slots did not follow temporal order. This was not a problem for the previous FW, but seems to be with the new FW. I cleared all the time slots. Re-entered the same values using the first two slots, and put them in ascending time order. That appears to have fixed the issues I was seeing.

I stopped "experimenting" once things worked, so can't say if it was the unused slot in the middle of the list, or not having them in time order that caused the problem (or both).

I did ask my installer about the problem, and he reported that Logan told him it is a bug in this FW.

[ClassicCrazy]

glad you got it figured out !
I am chasing software issues - not related to Midnite stuff .  But it seems to be the time of season for staying late .
Larry