Dear all, this is a question out of curiosity to understand the system better.
I have a 3 phase system with three MP II 5000/70, a Cerbo and 6 Pylontech 5000 batteries connected through 3 Lynx Distributors (in series).
When my SolarEdge system has “filled” the batteries (SoC = 100%) the difference between the highest and lowest battery is 40 mV and stays that way for some time while the systems directs the surplus power to the grid.
On the 14th of September i saw a new phenomenon. Both the highest and lowest voltages dropped and where only 30 mV apart and started to increase again like a stair case. During that time the surplus power was still floating to the grid (SoC = 100%). At a certain we started cooking (SoC < 100%) and the Voltage dropped …
What is the battery doing while it is in staircase mode and is this mode necessary to keep the batteries in balance ? Should i prelonge the period the batteries are at 100% SoC
Most likely that curve is formed due to the data aquisition/update interval, which on VRM is 1min at its fastest setting. You can see the same steps everywhere else in the graph. In this particular case it simply means that the ceel voltage climbed at a steady rate of about 0.025V over four minutes (guessing every step is a minute as we dont see the time axis) which means that every data collection intervall sees a voltage increase of 0.00625V compared to the previous one.
Now the points inbetween could get interpolated, to create a smooth line instead of a staircase, but the way it is now you can at least you can easily see each individual datapoint instead of imaginary filler.
Theres a small possibility that the cell voltage indeed climbed like this, but its rather unlikely. The only way to clarify would be to use a data aquisition that is faster than VRM.
@chrigu thanks for the swift reply, i have enlarged the picture and do not think it is because of the data acquisition interval as the periods are pretty long compared to the 1 minute of the VRM.
I am with you in case the accuracy is limited to 10 mV and the readings are rounded to that number, so the way the voltage is increasing is like you describe over a period of time ..
I just wonder why the system changes from the higher voltage all of a sudden to this (sort of) staircase interval as nothing changes (SOC = 100% and remains that way) … what is the system doing ???
And thanks for pointing out that i have not clipped the picture correctly … now updated with time stamps visible … sorry to have put you on the wrong path …
Then it might be the result of the BMSs ADC converting the analog cell voltage into binary data, and those 0.00625V are smallest difference between two bit values. So its the smallest increment the BMS can work with.
VRM only uses the 1min intervall of you set it to that, by default its 15min afaik.
Load changed, charge changed, BMS switched balanceing on/off? Add battery current and DVCC signals to the graph.
If a DVCC signal changes at the same* time then the BMS instructed a change in charge current/voltage. (In theory you can also conclude whether the DVCC change caused the voltage drop or vice versa. If the CVL dropped and the cell voltages dropped, then the cell voltage drop was likely caused by the CVL drop. If the cell voltage dropped but CVL went up, then CVL was reacting to the voltage drop.)
If the charge current changed, but without DVCC changes, at the same* time then the load/charge changed on its own.
If neither charge current nor DVCC changed at the same* time then its likele the BMS balancing.
(*) Data collection/aquisition timing between different devices is not guaranteed to be synchronous. The 1min VRM intervall is likely way too coarse for that to be significant. But you can get an idea on differences by comparing for example battery voltage collected from different devices
@chrigu , many thanks for your explanation and i found the reason with comparing different graphs like you suggested. The water cooker was turned on and for any reason the sun decided on that moment also to shine a lot less … the battery was activated but not shown in the SoC and when the water was hot the amount of solar power was so low (~1000 watt) that the charging of the battery was slow, hence the staircase …
Once more, many thanks for your guidance and questions and pointers, i have learned a lot on how to use VRM to check these “anomalies” (that are normal things) …
