diy solar

diy solar

Is this related to LiFePO4 chemistry?

jailcee

New Member
Joined
Feb 11, 2021
Messages
22
I have a 8s 280ah battery made of 8 cells with an overkill BMS. I'm seeing new behavior that I'm trying to understand and wondering if it is related to chemistry. Cell 7 (7th cell in the 8S config) of my 8 cells is reading at higher voltage (up to 3.65 and triggering cut-off) when charging, at normal voltage (similar to the other cells) when idle, and at much lower voltage than the other cells when discharging. I had not seen this behavior before. Has that cell suddenly gone bad on me or is this something temporary relating to the LiFePO4 chemistry? The cell has been treated exactly the same as the others.
 
I have a 8s 280ah battery made of 8 cells with an overkill BMS. I'm seeing new behavior that I'm trying to understand and wondering if it is related to chemistry. Cell 7 (7th cell in the 8S config) of my 8 cells is reading at higher voltage (up to 3.65 and triggering cut-off) when charging, at normal voltage (similar to the other cells) when idle, and at much lower voltage than the other cells when discharging. I had not seen this behavior before. Has that cell suddenly gone bad on me or is this something temporary relating to the LiFePO4 chemistry? The cell has been treated exactly the same as the others.
Have you tried top balancing the cells?
 
Right out of the gate I top balanced the cells by having them in parallel with a 3.65v charging current. That was a few months ago and the cells have been at work in my off-grid system for the past couple of months. I've not seen this behavior until now.
 
Unless you find a poor connection I would assume that cell is losing capacity. If the trend gets worse I think I would be looking for a single replacement.
 
I would mark that cell so I didn't loose track of it .... put them back in parallel and top balance again .... then clean all connection points and apply noalox or similar .... re-assemble with that cell in a different position .... see what happens.
 
Cell 7 (7th cell in the 8S config) of my 8 cells is reading at higher voltage (up to 3.65 and triggering cut-off) when charging,
Just out of curiosity, how high is the battery voltage and what are the voltages of the other cells?

If you're charging you battery really high, like to 14.4V or something like that, its much easier for a single cell to go rogue because at that level just a little energy causes wild voltage swings.
You may need to adopt more conservative charging voltages (like 14.0V or even 13.8V) to keep your cells in a voltage range where they remain similar. Nearly all of your capacity is under 13.8V so you wouldn't be sacrificing anything by adopting a more conservative charging strategy.
Similarly when discharging, this lower capacity cell may drop significantly more than the rest before some point, perhaps below 2.8V or maybe 3.0V.

This is not a particularly uncommon situation or solution.
 
In summary my understanding:
  • no current cell 7 is same voltage as the other ones
  • charging its voltage is higher than the other ones
  • discharging its voltage is lower than the other ones

That would point to a cell or connections with a resistance higher than the other ones. Where are the BMS cell taps connected?
  • if it is directly on that cell posts, it is the cell internal resistance that is higher
  • if it is on a cell link or the post of a different cell, it could be a bad connection (not tight enough or dirty)
You can get an idea of the internal resistance with two measures (easier to do when discharging):
  1. record cell voltage (V1) when current (I1) is between 1A and 10A
  2. record cell voltage (V2) when current (I2) is at least 30A above I1
Cell Resistance = (V1 - V2) / (I1 - I2) current is negative if discharge
It will not give you the same resistance value as the one measured with proper equipment (pulse measure), but good enough as a first investigation to compare cells resistance.

While you wait for a new cell, reducing charge and discharge current will reduce the difference in voltage and the risk of cut-off.
 
There's a bunch of great suggestions here, thank you all!

I will:
- mark the cell to be able to ID it easily
- reassemble the battery with the cell in a different position (say #4?) using an anti-oxidant compound between the cell terminals and bus bars
- Check behavior
- if still happening lower charging voltage from 28.2v to 28v (this is a 24v system)

At the same time I will attempt to measure cell resistance. While I'm at it I think I'll check it on all the cells.
 
I've marked the cell in question, moved it to position 5, and reassembled this time using a dab of Ox-Guard on the terminal of that cell. I'm currently there and getting full sun on my panels pulling around 1200w. I'm not getting the same behavior with the cell tripping over voltage, in fact I can see that it is unbalanced compared to the others (lower voltage) but under about 43A charging current (at 28.2v) it is no longer spiking.

In other words those steps resolved my issue.

I'm now thinking that it would make sense to redo all the cell-to-busbar connections with a dab of anti-oxidant OX-Guard, both because it seems to make a much better connection and as a preventative measure from this happening with any of the other cells. Any reason I shouldn't do this?
 
Back
Top