Balancing a troublesome CALB cell

mldee

New member
Joined
Mar 2, 2018
Messages
14
I have been using a 8S1P new CALB 100Ah LiFePO4 27Varray with Batrium wmon4 BMS for about a year now to power my home waterpumps etc. Charging is via the PWM charger in the PIP2424 inverter, set to 27.9 V charge.

I noticed that cell 6 has always been a little reluctant to stay inbalance with the other seven cells, usually less than .01V difference.

However lately cell 6 is getting into wider swings as seen in the 11am charge pic below:



image_qfpcqs.jpg


I realise that this is cooking the mon. Eventually, in a couple of hours all the cells will balance and then cell 6 may even finish up .01v lower than all the others.

Looking for suggestions oh how to rebalnce cell 6 to the others, (high or low)or is it a potentially bad cell?
 

Korishan

Moderator
Staff member
Joined
Jan 7, 2017
Messages
6,329
You may need to pull the pack out of service and do a capacity test on it specifically, or check the fuses on the cells (if you have these). It's possible you have a blown fuse, or 2, or 3.

Considering that pack 6 is 'higher', I would lean more towards blown fuses in that pack. And maybe, you are getting more and more popping as the loads are are balanced across the whole pack anymore.
 

mldee

New member
Joined
Mar 2, 2018
Messages
14
Thanks for the prompt reply.

However, these are 8 xCALB 3.2V 100Ahprismatic cells in series, not 18650 packs, so there are no fuses involved.

I have also just noticed that the high level on this cell only occurs when there is no load on the system , causinghigh solar charging current, about 20+ amps.

As soon as a 220VAC water pump or such is running, the charge current drops back and feeds the pump via the inverter, and the cell 6 voltage drops back to normal level. Pumps stops, cell 6 goes high again.

Also noticed that when there is a high load put on the system (eg, 900W heat pump hot water system) cell 6 drops significantly lower than the other cells, will try to get a pic when the HWS heat pumpstarts at 1pm.


Here is pic of system with Heat pump on, so small charging current. Cell 6 now sits lower.

image_qiuitw.jpg
 

Sean

Member
Joined
Oct 8, 2016
Messages
946
Cool the mons, or set a lower charge target when in bypass, and set a lower bypass current so you aren't cooking the mon - cell voltage will rise irrespective of what you set the bypass voltage to when the mon is in thermal bypass.

.... and check cell 6 connections.

........ and confirm, with an calibrated voltmeter, that the mon is reporting the cell voltage accurately.
 

mldee

New member
Joined
Mar 2, 2018
Messages
14
And here is the final 4pm fully charged, basically no load pic. Cell 6 sits about .15V below the rest. It may improve over night as all cells discharge to about 50V with night intermittent pump loads.Will look again tomorrow morning.

I've checked connections and voltages, all is well.

I'd really like to fix this.


image_ljrvwq.jpg
 

daromer

Moderator
Staff member
Joined
Oct 8, 2016
Messages
5,393
Since it goes high during charge and low directly during load you have a weak cell. You cannot balance away that fact.

Retest all cells for capacity and even IR and you will see whats going on...

You can potentially top balance it and leave it as such but cell 6 will drop faster than all other no matter what you do and most likely due to high IR have issues during charge and during last % of charge go way high. To play with this you need to reduce the charge during the last & of SOC.... With that said i would recommend to swap that cell out.
 

Sean

Member
Joined
Oct 8, 2016
Messages
946
mldee said:
I'd really like to fix this.


image_ljrvwq.jpg

What method are you using to determine that each cell is properly fully charged ?
 

mldee

New member
Joined
Mar 2, 2018
Messages
14
Yep. I tend to agree with you.

This is the status at 7pm, all looks great!


image_sqpuwr.jpg


Little load, no charging current, everything's happy.

I'll contact the supplier and try for a replacement as thecell is only 15 months since purchase new.

Thanks for your insights
 

Redpacket

Well-known member
Joined
Feb 28, 2018
Messages
1,239
+1 to Sean's suggestion ie check the connections for cell 6, eg clean the cells terminals & cell-cell straps, plug/unplug the cellmon a few times, etc.

You might also try backing of the charge CV voltage a bit eg from 27.9 (~3.48V/cell) down to say 27.6V (3.45V/cell). I run my LiFePo4 packs at CV = 3.41V/cell
If the issue persists, the cell may be on the way out, the more it's used in the pack, the weaker it will get, the more range of voltages you'll see on it.

Are there any signs of swelling?

Also do you have auto balance enabled in the WM from ~90% SoC?
 

mldee

New member
Joined
Mar 2, 2018
Messages
14
Sean said:
mldee said:
I'd really like to fix this.


image_ljrvwq.jpg

What method are you using to determine that each cell is properly fully charged ?

Since these are sealed LiFePo cells, not really much I can do to ascertain charge except depend on the Batrium BMS, which has been quite reliable and also shows the state of charge of the other cells plus the overall state of charge.
 

Sean

Member
Joined
Oct 8, 2016
Messages
946
Voltage is not the determining factor for charge termination, tail current is - meaning the cell should be kept in the CV mode until the current has dropped to a level as specified by the manufacturer - typically a few percent of its C rating.

So before you condemn cell 6, make sure it's being correctly, fully charged.


image_tvvxag.jpg


Ignore the voltages depicted, the concept I'm illustrating is the CV phase and decaying current until charge termination.
 

mldee

New member
Joined
Mar 2, 2018
Messages
14
And just to finish things off, here's the pic from 7.45am next morning at 46% charge. Looks good again.


image_aqnjzc.jpg


Cells all look in perfect condition, no swelling etc. I think I've got a dying cell.
 

mldee

New member
Joined
Mar 2, 2018
Messages
14
I dropped the CV charge point to 27.7 and will see if that helps, if so, might even go down to 27.6

Sean, thanks for the advice on charging, especiallythe graph,I'll take your advice and see how things go. In all honesty, I think my primary concern at the moment is trying to stop cooking the mon!

it's now 10.45am and cell 6 is starting to sneak up, .03V higher.

I have not looked at the WM4 auto balance setting, will have to go and study up how to dothat

Bit cloudy here today, so cells are charging a little slower than normal. 11am 63% chargepic:


image_uoirxc.jpg
 

Geek

Well-known member
Joined
Aug 15, 2017
Messages
934
What Sean is trying to say, is that lithium batteries have a fairly flat discharge curve. LiFePo4 especially. State of charge is impossible to predict based on voltage alone. It is most likely the cell is simply not reaching full charge. By lowering your float voltage, you are not going to improve the situation.

As for the mon - cook it - its cheaper than a battery. Put a fan on it to help cool it instead. Maybe even buy another. You can put several mons on one cell, to help balance it faster. Perhaps even put a dummy load on it - such as a light globe or a resistor, to help the mon out.

You probably do have a slightly crook battery, but with LiFePo4 cells, it is going to be hard to do much about the situation.
 

daromer

Moderator
Staff member
Joined
Oct 8, 2016
Messages
5,393
But if the charge state would have been a problem the cell wouldnt have switched being over voltage and then suddenly under voltage comparing to the other cells just by a little charge or discharge.
That clearly show signs of an issue with the cell or bad contact on cables/junction points.


Auto level as such wont do that much in this case since its passive balancing. It still require that the cell perform the same in most of the work area.

Note that 0.03V is nothing.... Im easily off by up to 0.05V or even more. But its not that off that its an issue. Ie same amount of during load as in during charge.
 

Redpacket

Well-known member
Joined
Feb 28, 2018
Messages
1,239
My money is the cell is failing - a cheaper stop gap may be to add a smaller LiFePo4 cell in parallel with your cell #6.
While not ideal, it'll postpone having to replace the cell & lighten it's load.
It'll also help to operate the pack so cell #6 isn't being pushed too high or too low, the more that happens, the faster the demise will be.
 
Top