Schneider LiFePo4 Build

[n2aws]

awesome, just so im crystal clear:

my cell target full charge with no load should be 3.4-3.5v

unlike with lead acid batteries i don't need my charger setting 2 volts above my pack target to get there---ie---my absorb/float setting (for this one time at least) should be 54.4v.

If i set my chargers at 54.4 and im still getting cell overvolts, reduce my charger setting to whatever the battery reading happens to be when the event occurs and maintain that till my delta reaches 10 mv or below, then creep up to 54.4 in .5v increments.

This is the point where you'll get different opinions, and as such, I'd direct you to the documentation from your cell manufacturer.

As an example: The manufacturer of my battery recommends 56.8v for the pack for bulk/absorb, and float.
My batteries are 16 cells, which works out to 3.55v per cell for "normal daily" cycling. (LiFePo4 chemistry)

I personally don't like to drive my batteries that hard, after looking at some of the various charts online. So I deviate slightly.
I still have 56.8 for bulk and absorb, but 54.4 for float. This gives me a good mix of "charges up so that top balancing can occur" and "cells are at a decent resting voltage" during float.

As for your last sentence, exactly. set your charger to maintain whatever voltage triggers the warning. Then wait until the cells balance and the warning clears. Again, make sure you are doing this with *no* load on the battery. Then add maybe 0.5v, and let it balance completely again. Repeat the "raise 0.5v and let balance" step until you reach the target voltage for the pack, and it is balanced with no alarms. As a purely "these are my settings" example:

If I wanted my pack voltage to charge up to 56.8v as my battery manufacturer recommends, but they were out of balance and I get alerts at 54.2v, I would do this:
charge to 54.2, let the warning trigger.
Stay at 54.2 until the cells are balanced and warning clears.
Add 0.5v, so set to 54.7. Let the cells balance (with or without an alarm)
Add 0.5v, 55.2, let them balance (regardless of alarm)
etc etc, until I finally got to 56.8v
At each stage, let the cells balance, and if there was an alarm, wait for it to clear.

 

[flpilgrim]

awesome, im finally down around 48 volts this morning and should have some sunshine today. ive changed my settings to be just a bit more conservative than yours and ill report back in.

I appreciate all the help, I feel like I'm getting close.

 

[400bird]

awesome, just so im crystal clear:

my cell target full charge with no load should be 3.4-3.5v

You are going to have a hard time balancing if you are only charging to 3.4 volts. You set balancing to start at 3.4 volts. While doing this balancing I'd run the cells up to 3.65 volts, that will give more time to balance.

unlike with lead acid batteries i don't need my charger setting 2 volts above my pack target to get there---ie---my absorb/float setting (for this one time at least) should be 54.4v.

If i set my chargers at 54.4 and im still getting cell overvolts, reduce my charger setting to whatever the battery reading happens to be when the event occurs and maintain that till my delta reaches 10 mv or below, then creep up to 54.4 in .5v increments.

54.4 is an average for 3.4 volts/cell.

 

[flpilgrim]

welp.
i dunno if i screwed something up last night when i was changing settings but it didn't balance at all today, charging or otherwise.

??

 

[42OhmsPA]

Are you able to change the balance delta to 1mV? 2 of my JBDs let me change to 1mV the other reverts back to 15mV.

Your settings look good otherwise, imo.
What voltage are you charging to and floating at?
Did you catch a screen shot of them above 3.4V?

 

[400bird]

Increase the cell full voltage from 3.45 to 3.65
Those settings are only going to balance from 3.39 to 3.45.
Balancing is going to be slow if you want the BMS to do it, like weeks. Give it all the time you can.

Alternatively you can manually discharge that one high cell.

 

[n2aws]

welp.
i dunno if i screwed something up last night when i was changing settings but it didn't balance at all today, charging or otherwise.

??
View attachment 188376View attachment 188377

Looking at the screenshot, it looks like something is still draying a little over 9 amps from the battery. You're going to want to let them sit, without any draw for this initial "get them back to heatlhy" top balancing. If it's at all possible, remove *all* load from them.

 

[SupraSPL]

In this case "cell full voltage" just resets your SOC display to 100% when that number gets hit (3.5V) during charging, but it will still charge above that and will continue to balance cells if needed.

There might be a little glitchiness with the buttons for "balance only when charging" so you might want to try that button, but you can turn that on and off without "writing to BMS" without resetting your SOC.

You're 80% state of charge voltage is set at 3.4 but that figure should normally be based on the resting voltage, so something along these lines (from Envision 305 datasheet)
100% 3.348V
90% 3.333V
80% 3.332V
70% 3.330V
60% 3.297V
50% 3.293V
40% 3.291V
30% 3.274V
20% 3.243V
10% 3.206V

 

[SupraSPL]

Are you able to change the balance delta to 1mV? 2 of my JBDs let me change to 1mV the other reverts back to 15mV.

I have seen some JDB BMS that allow .005V but most force .015V and above. I think the .005s I have seen are mostly on 4S BMS?

 

[flpilgrim]

What i mean is, the bms balancer never actually kicked on. i had the cells in range at 3.4 and above but the bms balancer didn't turn on.

I'm not sure it matters though, the cells actually had less variance when it wasn't active. Maybe i just need to stop worrying about it so much.

 

[42OhmsPA]

I have seen some JDB BMS that allow .005V but most force .015V and above. I think the .005s I have seen are mostly on 4S BMS?

The older 200A 7s-21s let me set to .001V, the newer 300a reverts back to .015V.
The 300a pack is the worst balanced but I didn't top balance...

 

[SupraSPL]

What i mean is, the bms balancer never actually kicked on. i had the cells in range at 3.4 and above but the bms balancer didn't turn on.

I'm not sure it matters though, the cells actually had less variance when it wasn't active. Maybe i just need to stop worrying about it so much.

Correct if it didn't kick on there must have been less than .015V difference between the cells. You might have to charge up to a higher voltage to reveal it but in some packs that can be an exercise in futility anyway so I don't bother going above 3.55V anymore.

 

[smr.ruby]

I personally don't like to drive my batteries that hard, after looking at some of the various charts online. So I deviate slightly.
I still have 56.8 for bulk and absorb, but 54.4 for float. This gives me a good mix of "charges up so that top balancing can occur" and "cells are at a decent resting voltage" during float.

I can't speak to Overkill but realize too that some BMSs recommend matching bulk/float/absorb supply settings to the battery because they will handle their own sequence after bulk. Not to agree or disagree with whether this is a good/bad idea but if I keep 56.8 applied constantly to my Seplos BMS after it hits 100% charge it will disconnect the charge mosfet and drop down to its own stable voltage for 100% charge no voltage applied to the pack itself. So in that case it isn't like a float voltage of 56.8 means a constantly applied voltage to the batteries - like it would for a traditional "dumb" lead acid battery. On my particular BMS even reducing the float to 54 wouldn't actually do anything since the battery is disconnected on the charge side from the inverter anyway.

I haven't seen this often discussed and but I'm curious if other BMS handle their own "post bulk" control differently.

 

[MisterSandals]

I can't speak to Overkill but realize too that some BMSs recommend matching bulk/float/absorb supply settings to the battery because they will handle their own sequence after bulk.

Just trying to interpret this for understanding.
Who is “they”?

As written it implies the BMSs but (most? Any?) BMSs don’t do this (handle sequences?).

 

[smr.ruby]

Just trying to interpret this for understanding.
Who is “they”?

As written it implies the BMSs but (most? Any?) BMSs don’t do this (handle sequences?).

I meant to say "they"=[some the BMSs] will handle bulk to float sequence regardless of supplied voltage and without an OV condition. This seems distinct from the general philosophy most hold that the "BMS should only be for last resort for protection." I say "some BMSs" because I don't know how many of them do or don't. Honestly my experience is more with the commercial high voltage systems, I know very little about how this sequence of operations is handled in the lower cost consumer BMSs (JK, JBD, etc) beyond observations my with Seplos . I do have another Trophy battery that has unknown BMS and it definitely has a different sequence of operation after it reaches full bulk or 100% SOC.

 

[MisterSandals]

some the BMSs] will handle bulk to float sequence regardless of supplied voltage and without an OV condition.

I wouldn’t say “handle”, that sounds like they are doing something.
A BMS is a switch, on or off. If a boundary condition is hit, it turns off.

 

[mkimid]

For long term use,
3.40V per cell is the enough voltage to charge LFP battery to full, the charging late is 1/10C rate.
3.45V per cell is the enough voltage to charge LFP battery to full, the charging late is 1/5C rate.
3.50V per cell is the enough voltage to charge LFP battery to full, the charging late is 1/2C rate.

Will show the voltage spike when charge LFP cell at 3.45V(1/5C rate), and some cell will hit the over-voltage protection voltage.

 

[flpilgrim]

I wouldn’t say “handle”, that sounds like they are doing something.
A BMS is a switch, on or off. If a boundary condition is hit, it turns off.

i thought we were talking about active balancing.

ive been watching the pack and even when my delta is out of range the balancer doesn't come on at all when that "bal only when charging" button is off.

for those of you who only actively balance occasionally, what's the trigger point for you?

 

[MisterSandals]

ive been watching the pack and even when my delta is out of range the balancer doesn't come on at all when that "bal only when charging" button is off.

I skimmed from the top and noticed that none of your screenshots show cells over 3.390V where your balancing is supposed to start.
Can you post one where you think it should be balancing? Maybe there is a clue someone can spot.

 

[flpilgrim]