Top balance
- Thread starter Dyotat100
- Start date
Like charging lead-acid to 14.4V sees them settle in the 12.8-12.9V range in 24 hours - LF is similar. Their voltage should settle. I'm surprised they didn't settle more, but folks see different things with different cells.
Is your goal to test capacity?
Is your goal to test capacity?
I believe that after reaching target voltage of 3.65V or 3.64V, you should disconnect the charger but leave the cells connected in parallel while they ‘settle’
if you had left your busbars connected while you let the cells settle overnight, they would be better balanced.
If it were me, I’d rebalance to 3.64V (should be relatively quick) and then leave the busbars connected at least overnight so the cells can continue to balance at low current levels.
An 0.5mV difference will still drive more than 3 amps through a busbar representing 0.15 mOhms of resistance (the point being that what’s going on between cells is well below your ability to detect or measure it).
The longer you leave the cells connected together in parallel, the better balanced they will be.
Please report back once you do.
I’m kind of in the same stage as you (balancing a 16S 280Ah pack) and I believe (but am not yet certain) that a fully-balanced pack will not vary much if bus bars are removed after all ‘settling’ is done.
If the cells settle all the way down into the end of the knee around 3.4V, you won’t be able to ‘see’ (measure) any additional settling or stray anyway, so it really comes down to self-discharge rates.
If cells only settle to 3.5V and self-discharge is small enough to hold that voltage/charge overnight, a perfectly-balanced pack should settle to 3.5V connected in parallel and should stray very little overnight with the parallel busbars removed.
If self-discharge is too great to hold 3.5V overnight, it’ll come down to how evenly-matched the self-discharge rates of the different cells are.
One idea I might try is to go through several top-off, settle cycles tracking the settling rate. As long as the parallel-connected cells settle at a higher voltage after the same settling time, you know they have taken more charge. Once cells settle to the same voltage after the same settling time, you know that the top-off charge you are adding when you top-off is exactly enough to offset charge lost through self-discharge over the settling time.
If all cells lost the same amount of charge through self-discharge, you should be perfectly, perfectly matched.
And if different cells lost different amounts of charge over the settling period, mismatches will build up over time anyway so you’re about as perfectly balanced as possible with the cells you have...
John Frum
Tell me your problems
- Joined
- Nov 30, 2019
- Messages
- 15,233
The delta is only 41 mv.
Personally I would put them in series and see how they hold together during a cycle to your target cutoff voltage.
If it was me I would discharge them to 3.0 volts at ~0.2c and see what the delta is at the bottom.
Then charge them up to 3.65 volts per cell at ~0.2c and record that delta at the top as well.
If the low delta is acceptable and the high delta is acceptable and the capacity is acceptable then the cells are ready for service.
What BMS are you using?
Personally I would put them in series and see how they hold together during a cycle to your target cutoff voltage.
If it was me I would discharge them to 3.0 volts at ~0.2c and see what the delta is at the bottom.
Then charge them up to 3.65 volts per cell at ~0.2c and record that delta at the top as well.
If the low delta is acceptable and the high delta is acceptable and the capacity is acceptable then the cells are ready for service.
What BMS are you using?
Chargery with 300 amp DCC
John Frum
Tell me your problems
- Joined
- Nov 30, 2019
- Messages
- 15,233
I would disable balancing for the first cycle just to see how the cells hold together.
After the first cycle if all is well then enable it to balance on during the charge cycle and only above full resting voltage(~3.4 volts).
Ok thanks. I'm a ways away from that. I have to make my bus bars still. It's up to 3.639 right now and I'm watching it.
John Frum
Tell me your problems
- Joined
- Nov 30, 2019
- Messages
- 15,233
560 x .02 = 112 amps
So 120 is very close to perfect.
Bob B
Emperor Of Solar
- Joined
- Sep 21, 2019
- Messages
- 8,645
The Chargery has very good balance capabilities .... It should be very capable of keeping the cells in balance at their current voltage levels.
I'd just go for it and put the pack together.
With my 2P4S pack using 190 ah cells .... I didn't top balance at all except to do a VERY slow charge at the upper knee and let the Chargery balance them. Mine have stayed in balance very well since doing that initial slow top charge.
it’s actually only 31mV (3.589 to 6.200) which I agree is pretty good (up that far into the knee).
But I doubt settling is ‘done’ with voltages close to 3.6V and in my experience, settling will continue to at least 3.5V if not lower.
Once cells have settled in parallel to a voltage close to 3.5V, the major transients are done and if the cells continue to hold voltages close to that level once disconnected, they are top-balanced.
How well bottom-balanced the cells are is largely immaterial.
The real test to know how well-balanced your pack is (and what effective capacity it will deliver you) is to:
1/ Connect top-balanced cells into an 8S battery.
2/ Discharge to your desired cut-off voltage (monitoring cells on the way if you don’t have a BMS).
3a/ if you reach your desired cut-off voltage and no cells have reached LVD of 2.1V (or whatever voltage your BMS will use to protect cells), great - this is the capacity your battery will deliver.
3b/ if one of your cells reaches LVD before battery has reached desired cut-off, that weak cell will determine usable battery capacity (and you will never reach your desired cut-off voltage).
4/ In either case, now charge to full pack is series using an 8S charger (or voltage source set to desired charge voltage), monitoring cells on the way in case you have no BMS.
5a/ If battery voltage reaches target (of ~28v or whatever) without any cells reaching HVD (of 3.65V or whatever), congrats - this should mean that all your cells should have reached ~3.55V around the same time and your battery is perfectly too-balanced.
5b/ one cell hits HVD of 3.65V before battery has reached fully-charged voltage of ~28V, top-balance for your pack leaves room for improvement.
Note that this test is much safer with BMS connected. If you proceed without BMS, cells need to be monitored very closely to assure no cell is discharged below LVD or charged above HVD (which means it is much safer to proceed at lower current levels, especially when any cell voltages are above the upper knee or below the lower knee).
I agree, if you’ve got a BMS with good balance capability, the importance of a close-to-perfect top balance reduces greatly.
My bare-bones BMS had a passive balance capability of only 76mA per channel active only above 3.55V - it’s next to useless at achieving any balancing when charging at 10A, let alone at rated maximum charge current of 100A.
What are the balance capabilities of the Chargery BMS? (Current/channel, active or passive, Turn-on voltage)
Bob B
Emperor Of Solar
- Joined
- Sep 21, 2019
- Messages
- 8,645
The Chargery will balance at 1.2A ... passive balance. They are working on an active balance design.
It can be set to charge only while above a specified voltage level..... or even in all modes. The settings are extremely flexible. I initially just set mine to balance above 3.4V and then very slowly .... only 2-3 amps .... crept on on full charge after reaching the knee.
No, sorry, when I saw 8 cells, I assumed a 24V battery at 8S.
You can do all of what I posted earlier at 4S / 12V and you can even do one better by best-matching each 4S pack.
I am making 2 8S batteries from 16 cells and can explain the steps I used to identify the 8 cells which were best-matched for each battery. Before doing that, help me to understand:
1/ what charger are you using to top-balance and what is it’s maximum voltage and current rating?
2/ are you using a BMS to charge and/or discharge or are you comfortable monitoring closely enough during charge and/or discharge to proceed unprotected?
3/ how are you planning to discharge at 12V and do you have any way to discharge at 24V?
