Why the need to top balance LiFePo4 batteries? Does the BMS not take care of that by doing a balance charge? Just trying to understand before I build my first 24v battery. Most of my previous experience has been with 18650 battery packs and I have a charger that balances them while charging. Don't know if LiFePo4 works the same.
Why top balance?
- Thread starter JTrigger
- Start date
Ampster
Renewable Energy Hobbyist
It is a personal preference. I have top balanced some packs and in other cases I have just put them in parallel for a week and let them settle to some semblance of equal voltage. I am going to be assembling a large pack and will top balance it because it will be 30 kWhrs and the one Watt balancing current that my BMS can deliver would take forever compared to a smaller pack. I generally turn off the balancing function of my BMS and really only use the BMS for monitoring my pack. Since I watch cell voltage differences I want them to be minimal at the top since it is likely that my pack will spend more time near the top than near the bottom. If one of my cell degrades and becomes a runner it will be easier to identify if I know I have the top balance as a baseline. Stay tuned, and you will hear lots of other opinions.
Some on this forum have damaged new cells trying to top balance and I do not see the need.
Greater risk than reward IMHO.
I bought 16 new cells that arrived well balanced and put a Chargery BMS 16t on it. I set up the Chargery to balance only when charging. I simply haven’t had any problems and will proceed this way again with my next new battery.
YMMV.
Greater risk than reward IMHO.
I bought 16 new cells that arrived well balanced and put a Chargery BMS 16t on it. I set up the Chargery to balance only when charging. I simply haven’t had any problems and will proceed this way again with my next new battery.
YMMV.
John Frum
Tell me your problems
- Joined
- Nov 30, 2019
- Messages
- 15,233
Lifepo4 batteries have such a flat discharge curve that its very hard to tell if the cells are balanced.
The cells generally arrive from the courier ~50 SOC.
Often but not always the cells arrive balanced.
So to make sure the cells are truly unified, folks balance them above the flat part of the curve.
When my 8 280Ah cells arrived the voltages varied +/- 4mV which sounds like OK balance. When I charged each cell, they varied by about 20Ah in charge needed to fill. After top balancing, the voltages in mid range were +/- 1mV. And they reach full charge within <1Ah of each other. Without top balancing, I would have left 19Ah of capacity on the table for the pack. Of course I could have just turned on BMS balancing and let the BMS slowly balance the cells over time. But without top balancing followed by capacity testing I would not know if I really had matched cells nor what my actual capacity was.
Now after top balancing the balance is so good I'm leaving the BMS balancing off just to see how long it goes without needing any more balancing. Balancing just burns power. Everything I've read says that with a good matched balanced pack, you should not need to balance more often then every 6 months, or longer. So I'll see how that goes... To me the top balancing was worth the effort and no I didn't damage any cells.
Now after top balancing the balance is so good I'm leaving the BMS balancing off just to see how long it goes without needing any more balancing. Balancing just burns power. Everything I've read says that with a good matched balanced pack, you should not need to balance more often then every 6 months, or longer. So I'll see how that goes... To me the top balancing was worth the effort and no I didn't damage any cells.
I am new to this but I am experiencing exactly why you need to top balance at least with LFP. Similar to Airtime, my 16 120 Ahr LFPs arrived at about 50% SOC and all within 4-5 mV. Built a 16S pack with a Daly Smart BMS. Hooked all up to my 6kW Growatt SPF600TDVM Hybrid inverter, and all was good. Right up until the point that it wasn’t. Took me a while to figure it out as I was not understanding what I was seeing (remember it was literally the first battery pack I have ever built). The Growatt was throwing a high battery voltage code and shutting down during charging from the built in MPPT charge controller. Long story short, it turns out the Daly BMS is doing its thing and shutting down the pack as one cell was starting to go exponential on voltage. This shutting down caused the battery voltage to go high during the transient.
It would end nice here if that was the end and I learned my lesson. Unfortunately it looks like I am also learning that the Growatt does not protect itself adequately from the BMS cutting off the battery several times in a row. I blew two boards in the inverter TWICE now. I’m assuming I’m at fault and there is not something else wrong. Still working the issue. Really ticked though as the BMS was working exactly as it should, so that should not have caused a problem with a well designed inverter.
So my advice is to ALWAYS top balance!
It would end nice here if that was the end and I learned my lesson. Unfortunately it looks like I am also learning that the Growatt does not protect itself adequately from the BMS cutting off the battery several times in a row. I blew two boards in the inverter TWICE now. I’m assuming I’m at fault and there is not something else wrong. Still working the issue. Really ticked though as the BMS was working exactly as it should, so that should not have caused a problem with a well designed inverter.
So my advice is to ALWAYS top balance!
ArthurEld
Solar Wizard
You might have one cell that is moving away from the others too fast for the balancer to keep up.
That's too bad about your inverter. I heard it is hard on the inverter when the BMS disconnects the charging.
DIY involves expensive mistakes.
Looks like we both need to learn how to turn off the inverter when the BMS finds a problem.
edit: I reread your post. I was assuming you had been using this setup for a while before this problem happened.
Last edited:
RCinFLA
Solar Wizard
- Joined
- Jun 21, 2020
- Messages
- 3,565
The majority of BMS's are just resistor dump balancer. The resistor dump balancers typically only pull 100mA to 200 mA during active balancing. In addition they only become active when a cell is greater then about 3.4v, near full charge.
This means if there is a moderate difference on SOC of cells and you are pumping a significant charge current into series connected cells the highest SOC cell will hit its high voltage limit before some of the other cells are charged up.
The best thing that happens is the BMS cell high voltage protection shuts down the BMS cutout switch preventing overcharging, But it still leaves you in a situation of a disconnected battery pack, Eventually the high SOC cell will bleed off by balancer and BMS will reconnect. The same process will continue to repeat until cells eventually get balanced. Not a great way to achieve balance. Without a BMS you can end up with an overcharged, bloated cell.
If you charge at a very low rate, below the resistor dump rate current, the pack will eventually balance out. At such a low charge rate it could take weeks to balance out large AH capacity cells.
Resistor dump balancers are really only for balancing out variance in cell internal leakage which is very small. If you have some capacity variation between cells and run battery pack over a wide SOC range the resistor dump balancers will not be able to keep up with balancing.
This means if there is a moderate difference on SOC of cells and you are pumping a significant charge current into series connected cells the highest SOC cell will hit its high voltage limit before some of the other cells are charged up.
The best thing that happens is the BMS cell high voltage protection shuts down the BMS cutout switch preventing overcharging, But it still leaves you in a situation of a disconnected battery pack, Eventually the high SOC cell will bleed off by balancer and BMS will reconnect. The same process will continue to repeat until cells eventually get balanced. Not a great way to achieve balance. Without a BMS you can end up with an overcharged, bloated cell.
If you charge at a very low rate, below the resistor dump rate current, the pack will eventually balance out. At such a low charge rate it could take weeks to balance out large AH capacity cells.
Resistor dump balancers are really only for balancing out variance in cell internal leakage which is very small. If you have some capacity variation between cells and run battery pack over a wide SOC range the resistor dump balancers will not be able to keep up with balancing.
Last edited:
RCinFLA That’s exactly where I’m headed. Bought an 80V 5-A power supply and I’m going to charge very slowly as I don’t want t disassemble the pack. And you are right on. The problem with the BMS is only that it can’t keep up when the voltage goes exponential on the cells that “top off” first. From what I’m seeing, the Daly BBS only balances at up to 35 mA. That is probably fine once your pack is balanced, but it’s not going to get the job done if you are not initially balanced.
Capt Bill
Sailing Options
I also believe Top Balancing is important. ... Re: (this aspect of challenge/ copied from post above:
" ... Built a 16S pack with a Daly Smart BMS. Hooked all up to my 6kW Growatt SPF600TDVM Hybrid inverter, ...
Unfortunately it looks like I am also learning that the Growatt does not protect itself adequately from the BMS cutting off the battery several times in a row. I blew two boards in the inverter TWICE now. ... The Growatt was throwing a high battery voltage code and shutting down during charging from the built in MPPT charge controller. Long story short, it turns out the Daly BMS is doing its thing and shutting down the pack as one cell was starting to go exponential on voltage. This shutting down caused the battery voltage to go high during the transient. ... I am also learning that the Growatt does not protect itself adequately from the BMS cutting off the battery several times in a row. I blew two boards in the inverter TWICE now." ...
This might be an option to consider (not sure); I like my Chargery BMS8T or 16T (w LCD display into lots of configuration options); PLUS hooking up the Chargery BMS's "high cell volt trigger" to cut off my Solar INPUTs to my 3x MPP LV2424s (plus cut off grid IN to Lv2424 via SSRs / because of my LV2424's grid charging options)! ... I like that set up option ... Because: ... I think that route lessens or Prevents the Higher Amp Battery OFF ON OFF ON cycling, .... that might happen when no one is there noticing that kind of stressful equipment event.
Also: ... For better cell voltage balance @ top end of charge: I configured my Chargery BMS8T's Balance setting to start at 3.2 volts on Charge cycles only. ...PLUS lowered my 29.0v or 28.8v absorption /bull (max) charge voltage to 28.4v / with float set at 27.3 (or 27.6v when I want more MPPT charge amps near top end); as that seems to charge my Eve 280Ah LiFePO4 cells to full without triggering any high cell volt spike events.
" ... Built a 16S pack with a Daly Smart BMS. Hooked all up to my 6kW Growatt SPF600TDVM Hybrid inverter, ...
Unfortunately it looks like I am also learning that the Growatt does not protect itself adequately from the BMS cutting off the battery several times in a row. I blew two boards in the inverter TWICE now. ... The Growatt was throwing a high battery voltage code and shutting down during charging from the built in MPPT charge controller. Long story short, it turns out the Daly BMS is doing its thing and shutting down the pack as one cell was starting to go exponential on voltage. This shutting down caused the battery voltage to go high during the transient. ... I am also learning that the Growatt does not protect itself adequately from the BMS cutting off the battery several times in a row. I blew two boards in the inverter TWICE now." ...
This might be an option to consider (not sure); I like my Chargery BMS8T or 16T (w LCD display into lots of configuration options); PLUS hooking up the Chargery BMS's "high cell volt trigger" to cut off my Solar INPUTs to my 3x MPP LV2424s (plus cut off grid IN to Lv2424 via SSRs / because of my LV2424's grid charging options)! ... I like that set up option ... Because: ... I think that route lessens or Prevents the Higher Amp Battery OFF ON OFF ON cycling, .... that might happen when no one is there noticing that kind of stressful equipment event.
Also: ... For better cell voltage balance @ top end of charge: I configured my Chargery BMS8T's Balance setting to start at 3.2 volts on Charge cycles only. ...PLUS lowered my 29.0v or 28.8v absorption /bull (max) charge voltage to 28.4v / with float set at 27.3 (or 27.6v when I want more MPPT charge amps near top end); as that seems to charge my Eve 280Ah LiFePO4 cells to full without triggering any high cell volt spike events.
Last edited:
Agree that a set up where the BMS would also cut off the solar input would have saved my Growatt from it’s own limitations. My system is running nicely now. Going on 3 months of uninterrupted ops! What works for me is staying well away from the min and max voltages for the 48V battery. That will greatly improve battery life and it gets me through almost every day with no utility power required (2 cloudy days in a row will send me to utility). Can’t help myself though, and I ended up ordering 16 Lishen 202 ahr LFP cells to build another battery. They will definitely get top balanced Before going into service!
