I know a lot of the threads here are based on electric vehicles but I'm sure some of you have the knowledge to help me out. I've been researching for over a year on building a battery pack and the uses I need it for in my van conversion. I saw great reviews for a company and decided to pull the plug. $4k later (US dollar) I have a serious battery bank capable of my workstation needs on the go. I'm having an issue as once I pass the 90% sate of charge the cells start to get out of whack. One cell would be 70% 3.33 or something and the other considerably larger. It only seems to be a variance between cells 1-2 and 3-4 which is strange to me. Cells 1 and 4 are the largest in gap with a 0.3+ variance.. I tried changing bus bars and pumping up the copper between cells 3 & 4 without luck. Did i just get bad cells? I've attached some photos of the setup, I hope someone has insight on this as I'm losing capacity and using top capacity I don't want to all in the same pack!01
LifePO4 bank issues
- Thread starter Leeford
- Start date
Apparently your pictures did not actually attach; you'll have to reupload them in the attachments tab on your post.
It may help if you give complete and exact details of the pack composition and construction, and it's usage scenario.
Without knowing what kind of cells, what chemistry, can't say much about the SOC levels you're referring to, as they don't sound like they match any of the ones I'm familiar with.
But if it's LiFePO4, then at 3.3v they're almost full. 3.2v is in the main part of the discharge curve, where most of the capacity is.
Your BMS should be able to balance the cells for you, usually this is during charging once the pack has reached "full". Most LiFePO4 packs with BMS balance at 3.6-3.65v. There's extremely little capacity between there and 3.3-3.4v, probably not even a percent, so the voltage rapidly drops once a load is applied.
It's possible your cells are not well-matched (this is common, unfortunately), so they dont' all have the same capacity. It's also possible one or more of them is damaged, or failing. You'd have to capacity test and test the internal resistance (Ri) of each one individually to find out for certain, but the simplest thing to do is just to fully charge the pack and leave it on the charger long enough for the BMS to finish balancing the cells.
That can take hours to days to weeks depending on the size (capacity) of the cells, the amount of capacity imbalance, and the amount of balancing current the BMS is designed to shunt/bleed.
Once the pack is rebalanced, then just make sure you always recharge it and leave it long enough to finish the balance cycle, and it'll stay that way if there's no problems with the cells, and they are reasonably close in capacity/performance to each other.
If there is no BMS, then you'd have to manually charge each cell individually to full, or do the normal bulk charge until one cell reaches full, and manually drain down that cell till it matches others, then repeat that cycle until they're all the same (this is what the BMS would do).
It may help if you give complete and exact details of the pack composition and construction, and it's usage scenario.
Without knowing what kind of cells, what chemistry, can't say much about the SOC levels you're referring to, as they don't sound like they match any of the ones I'm familiar with.
But if it's LiFePO4, then at 3.3v they're almost full. 3.2v is in the main part of the discharge curve, where most of the capacity is.
Your BMS should be able to balance the cells for you, usually this is during charging once the pack has reached "full". Most LiFePO4 packs with BMS balance at 3.6-3.65v. There's extremely little capacity between there and 3.3-3.4v, probably not even a percent, so the voltage rapidly drops once a load is applied.
It's possible your cells are not well-matched (this is common, unfortunately), so they dont' all have the same capacity. It's also possible one or more of them is damaged, or failing. You'd have to capacity test and test the internal resistance (Ri) of each one individually to find out for certain, but the simplest thing to do is just to fully charge the pack and leave it on the charger long enough for the BMS to finish balancing the cells.
That can take hours to days to weeks depending on the size (capacity) of the cells, the amount of capacity imbalance, and the amount of balancing current the BMS is designed to shunt/bleed.
Once the pack is rebalanced, then just make sure you always recharge it and leave it long enough to finish the balance cycle, and it'll stay that way if there's no problems with the cells, and they are reasonably close in capacity/performance to each other.
If there is no BMS, then you'd have to manually charge each cell individually to full, or do the normal bulk charge until one cell reaches full, and manually drain down that cell till it matches others, then repeat that cycle until they're all the same (this is what the BMS would do).
John in CR
100 TW
What's this 800ah you mention in the other thread title? How many actual cells do you have. BTW, you can't really look at voltage of LiFePo4 to know the status of charge unless it's totally full and recently charged or totally empty. Also, under load 3.33V is certainly far more than 70% full.
Like AW stated, you first need to fully balance charge all cells. With Lifepo4 I like to put them all in parallel and fully charge them and let them rest in parallel for at least a day, and then again hit them with the single cell charger.
After a proper balance charge, if you want to compare cell health, you need to do a full discharge and see if any cells are dropping below 3.0V much earlier than the others. A pack is only as good as its weakest cell, or weakest group of parallel cells. Lastly, assuming the 800ah is accurate you must have more than 4 cells, and all cells of equal level must be connected in parallel, and then each parallel group you can then treat as if they are single cells and connect the 4 groups in series.
If you don't already have a quality BMS, then I strongly suggest buying a good RC charger with which you can do single cell charging as well as bulk or balance charging along with useful battery and/or cell testing. My iCharger has been an invaluable tool for me in this hobby for 8 years. I've built many packs and tested thousands of cells. I've yet to use a BMS other than one that came attached to a pack I purchased 11 years ago, and quickly killed the pack due to a BMS failure. I've proven to be a 100% reliable and conservative BMS for my packs since then.
Like AW stated, you first need to fully balance charge all cells. With Lifepo4 I like to put them all in parallel and fully charge them and let them rest in parallel for at least a day, and then again hit them with the single cell charger.
After a proper balance charge, if you want to compare cell health, you need to do a full discharge and see if any cells are dropping below 3.0V much earlier than the others. A pack is only as good as its weakest cell, or weakest group of parallel cells. Lastly, assuming the 800ah is accurate you must have more than 4 cells, and all cells of equal level must be connected in parallel, and then each parallel group you can then treat as if they are single cells and connect the 4 groups in series.
If you don't already have a quality BMS, then I strongly suggest buying a good RC charger with which you can do single cell charging as well as bulk or balance charging along with useful battery and/or cell testing. My iCharger has been an invaluable tool for me in this hobby for 8 years. I've built many packs and tested thousands of cells. I've yet to use a BMS other than one that came attached to a pack I purchased 11 years ago, and quickly killed the pack due to a BMS failure. I've proven to be a 100% reliable and conservative BMS for my packs since then.
I see you've started a whole new thread instead of replying to those trying to help you.
Hopefully the mods will move that into this existing thread.
(it's quoted below in case you just delete it instead)
And then hopefully you will reply to those already trying to help you here, instead of continuing to make new threads for the same thing, so you can get the help you need.
Until you reply to us with the necessary information, we can only make guesses.
Hopefully the mods will move that into this existing thread.
(it's quoted below in case you just delete it instead)
And then hopefully you will reply to those already trying to help you here, instead of continuing to make new threads for the same thing, so you can get the help you need.
Until you reply to us with the necessary information, we can only make guesses.
