Non-bms charging lifepo4 200ah 8s
- Thread starter bennymack
- Start date
With a UP1350W (have to google it, I can't seem to find the manufacturer official website) which supports LiFe chemistry.
I essentially intend it to be an "interactive" only battery. I'm not going to stick it in a trailer and forget about it while solar charges it and some load discharges it. I'm going to manually charge it up from time to time and discharge it under supervision.
I can't find a manual for the UP1350W, but the description specifies an input voltage of 11-30V. What I understand from your OP is that you are feeding it with a 10 volt power supply. That could cause problems.
It also lists Lipo/LiFe/LiIon/LiHv all together which would make me wonder if the manufacturer considers LiFe as different than LiFePO4. I can't find any charging voltages listed so it's hard to tell.
The bottom line is you need to have the proper charging equipment.
It also lists Lipo/LiFe/LiIon/LiHv all together which would make me wonder if the manufacturer considers LiFe as different than LiFePO4. I can't find any charging voltages listed so it's hard to tell.
The bottom line is you need to have the proper charging equipment.
The configured voltages for LiFe look good.
The 10.0v of the power supply was with regards to top charging all in parallel to 3.6v, on the top end it goes to 30v, 1200w.
What would be the proper charging equipment for this use case if money were no object?
Something like this: https://www.amazon.com/Kungber-Adju...ords=bench+power+supply&qid=1621027430&sr=8-3 with upgraded leads - 10AWG, ring terminals.
It's only 10 amps, but that should be fine for what you are doing. Just set it for 3.65 volts and wait for the current to drop to almost nothing for each cell or put them all in parallel and wait a few days. It is possible to charge faster in series, but you have to watch each cell voltage like a hawk.
Of course you can spend more if you want to.
It's only 10 amps, but that should be fine for what you are doing. Just set it for 3.65 volts and wait for the current to drop to almost nothing for each cell or put them all in parallel and wait a few days. It is possible to charge faster in series, but you have to watch each cell voltage like a hawk.
Of course you can spend more if you want to.
I thought of a way to slightly rephrase/clarify my question. Probably comes down to the details of how a typical BMS works and my lack of knowledge of them.
Say you're using solar power to charge a bank of 8 lifepo4 200ah packs in series. When the first cell hits 3.65v, the bms starts burning off current from that highest cell (at least this is my understanding). If this is how it works, is there like a top tier BMS that does the balancing in a more sophisticated manner? You're using hard won solar to charge your pack, you don't want to waste some amount of it as heat to balance the cells, right?
Say you're using solar power to charge a bank of 8 lifepo4 200ah packs in series. When the first cell hits 3.65v, the bms starts burning off current from that highest cell (at least this is my understanding). If this is how it works, is there like a top tier BMS that does the balancing in a more sophisticated manner? You're using hard won solar to charge your pack, you don't want to waste some amount of it as heat to balance the cells, right?
shortyjacobs
New Member
- Joined
- Mar 8, 2021
- Messages
- 21
That's not how a BMS works. The batteries are in series, the BMS can't just turn a cell "off" as that would mean the remaining pack voltage would be lower. When any given cell hits max volts, the BMS shuts the whole damn battery down. When any given cell hits minimum volts, the BMS shuts everything down. In the meantime, the BMS is seeping current between cells to try and balance them, but it's not working too hard at it. The voltage/SOC curve is so damn flat that highly unbalanced cells can still have extremely close voltages.
That's why you top balance. Get them all to the same max, and the chemistry is so nice and stable that they will stay that way generally forever, with the itty bit of balancing provided by the BMS.
Also, we use BMS HVC and LVC (high voltage cutoff and low voltage cutoff) to super-max charge and super-discharge, but generally speaking we don't like to rely on the BMS for everyday charging/discharging duties - it's a last resort if something else fails and a cell starts to overcharge or undercharge. Once you have the cells top-balanced and you know the capacity by doing a good HVC to LVC discharge/capacity test, you set your charger voltages so that you don't actually hit the high voltage cutoff any more....again, flat voltage/SOC curves except at extremes mean you don't need that last millivolt to actually have full capacity.
@bennymack - If you are not going to invest in a BMS and use top-balancing techniques, your casual use might be better served doing a bottom balance, and simply cutting off charge when a "pack level" voltage of any ONE cell reaches the trigger point.
Will discusses it here:
While Will is right that top balance, once achieved, is not much of an issue down the road with cells that are perfectly matched in both capacity and internal resistance.
But you don't know that with yours. And you don't have any balancers that can handle balancing in a timely manner.
So - since solar is part of your plan, you might simply bottom balance, and once achieved, use your controller's HVD, or high voltage disconnect when a pack / bank level voltage has been reached. This is determined initially by YOU monitoring charge after a bottom balance, taking note of the first cell to reach 3.6v upon recharge, and then measuring overall bank voltage as the HVD trigger to stop. This overall bank voltage is now what you set your HVD to in your controller to disconnect at.
But please - invest in a GOOD quality multimeter. Do you absolutely trust your fancy screen? What if it is mis-calibrated? You got a lot on the line, so sorry - don't be penny-wise and pound-foolish with those cells, otherwise this is all just academic arguing with a reddit cross-post, and not anything practical.
Achieving initial bottom balance in practice with cells that already have a good state of charge in them can be sped up by doing a bulk discharge like normal until you get near the bottom - but keep an eagle-eye on individual cells during that bulk discharge! Most people don't let any individual cell go lower than 3.0v. After you get near to where any individual cell is close to 3v, disconnect all the cells, and individually start discharging each one to get to the near 3v low voltage. Sometimes 6V rv headlamps are used in a pinch as a load. Sometimes resistors. Once achieved, attach in series, and perform your initial watch for any individual cell reaching 3.6v first, and make your HVD settings.
Hard core bottom-balancers may end up letting cells rest for a day after doing the initial discharge, since cells tend to recover a bit, and discharging again until it reaches 3v individually. And then charging and watching to determine their HVD based upon the first cell to reach 3.6v first. But unless you are in an EV type of application, just doing the bottom discharge once may suffice.
Sorry about being so repetitious - just trying not have you damage cells by taking them lower than they should go if you do a bottom balance.
Choose a strategy and let us know how it works out.
Will discusses it here:
While Will is right that top balance, once achieved, is not much of an issue down the road with cells that are perfectly matched in both capacity and internal resistance.
But you don't know that with yours. And you don't have any balancers that can handle balancing in a timely manner.
So - since solar is part of your plan, you might simply bottom balance, and once achieved, use your controller's HVD, or high voltage disconnect when a pack / bank level voltage has been reached. This is determined initially by YOU monitoring charge after a bottom balance, taking note of the first cell to reach 3.6v upon recharge, and then measuring overall bank voltage as the HVD trigger to stop. This overall bank voltage is now what you set your HVD to in your controller to disconnect at.
But please - invest in a GOOD quality multimeter. Do you absolutely trust your fancy screen? What if it is mis-calibrated? You got a lot on the line, so sorry - don't be penny-wise and pound-foolish with those cells, otherwise this is all just academic arguing with a reddit cross-post, and not anything practical.
Achieving initial bottom balance in practice with cells that already have a good state of charge in them can be sped up by doing a bulk discharge like normal until you get near the bottom - but keep an eagle-eye on individual cells during that bulk discharge! Most people don't let any individual cell go lower than 3.0v. After you get near to where any individual cell is close to 3v, disconnect all the cells, and individually start discharging each one to get to the near 3v low voltage. Sometimes 6V rv headlamps are used in a pinch as a load. Sometimes resistors. Once achieved, attach in series, and perform your initial watch for any individual cell reaching 3.6v first, and make your HVD settings.
Hard core bottom-balancers may end up letting cells rest for a day after doing the initial discharge, since cells tend to recover a bit, and discharging again until it reaches 3v individually. And then charging and watching to determine their HVD based upon the first cell to reach 3.6v first. But unless you are in an EV type of application, just doing the bottom discharge once may suffice.
Sorry about being so repetitious - just trying not have you damage cells by taking them lower than they should go if you do a bottom balance.
Choose a strategy and let us know how it works out.
Last edited:
The first discharge/charge cycle didn't go very well. When I started the balance charge, one cell jumped up in voltage way before the others causing the charge to slow to a crawl. I figured it was due to the resistance of the crappy bus bars that came with the packs. So I ordered two 36"x1"x1/8" copper flat bars and made my own. Definitely improved the situation resistance wise. Now I'm back to top balancing them and will try another discharge/ charge cycle.
Ow, that's going to take a long time, so I'll cut to the chase with the strategy based on what you have and a warning in this situation.
1) Because your charge current is so low, you'll be fully charged by the time the bank reaches 3.45v as measured at the battery terminals. Do NOT try to take them up to 3.65v. If you were to perform a capacity test stopping at 3.45v due to the puny charge current, you'll find you have already reached full capacity. Trying to take these cells higher in voltage to 3.65v would be considered OVERCHARGE.
2) Depending on the existing SOC state of the cells when you began, this might take a looooong time. And this is what hurts. Taking too long to do the job. So you aren't doing the cells any favors, but don't be surprised if your first capacity test comes in under what you expect.
This is a trend I'm seeing elsewhere, and in fact even in Will's top balance video that most people miss with his huge string of cells laid out in parallel across the bench: (He gives a major hint at about 4:23 into the vid, but who's paying attention?)
He is already starting from cells that have been bulk-charged to a high SOC state, and is just doing a quick top-balance top-off!
However, like most who get these cells off the boat at 30% SOC or less in shipping, and rush to the shed to charge, do so with puny charge current taking FOREVER to charge, and in fact may not notice that their cells initially seem to stall well before 3.65v. (because they are already fully charged!) If they leave them on long enough and DO reach 3.65v, that is from overcharge secondary reactions!! Ie, no current flowing, but voltage is rising!
All due to insufficient charge current. You are "trickle-charging" these cells at your rate (anything much less than 0.2C) and as such please simply STOP AT 3.45V, nevermind the bit of harm taking forever to do so.
The devil's in the details. Whatever you do, please, just stop at 3.45v due to starting out with your tiny charge current on relatively low SOC cells.
1) Because your charge current is so low, you'll be fully charged by the time the bank reaches 3.45v as measured at the battery terminals. Do NOT try to take them up to 3.65v. If you were to perform a capacity test stopping at 3.45v due to the puny charge current, you'll find you have already reached full capacity. Trying to take these cells higher in voltage to 3.65v would be considered OVERCHARGE.
2) Depending on the existing SOC state of the cells when you began, this might take a looooong time. And this is what hurts. Taking too long to do the job. So you aren't doing the cells any favors, but don't be surprised if your first capacity test comes in under what you expect.
This is a trend I'm seeing elsewhere, and in fact even in Will's top balance video that most people miss with his huge string of cells laid out in parallel across the bench: (He gives a major hint at about 4:23 into the vid, but who's paying attention?)
He is already starting from cells that have been bulk-charged to a high SOC state, and is just doing a quick top-balance top-off!
However, like most who get these cells off the boat at 30% SOC or less in shipping, and rush to the shed to charge, do so with puny charge current taking FOREVER to charge, and in fact may not notice that their cells initially seem to stall well before 3.65v. (because they are already fully charged!) If they leave them on long enough and DO reach 3.65v, that is from overcharge secondary reactions!! Ie, no current flowing, but voltage is rising!
All due to insufficient charge current. You are "trickle-charging" these cells at your rate (anything much less than 0.2C) and as such please simply STOP AT 3.45V, nevermind the bit of harm taking forever to do so.
The devil's in the details. Whatever you do, please, just stop at 3.45v due to starting out with your tiny charge current on relatively low SOC cells.
Last edited:
I should have mentioned that this is the end of the whole top-balance process. I've charged them all up in pairs then only connected them all in parallel to finish it off. I shut it down shortly after the post. They are all within 0.01v so I consider that a success. Now I will discharge them and charge and see how that goes.
Quick update on this project. I decided to go with a power labs pl8 charger since it can do balance only charging. I charge with the banana leads until one cell is nearly full then switch to "accurate" charge with banana output disabled. I believe the cells are staying well balanced with my massive bus bars because the accurate charge only puts 1.5ah to 2ah in to finish it off.
I picked up a drok hall current sensor to get an idea of the amount it's discharged. Unfun fact, the conversion efficiency of lifepo4 to chevy volt drive batter is about 65%
I also went ahead and made a better home than the flimsy milk crate I had been using. Goal was to allow less movement of the cells during transport so as to not stress terminals. I think I accomplished this mission! I used two sheets of 0.75" foam all around and this provides just enough to make them snug in a husky 20" tool box. When the top is latched l, it's very rigid. The foam strip on top is to add a bit more rigidity as it gets slightly compressed when latching the lid.
I picked up a drok hall current sensor to get an idea of the amount it's discharged. Unfun fact, the conversion efficiency of lifepo4 to chevy volt drive batter is about 65%
I also went ahead and made a better home than the flimsy milk crate I had been using. Goal was to allow less movement of the cells during transport so as to not stress terminals. I think I accomplished this mission! I used two sheets of 0.75" foam all around and this provides just enough to make them snug in a husky 20" tool box. When the top is latched l, it's very rigid. The foam strip on top is to add a bit more rigidity as it gets slightly compressed when latching the lid.
