Sinking in Regret

[WilliamFR]

So, i'm having a bit of bad luck with top balancing my cells so far. I have 16x Docan Power 280AH to make a 48v bank. They call came at 3.29V, so i batched 4 together in series and charged at 14.60v on my dc power supply (the one Will recommends, here. At first, it was charging rather slow, and with all the videos I've watched, pretty normal. At this point, the red light is on for C.C. on the charger. After a bit, can't say exactly how long it was, each cell was hitting about 3.35V. I turn my back, and a few mins later i go check voltages again. It changed to green C.V. light, and one of the cells has jumped to 4.4v. The cell in question had the power lead on it while charging..just to note. I burn off some of the charge with a halogen light, and swelling ensues the next day. I'm not really sure why one cell, within a blink of an eye has jumped sky high.

At this point, i'm a little freaked out. I decide to take four more, and try again..surely the first was a dud. The same thing happened, and it happens SO fast that i can barely catch it in time. It's random..it can charge for hours and then all of a sudden jump through the roof. Luckily i caught this one much sooner.

Now i'm terrified. I decide to parallel them and slow charge. Within 24 hours i want to say i gained a tenth or two of volts on all cells. That might literally take a month. So i take four more cells..i think..no damn way this is going to happen again right? and if it does, i'm going to watch it like a hawk. Sure a damn nough it happened AGAIN. This time on the negative lead cell...so i guess that has no relevance. They were all around 3.35, then bam one cell went to 3.72. So i cut the power and let it chill for a few minutes and the high cell dropped to 3.52.

Either way, i'm not sure what the hell is going on here. Am i stupid? Am i doing something wrong? Clearly, i don't understand charging cell behaviors. Help me out please before i burn all these suckers up. I feel like i should have just bought premade banks at this point.

 

[sunshine_eggo]

So, i'm having a bit of bad luck with top balancing my cells so far. I have 16x Docan Power 280AH to make a 48v bank. They call came at 3.29V, so i batched 4 together in series and charged at 14.60v on my dc power supply (the one Will recommends, here. At first, it was charging rather slow, and with all the videos I've watched, pretty normal. At this point, the red light is on for C.C. on the charger. After a bit, can't say exactly how long it was, each cell was hitting about 3.35V. I turn my back, and a few mins later i go check voltages again. It changed to green C.V. light, and one of the cells has jumped to 4.4v. The cell in question had the power lead on it while charging..just to note. I burn off some of the charge with a halogen light, and swelling ensues the next day. I'm not really sure why one cell, within a blink of an eye has jumped sky high.

Because it's full. The guide is very specific. You NEVER series cells unless you have the BMS installed. period.

At this point, i'm a little freaked out.

Good. Wish you had posted sooner.

I decide to take four more, and try again..surely the first was a dud. The same thing happened, and it happens SO fast that i can barely catch it in time. It's random..it can charge for hours and then all of a sudden jump through the roof. Luckily i caught this one much sooner.

Zero surprise it happened again.

Now i'm terrified.

Good. wish you had posted at this point.

I decide to parallel them and slow charge.

finally a touch of sanity.

Within 24 hours i want to say i gained a tenth or two of volts on all cells.

That's a lot. You're lucky.

That might literally take a month.

See below.

So i take four more cells..i think..no damn way this is going to happen again right?

Oh shit.

and if it does, i'm going to watch it like a hawk. Sure a damn nough it happened AGAIN.

Coulda warned you had you posted.

This time on the negative lead cell...so i guess that has no relevance.

None whatsoever. Exactly the same amount of current passes through the series cells. You're just finding the ones at higher states of charge than the others.

They were all around 3.35, then bam one cell went to 3.72. So i cut the power and let it chill for a few minutes and the high cell dropped to 3.52.

Either way, i'm not sure what the hell is going on here. Am i stupid?

I don't know you well enough, but I can confirm you are doing stupid things.

Am i doing something wrong?

Yes. You didn't follow the top balancing guide at all.

Clearly, i don't understand charging cell behaviors. Help me out please before i burn all these suckers up. I feel like i should have just bought premade banks at this point.

It's like watching a crazy person. Please follow the guide!

16 * 280 = 4480Ah

As received could be as low as 30% SoC

0.7 * 4480 = 3136Ah needed to charge.

3136Ah / 10A = 314 hours = 13 days. more if you don't get the full 10A.

DAYS may pass without you seeing any increase in voltage at all. you might even see it drop .01V.


READ THE ENTIRE GUIDE BEFORE YOU DO ANYTHING ELSE!

 

[Horsefly]

I'm not going to beat the truth into you, as I think @sunshine_eggo has done a pretty good job of that.

If you have at this point a few cells that went up to around 4.4V (I can't tell how many) but didn't stay there for any amount of time and did not swell, they are probably fine. Perhaps you have learned a lesson here. Do what @sunshine_eggo says and follow the top balance guide in the resources.

Charging LiFePO4 cells takes some getting used to. They have a very flat charge curve until they are getting up over 90% SoC, and then they go up very quickly. Never charge in series without a BMS. When charging in parallel, never adjust the voltage you set before you connected the cells, and be patient.

 

[John Frum]

@WilliamFR is there some bad doco out there that needs to be deleted?

 

[TomC4306]

I believe the biggest mistake you made was adjusting the power supply while the batteries were connected.
Don't do this.
Disconnected from everything, turn on the power supply.
Set the voltage.
Turn off the power supply.
Connect your battery or cells.
Turn on the power supply.
DO NOT ADJUST THE VOLTAGE.
If you set the MAXIMUM voltage this way, the cells will not overcharge.

 

[WilliamFR]

Because it's full. The guide is very specific. You NEVER series cells unless you have the BMS installed. period.



Good. Wish you had posted sooner.



Zero surprise it happened again.



Good. wish you had posted at this point.



finally a touch of sanity.



That's a lot. You're lucky.



See below.



Oh shit.



Coulda warned you had you posted.



None whatsoever. Exactly the same amount of current passes through the series cells. You're just finding the ones at higher states of charge than the others.



I don't know you well enough, but I can confirm you are doing stupid things.



Yes. You didn't follow the top balancing guide at all.



It's like watching a crazy person. Please follow the guide!

16 * 280 = 4480Ah

As received could be as low as 30% SoC

0.7 * 4480 = 3136Ah needed to charge.

3136Ah / 10A = 314 hours = 13 days. more if you don't get the full 10A.

DAYS may pass without you seeing any increase in voltage at all. you might even see it drop .01V.


READ THE ENTIRE GUIDE BEFORE YOU DO ANYTHING ELSE!

Well, I think my biggest mistake was just simple watching Will's videos instead of reading a guide. The information is a bit spread out, and i'm not sure he hits on everything. One in particular, he suggests grouping the cells in 12v series and charging them up before balancing if they are at a low SOC. Maybe i missed the part about the BMS needing to be connected first. Regardless, feel free to use my thread as an example of what not to do. I do appreciate the feedback and the tid bits. If i were to put them in parallel and just wait it out, whats the biggest discrepancy in voltage i can put them in parallel with? I have one at 3.60 and one at 3.35 for example. Will that fry anything or is that an acceptable range?

 

[WilliamFR]

I believe the biggest mistake you made was adjusting the power supply while the batteries were connected.
Don't do this.
Disconnected from everything, turn on the power supply.
Set the voltage.
Turn off the power supply.
Connect your battery or cells.
Turn on the power supply.
DO NOT ADJUST THE VOLTAGE.
If you set the MAXIMUM voltage this way, the cells will not overcharge.

if there was one thing i didn't do, it was change the voltage on the power supply while the batteries were connected. the way things are going though, i might as well have lol.

 

[WilliamFR]

@WilliamFR is there some bad doco out there that needs to be deleted?

no, i just didn't follow the guide as i should have. it was a costly mistake. i'm sure i have two that are a little swollen but i'm unsure the extent of damage at this point.

 

[jberger]

Well, I think my biggest mistake was just simple watching Will's videos instead of reading a guide. The information is a bit spread out, and i'm not sure he hits on everything. One in particular, he suggests grouping the cells in 12v series and charging them up before balancing if they are at a low SOC. Maybe i missed the part about the BMS needing to be connected first. Regardless, feel free to use my thread as an example of what not to do. I do appreciate the feedback and the tid bits. If i were to put them in parallel and just wait it out, whats the biggest discrepancy in voltage i can put them in parallel with? I have one at 3.60 and one at 3.35 for example. Will that fry anything or is that an acceptable range?

I don't think it's at all fair to blame Will for any of this, you didn't do it correctly, it's on you.

 

[acdoctor]

Don’t connect those two in parallel. If the 3.6 volts has been off the charge for hours you need to reduce it to at least 3.5 volts. I would not parallel cells with greater than 0.1 volt difference. Opinions may very.

If I were you I would bleed some voltage off of the high cells and STOP and study and understand what you are dealing with before you get hurt.

 

[WilliamFR]

I don't think it's at all fair to blame Will for any of this, you didn't do it correctly, it's on you.

i think what i said was fair, along with my following comment saying i should of read the guide. i think i've already taken blame. at this point, that part doesn't matter, just trying to make the right decisions moving forward really

 

[John Frum]

i think what i said was fair, along with my following comment saying i should of read the guide. i think i've already taken blame. at this point, that part doesn't matter, just trying to make the right decisions moving forward really

I think you should do a capacity/round trip efficiency test on each cell individually.

 

[sunshine]

whats the biggest discrepancy in voltage i can put them in parallel with? I have one at 3.60 and one at 3.35 for example. Will that fry anything or is that an acceptable range?

You can always put a light bulb in series if there is a mismatch.
Thanks for your post!

 

[sunshine_eggo]

Expedited charging:

READ THE GUIDE BEGINNING TO END. I makes several recommendations that are key to the process, but you need to read it all.

Build your 48V battery and attach BMS. Cell 1 is at the (-) where the BMS attaches.

charge cells 1-8 with the power supply attached to the BMS output and the 8th cell.

Once the BMS trips, swap cells 1-8 and 9-16 so that 9 is in position 1 and 16 is in position 8.

Charge cells 9-16 (now in the 1-8 position) with the power supply attached to the BMS output and cell 16 (now in position 8).

Once the BMS trips, parallel all cells and top them off to 3.65.

 

[WilliamFR]

I think you should do a capacity/round trip efficiency test on each cell individually.

well, i'm deep into the balancing guide now..better late than never. i tried searching for a capacity test doc, but i may be typing the wrong thing. do you have a link?

 

[WilliamFR]

Expedited charging:

READ THE GUIDE BEGINNING TO END. I makes several recommendations that are key to the process, but you need to read it all.

Build your 48V battery and attach BMS. Cell 1 is at the (-) where the BMS attaches.

charge cells 1-8 with the power supply attached to the BMS output and the 8th cell.

Once the BMS trips, swap cells 1-8 and 9-16 so that 9 is in position 1 and 16 is in position 8.

Charge cells 9-16 (now in the 1-8 position) with the power supply attached to the BMS output and cell 16 (now in position 8).

Once the BMS trips, parallel all cells and top them off to 3.65.

reading it now, thanks for the help. still waiting on my bms to come in. going forward i'm taking everything nice and slow lol

 

[sunshine_eggo]

reading it now, thanks for the help. still waiting on my bms to come in. going forward i'm taking everything nice and slow lol

Then 16P.

Fabricate new leads if needed as described in the guide.

Set power supply to 3.65V.
Set current to 10A.
Attach power supply.
Check it occasionally.
Stop when CELLS measure 3.65V with a meter and power supply reports 0A. Power supply voltage reading will always be higher.

 

[John Frum]

well, i'm deep into the balancing guide now..better late than never. i tried searching for a capacity test doc, but i may be typing the wrong thing. do you have a link?

I retract my suggestion.
Just top balance in parallel as described by @sunshine_eggo
You can capacity test them as a 16s battery as part of the commissioning of the battery.

 

[740GLE]

Wow I’m impressed!!!

 

[toms]

The mistake you made was trying to top balance your cells before you completed your build.

It is a common error, responsible for destroying and shortening the life of more LiFePO4 cells than anything else.

Following the top balancing guide as mentioned will accomplish what it says, but is a poor way of achieving the desired result of avoiding cell level high voltage disconnects in most cases.