diy solar

diy solar

Top Balancing using a PSU: is this exactly the other way around?

powerB

New Member
Joined
Feb 17, 2021
Messages
6
I've got a simple 4S 100Ah battery sitting on my table, waiting to be top balanced for the first time. I am a novice when it comes to this stuff, but I've done a lot of reading and researching and have downloaded the Tutorial about balancing LiFePO4 with a 10A power supply. I think I have a good grasp of the important concepts but I am not sure if I'm doing everything right, though...??

The individual cells each measured in at 3.30v and I've hooked them up in parallel. I've decided not to do the pre-charging but instead go straight to the balancing. So I've taken the following steps:
- set the max voltage of the PSU to 3,65v (nothing connected)
- attached the positive (+) and the negative (-) clamps together to then dial in the amps (set to 10A)
- only after that is done clamp to and connect the battery terminals

...the PROBLEM (if it is one): the PSU display shows that the PSU is in the CV (constant voltage) mode, the display shows a permanent 3,65v and the amperage has dropped to around 4,0-4,2A and seems to slowly slowly decrease over time.

Now I'm really wondering whether this is correct?? I thought LiFePO4's needed to be charged up to 3,65v with a constant current??
Is charging at a constant voltage bad?
In another thread one person's response to a similar situation was: the PSU leads could be the problem (because they're way too thin)?
Am I misunderstanding?

Here are three pictures to show whats going on:
lifepo4_detail.jpg Step 1: set the max voltage to 3,65v

lifepo4_detail2.jpg Step 2: set the max current to 10A (by clamping the two leads to each other, as per PSU manual)
lifepo4-detail3.jpg Step 3: connect the cell and watch this happen: the PSU is in the CV mode, the amps are down to 4,377A and slowly seem to be dropping
 
Being in CV is normal..... nothing wrong with that.

However, it looks like you are only charging a single cell. To ballance the cells, they need to be all charged together (in parallel)
 
Exactly what Filterguy said. If you were to charge in CC, you have to watch the voltage at the battery terminal VERY close, because when the batteries reach a full state of charge, the voltage will spike beyond 3.65 very quickly. It's best to do the parallel top balance in CV 3.65. But will take a long time if the SoC is low. That's where you can do a pre-charge in Seires with a BMS attached using CC mode. When 1 battery reaches the limit of 3.65, then it shuts off. Then rewire to parrallel and do the top balance with CV 3.65. As the battery voltage approches 3.65, the current will approach 0. each sell will stop charging when it matches the CV. So once it hits 0 (or near 0), then the top balance is done.
 
I'm on day 3 parallel charging. These 280 amp hour batteries take a long time. Been charting the numbers since yesterday at 110 pm. At that time the amps were flowing at 4.394 and the cell volts were 3.355. It's now 9pm (the next day) and the current is down to 3.968 and the cell voltage is 3.379 volts. Very slow progress. All charging at 3.65 volts.
 
I'm on day 3 parallel charging. These 280 amp hour batteries take a long time. Been charting the numbers since yesterday at 110 pm. At that time the amps were flowing at 4.394 and the cell volts were 3.355. It's now 9pm (the next day) and the current is down to 3.968 and the cell voltage is 3.379 volts. Very slow progress. All charging at 3.65 volts.
That all sounds normal.... particularly if you don't do a pre-charge.
 
Some quick numbers.

You have 4 100Ah cells. Each Cell has 100Ahx3.4V=340Whrs for a total of 4x340Wh=1360Wh.
Let's assume the cells started half charged. That means they need 1360/2= 680Wh to get charged.

You are charging at 3.36Vx4A=13.44W.

That means you need 380Wh/13.44W=50 hr to charge. (or 100 hours if the cells were totally empty to start)
 
Hi everybody,
well I've been making slow but steady progress and reached the end of the process. Is it normal that when you let the battery rest for a while that the voltage drops? The power supply has spent about two hours hovering around the 3,65V and 0.02A mark, very very slowly decreasing. When it got close to 0.01A I stopped the charging, measured the voltage of the cells right then (3.65v) but then this morning the meter reading only gave me 3.59v... Should I keep charging, literally all the way down to zero? Or can I consider the cells balanced even at 3.59 and now move on to the series connection with the BMS?

Thanks!
 
You should be ok. They always settle. Don't try to get them to stay at 3.65.
If you do each cell the same way they should be close enough to each other to connect serial when you are done.
I thought it was better to place them all in parallel first, then top them off; rearranging them into series packs after. This is what I plan on doing, anyway. I won’t worry about it till i’m close to having BMS’ though.
 
I thought it was better to place them all in parallel first, then top them off; rearranging them into series packs after. This is what I plan on doing, anyway. I won’t worry about it till i’m close to having BMS’ though
Balancing in parallel is certainly the 'traditional' way of doing the balance. However, some folks on the forum feel it is better to charge the cells individually to a common voltage. Theoretically, if you charge all the cells individually to the exact same voltage and let them all go to zero current, it is identical to doing them all at once in parallel.

I will stay with parallel balancing, but others choose a different path.
 
Balancing in parallel is certainly the 'traditional' way of doing the balance. However, some folks on the forum feel it is better to charge the cells individually to a common voltage. Theoretically, if you charge all the cells individually to the exact same voltage and let them all go to zero current, it is identical to doing them all at once in parallel.

I will stay with parallel balancing, but others choose a different path.
Don’t misunderstand my post. I was responding to Arthur. I charge up to a high state of charge first, then parallel connect for top balance, then series connect them. His post said you’d be ok to skip the parallel part. While that may be true, seems better to parallel them first (before series) so they can top up and equalize together.

@smoothJoey why the frownie face?
 
I thought it was better to place them all in parallel first, then top them off; rearranging them into series packs after. This is what I plan on doing, anyway. I won’t worry about it till i’m close to having BMS’ though.

Balancing in parallel or individually is fine. Conceivably, parallel will make them closer, but once in use any passive BMS ought to finish the job eventually.

Don't have a BMS yet? OK, so long as you don't even think of charging or using in series until you do. To protect against temptation, don't even connect in series until then.

If BMS will come later, charging with 3.65V (or lower) power supply sounds like a good way to stay busy for a while.
Don't want to leave them in topped condition, so lower like 3.5V might be better if "the BMS is in the mail".
 
Balancing in parallel or individually is fine. Conceivably, parallel will make them closer, but once in use any passive BMS ought to finish the job eventually.
makes sense
Don't have a BMS yet? OK, so long as you don't even think of charging or using in series until you do. To protect against temptation, don't even connect in series until then.
too late! I did a 2s discharge to bleed off some ah. All good though. I’m quite learned on my plan and approach.
If BMS will come later, charging with 3.65V (or lower) power supply sounds like a good way to stay busy for a while.
Don't want to leave them in topped condition, so lower like 3.5V might be better if "the BMS is in the mail".
It’ll be some time till I have my 24v BMS, but I’ll have a couple 12v BMS in a week or so. I will be able to do some additional testing with my panels and SCC.

Thanks for your input.
 
too late! I did a 2s discharge to bleed off some ah. All good though. I’m quite learned on my plan and approach.

Not a problem so long as you stopped before over-discharging a cell.

You would get in trouble charging near the upper knee, or discharging near the lower knee.
 
Not a problem so long as you stopped before over-discharging a cell.

You would get in trouble charging near the upper knee, or discharging near the lower knee.
Nah. I barely pulled off 2 ah or so. Maybe it was 3.

I don’t know from experience, but I imagine they start diverging a good bit in either of those areas, which is what makes it unsafe unless you’re attentively monitoring the cells yourself. Based on what I’ve learned, if I were to try this I would be constantly rotating my monitor on the cells. I’d only do it in an emergency. It’s just a lot more simple and safer to have a BMS. Perhaps a danger zone post about doing this would be useful?
 
@smoothJoey why the frownie face?
Because I think myself and others have demonstrated its unnecessary to parallel the cells.
Even if topping off the cells individually does not produce quite as good a result, a properly configured BMS should get things sweet in 1 or 2 cycles.
That is from my own experience.

Think of it like 80 grit paper for the first pass and 200 grit paper to finish.
I'm a lousy carpenter but I think the analogy works.

The reason to individually top cells is because you don't have to break the pack.
The reason to avoid breaking the pack is those tapped aluminum cells are delicate.
The bonus is its much faster.
 
Actually let me be more honest.
SmoothJoey is an ironic nickname.
My ability to focus on repetitive stuff is terrible and I'm as clumsy as a bull in a china shop.
The less my hands are in proximity to the cell terminals the better.
 
Back
Top