Parallel charging a large array of high capacity prismatic LiFePO4 cells

Thread Starter

OneMist8k

Joined Aug 4, 2020
12
I have 48 prismatic LiFePO4 cells that are perfectly balanced. They have been sitting in parallel for days, and each cell is within .001 volt of each other. Each cell nicely measured 2.800 volts.

While the cells were in parallel, I went to charge them using a bench power supply. The positive and negative leads were attached to opposite ends of the very large string of 48P. The charge voltage was 3.6, the CC started charging at about 16.4 amps. All seemed okay. I'm guessing these large 150a prismatics will take days to fully charge.

HOWEVER, after about 20 or 30 minutes I measured the voltage of the cells. The two cells at the ends had a higher voltage than the cells in the middle. Remember, the end cells is where the power leads were attached, (+) on one end, (-) on the other. The difference was more than .100 volt. I freaked out a bit and stopped charging. Right now the cells are all in parallel, getting in balance.

THE QUESTION: During a parallel charge, shouldn't all cells be at the same voltage? Is a 48 cell bank simply too large to charge this way? Should I reconfigure the cells to charge in series? Everything I've seen on parallel charging uses a charge board. Obviously there is no charge board for a large array of large capacity cells.

Am I being too cautious / too paranoid?
 

WBahn

Joined Mar 31, 2012
26,841
I don't know the specific concerns/issues with this type of cell, so these are generic observations.

A large number of things in parallel are pretty much never truly in parallel. There are all kinds of things that mess this up, ranging from interconnect issues to device issues. Again, not knowing anything about these kinds of cells, one thought that comes to mind is whether thermal effects are resulting in the imbalance of the cells. No two cells are going to be truly identical, so even if they are adequately in parallel at the terminals, they are going to charge at different rates, resulting in different thermal profiles, which might further drive differences.

As for being too cautious or too paranoid, that will depend on the consequences involved. Some battery technologies are pretty tolerant of charging abuse with the primary consequence being reduced life, while others are very sensitive with the primary consequences being energetic fires and/or explosions. My very limited understanding of this battery technology is that it is on the more-tolerant side. So it comes down to what are the potential consequences if they aren't charged properly and how much are you willing to accept the risks associated with not doing so.
 

Thread Starter

OneMist8k

Joined Aug 4, 2020
12
As for being too cautious or too paranoid, that will depend on the consequences involved. Some battery technologies are pretty tolerant of charging abuse with the primary consequence being reduced life, while others are very sensitive with the primary consequences being energetic fires and/or explosions. My very limited understanding of this battery technology is that it is on the more-tolerant side. So it comes down to what are the potential consequences if they aren't charged properly and how much are you willing to accept the risks associated with not doing so.
I am primarily concerned if damaging the cells is possible.
 

Alec_t

Joined Sep 17, 2013
13,041
I have absolutely no experience with these cells, but like WBahn I wonder if thermal effects play a part. Think penguin huddle. The cells at the ends are likely to be a tad cooler than those towards the middle of the array.
 

LowQCab

Joined Nov 6, 2012
2,315
First, don't confuse everyone here by calling them "Prismatics".
Is there a difference in the performance of
a "Prismatic"-LifePo4-Cell, vs, a generic LifePo4-Cell ????

A short Search turned-up no information, so we must "guess" about what You may have.

You're probably going to have to add Balancing-Circuitry.
.
.
.
 

Ya’akov

Joined Jan 27, 2019
6,260
First, don't confuse everyone here by calling them "Prismatics".
Is there a difference in the performance of
a "Prismatic"-LifePo4-Cell, vs, a generic LifePo4-Cell ????

A short Search turned-up no information, so we must "guess" about what You may have.

You're probably going to have to add Balancing-Circuitry.
.
.
.
There is a difference between prismatic and cylindrical cells. They trade off between capacity and discharge rate. I am not certain but I think that also affects charging current.

Prismatic is conventional terminology.
 

Thread Starter

OneMist8k

Joined Aug 4, 2020
12
First, don't confuse everyone here by calling them "Prismatics".
Is there a difference in the performance of
a "Prismatic"-LifePo4-Cell, vs, a generic LifePo4-Cell ????
There might be a difference... I wouldn't presume either way, that is why I qualified the cell type.
 

dl324

Joined Mar 30, 2015
14,897
I have 48 prismatic LiFePO4 cells that are perfectly balanced.
I very much doubt that that's the case. Even if it were true, they won't stay matched for long.
They have been sitting in parallel for days, and each cell is within .001 volt of each other. Each cell nicely measured 2.800 volts.
With the batteries in parallel for days, they should all be at the same voltage. That doesn't mean that all of them would be at that voltage if they were removed from the parallel configuration.

Charging batteries in parallel isn't a good idea because you can't measure the actual voltage of any of them.
 

Irving

Joined Jan 30, 2016
2,979
Parallel charging, and storing, close-coupled LiFePO4 cells, whatever their construction, is perfectly valid - to a point. 4P OK, 8P OK with very low impedance interconnect, but 48? Asking a lot. You won't damage them as such, but the ones nearer the charger will determine the cut-off point so the far ones will be under-charged, which means they are prone to being over-discharged if you take them close to 100% discharged. In the short term break them into groups of 6, make sure the interconnects are low impedance (12 - 10AWG) and feed each group with its own pair of heavy wires (connected to the centre of the group) from the charger, keeping them all the same length.

Personally I'd charge them as 8S 6P using a really good 8S charger such as the Revolectrix PL8, sadly now discontinued (I do 8S 5P at 40A), and interconnects of 10AWG minimum. Most hobby chargers aren't accurate enough to do this reliably.

How do you plan to use these cells?
 

Thread Starter

OneMist8k

Joined Aug 4, 2020
12
How do you plan to use these cells?
EV for a boat. 3P16S for a 48v motor. Once in this configuration that is how they will be charged and discharged. The motor 350kW.

As for parallel charging, I am now thinking of charging each one individually (48...egad!) so I can measure and record the charge time for each one individually. Then after they've been rebalanced, connecting them as P16S banks (meaning each individual cell is in series) for a discharge test where I can monitor each cell with a BMS. (I may charge them in series with a different 48v charger, still thinking about that.) I'll have to do the discharge test 3 times, though. In the ultimate 3P16S configuration I won't be able to see individual cell performance over time, only down to the 3P cell. However if there are any marginal cells in the batch keeping records of each cell with these tests will let me know which ones to keep an eye on.

All of this will take over a month for sure, but I'm not going anywhere. The plan is still evolving. :)
 

Thread Starter

OneMist8k

Joined Aug 4, 2020
12
Resistance in the wires between cells? Are they paralleled with massive copper bus bars or with skinny wire?
12 gauge AWG.

First, don't confuse everyone here by calling them "Prismatics".
Is there a difference in the performance of
a "Prismatic"-LifePo4-Cell, vs, a generic LifePo4-Cell ????
I wouldn't presume, which is why I qualified the type in my question.
 

Thread Starter

OneMist8k

Joined Aug 4, 2020
12
? Should get the boat airborne in no time!
Oops! Sorry, the motor is 20kW @ peak (2 x 10kW motors) LOL!!!! I was just working with the ampacity requirements for the cabling for each motor and had the 350A number on my mind. Totally my mistake!!!
 
I have 48 prismatic LiFePO4 cells that are perfectly balanced. They have been sitting in parallel for days, and each cell is within .001 volt of each other. Each cell nicely measured 2.800 volts.

While the cells were in parallel, I went to charge them using a bench power supply. The positive and negative leads were attached to opposite ends of the very large string of 48P. The charge voltage was 3.6, the CC started charging at about 16.4 amps. All seemed okay. I'm guessing these large 150a prismatics will take days to fully charge.

HOWEVER, after about 20 or 30 minutes I measured the voltage of the cells. The two cells at the ends had a higher voltage than the cells in the middle. Remember, the end cells is where the power leads were attached, (+) on one end, (-) on the other. The difference was more than .100 volt. I freaked out a bit and stopped charging. Right now the cells are all in parallel, getting in balance.

THE QUESTION: During a parallel charge, shouldn't all cells be at the same voltage? Is a 48 cell bank simply too large to charge this way? Should I reconfigure the cells to charge in series? Everything I've seen on parallel charging uses a charge board. Obviously there is no charge board for a large array of large capacity cells.

Am I being too cautious / too paranoid?
1m of 12 gauge wire at 16 amps give 0.08V, close to the 0.1V you mention... I do not know how you connect all the cells, but one idea could be to use smaller wire and connect each cell to the power individually; or as mentioned above in groups. Individual connections will give you some "balancing" with the wire resistance
 

Thread Starter

OneMist8k

Joined Aug 4, 2020
12
I do not know how you connect all the cells, but one idea could be to use smaller wire and connect each cell to the power individually; or as mentioned above in groups. Individual connections will give you some "balancing" with the wire resistance
I'm guessing that is how charge boards work for smaller cells?
 

timm27

Joined Dec 11, 2020
9
My feeling is that charging in parallel is a bad idea. In series each cell will get the same current and you can easily monitor the voltage across each cell. You would not sensibly try to build a high current accumulator from LiFEPO4 cells in current. If you need to charge in parallel you - such as to charge separate battery packs then you need to design a multi-channel regulator so each cell gets an appropriate current and charging can be stopped as each cell/battery pack reaches its target voltage.
 

du00000001

Joined Nov 10, 2020
106
While parallel charging is not bad in itself, given the number of cells it may turn out problematic.
As others already mentioned you'd better go for small(er) groups of cells, connected with equal-length wires to the points where the power is connected to.
(OTOH, once the "edge cells" have a somewhat higher cell voltage, the current will go to cells farther "away". Given enough time and keeping the max. charging voltage a constant, I'd expect all cells coming out of charging with the same cell voltage. But that will take its time...)
 

Irving

Joined Jan 30, 2016
2,979
For your 16S3P arrangement you must connect all cells in one group to all the cells in the adjacent group... its tempting and easy to connect like the top diagram, but that's wrong. You should do it like the lower diagram, keeping the wires as far as possible the same length.
1660311401061.png

Here's why...

If you connect cells in a group together but only have a single connection group to group from the middle cell then the currents are unbalanced and the middle cell will discharge faster than the outer ones.

1660309311116.png

By using a star connection the load is shared equally across cells. This also benefits charging too. Your balance wire should be connected to the star point.

1660311901595.png


An alternative to pigtails is nickel-plated copper busbar. I don't recommend these for prismatic cells as its hard to keep the cell fixed with all the vibration and shock and I've had cells break internally. They are great for the cylindrical cells in the orange holders. Also the ones supplied by the cell vendors are something like 0.75mm thick and have less square mm of copper than 10 or 8AWG so you have to stack them which leads to other issues.
 
Top