Easy Bottom Balancing

[Ampster]

No cells are connected to each other at the moment. I wil discharge those two cells a bit and then connect all of them in parrallel?

No need to charge or discharge them. They will all come to the same voltage when you put them in parallel. The High charge in some will flow to the other cells. Then charge the group in parallel. If you want to be safe overnight set your power supply to 3.55 over night. That way in the morning you can take them to 3.60 or 3.65 in parallel.

 

[Ampster]

If you are uncertain read the post about parallel top balancing in the Resource section.

Top Balancing LiFePo4 Cells using a low cost benchtop power supply.

To get the paper, click on the orange at the top of this page. This is a tutorial with detailed steps on how I top-balance my LiFePO4 Cells using a low-cost benchtop power supply. To get the tutorial, click on the orange "Download" button...

diysolarforum.com

 

[Ampster]

This started out as a question about bottom balancing. I should have posted that link earlier.

 

[fafrd]

I’m just learning about this now, so take any advice from me with a grain of salt.

I’m in the middle of charging up my 90Ah 8S battery for the first time.

If I believe what the BattGO tells me, I got some bad cells.

But if ignore the BattGO and read the cell voltages with the same multimeter, all of my cells are at exactly 3.310V, meaning ~80% SOC.

So now I’m going to add ~5% / 4.5Ah at a time and see whether the voltages start to stray away from each other as I start getting into the ‘hockey stick’ above 3.315V...

Obviously, getting a 3.65V charger and charging all of your cells in parallel with large common busbars is the ‘best’ / recommended way to top balance. I just plan to see how close I can get using my 20A 28.6V charger, my BMS, my BattGO, and my multimeter before making that investment.

Will report back as my experiment unfolds...

The BattGO turns out to be all but useless for top-balancing, at least in my case. It’s so inaccurate I can’t trust what it says to truly get balanced. I’m exchanging it for another one in case I just got a bad unit.

Charging until my BMS cut’s off the high cell at 3.75V then using a 2.5ohm 50W resistor to drain the high cells down to the low cells has been much more effective.

My last run has all 8 cells charged to 3.561V (all measured with the same multimeter).

I’ve noticed that there is always some settling after either charging or discharging and also see that the EVE datasheet includes a 30 minute (or in some cases 60 minute) rest for many of their electrical performance criteria, so I’ve done the same to get these results:

-Charge and then test 30 minutes
/Measure all cells and drain Hugh cells to low cells using 2.5ohm resistor (timed).
-Repeat cycle until high-low is within target.

It took about 10 cycles to get the 8-cell battery perfectly top-balanced (within the +/-0.5mV accuracy of my multimeter).

Now that I’m ‘perfectly balanced’ I’m going to leave it overnight before calling the job completely done.

This was probably more trouble than using a 3.65V charger and parallel bus bars, but it wasn’t too bad and allowed me to use the 28.6V LiFePO4 Charger I had rather than spring for another charger I’d only use once.

If I don’t need to screw around with more balancing cycles after resting overnight, I will finally perform a discharge test to see what capacity I’m actually getting and how closely-matched the cells still are after being discharged.

 

[newbostonconst]

I top balanced my cells and then assembled them in a 2P16S pack. As I began exercising them some started moving toward 3.65 per cell so now I am manually top balancing the runners with a one Ohm 50 Watt resistor. The issue with a 2P pack is you don't know which of the two cells is the weakest. I have already removed one pair and replaced them with a spare pair that had to be adjusted to the rest of the pack. Now the task of individually capacity testing each of those two at the rate of 40 Amps, which for a 280 Amp cell is going to take 14 hours round trip.
My point is with new cells you never know. It may take several weeks of exercise before a weakness reveals itself. Fortunately I ordered spares so I cans substitute and not have to wait for my vendor to send me replacements.

This took me a week and a half and am now waiting on a board from outback.

I really thing we need to start pushing for bigger and bigger BMS's or Active Cell Balancers. My 5 amp active cell balancer still isn't enough when charging and discharging everyday( I have to take a 1/2 ohm resister to bring my one cell down like you describe). I am wondering if the internal resistance is off on my one new cell I got. I am thinking about paralleling a second 5 amp balancer.

The new 280 Ah cells are huge and I really don't think the current BMS's aren't big enough. All good info.

 

[Luthj]

There are two factors which influence balance. Capacity, and internal resistance. Internal resistance typically only comes into play at higher current rates. P= I^2 * R

Assuming a top balanced pack (which is typical of any low C rate storage type), you will see an apparent imbalance if one cell has a lower capacity, and you discharge into the knee area. That low cell will drop out sooner. The solution is to replace the cell, or to de-rate the pack for the lowest capacity cell.

Internal resistance differences can be used to roughly work out how long a balance will take.

Take a Hypothetical 100AH pack, the high cell is 1mOhm, and the low cell is 0.5mOhm. Discharge at 1C, charge at 1C.

Estimate power dissipated in the high cell.
100^2 * 1mohm =10W

Energy lost to heat during charge and discharge.
10W * 2hr = 20W-hr

Repeat for low resistance cell.
100^2*0.5 = 5W
10W-hr

The difference will need to be dissipated, by bleeding 10w-hr off the low resistance cell.

Assuming a 1A balance starting at 3.4V
1A * 3.4V = 3.4W balance power
10W-hr/3.4 = 3 hours.

Now this is a pretty extreme example, as 1C and a 2x difference in resistance is not typical. Lets do an example with more typical rates.

100AH battery high cell is 1mohm, low cell is 0.75mohm, 0.2C charge/discharge

20^2 *1mohm = 0.4W
20^2 * 0.75mohm= 0.3W

Energy dissipated
0.4 * 10 =2W-hr
0.3 * 10 =3W-hr

Difference 1W-hr
1W-hr/3.4 = 0.3 hours or around 18 minutes.


Obviously these are rough estimates, as the internal resistance will vary with SOC, and charge/discharge. Typical cell internal resistance should be within 15% for a production batch, often lower for higher quality cells (cough cough, Tesla). At least for smaller cells. I haven't seen any recent data on the larger prismatics.

Another factor which can influence pack balance significantly, is bus bar and contact resistance. Its not hard to get several milliohms of resistance from bad contact, crimps, corrosion, etc. These can mimic a poorly matched cell. Typically they can easily be identified with a volt meter, and a decent sized load.

 

[BobF]

Hi Everyone. I believe there never will be an end to the discussions as to what is best: Top or Bottom balancing. It probably comes down to what work's best for you. I choose to bottom balance. I have 16 400ahr cells from GBS, and running my boat on 24V. I use the basic Chargery unit. I bottom balanced all cells using a Power Lab 8, which discharged them into a LA battery. They were about 50% SOC direct from the factory in China. Each came with a printed sheet showing all cell's voltage when they left the factory. I do NOT, nor ever have balanced the cells in the past 3 years. That means about 1,000 cycles or more. Yesterday all cells were 8mv apart. Sometimes they go up to 25mv. I am now in my 3rd year of living fully off grid on my boat, and those cells have remained as quoted above. Daily discharge is 7.18kwhr. Lowest discharge during all this time was 417ahr, and that was 2 days ago when a typhoon came into the Philippines where I am moored. So, I have no balancing, active or otherwise. The cells take care of themselves. The Victron inverter, and my Victron SCC are my first line of defense. Two hi amperage, low coil draw relays handle the Hail Mary HVE and LVE. The Chargery and Victron equipment consist of my Battery Monitoring System, not necessarily a Battery Management System. I look at the Display on the Chargery when ever I am in the back of the vessel where the batteries are. For almost a year, upon first start up, I would write down all the cell voltage every day at the same time. After a while, I got very confident in the system and ceased writing a logbook. The BMV 700 is connected to my laptop via USB and since I use it every day, I can always monitor pack volts and SOC at a glance. One of the SCC is blue tooth capable, and has that wonderful 30 history via the Vic App. As you can see, it works for me. I don't fuss about balancers nor worry about the last MV. Other than the huge expense of the batteries (by sea freight, and importation fees), my components are relatively cheap. The Chargery when I bought it was about $95. The relays via eBay about $25 each. I have a 250/70 and 150/85 SCC's and 4 x 450W panels and 6 x 360W panels. Since January, the generator has only run about 20hrs. Hope this might help some in their decisions as which way to go. Hard to argue with the results noted above!

 

[Sojourner1]

@BobF starting with proven quality cell from an established company is step ahead than what's the cheapest cell that can be purchased. When the time comes for an upgrade or replacement GBS will be purchased again, just in a larger cell format. But than I should still have some years before that's needed. Being on water away from civilization can be more life threatening than on land.

 

[Ampster]

Being on water away from civilization can be more life threatening than on land.

And that is why that use case is a good example of when bottom balancing makes sense.

 

[wholybee]

This took me a week and a half and am now waiting on a board from outback.

I really thing we need to start pushing for bigger and bigger BMS's or Active Cell Balancers. My 5 amp active cell balancer still isn't enough when charging and discharging everyday( I have to take a 1/2 ohm resister to bring my one cell down like you describe). I am wondering if the internal resistance is off on my one new cell I got. I am thinking about paralleling a second 5 amp balancer.

The new 280 Ah cells are huge and I really don't think the current BMS's aren't big enough. All good info.

If you think that you need more than a 5 Amp balancer, either your cells are not balanced as you think they are, or you have a problem with one or more of your cells.

You should be able to go quite a while with no balancing whatsoever. Even with cheap commodity cells I would hope weeks to a month. With the best quality cells you can go years without balancing again. In either case, the small mA balancing a BMS does should be enough to keep them in balance forever, and 5A should cover the outside case with badly mismatched cells.

But my first guess is that the cells just were not completely balanced to start with.