1 - 12 of 12 Posts

#### dtbaker

·
##### Registered
Joined
·
4,326 Posts
Discussion Starter · ·
I am fine tuning the balance on cells now that they are all installed and have been through a couple charge cycles. I have decided to start with no BMS, top-balance, and my charger is set to CA->CV at average cell voltage of 3.65*38= 139v for my 38 cell nominal 120v pack. If all cells were PERFECTLY balanced, they should all be 3.65v at end of charge.

Most of the cells in my pack are very close at end-of-charge, but I have a few I would like to pull down a little to even things up a little better before I button up the lids and consider myself at a baseline start....

I have wired up two 50watt resistors, and have been experimenting a little to figure out how much time 'brings down' the end voltage by .01, but so far is is a little frustrating as the voltage bounces back a little after discharge, so I haven't come up with a good predictor yet. I also have a power supply I can set to either CV or CA and add juice to single cell if required. I keep over-shooting with the resistor/power supply. My clamp ammeter shows that the resistors pull about 6amps from a single cell when connected.

I am hoping some others may have some tips on manual balancing?

If and when required, have you come up with any good predictors for how much discharge is required to change the 'end of charge' voltage?

#### major

·
##### Registered
Joined
·
7,791 Posts
I am hoping some others may have some tips on manual balancing?

If and when required, have you come up with any good predictors for how much discharge is required to change the 'end of charge' voltage?
Hey dt,

I have done it a few times. Same way as you're talking about. I use 3 decimal places for the voltage. Doesn't your new meter go there? And when you put the resistor across the cell and draw 6 amps, the cell voltage drops by I * Ri (internal resistance of the cell). Say the cell Ri = 0.0005Ω. Then when you connect the resistor and draw 6 amps, the cell voltage will immediately drop by 0.003 volts. Remove the resistor and the cell voltage will immediately rise by 0.003 volts. This is expected and can be foreseen. However, there always seems like there is more bounceback, sometimes less in the minute or two after you stop discharge or charge. All I can say is you kinda get the hang of it after a while

major

#### dtbaker

·
##### Registered
Joined
·
4,326 Posts
Discussion Starter · ·
All I can say is you kinda get the hang of it after a while

major
yeah, my first pass I was kinda going seat of the pants, and overshooting, but am narrowing the 'predictable' effect.... hoping to hear from others that may have tried.

I'll post my results....

#### tomofreno

·
##### Registered
Joined
·
2,318 Posts
Hi Dan, I've done the same, but use a 3 Ohm R to only pull a bit over 1A. You get a feel for it after you record time and voltage change on a few cells. It doesn't take much charge to change V of a cell by 0.01V on that part of the curve - I think around 1/5 Ah or so in my experience.

#### dtbaker

·
##### Registered
Joined
·
4,326 Posts
Discussion Starter · ·
Hi Dan, I've done the same, but use a 3 Ohm R to only pull a bit over 1A. You get a feel for it after you record time and voltage change on a few cells. It doesn't take much charge to change V of a cell by 0.01V on that part of the curve - I think around 1/5 Ah or so in my experience.
hhhmmm, interesting, maybe I will make a 'fine' resistor at lower wattage to get the last little bit. I really would be happy getting them within +/- .02v at end of charge and then running for a while to see if they drift apart or stay at the same relative differences for significant time.

other option is to do a full charge, knock any high ones down below 3.6, then go thru EVERY cell with power supply set to 3.70 and let it go until a=0.0 before moving on to next cell. Not sure if this would work better, or not. Basically a cell level balance post install.....

#### dtbaker

·
##### Registered
Joined
·
4,326 Posts
Discussion Starter · ·
one thing that makes this a bit of a moving target is that when my pack voltage hits the CA-CV switch point at 139v, it only sits there for a couple minutes before amps drops to 1.0 and the system flips into hold mode.... so far it looks like there is a reasonable window of time where the pack drops to 134-135v and hangs there for a while, pulling just a trickle of juice from the wall.

Approach tonite will be to attempt some fine tuning to balance after end-of-charge while the pack is in 'hold' voltage.

#### gdirwin

·
##### Registered
Joined
·
308 Posts
One idea I was playing with was a welded "contraption" that you drop onto the batteries (while they are installed) to put them all in parallel - ie you would remove all of your copper inter-battery busbars, then drop a metal grid on top of them for a few minutes to put them all in parallel and then re-connect all of your busbars - perfect balancing...

This saves the work of having to remove the batteries (at which point only 2 simple bars of steel are required if all of the terminals are line up).

Has anyone done this? What would it look like? Is it worth the effort?

#### dtbaker

·
##### Registered
Joined
·
4,326 Posts
Discussion Starter · ·
One idea I was playing with was a welded "contraption" that you drop onto the batteries (while they are installed) to put them all in parallel - ie you would remove all of your copper inter-battery busbars, then drop a metal grid on top of them for a few minutes to put them all in parallel and then re-connect all of your busbars - perfect balancing...

This saves the work of having to remove the batteries (at which point only 2 simple bars of steel are required if all of the terminals are line up).

Has anyone done this? What would it look like? Is it worth the effort?
the last thing I would want to do is un-torque and re-torque my connector nuts.. very time consuming and prone to cause issues eventually. I would contemplate installing some 16ga wires run to a centralized terminal block for easy access monitor/balance.... but plan t wait to see how often I actually HAVE to do any re-balancing before investing the time, money, and weight of a couple hundred feet of wire.

besides which few people have linear packs of cells, or in one location.... so switching from serial to parallel connections is a pretty major pain.

#### ElectriCar

·
##### Registered
Joined
·
1,901 Posts
Some have suggested using a head light to drain them one at a time.

#### brainzel

·
##### Registered
Joined
·
341 Posts
I just took my 38 batteries (CALB) out of the car to balance at the bottom.
The voltages were quiet together at 2,85V @ 27Amp load (heater).

After that I wired all together in parallel and waited an hour or two.
Then I took five cells in series, wired a constant voltage DC power supply, adjusted to 16,25V and looked at the current.
At 1,5A i stopped the charge and went to the next five cells.
My thought: if I stop at the same chargepoint (16,25V @ 1,5A) every pack must have the same level.
And it worked.

Batteries back in the car, charged up and everything looks great.
Yesterday I got seven additional cells, so I have to do it again the next days.

Why doesn't charge to 16,25V until current is 0A?
It would take a brazillion hours

Why don't use a resistor to drain the battery?
The voltagesag is depends on the load, so at 1Amp the cell sags ... hardly.
At the lower knee, there are quiet a few amphours in there, so you have to wait and drain a lot of time.

table: lowest cell voltage of 38 cell pack; amphours counted; load
3,00V @ 162,1Ah @ 27A
2,98V @ 163,5Ah
2,97V @ 163,9Ah
2,96V @ 164,8Ah
2,95V @ 165,3Ah
2,94V @ 166,0Ah
2,93V @ 166,4Ah
2,92V @ 167,0Ah
2,91V @ 167,6Ah (3,05V highest cell voltage)
2,90V @ 168,0Ah (3,04V highest cell voltage)
2,89V @ 168,5Ah
2,88V @ 169,0Ah (3,03V highest cell voltage)
2,87V @ 169,4Ah
2,86V @ 169,9Ah
2,85V @ 170,2Ah (3,01V highest cell voltage)

#### Attachments

• 55.9 KB Views: 116
• 13 KB Views: 159

#### dtbaker

·
##### Registered
Joined
·
4,326 Posts
Discussion Starter · ·
New day, new problems!
I have to take myself back: the next balancing went out of control and I don't have any idea why.

my guess would be that since you bottom balanced, and in sections, you may not have started off as balanced as you thought AND if you hit your end of charge and any shunting happened (do you have a shunting BMS) it would send your system totally out of balance.

My gradual top balancing is bringing things into line nicely at end of charge, I am just about to button the lids on. I am no longer worried that I have any cells sticking out at the high end that might go ballistic at end of charge. I have been checking after end-of-charge when pack voltage settles from its high of 139 down to 135.... and making minor tweaks with power supply/resistor to the highest and lowest. Cells are all coming into line closely at the very top now, and almost identical once off charge and settled to operating voltage.

1 - 12 of 12 Posts