Trouble getting 13s pack to balance

[cdb0ewm]

Can anyone give me a suggestion re gettiung my 13s to balance. I have charged the pack for 8+ hours and 2 groups of cells go to 3.6 which the other 11 stay at 3.3. See the info below.
Thanks

[steves1977uk]

Could be loose bolts, bad soldering on the wires/ring terminals... Could also be the Hyperion starting to fail. If so, replace with 2xPL8's. BM can advise you better on this than I can.

Steve

[cdb0ewm]

Checked all connections but everything is fine
Thanks

[Burgerman]

You have 2 sets full. The hyperion only balances at 300mA. So its going to take a long time. So set it to charge and just leave it charging. Be patient. Go to the pub, sleep... This is why you must make sure every group has the same natural self discharge, and that all should be fully charged before assembly.

Once a set of cells reaches 3.6v the charger has to slow down to 300mA charge approx. Because the full cells must be pulled down, at 300mA to stay within limits. For what its worth a BMS is typically 50mA to 100mA, and the PL8 is 1000mA. So 3x slower, and 3+ faster.

[cdb0ewm]

thanks, just fired them up again

[Burgerman]

Once balanced. Change to 3.65v and top up again, and leave it in a 3.65v state on charge, after its balanced, for another hour. Then set everything back to 3.600V.

[Burgerman]

Is it balanced yet?

[cdb0ewm]

Hears 15 hours of charge

Should I try to charge of the 11 packs individually

[Burgerman]

No just leave it. Says a few mins? Will get there in the end if healthy. Back to pub. Why did you not charge all full to start with?

Or if you are desperate to speed it up, take a couple of car headlamp bulbs about 100 to 150 watts, to use as a resisive load, and put them across the two higher groups. To remove charge from the high ones, it will help the balancer.

That will allow the charger to charge at a high rate for all the other cells. Keep charging as you do it and watch the screen. The bulbs will not illuminate... Much. Connect a few more if you have any.

[cdb0ewm]

Thanks

Not in a rush.

Could I isolate the 2 packs and use the Hyperion discharge function?

[Burgerman]

How could you know how much?

[cdb0ewm]

Well, everything is back to full charge. It did take a while.

Thanks for your help

[cdb0ewm]

OK...I seems I'm still have a balancing issue. I charged the pack again, all 13 cells started out around 3.2v. One cell went to 3.6v within 20 minutes and the other cells moved to 3.3v within an hour. After 11 hours of charging the one cell remained at 3.6 and the other 12 cells got to around 3.4v.

Any suggestions and what am I doing wrong.

Thanks

[Burgerman]

Bad connection somewhere lets the full one seem full because its only charging 1 to 12 cells... And thats likely it.
Bad cells somewhere in all the rest...
Wired incorrectly.

Only options are above. KEEP charging till all are full. Measure volts on each individual parallel group by cell. All should be the same. If some are lower you have a bad connection.

Dont know what you did, history of pack, care in assembly, but all must be 100%. Are you using this? Is anything connected to it?

Charge and capacity test each parallel group indipendently. Check individual cell voltages on this group as you do. All must be the same or you have connection problems.

A PL8 will balance it 3x faster, but you need 2.

[Burgerman]

Just looked back in history. A mix of rescued cells, new cells. From a badly built pack that was allowed to go dead.

It would seem that you need to rebuild it properly, cell by cell, testing very carefully by capacity, impedance, self discharge etc. To find all the less than good ones. But first, once its FULLY charged, do a discharge test.

Set discharge amps to max. And post the discharge curve. It will take a day or more. But only do about 5 or 6 hours for now. To look for obvious issues. And then recharge and post that graph too.

[cdb0ewm]

ok will let you know

thks

[Burgerman]

How long since last charge?

[robert97]

had an "infant mortality" problem with a newly built pack.
initial testing on all seventy cells showed all good.
then this charge problem showed.

problem balance

took the second column apart and found this leakage...

corrosion

replaced it and all looks good again.

[shirley_hkg]

Is it cell #2 ?

[Burgerman]

Thats the 1st actual bad cell that want caused by under or over voltage I have seen on headways. Looks like it was bad from manufacture.

[robert97]

of about 150 headway 12Ah cells, my only problem cell.
and it failed/leaked before the pack was in service.

to shirley: it was in cell group 2. actually the end one #4 in pic.

cell#4 leaked in this pack

[Burgerman]

Those crimped balance wires will fail. They will bend back and forth against the crimped terminal and fall off. The far left and far right one already look dodgy. Crimping sucks. I crimped the OUTER cable covering gently, and soldered the copper. And added a short heat shrink.

[robert97]

thanks for the note.
the previous pic: "cell#4 leaked in this pack" shows a temporary balance cable used during build.

here is the good one:

actual pack with soldered/heatshrink wires

[Burgerman]

That looks better!

[cdb0ewm]

When you say "Set discharge amps to max" I'm assuming you mean to change the Hyperion profile amp field to as high as it goes or is there a specific amp value to use?

Just checking because I don't want to screw up the pack...again

[Burgerman]

Max. But thats only going to be 80 watts. So under 2 amps. Set it to 2 or more. I just want to see the first few hours discharge. After it finally charges.

You cant screw it up. It can do 10C so 72Ah x 10 or 720Amps continuous!

[cdb0ewm]

OK...i'm still having a problem with 2 of the packs. This graph shows the 13 packs after 15+ hours of charging (I know the time shows less, I stopped and restarted when I had to leave) There are 2 packs that just won't move above 3.36+.

As you directed I put the whole 13 packs on discharge. See the graph below

Any suggestions as to next steps (I checked all connections and wiring)
Thanks

[Burgerman]

Was it FULL on all cells BEFORE you discharged it? If not then its no help. You must charge ALL parallel cell groups to 100% first.

You must charge till all cell groups are full. As it is you keep stopping. 15 hours is terrible, but if it takes longer it just does. It can only balance at 300mA so if a cell group is still 6Ah out, it will take 18 hours more.

Charge till its just got those two groups left behind, then charge them alone at 3.600V to catch up with the others. Or just wait.

[shirley_hkg]

I've seen cells that are still good at 90% capacity , but they can't hold the charge above 3.37V . They induce 500mA self-discharge current @3.5V . SMALL balancing current can never bring them up to 3.6V.
Besides , they will drop to 3.37V in minutes if charge stopped , and stay there as normal .

That might not reduce usage much , but will hinder a normal balance certainly .

If that two cells of yours have been fully charged before , I'd suggest to dismantle them . Hook them up as 1PXS to see how they are charged individually . Hope you can spot the bad guys .

[Burgerman]

Just charge the two low groups fully! Seperately if needed. Only then after charging as a full pack and letting it balance, can you do a quick discharge and recharge to see if it all behaves correctly when recharged again.

I warn everyone that building something as complex as a BM3/Roboteq/lithium chair isnt simple. And cannot be farmed out to "engineers". And you must understand it all. And you must get it right. All of it, from the start. Nobody listens. So the result is things like this. I could sort out that pack, have it balanced, repaired if needed, and up and running easily. And this is the easy bit.