Is this normal balancing behaviour? (Junso Charger)

[Toombaa]

Dear e-rollers,

First off all thanks to the strong stakeholders of this forum...the info in this forum (and the grin technolog website) is very awesome for a chemist, electronics, mecanics and finially bike lover such as me.
So just finished I constructing my first (already know it's not going to be my last) ebike.

I made 2x9S packs hanging ons resp. left and right side bike bags next to the back wheels for a low gravity and very steady bike.

This morning however something unexpected happened, had to frown when seeing the strange shape of the charging curves I obtained with me 2x Junso 1010B+ chargers this morning;

balancing.jpg

(136.75 KiB) Downloaded 2 times

After the first cell reached limiting voltage (3,65V) the others start to loose charge and the voltage differance between them gets again much larger??? also the input power drops very rapidly to +/- 10W...
If you watch cell voltages on the charger some are blinking others not...seemingly randomly changing from blinking to non-blinking state with a 1 minute ritmus.

Anyone shed some light as to why the cells suddenly start loosing energy? Have others observed the same? Is there anywhere a description to be found for the balancing behaviour to be found?
Is this considered heathly for the batteries? (seen the very limited amount of energy going in in this last phase I am tempted to stop the charging process earlier or use a lower end voltage)

(charger is set for a constant balancing behaviour (frmo the start of the charge on) / without trickle balancing / normal balancing speed)

Thanks for sharing with this lost Belgian,

Tom

[Sunder]

This is just a guess, but I've heard most chargers charge through the discharge cable, then drain the highest cells through the balance cables and continue to charge through the discharge cables until all cells match at the cut off point.

BUT - some esoteric chargers do the rapid charge through the discharge cables, then when one cell reaches peak, they trickle every other cell through the balance cables.

Could your charger take the second method? does your manual say anything?

[Truusje79]

Hi Toombaa,

It seems from your graph that your charger starts balancing all the other cells when 1 cell reached the desired voltage ( 3,65 volt ) and that takes ages.
I have a Icharger 106B and it normally starts balancing the cells right from the start, for instance my slightly sick and abused 3s lipo pack starts with 3,10 3,56 2,96 ( oops ) cell 1 and 3 charging, cell 2 blinking, charging, blinking etc etc... this goes on for 5 - 10 minutes or so and before good cell number 2 reaches 3,7 volt cell 1 and 3 have caught up to cell 2 and will now charge to full 4,15 volt together.

This is only a small lipo pack 1800mAh, but I can imagine larger capacity Headway cells take a lot longer to balance.

Do you have option Li Balance ON
ALWAYS standing on "always"
You can also choose "storage voltage" or " CV phase " but I have it on "ALWAYS"

Hope this helps......

[nlc]

Hum, everything seems normal for me :

When the first cell reach the 3.65V, the charger pass in low current charge mode, approx 10W thus 10W / 3.6V average x 9 cell = ~300mA, IE approx the balancing current (because if charge current is higher than balancing current the first cell will still be overcharged !).

Because the current moves from approx 2.6A (85W/3.6Vx9 cell) to 300mA, the other cell are charged under a very low rate thus their voltage goes down (but they don't loose power, it's only the voltage which goes down because the current is severely reduced !). After their voltage stabilized to the voltage corresponding to the applied current, it begins to slowly goes up, but charge end before all cell reached the 3.65V value.

What we can conclude is that your pack's cell have high capacity unbalancing, and the green cell have a capacity really lower than other !!

These curves are also really interesting because we can really see that after 3.5V the voltage goes up very quickly, which proves that charge to 3.65V is not very useful and just add 5-10% capacity. IMO it can be better to set the HVC to 3.5V for the cell health and longevity

[999zip999]

I have 4 single cell charghers 2a. from a headway guy. He used them to bring up low cells in a large pack. They stop at 3.64-5v. The ping ones stop at 3.8v. The headways can have a balance problem. One or more cells can just be a bit lazy. How's your pack's ah ? You could get 2 single cell charger's to bring up low cell's faster.

[scoot]

Sounds pretty normal to me too... that is, depending how new the packs are, and depending on the actual cells you are using/charging, I think it may be worthwhile to to wait and see how they all behave after some more discharge/charge cycles. As mentioned previously regarding Headway cells), my packs have all needed some break in periods before the cells settled down and acted in unison.

If it were me, I would disconnect from the motor controller when charging for while (just in case there was any current being diverted), and when I have been in a hurry, the single cell top off technique has actually speeded things up for me (also previously mentioned here). BMSs and even balance chargers can be take a long time to completely balance some new packs... at least that's what I have observed in my packs. After that they been better and better as time (cycles) goes on. If you find that after 20 cycles or so, that their is a straggler or two still, then I would try to get a discharge test with a CellLog to see if there are significant differences in discharge characteristics.

[Toombaa]

Thanks for studying my case and your usefull replies!

@truus: yep, de balancing staat op ALWAYS. I guess it's balancing from the beginning, probably the balance current (300mA

according to nlc) is proabably to small to be notice on the graph.


@nlc: Very clear explanation, feel stupid not to have realized earlier myself...the drop @3,65V is off-couse a voltage drop

due to current reduction...stupid me.

I can see Cell 9 is earlier full indeed, possibly it's one if the 2 extra cells I had ordered in a second batch (from evassemble.com whereas first 16s comes from bmsbattery) as backup cells, but ended up adding them to the setup as it was rock steady at 16S....so ironically the 2 spare cells that were ment to keep the pack in good condition no thew make it less balanced ...anyway when reading truusjes values I guess my packs are still not running for the title of most unbalanced packs on the forum.

Will try to do some discharge logs with celllogs and research on the single cell top off technique. (thanks Scott - also packs are very new indeed and were disconnected from controller for charging, will research on the single cell top off technique...but if the diagnosis of less capacity is correct (seems to be supported by the latest charging graph) not sure this is a durable solution).

In an attempt to end the charging process sooner I switched to FAST setting for balancing (stops charging at ICC/5) and charging to just 3,4V. Interesting to see cell 5 (probably the best cell)reaching 3,4V first because of better state of charge at the start and again cell 9 reaching "real" saturation (due to full charge) 3,4 V probably earlier in a second stage. But now a new mystery (at least to me appeared)...What can be the origin of the power blackout (but not fully black...) spikes at the beginning of the graph?

Again, thanks to all, reassured all's pretty normal, can read the curves better already,

Tom

[chilledoutuk]

Out of interest did they have numbers on the two different sets of headway cells you got as they normally state what there capacity range is on them.

Honestly though when building packs with large cells its important as you now know to get the cells all from the same batch.

however if you are running your cells in 2p configuration its possible you could test the capacity of your cells and match them up to keep the capacity as consistant as possible over all the packs.

Failing that you could get yourself a strong balancing charger like the Cell Pro 10s its very accurate and balances from the beginning of the charge process and also can charge through the nodes only which is pretty cool.

I am not sure if all the ichargers balance from the beginning of the charge if they do then those would work well also.

Otherwise you could just leave it going for quite a while to balance up your packs nicely