General > General Technical Chat
LiPo Battery on Holding/Float Charge
Siwastaja:
That is not my intention but I'm not always very good at expressing my intentions. I enjoy the discussion and truly hope you enjoy it too. I have been "accused" of sounding condescending before so it's likely a true observation; but in reality, I want as many as possible to be able to reach as deep understanding as possible; if that wasn't my intention I would write shorter replies only arguing about disagreements, not providing as much information. Or if I felt so superior I wouldn't bother to reply at all to mere mortals, instead sitting comfortable in an ivory tower.
And I'm not saying you are contradicting yourself, I'm saying the Samsung datasheet is in contradiction if interpreted like you did even though your interpretation would be very sensible as it is.
Because in my view, ending up below 3.0V is either because of a faulty cell or violating the specs, but I do admit there can be some interesting corner case I'm not thinking about. For example, this might be a case of me saying a one-in-thousand heavily-used-but-not-abused-for-3-years "cell is faulty" but Samsung saying "it's still barely acceptable". Maybe that imaginary cell increased its self-discharge at 0% SoC from basically zero to say 2%/month which still isn't much but requires precharging step if the pack designer didn't add margin on bottom.
And of course, the fact they are allowing 10µA still at 3.0V in their BMS quiescent current requirement means it can go below 3.0V due to BMS, just sitting on the shelf. But IMO it is just so stupid to first allow slight overdischarge by BMS then allow a recovery precharge. Why not just require the BMS to be better designed not to do that in the first place. Especially because they require the BMS to have Iq < 1µA below 2.5V, if you can achieve this by design you can equally well achieve that at over 3.0V, too! That would solve the whole issue, they are making it more complex than it needs to be.
Also, they allow much higher BMS Iq when in "sleep" mode whatever that is, without defining the criteria for exiting the "sleep" and entering "shut down". Maybe a compliant system is allowed to pull 250µA even at 3.0V when a user switch is in "power on" state or something stupid like this. 250µA would overdischarge an empty cell in no time.
In any case, this doesn't matter so much because I think that your interpretation of the precharge step is better than mine and I would like Samsung to more clearly define it like you interpreted, with 2.5V initial acceptance limit. It's just the I don't believe they meant it that way. In fact that would be even stricter I would like, I personally use something like 1.5V, C/50, 1-2 hour timeout to 3.0V, when I allow the preconditioning at all, that is.
Navigation
[0] Message Index
[*] Previous page
Go to full version