In reference to post #350, I'm referencing the entire post but don't want to quote the whole thing, since it is rather long
Quote:
Originally Posted by crx_rogus
But the BCM can only look the cells 12 at a time; I hope it can detect the voltage collapse of a single cell in each of those 12-cell strings.
...
Fast Charger:
Charge C-rate: 1C
Charge time: 1h+
Temp range: 50 - 113 deg.F (10 - 45deg.C)
Termination method: NdV responds to higher settings; uses dT/dt, voltage plateau, absolute temperature and time-out-timer.
If the Insight's cells are 6.4Ah rated, 1C = 6.4A, 0.5C = 3.2A, 0.3C = 1.92A, 0.1C = 640mA unless I'm clueless.
Before the table, it mentions "NiMH batteries should be rapid charged rather than slow charged. ... Because a NiMH battery does not absorb overcharge well, ... The recommended trickle charge for the NiMH battery is a low 0.05C [320mA in our case if I have my numbers right]."
That suggests our packs shouldn't even see the 350mA as a real charging current, but that defies observed reality. It also suggests that beefing up the chargers and making them temperature-sensitive would be a good idea. Beefing them up to where the cells overheat even with a full 12V to the fan would however not be a good idea. As weather changes, I suspect acceptable sustained charger power levels should also change.
|
The only issue that I have with fast charging these that I've come across is in order to charge 1C, there are some cells on a stick that get hotter than others. I just charged a stick at slightly under .5C of it's discharged capacity, at 3 amps the two far outside cells didn't even get to 30C, two of them got to 40C and one to 45C which after I noticed this and terminated the charge, it went to 50C within about 10 minutes. I've charged this stick at a lower rate followed by a trickle charge and I've manually measured voltages and they all experience voltage drop and at lower rates they seem to raise temperature in a uniform fashion within the same stick to where if I stop charging with the coolest stick at 35C, the hottest is at 40C with the odd part of this being that every time I cycle it's not always the same cell(s) that ends up the hottest either.
The cells are rated at 6.5Ah. They seem to take a 2C charge pretty well at the stick level but grid charging that at the pack level without temperature monitoring can get them hot enough to hit 45C if a charger doesn't terminate through dV, which with an unbalanced pack will not happen in time to prevent an overheat with the cells at a lower SOC since they are still rising in voltage and dV isn't pronounced enough at lower currents either for effective detection, which is where we are sitting at around C/20. I've read that they can handle the overcharge through recombination with C/20, ideally it shouldn't be done without a purpose for longer than it needs to be done as you eventually reach the point where all cells are topped off and energy is being wasted as heat. It would be nice to have a 1 amp charge with an end of charge detection scheme but I think that the closest that we can get easily without popping over our packs and throwing temperature sensors all over and to monitor things or wire into the PTC strips(which would likely be much easier) is to have the charger trip off with Mike's design at a predefined voltage that occurs somewhere near a full charge.
Mike, The switchover point of your charger being lower than the ending final voltage of the pack at 350mA might be too low considering that the higher rate would have a higher terminal voltage, we might need to find a way to calculate an ideal voltage for that switch-off of the higher current to occur, granted after they are balanced would be the safest time to do that but if it's done too early, it might take a significant time for the ending top-off. I'm not sure how exactly to measure this other than playing with my hobby charger to find the numbers but many(if not most) of my sticks are damaged, worn out, or below capacity, and likely wouldn't produce a reliable result in testing. ...although the average Insight pack might have its fair share of bad apples even if it performs as if it were perfectly healthy. I figure that once the 'full pack voltage at 350ma' is figured out and the pack trickled a few times for a long enough duration to have a pretty safe idea that things are balanced that raising the voltage of the switch over to lower current could be done. If the full pack voltage at 350mA sticks at, say, 170 volts or so, if we end at 170 volts when charging at 1050mA, I think it's got some distance to go with charging once it drops down to 350mA. I'm just not sure how much though. I'm looking to find out however since if I go with a NiMh buddy pack instead of a LiFePO4 pack replacement, I'll want to be able to get a full charge within a reasonable period, I figure if I can manage a full charge, or close to it in about 10 hours I would be set, I think this is pretty easy to achieve though, 1 amp per pack, if not completely empty would be about 6Ah, charging to a termination point where it has all but around 1Ah in, drop to 350mA and cut the charge after about 5 hours of that would likely ensure a full charge in 10 hours or so, right? That's considering that when slow charging that the factor is 1.6 or 16 hours to fully charge at .1C since the charge efficiency drops with lower currents, not sure where things stand at .05C though.
I'll be thinking about this, I'll eventually get the time to play with the concept of finding a decent termination voltage that leaves about 1Ah left over.
