Thread Rating:
  • 1 Vote(s) - 5 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Top vs bottom balancing?
#31
This might be an obvious point, but for those of us who prefer full SoC batteries, bottom balancing means you'd have (mostly) empty batteries to test/balance your pack.
Active top balancing means you can have full SoC while balancing.
Running off solar, DIY & electronics fan :-)
Reply
#32
You can set any low voltage threshold you like to bottom balance the cells. If you don't want 100% DoD you can go for 80 or 50 with repectively higher end of discharge voltages. You could also do it only once to 100% DoD and then cut off earlier at subsequent cycles. Bottom balancing works on the principle of putting in and taking out known amounts of stored energy. If you work with these numbers you can make it work for any voltage range you like for your cells.
Reply
#33
I am no expert, and I am still learning so take this with a grain of salt.

I have recently been researching LiFePO4 chemistry because it seems to be a much better fit for a small 4S / 12V "suitcase" power-pack. Depending on the components someone might use, 3S or 4S 18650 Li-Ion's might be OK, but regardless...there are a LOT of people still using LiFePO4 because of the perceived inherent safety of these batteries, so there are a lot of useful and helpful youtubes. this is where I started to see a lot of advice from experienced pack-builders to "bottom balance". I still don't have a solid answer as to why it might be better for LiFePO4, but it may be worth continued research.

I have known for a while that any lithium chemistry cell will last much longer if it is not stored at the fullest possible voltage. Our beloved 18650's using NCA/NCM chemistry "can" be charged to 4.2V, but...I now only charge them to a max of 4.1V, and for long-term storage (winter ebike) I leave them at 3.7V.

Since I recently bought the largest 18650 battery I could find (24-Ah) I only charge this one to 4.0V max. My point is...if you do not charge to the max 4.2V, then you have some headroom to have the cells slightly out of balance with no harm (if you are bottom balancing.). If all my cells hit 3.0V at LVC at the same time, then I charge a 7S pack to 4.0V per cell (7S X 4.0V =) 28.0V...one P-group can be at 4.2V and another at 3.8V (wildly out of balance) but the P-string that was at 4.2V would not be harmed. From everything I've seen, none of the cells we are using would ever go that far out of balance.

To transfer this thought over to top-balancing (which is what our BMS's do)...when one P-group drains to the LVC, the others will be close in voltage, but will start to sag, so the pack voltage will start to sag, and the pack-LVC will trip. At the bottom, one P-group will be a little off compared to the next P-group. Once the pack is being bulk-charged, the P-groups would get closer in voltage as they get closer to the full voltage, and once there, the BMS would perform a quick balancing act.

The way I understand it, if the pack is top-balanced, then...you should be able to remove the BMS and you could then cycle the pack several times, with the top voltages always arriving at a close end-result, and the bottom voltages always being a little out of whack (dependent on P-string resistances being different).

Until I get more info, I will bottom balance LiFePO4, and top-balance Li-Ion NCA/NCM 18650's, and neither will be charged to over 90%
Headrc likes this post
Reply
#34
CC CV Top balancing is a no-no. All depends on your charger current control.

Excess charge above 4.2V causes more damage than a lower discharged state and a lot of the top chargers have pre-set balancing voltages and therefore you can easily end up balancing above 4.2V.

If your charger is pushing out 2A and the balancing dump for a cell is only 1A then your cells can be over charged with 1A while balancing. This is why your charge current needs to decrease before getting close 4.2V and decrease below your balancing dump load current as you get to 4.2V. The effect of this is also it takes longer to top balance.

Say you have 4 cells in series 4.1V, 4.1V, 4.1V 4.2V.
Overall 16.5V for the pack
Full charge is 16.8V so your CC CV charger is pushing out 2A
One cell kicks in the balancing load of 1A - that cell is still seeing +1A being pushed into it and the goes over 4.2V until the CV limit is reached and the current falls below 1A and then you still have an overcharged cell while the balancing load brings the voltage back down.

For top charging the charger should either charge at the rate of your balancing current OR reduce current before getting close to the CV level.




This does not occur to the same damaging effect with bottom balancing and the cells can be balanced quicker.
Reply
#35
spinningmagnets:
It's not about what is better for the cells. It's also not about LiFePO. It's simply another method of balancing. Top balancing is usually the standard method as it is done by many BMS and by all balance chargers. Bottom balancing is an alternative method and the aim is to get away without balancing on every cycle, especially with used cells that are prone to drifting.

completelycharged:
That's not how balance chargers work. CCCV top balancing is fine and it is actually the default. You always charge lithium cells CCCV, no matter if top balancing, bottom balancing or not balancing at all. Charging method has nothing to do with the balancing method.
Reply
#36
The cheap BMS units are passive and not active and only rely on a small bypass load resistor before disconnecting compeltely if the voltage goes too high.

I would agree with you for a switch inductance active BMS...

Look at the BMS board here and this guy has even measured 4.6V on a cell....
Left first row is the bypass resistors, 2nd row is the switching, then on the right the larger overall BMS switch disconnect FETS.

https://forum.arduino.cc/index.php?topic=487113.0
Reply
#37
You said charger so I wasn't thinking about a cheap BMS Wink
But this is a perfect example for bottom balancing. In case of a crappy BMS it is probably better to have no BMS at all (just a low voltage cutoff maybe) and balance the battery by bottom balancing.
Reply
#38
True, my english is c**p at times Big Grin

Yeah, the cheap BMS units are ok for 1-3p with a long floats, with the typical setup that I have seen on the site so far the only thing the cheap BMS boards provide is over discharge protection. That is potentially more valuable than the lack of decent balancing... I had 2 lead acids 110Ah that I accidentally flat lined for about 6 hours, lost about 60% capacity and top V was 12.8V... (no emogi for head butting a wall).
Reply
#39
The cell balancing questions posed above appear frequently - likely because good documentation is often lacking with cheaper balancing solutions. Below are a couple of good introductions to most all aspects of balancing. They answer these questions and also discuss some more advanced topics often omitted or glanced over even in better manuals. Both are well-worth bookmarking.

Beware there that are frequent misunderstandings around these matters (some repeated in various posts above). These will be quickly clarified after reading the expositions below.

Barsukov and Qian. Cell-Balancing Techniques: Theory and Implementation. 2013 (This is chapter 4 in the book linked here, and I recommend that you get the entire book since then you will also have introductions to prior topics on which the exposition depends).

Omariba et al. Review of Battery Cell Balancing Methodologies for Optimizing Battery Pack Performance in Electric Vehicles. 2019.

Below is a handy table excerpted from the prior paper.
[Image: 6Lmm9.png]
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)