Getting cozy with LiPo

[hillyterrain]

So I've been using a 12s 5Ah LiPo pack now for a little while and just want to make sure I'm doing everything right, just kinda piece of mind thing

1. I should stop using the LiPo pack when the cell-voltage drops under 3.7V UNDER LOAD, what the voltage without load is is basically irrelevant, is that correct? So a pack-voltage of 44.4V should be my cut-off?

2. When I connect the batteries to the charger all cells are in balance down to 10mV, say 3.81xV, but one cell always comes in 40mV low, e.g. 3.77xV, that cell catches up fast during charging, probably lower internal resistance? Should I replace that cell, is this cell of potential danger or is everything fine here?

3. I'm using the pack with a 20amp controller, so at 4C, Turnigy rates them at 30C cont. I charge them to 4.15V (is that too high, should I just do 4.1V?) And discharge to ~45V pack voltage, I get around 2.4 - 2.6Ah out of the pack (rated 5Ah), is that what I should expect? Would I get more Ah if I run a lower C-rate?

Thanks for sharing your insights!

 

[Ykick]

3.7v under load tends to leave significant capacity on the table. I don't mind 3.3-3.5V per cell UNDER LOAD as long as I can be reasonably certain the balance is good enough and/or cell level LVC solution is employed.

40mV differential doesn't worry me much but certainly a cell I would keep an eye on. Do yourself a favor and verify the cell meaurement with some other meter. Balance charger tolerances can and do drift around over time in my experience.

Cell replacement is a way to extend pack usefulness but a serious undertaking.

 

[wesnewell]

1. No. Set LVC to 42V. That's 3.5V per cell under load. With only a 5ah pack, that should keep resting voltage above 3.6V per cell. Resting voltage is what matters Loaded voltage can vary greatly depending on load . an LVC of 44V is ultra conservative.
2. No, it's fine.
3. 4.15V per cell is fine. You lose a little ah by not going to 4.2V, but it's minimal. Where you are losing the most is on the low end. A 4C rate is nothing to 20C cells. The reason you only get 2.6ah is because you aren't fully charging them, and mostly because you are stopping way before they are down to 0% soc. Setting LVC to 42V will allow you to use down to ~4% soc.

 

[dogman dan]

It's stop at 3.5 resting IMO. Only you know how much sag under load, and that will vary as the cells get really old. But 42 sounds good for an lvc if you don't need that last half mile of range. 40v is also reasonable, since you could have 2v of sag if you need to do an uphill start at the end of a ride.

Nothing wrong with stopping at 3.7v, if you don't need to go further. I think it helps a pack stay balanced. I don't know if it really matters with lipo to leave some in the bottom for longer life span. I think of lipo as 2 years anyway. Maybe longer in cool climates, but 2 years where I live.

I would never start pedaling with the motor off because my packs were at 3.6v, that's for sure. But I'll start trying to use just 70w of motor power. Don't hesitate to use it all if you have to. Just adjust your ride speed next time to avoid doing that every ride.

Charge to 4.1 if you don't need the range, 4.15 or 4.2 if you do. 4.15 is a good compromise number. If you are bulk charging, 4.1 leaves more headroom for safety of course.

Your packs are what I call balanced enough. I wouldn't sweat it. I do like to make a note on the pack though, if one cell shows consistent difference immediately after it has been balanced. When it's getting consistently .15 off, I might chuck it. In the meantime, it should be safe if you are stopping early for sure. But if you run till cutoff, then do a quick check of that marked cell to be sure it's not gone below 2.7v.

 

[hillyterrain]

Thanks for all your replies, I do check with a cell-log to not only rely on the hyperion and balance charge every time.

I have the hope that if something goes wrong with a cell, say one puffs, the hyperion detects a voltage change and stops charging, is that assumption correct?

 

[dnmun]

no, that assumption is not correct.

lipo is considered to be totally discharged at 2.7V and fully charged at 4.2V. most manufacturers use 3V as the minumum safe voltage, but the BMS manufacturers use 2.5V so that the sag under load allows the cell to go to essentially full discharge and it will rebound to about 3V when the load is removed.

 

[wesnewell]

Lipo is totally discharged at ~3.4V resting. Taking it lower than that is just killing it quicker.
http://endless-sphere.com/forums/viewtopic.php?f=14&t=47294

 

[liveforphysics]

lipo is considered to be totally discharged at 2.7V and fully charged at 4.2V.

If you want to technically extract all the electrical energy available in the cell, most lithium ion based cells will survive a charge to ~>4.6vdc and then you can pull it down with a joule-thief circuit and hold it there until it all decomposes away in unreversable electrochemical reactions over a week or so ending up at 0vdc. You might get double or triple the energy.

However, most of us aren't building 1-time use packs.


For special extreme range record breaking or energy/space limited racing where the pack only needs to live a small number of cycles and performance is the critical concern, cycle from 4.2vdc to 2.7vdc if you want, they generally don't just burst into flames using those values if you're lucky enough to have perfectly capacity matched cells to keep some cell from reversing (which is nearly impossible to do with RC cells poor capacity binning/screening).


If you want a battery to use for something like commuting or recreation, 4.2v to 2.7v gives you another ~10-15% capacity in exchange for destroying your battery in perhaps as quickly as 1/10th the time (assuming it doesn't just die from the weakest cell getting reversed on the first attempt of discharging to 2.7v) of cycling your cells from 4.1 to 3.7 (depends greatly on the cell obviously, this is just a very crude ballpark).

 

[liveforphysics]

If that response was overly complex to follow, target 4.1v to 3.7v That leaves a noob a little buffer of safety room for the various random shit that happens in real life.

If you're a pro and you understand what's going on, do whatever the heck you want to do, just be aware you will be buying new cells sooner than someone who keeps the HVC/LVC limits in a milder place.

 

[hillyterrain]

3.7V is resting or under load?

Thanks for the detailed answer!

 

[ZOMGVTEK]

Resting.

If the loads are sane don't pay much attention to what voltage the cell is at while loaded, from a cell perspective. Obviously the IR needs to otherwise be accounted for electrically.

 

[hillyterrain]

So if I only want to charge to 4.1V on the Hyperion I can't use the LiPo setting but have to use the LiIo setting, do they have different charging profiles or is it just fine?

 

[liveforphysics]

So if I only want to charge to 4.1V on the Hyperion I can't use the LiPo setting but have to use the LiIo setting, do they have different charging profiles or is it just fine?

Amps and volts don't have preferences as to the cell type, you're golden.

 

[migueralliart]

If that response was overly complex to follow, target 4.1v to 3.7v That leaves a noob a little buffer of safety room for the various random shit that happens in real life.

If you're a pro and you understand what's going on, do whatever the heck you want to do, just be aware you will be buying new cells sooner than someone who keeps the HVC/LVC limits in a milder place.

I tried suggesting this to methods for his lvc hvc boards and got mixed responses. I guess everyone has a different story.