safety Safety - charging overnight

[Keith]

All,

I have noticed a few times, most recently this week in this discussion: http://forum.electricunicycle.org/topic/1892-tg-t3-advice-wanted/?do=findComment&comment=24160, advice to "try charging it overnight" or "I charge it overnight so it is always ready the next morning" etc. The former in reference to a battery, charger, or control board that had a problem the user was trying to diagnose.

Now a good quality charger, BMS and cells should be perfectly safe to do that providing none of the electronics fails in any way. However even good electronics might fail, and quality isn't always good. Above all, even if the risk is a tiny fraction of 1%, the energy in these cells if a fault does occur is high, any fault that allows even one cell in a lithium Ion pack to significantly charge above 4.25v is a potential fire risk. 

Please can I respectfully suggest you do not leave chargers unattended for long periods and certainly not overnight unless you are absolutely certain it is in a fireproof area with good ventilation (like the middle of an open garden space). 

i do not wish to be alarmist, I've been using much more easily damaged Lithium Polymer batteries for years and I've never had a problem because I am careful with them, but I have burnt a hole in a sofa charging a laptop because I didn't give charging a piece of everyday consumer electronics a second thought.

Ideally you should try to disconnect a charger as soon as possible after it indicates charging is complete. Lithium batteries do not trickle charge like NiMh's so there is no benefit in leaving them any longer, only a still present risk.

[Jason McNeil]

For the principle of better-safe-than-sorry, agree with Keith. Although it's a shame that a few lazy back-alley producers are causing general paranoia & uncertainty about perfectly safe technology/products; the challenge is to outlaw the bad producers.

Here is an example test report with the Samsung 25R battery cell, their definition of over-charging is 4x more current & nearly 4x more voltage than the specification, it's obviously irreversibly damaged but no battery fire or other collateral damage. 

[zlymex]

When the indicator turns to green, there is still a charging current of about 250mA to 300mA so it is not charge complete indication.
Even one hour after turns to green, there is still a charging current of about 60mA(for one pack) and remain this current for as long as it plugs in. This is not trickle charge current, rather, it's the balance current go through balance resistors in the BMS when the cells are perfectly balanced. In order for cells to be balanced, my suggestion is to continue charge 3 hours after the charger LED turns green once every week.

Every BMS in a EU has the balance function build-in. If always disconnect a charger as soon as the indicator turns to green, the cells will never have chance to balance.

[SlowMo]

Is the theory of not charging to full capacity to prolong the useful life of the batteries as stated by others not correct?

[Jason McNeil]

11 minutes ago, SlowMo said:

Is the theory of not charging to full capacity to prolong the useful life of the batteries as stated by others not correct?

It's correct, the practice is common practice in the EV world to maximize cell lifespan. 

[Blackwheel Jack]

42 minutes ago, SlowMo said:

 

Is the theory of not charging to full capacity to prolong the useful life of the batteries as stated by others not correct?

 

25 minutes ago, Jason McNeil said:

It's correct, the practice is common practice in the EV world to maximize cell lifespan. 

I'm very confused by this exchange - 

On the one hand Zlymex is saying we should charge fully and leave the charger connected - that is also what appears in the Airwheel instructions - so that cell balancing takes place.

On the other hand, Jason, you're confirming that it is good practice to NOT fully charge in order to prolong battery life. (excuse the split infinitive...)

What am I not understanding?

 

[Jason McNeil]

Both claims are correct: ideally the BMS would have a setting that would allow the owner to set the level at which they are charged, problems occur when cells are topped-up at different levels which is what Zlymex was referring to.  

[esaj]

6 minutes ago, Blackwheel Jack said:

I'm very confused by this exchange - 

On the one hand Zlymex is saying we should charge fully and leave the charger connected - that is also what appears in the Airwheel instructions - so that cell balancing takes place.

On the other hand, Jason, you're confirming that it is good practice to NOT fully charge in order to prolong battery life. (excuse the split infinitive...)

What am I not understanding?

 

From my understanding, they're both right   Charging the cells to full voltage stresses them more than charging to a lower voltage, and in the long run, starts to affect the lifetime of the cells (by causing a drop in total capacity over time):

In terms of longevity, the optimal charge voltage is 3.92V/cell. Battery experts believe that this threshold eliminates all voltage-related stresses; going lower may not gain further benefits but induce other symptoms. (See BU-808b: What causes Li-ion to die?) Table 4 summarizes the capacity as function of charge levels. All values are estimated.

Charge level(V/cell)	Discharge cycles	Capacity at full charge	Table 4: Discharge cycles and capacity as a function of charge voltage limit Every 0.10V drop below 4.20V/cell doubles the cycle but holds less capacity. Raising the voltage above 4.20V/cell would shorten the life. Guideline: Every 70mV reduction in charge voltage keeps 10% of usable capacity vacant
[4.30] 4.20 4.10 4.00 3.92	[150 – 250] 300 – 500 600 – 1,000 1,200 – 2,000 2,400 – 4,000	~[114%] 100% ~86% ~72% ~58%

Most chargers for mobile phones, laptops, tablets and digital cameras bring the Li-ion battery to 4.20V/cell. This allows maximum capacity, because the consumer wants nothing less than optimal runtime. Industry, on the other hand, is more concerned about longevity and may choose lower voltage thresholds. Satellites and electric vehicles are examples where the importance of longevity surpasses harvesting maximum capacity.

Source: http://batteryuniversity.com/learn/article/how_to_prolong_lithium_based_batteries

Also the depth of discharge (how empty you ride them before recharge), use and storage temperatures and charging & discharging currents affect the overall lifetime of the cells.

On the other hand, it seems that the BMSs used in the packs don't start balancing until near the full voltage. Here's some random BMS spec sheet I've used as an example before:

Note that it says "Balance voltage:  4.20 +- 0.035V", so I suspect it won't start balancing until near 4.2V per cell. As the balancing can take a while (it happens with low currents), you need to keep it charging still after the light turns green. Here's a charging graph from my original 264Wh (nominal) Firewheel batteries charged all the way to full (until Charge Doctor showed somethin like 10mA or less):

Don't remember the specific values, but the charging light turned green around 250mA charging current, which occurred somewhere around the 150-minute mark (2.5 hours). Yet, the charging still continued over an hour before the charging current (pretty much) completely died off.

I could be wrong, but I don't think it's dangerous or very taxing on the batteries to do a full charge every once in a while to get the cells balanced. I think it's more important not to leave the batteries fully charged, especially for longer periods (weeks, months?) and in elevated temperatures.

[Blackwheel Jack]

OK - I asked for that.

Going for a lie down now.....?

[Jason McNeil]

Following on from @esaj (as always) excellent post, I'm in contact with one of the most intelligent & knowledgeable suppliers I've worked with to incorporate a toggle 60v/67v toggle switch. Will need some extensive testing, but the concept will be to offer 60v 'regular' daily charging, then once a week recommend that owners use the 67v setting to top-up the cells. 

[DangerousDick]

Trouble with not 'topping' EVERY cell off is that the imbalance could get worse over a period of time.  If one cell in the pack drops too low to recover when the rest are half way that one cell will be destroyed and the pack no use.

By charging - and then allowing all the cells to reach the same voltage by the electronics 'bleeding' off excess voltage from the full cells to allow all the weaker ones to catch up - you ensure that you start every cell from the maximum, and theoretically at least, they'll all discharge to the same level.

More complex than I can explain and the above posts from esaj and Jason are spot on, just needs a bit more understanding to get the most out of the knowledge provided here .

I suppose ideally you'd give a full 'balance' charge every 10th or 15th charge, and the rest just short of full to really get the most out of a pack.  Then again, you wouldn't want to remove that charge as quick as we do to extend as well - that means not necessarily pushing the top speed like most of us do, and avoiding high inclines.  Then there's keeping them out of the cold and not bouncing them up and down curbs too but how many of us appreciate all these little things which combine to reduce the possible life a battery can offer and just go out to enjoy ourselves .

Your battery won't last forever, bank / budget etc. on replacing it every couple of years if occasional 'leisure' users, or may be even more than once a year if regular 'commuter' users?  At least if you're not expecting too much, then most of the time you'll be more than satisfied?

[Colestien]

  General Paranoia, yap that's it .  Don't care what it is called.  After multiple reports of Hoverboard fires, I will stop charging when I'm not around/asleep.  I've got a lazy (cold) day today, so I guess this a "balance" the battery day.  Thank you all for the insight into the charger.  I've had a habit of hovering over the Wheel, waiting for the charger to go green.  Green is for GO. That's exactly what I've been doing.  Now I now it OK to let it set green for a while every so often.  

[Cranium]

I was under the impression that the BMS balancing of lower voltage cells in was separately controlled from the charging and thus continuous in operation.

I agree that if you truly want a fully charged battery, leave the charge plugged in a while after the light goes green; however, the difference will be very minimal.  On the Ninebot, the charger light turns green when current drops below 120mA.  This means that over a 3 hour period as the current continues to drop, you would likely receive ~ 0.25AH which is negligible in comparison to the battery size.  

[esaj]

42 minutes ago, DangerousDick said:

Trouble with not 'topping' EVERY cell off is that the imbalance could get worse over a period of time.  If one cell in the pack drops too low to recover when the rest are half way that one cell will be destroyed and the pack no use.

By charging - and then allowing all the cells to reach the same voltage by the electronics 'bleeding' off excess voltage from the full cells to allow all the weaker ones to catch up - you ensure that you start every cell from the maximum, and theoretically at least, they'll all discharge to the same level.

More complex than I can explain and the above posts from esaj and Jason are spot on, just needs a bit more understanding to get the most out of the knowledge provided here .

I suppose ideally you'd give a full 'balance' charge every 10th or 15th charge, and the rest just short of full to really get the most out of a pack.  Then again, you wouldn't want to remove that charge as quick as we do to extend as well - that means not necessarily pushing the top speed like most of us do, and avoiding high inclines.  Then there's keeping them out of the cold and not bouncing them up and down curbs too but how many of us appreciate all these little things which combine to reduce the possible life a battery can offer and just go out to enjoy ourselves .

Your battery won't last forever, bank / budget etc. on replacing it every couple of years if occasional 'leisure' users, or may be even more than once a year if regular 'commuter' users?  At least if you're not expecting too much, then most of the time you'll be more than satisfied?

Yeah, I think you can become a bit too cautious over the battery if you think about it too much... the wheel's not much fun if you only most of the time try to keep using only down to, say, 40% and charging up to 80%, especially if you have a small battery, and ride it at walking pace to make the pack last for as long as possible. Especially for the lower quality wheels, I wouldn't be surprised if something else breaks before the battery any way.   But then again, for large multi-packs and more expensive wheels, you can probably pull high currents from the pack without any care, and doing the "don't ride to empty and don't always charge to full" -routine should keep your pack healthy for a much longer while.

About the voltages being off, I'm not sure if it has much other effect (cell reversal maybe in extreme situations?) in series connection except lower than normal voltage, but at least between parallel cells or batteries, this is what happens (at least to best of my knowledge):

The batteries (or cells) in parallel form a closed circuit. Current flows (according to the convention) from more positive potential towards more negative potential (and in reality, in the opposite direction, blame Ben Franklin, but the actual direction of the current is not that much of the issue here). If both the cells/batteries in parallel are at the same voltage, the potential difference is essentially 0V and no current will flow. However, if the voltages are off and there's a closed circuit, current will begin to flow due to the difference. The voltage difference will settle once the fuller cell/battery discharges itself enough to charge the other until an equilibrium is reached, but just as a somewhat radical example, here's what happens if you connect two or more 16S-batteries in parallel with one at 60V (3.75V per cell on average)  and the rest at 67.2V (full 4.2V per cell):

I have marked the conventional directions of the currents as arrows. In the first example, Bat1 is at 67.2V, and Bat2 is at 60V when they're being connected in parallel. Both have internal resistance of 0.48ohm (16 cells in series, each cell with 30milliohm = 0.03ohm resistance). How much current flows through the circuit? I hope I got this correct (I did use LTSpice-simulation to check at least the currents and dissipation powers ): I = U/R, where U (the voltage) is the difference between the voltages of the batteries: 67.2V-60V = 7.2V and the total resistance of the circuit formed by both batteries is 0.96ohm. Here we get: I = 7.2V / 0.96ohm = 7.5A. As the current does nothing else except heat up the batteries, how much power is being dissipated as heat? The voltage DROP to the 0V -ground potential over the lower voltage battery is (V_bat1 + V_bat2) / 2 = (67.2V + 60V) / 2 = 63.6V. The power being dissipated as heat is P = U*I, in this case 63.6V * 7.5A = 477 watts. Not that much if you compare to about how much the motor can draw (especially during peaks), but then again, in that case all the power isn't being dissipated only by the battery, but by the motor, mosfets etc...

Then we have the lower example, this time with three similar batteries at 67.2V (Bat4-6), and a fourth (Bat3) with 60V voltage added. This time, you actually have more like 3 resistances (Bat4-6) in parallel and one in series with them. What happens to paralleled resistance? The resistance drops, and the current becomes even higher. In this case, the total resistance of the circuit is R_Bat3 + 1 / ((1/R_Bat4) + (1/R_Bat5) + (1/R_Bat6)) = 0.48 + 1 / ((1/0.48) + (1/0.48) + (1/0.48)) = 0.64Ohm. So the current will now actually be... 7.2V / 0.64Ohm = 11.25A. The voltage drop over Bat3 is (60V + 3 * 67.2V) / 4 = 65.4V, so the power dissipated as heat is 735.75W. Time to duck and cover, I think.

And that's why you should charge your batteries to similar voltage before connecting them in parallel.

Please correct if I made any mistakes above.

[Cranium]

@esaj  The problem I have with many of the charts and test that are on battery university is that they don't state all of the test conditions for the results.  And the actual research papers that they get their information from aren't always available without a subscription to the site where the paper is hosted or a single paper purchase option.  A couple of questions I didn't find clear answers to was 1.  What state was the battery in at the beginning of the charging cycle?  2.  What constitutes a discharge cycle in this testing?  3. What is the correlation between charge capacity and depth of discharge?  4. What is the final state of the battery at the end of the rated discharge cycles (percentage of battery decrease)?

The chart for charge levels you referenced is very surprising and I don't know of any manufacturer or charging system I've used for RC LiPo batteries that recommends charging at <100% of the capacity to prolong battery life.  But until I find something that contradicts the table, I will assume it to be accurate.

Charge level(V/cell)	Discharge cycles	Capacity at full charge	Table 4: Discharge cycles and capacity as a function of charge voltage limit Every 0.10V drop below 4.20V/cell doubles the cycle but holds less capacity. Raising the voltage above 4.20V/cell would shorten the life. Guideline: Every 70mV reduction in charge voltage keeps 10% of usable capacity vacant
[4.30] 4.20 4.10 4.00 3.92	[150 – 250] 300 – 500 600 – 1,000 1,200 – 2,000 2,400 – 4,000	~[114%] 100% ~86% ~72% ~58%

Also as the article states:  If you are recharging your EU after each ride and not riding until the battery is dead, the longer the battery will last. One should try to avoid full discharges and charge the battery more often between uses. Partially discharging Lithium batteries is fine. 

Depth of discharge	Discharge cycles	Table 2: Cycle life as a function of depth of discharge A partial discharge reduces stress and prolongs battery life. Elevated temperature and high currents also affect cycle life.
100% DoD 50% DoD 25% DoD 10% DoD	300 – 500 1,200 – 1,500 2,000 – 2,500 3,750 – 4,700

If you look at both of these, it appears that the Depth of discharge has a greater effect on the life of the battery than the charge level.  For instance if I charge to ~75% of the battery capacity, I'm getting between 1000-1200 discharge cycles.  But if I charge to 100% and then discharge to 25% (using the same 75% of the capacity), I'm getting 2,000-2,500 discharge cycles.  But the article doesn't really state the starting charge OR state the correlation between the two tables so this has some conjecture involved.

@Jason McNeil, your idea of only partially charging your pack is interesting and would certainly prolong life but why limit the range of your EU?  I certainly wouldn't want to become stranded somewhere because I chose not to fully charge my pack.  Perhaps with the really large packs it would not be as big of a concern.  

If you put aside $0.50 each time you charge the batteries you will always have more than enough money to purchase new batteries.  This would be much more appealing to me than knowing that I'm almost always operating at 75% capacity with the benefit of only having to put aside aside $0.20 each time I charge it for new batteries.

[esaj]

10 minutes ago, Cranium said:

@esaj  The problem I have with many of the charts and test that are on battery university is that they don't state all of the test conditions for the results.  And the actual research papers that they get their information from aren't always available without a subscription to the site where the paper is hosted or a single paper purchase option.  A couple of questions I didn't find clear answers to was 1.  What state was the battery in at the beginning of the charging cycle?  2.  What constitutes a discharge cycle in this testing?  3. What is the correlation between charge capacity and depth of discharge?  4. What is the final state of the battery at the end of the rated discharge cycles (percentage of battery decrease)?

I cannot tell, as the article isn't that clear on them, but I'd guess that the voltage at the beginning of the charge test would be the "cut-off" voltage (whatever that may be for the cell in their example, 2.5V?), a discharge cycle would be from full to empty (full being the voltage the battery was charged to, so in terms of mAh, less than for a more fully charged cell?), depth of discharge tests would use full 4.2V voltage as starting point and for the last one 

Table 2 compares the number of discharge/charge cycles Li-ion can deliver at various DoD levels before the battery capacity drops to 70 percent. 

But, I'm mostly just guessing here. Battery University does have some discrepancies and out-dated information here and there, but it's the best source for general battery information I've found in concise form without going too much into the details like specific electrochemical reactions (which go way above my head ). Also, the company (Cadex) that maintains the pages sells stuff like battery testing equipment, chargers and (I think) batteries, so they sometimes might have some hidden agenda... Now I need a tinfoil hat  

[Cranium]

2 hours ago, esaj said:

I cannot tell, as the article isn't that clear on them, but I'd guess that the voltage at the beginning of the charge test would be the "cut-off" voltage (whatever that may be for the cell in their example, 2.5V?), a discharge cycle would be from full to empty (full being the voltage the battery was charged to, so in terms of mAh, less than for a more fully charged cell?), depth of discharge tests would use full 4.2V voltage as starting point and for the last one 

If we are going to assume the charge cycle in the table above is from empty and the discharge cycle in the table above is to empty then for the vast majority of people charging should not ever really be a worry.  They typical 1000+ charge cycles would apply.  

Charge it up as much as you like and recharge after your rides!

Storage is another story.  With RC LiPo batteries, if you are going to store them for long periods without use (>1 month) then it is best to store them at 50% charge.  My LiPo chargers have a storage setting that will only charge them to 50%.  I don't know what the recommended storage method is for these EU batteries.

 

[WakefulTraveller]

So I have this electric outlet timed cutout thing (the power will turn off 1/2 hour / 3 hours / 6 hours after pressing the button) and I'm wondering if I can use that to avoid over charging the EUC. It won't damage anything to just cut the power like that, will it?

[Gimlet]

Out of interest, as I understand it the balancing only takes place whilst the charger is connected because it is needed to complete the circuit through all the cells in the battery packs.

Will this not still happen if the batteries are only charged to 60V but left connected for a few hours for balancing?

The circuits would still be complete to allow the flow to the weaker cells but no further power would be added once the charger voltage and battery voltage were both equalised at 60V.

[esaj]

39 minutes ago, WakefulTraveller said:

So I have this electric outlet timed cutout thing (the power will turn off 1/2 hours / 3 hours / 6 hours after pressing the button) and I'm wondering if I can use that to avoid over charging the EUC. It won't damage anything to just cut the power like that, will it?

AFAIK, no, it shouldn't cause any damage. Rationally, if the pack or charger could  be somehow damaged by cutting the power, how would you ever turn it off, since there (usually) is no on/off-switch, you just pull the plug from the wall outlet?

15 minutes ago, Gimlet said:

Out of interest, as I understand it the balancing only takes place whilst the charger is connected because it is needed to complete the circuit through all the cells in the battery packs.

Will this not still happen if the batteries are only charged to 60V but left connected for a few hours for balancing?

The circuits would still be complete to allow the flow to the weaker cells but no further power would be added once the charger voltage and battery voltage were both equalised at 60V.

I guess some BMSs could be "smart" enough to handle the balancing "properly", but from what I've gathered, many use some form of a "bypass" for the cell, so that once the cell reaches the full 4.2V or so, the current won't pass through the cell anymore, but through something in parallel with the cell (like a 4.2V zener or something). It does (sort of) balance the cells (all get charged to 4.2V) when charging to full voltage, but nothing will happen unless you charge it all the way through to around 67.2V (and then keep in charge for some time until the current drops to some really small value, ie. all the cells have reached 4.2V and it's only the current flowing through the bypasses). But, I could be wrong

[UKJ]

I am trusting my "charge doctor" more and more to turn off at 90% but also noted that to balance takes a few hours after led goes green and do this about every 10 or so.

ukj

[Keith]

My understanding is as @esaj has stated above. Some BMS may be sophisticated enough to keep all of the cells at the same voltage at all points during charge but many seem to simply shunt the charge around cells once they reach 4.2V, and therefore won't completely balance the pack unless the charger is left on for a few hours after charge shows complete. (And will not balance at all if a lower charge voltage is used) However, if the cells go significantly out of balance after less than 10 lower voltage charge cycles the pack is probably on its way out anyway. Cells are damaged by going to too low or too high a voltage, the BMS should protect from a cell getting too high. However a cell that is out of balance low is it most risk of harm if the battery is discharged near fully, so balancing is probably most important before a long run.

Do not lose sight of my original post though. Whilst the risks may well be very, very low, regularly leaving the charger on all night is betting your life that nothing in the charger or BMS will fail while you are asleep. If you must do it in your house at least make sure you have a good smoke detector!