How often should I balance my batteries?

[Denny Paul]

I tried searching, but maybe my search terms are off and I couldn’t find anything readily. 

So I live at the very top of a hill; the only way to get anywhere is to go downhill. Charging to 100% is a pain cause I’ll get tilt back wherever I go. I bought a fast charger from ewheels that has the option to charge to 80% or 90% and love that functionality. So much so that I’d like to use it exclusively, forgoing charging to 100% ever altogether.

However, i think I read somewhere that to balance the battery cells you need to charge it up to 100% occasionally. So two questions: 1) is it really necessary? Providing a guess as to how long it would take for damage to set in would be helpful in assessing this. 2) if it is necessary, how frequently should I charge to 100%?

Edited February 23, 2020 by Denny Paul

[Nic]

If you really want to look after your batteries then you should try never to let then fully discharge as this is when they tend to go out of balance and later fail. If you aim to keep the batteries charged between 20-80% whenever possible they should last many years When your wheel says its low on power then don't just keep riding it on tilt back as this is not going to do the batteries much good because the cells will all have slightly different capacities due to manufacturing tolerance and there might be some cells that are already fully drained. If you are able to check the cell voltages on your EUC App then you will know when/if they need balancing. I suspect many will balance their batteries just out of routine even if it isn't required. I'll let others with more experience with EUCs and balancing advise as they are better placed to answer your query.

[Chriull]

1 hour ago, Denny Paul said:

However, i think I read somewhere that to balance the battery cells you need to charge it up to 100% occasionally. So two questions: 1) is it really necessary? Providing a guess as to how long it would take for damage to set in would be helpful in assessing this. 2) if it is necessary, how frequently should I charge to 100%?

There seems to be no real thourough tests/experience.

Imho it mainly depends on the selection/state of the cells. If they are within a small tolerance they will stay quite balanced by themselves and will need seldom to (almost) no balancing.

If one has bad luck and some cells different to the others one should balance every time - and even that could be not enough to keep the cells in sync...:(

One easy way to get a feeling how ones pack behaves over time is to charge it full and not the reached voltage. If it gets lower it could be/is a sign of not so good balanced cells.

[Rywokast]

its impossible to say.. it will vary wildly from wheel to wheel, different setups, capacities, hardware/firmware, manufacturers (of the batteries as well as the wheel itself). in reality, probably 99% of the time for the expected lifespan of a PEV you wouldnt even need to do it at all, its merely good practice and BETTER for the health of the packs, anyone can get a dud and anyone can get lucky for me i just do it about every month (at least, i will charge to 100% if i have a long ride planned)... for me, i have had one single wheel that has been with me from day 1.. it gets used nearly every day, and every single time i charge the battery to 100%, i have noticed no degradation in battery life, nor has it suffered in any way known to me, and this is supposedly the worst practice lol... these things arent meant to last a life time.. i wouldnt put too much thought into it.. perhaps a few times a year you could do it and walk down the hill lol. now the time you should absolutely be worried is if you see that the battery life has suddenly significantly decreased, or it isnt possible to attain a full charge, that could indicate some sort of fault that could be dangerous to you.. normally this would happen quite suddenly and be very noticeable, a high number of charge cycles will very slowly over time degrade the battery, but this isnt what you should be worried about

[null]

I find going by the Km is easier to pace, 1000 km is about 10 cycles on my 18XL. (How often and how long should be is an unknown, but I find this a good way to time it)

[Planemo]

The Z10 is one wheel where I would advise going against the 'every 10 cycles' suggestion. Because the 2 packs on the Z are individually monitored, a small variance between the two packs (even though the cells in each pack may be level) could cause problems which wouldn't occur on other wheels where the BMS(s) don't see 2 or more packs.

I charge my Z to green light status (and leave it there for a minimum of 3 hours) at least every 3 charges. Constant monitoring of individual cell voltage (via NineBatt app) since new shows they are all very close to each other so I will keep doing the same as I am now.

 

[alcatraz]

I would love to be able to see individual voltages from my phone. It seems that this functionality is only available with more expensive BMSs.

To the OP. If you're worried you could attach balance leads to the battery pack and monitor/balance that way.

I have an older v8 DIY pack where I have two cell groups with accelerated self discharge (bms balancing appears broken or is not available). It means that given a long enough time those groups will drop out of balance. I simply hook them up to a single cell charger (usb 3A current phone charger) and set a timer for lets say 15min and they're back in balance. The other cells have only minor imbalances. Those smaller imbalances I even out with a balance charger every 6 months.

[Planemo]

To be fair even the 'smart' BMS in the Z10 doesn't monitor (or at least, NineBatt doesn't pick it up) individual cells. It only monitors strings of 3 cells. Better than most but certainly, being able to monitor every single cell would be ideal. You could also then add detailed 'smart' monitoring over time (ideally done by and within the app) to pick up any cell which appears to be under-performing and notify the user so they can deal with it before it brings the other cells down, or at least before it takes out another 2 cells (as in the Z).

[Chriull]

29 minutes ago, Planemo said:

t only monitors strings of 3 cells. Better than most but certainly, being able to monitor every single cell would be ideal.

You mean the 3 cells in parallel? The Z10 has 2 packs of ?15s3p?

If so, they cannot be monitored separately - the connection would have to be removed to do so.

The parallel cells will always die/age together...

[Planemo]

2 hours ago, Chriull said:

You mean the 3 cells in parallel? The Z10 has 2 packs of ?15s3p?

If so, they cannot be monitored separately - the connection would have to be removed to do so.

The parallel cells will always die/age together...

14s3p x2 in the Z.

Yeah, we spoke about the connection between 3 cells before, I remember saying something along the lines of there must be a way of isolating cells for the purposes of monitoring but that it hadn't been done yet.

[alcatraz]

9 hours ago, Planemo said:

14s3p x2 in the Z.

Yeah, we spoke about the connection between 3 cells before, I remember saying something along the lines of there must be a way of isolating cells for the purposes of monitoring but that it hadn't been done yet.

Luckily we only have 2-3 cells to check in each group. On big electric vehicles there could be dozens. If you do have a group that's bad you could take the 2-3 cells out and measure their internal resistance which takes a few seconds. The bad cell would most likely have a very different value from the others.

It would be nice if cells could be easily disconnected and removed from a pack. I guess the safety of a good low resistance connection is more important right now. Maybe there will be smaller cells with screw on terminals later.

This together with improved BMS technology with higher balancing currents could keep a pack alive for a human lifespan.

[alcatraz]

In china there are special passive balancing boards now with 2A currents that could fit inside an EUC. Pretty cool.

That's 20-40 times the balancing capability of common bms.

[someguy152]

those boards are cool but now i'd have to remove the shrinkwrap to solder them on

[mrelwood]

My experience consists of a 16S and the MSX, first on which both packs failed (a cell group at 0V), first at 4000km and second at 8000km. This was despite my best efforts to charge to 90% and balance about every 10th charge. The MSX gets balanced at almost every charge, and still going strong at 11000km.

I’m aware that they use different cells and balancing procedures, but at least charging to 90% or less to prolong the battery lifetime seems a moot point.

On 2/24/2020 at 12:01 PM, null said:

I find going by the Km is easier to pace, 1000 km is about 10 cycles on my 18XL. (How often and how long should be is an unknown, but I find this a good way to time it)

This is a good reminder. If constantly charged to 80% instead of 100%, the battery is expected to show noticeable wear only at 1500 cycles instead of 300-600 cycles. By the quote above, even the 300 full cycles would make for about 30000km. Compared to my 4000km and 8000km, battery degradation is a non-issue.

I might take the situation as a catalyst to learn to idle on the EUC, and just start every ride with a minute of back and forth. 

Whatever you end up doing, I’d often monitor the charge voltage and current before removing the charger. Any app should show them. If current is at 0.00V or voltage below the rated max voltage minus 1V, likely the BMS has cut off the charge because of a bad cell inbalance.

[Planemo]

This talk brings me onto another benefit of having wheels with big (1850wh+) packs. Although I usually use most of my pack on a ride, there are times when I don't and may have 50% or so left. This allows me to go out for a little ride with my daughter without having to charge again, unlike my V5F which had to be charged every time I wanted to ride, and for most of the time so did the Z10. It just allows a little more flexibility and in general the bigger packs just don't see as many charging cycles. So I just wanted to point out that it's not all about the added range available per trip from bigger packs, they also let you manage your charges more effectively which ultimately reduces the number of cycles

 

[Rywokast]

31 minutes ago, Planemo said:

This talk brings me onto another benefit of having wheels with big (1850wh+) packs. Although I usually use most of my pack on a ride, there are times when I don't and may have 50% or so left. This allows me to go out for a little ride with my daughter without having to charge again, unlike my V5F which had to be charged every time I wanted to ride, and for most of the time so did the Z10. It just allows a little more flexibility and in general the bigger packs just don't see as many charging cycles. So I just wanted to point out that it's not all about the added range available per trip from bigger packs, they also let you manage your charges more effectively which ultimately reduces the number of cycles

absolutely, a larger battery is better in every single way besides weight, which.. the 99.9999% of the time that you are riding and not carrying it, makes very little difference

[Nic]

7 hours ago, mrelwood said:

My experience consists of a 16S and the MSX, first on which both packs failed (a cell group at 0V), first at 4000km and second at 8000km. This was despite my best efforts to charge to 90% and balance about every 10th charge.

Do you regularly run the pack down to empty?

If a cell group is at 0v then it would most likely have failed when discharged, rather than when charged because that energy would have had to go somewhere. It is probably a good idea to stop riding when you get a low battery warning and not try and limp home.

[mrelwood]

The 16S does lower the riding speed enough for me to avoid the lowest percentage ranges, so it was not something I’d commonly do.

I don’t think a cell pair failure generally happens very suddenly though, but that it would require a longer period of increasing imbalance that the balancing of the charging process just can’t remedy.

[Planemo]

9 hours ago, Rywokast said:

absolutely, a larger battery is better in every single way besides weight, which.. the 99.9999% of the time that you are riding and not carrying it, makes very little difference

Exactly. And with the difference between (for example) a 1600wh and 1860wh msx being only 1.2kg, its a no brainer for me given the total weight of the wheel.

I will add though that I dont carry my wheel anywhere generally

[WI_Hedgehog]

The Battery Management System monitors each group of 3 cells wired in parallel. Each cell in that group will age a bit differently, and kind of average each other out, so the important thing to watch is each resultant group of 3 cells. (The only time per-cell information is important is when one is prematurely failing.)

Each pack should have its own BMS.

As the packs age and with use, the charge states will get more out of balance with each other: some groups will not be as strong as others and will not store as much energy. (I apologize, the rest of this is incorrect, as pointed out by @mrelwood.) When this gets way out of balance the BMS will detect 1 group running low and start to speed limit the EUC early (or tilt-back early)--that's the sign the packs need to be balance charged.

Cells heat up much more when they're low on charge and trying to put out energy, and heat causes much faster cell degradation, so to extend their life, when the BMS causes premature tilt-back, balance-charge the packs.

If one set of cells is fully charging, you'll notice tilt-back when starting out, going down-hill. (time to balance charge)

If one set of cells is getting near low limit during use, you'll notice tilt-back when they get low. (time to balance charge)

If you are storing the ECU for a longer unused perious, balance charge (then bring the battery packs down to storage charge levels if you have a charger that can do so, which in the EUC world is usually not the case).

Edited March 30, 2020 by WI_Hedgehog

[mrelwood]

 

10 minutes ago, WI_Hedgehog said:

When this gets way out of balance the BMS will detect 1 group running low and start to speed limit the EUC early (or tilt-back early)--that's the sign the packs need to be balance charged.

With the exception of the Z10, the BMS on EUCs doesn’t communicate with the mainboard, and can not enforce a tilt-back. Tilt-back is caused by the battery only if the total battery voltage is too low or too high.

10 minutes ago, WI_Hedgehog said:

If one set of cells is fully charging, you'll notice tilt-back when starting out, going down-hill. (time to balance charge)

If the cells are out of balance enough for the BMS to stop the charge before the pack reaches 100%, regenerative braking can increase individual cells past dangerous levels before the overall voltage gets high enough for the firmware to even recognize the overvoltage.

The balancing current on an EUC BMS is likely too small to fix the pack if the imbalance is bad enough to show actual symptoms. If the wheel doesn’t charge to 100%, the charging gets interrupted before any notable balancing can even happen.

That’s why balancing should be constantly taken care of.

[Planemo]

29 minutes ago, WI_Hedgehog said:

so the important thing to watch is each resultant group of 3 cells.

Not sure how you propose to 'monitor' each group of 3 when the only wheel that can currently send this data is the Z10.

[WI_Hedgehog]

1 minute ago, Planemo said:

Not sure how you propose to 'monitor' each group of 3 when the only wheel that can currently send this data is the Z10.

That completely baffles me. I've never seen something that depended so heavily on lithium cells not monitor them (responsibly).

The more time I spend here the more I understand how so many unexpected face-plants occur. Thank you for the information.

[Planemo]

Many of your recent comments, whilst somewhat worthy, simply do not apply (yet) to eucs.

And dont get too hung up about faceplants caused by out of balance cells. The vast majority are nothing to do with the batteries.

[RagingGrandpa]

On 2/23/2020 at 11:35 AM, Denny Paul said:

Denny Paul

• EUC: Ks14c, Mten3, Tesla

Gotway is conservative and will give you beeps and severe tiltback at 3.3V/cell.

Kingsong runs them far deeper.

Continuing to ride after overdischarge is very bad (leads to voltage reversal for the weak cell) but here we're talking about <3V.

I recommend riding the Gotways till they tilt you off, and be more conservative with the Kingsongs...

Recharge including a few hours of balancing before a long ride; store the wheel at low SOC.

Don't sweat the small stuff, if there are cell defects it's bad luck and not your fault.