Video on 80% charging, "Good or Bad?"

[Tawpie]

Why, at my house of course! I'll tend to it, and won't wreck it too badly, but a wheel does require regular "testing" to ensure it hasn't gone bad. You know, just to be sure.

 

 

 

(indoors, under 60% state of charge, 40% is better. Unless it's a Z10, then you must must must charge it every month. it's a good idea to power it up every month just to be certain it's not self discharging, it shouldn't drop much at all though)

Edited April 22, 2022 by Tawpie

[supercurio]

@Dosingpsychedelics uh-ho. Which wheel model?
I'm asking because some wheels are known to drain a bit or a lot when off, some don't.

[Begodecrashtestdummy]

1 hour ago, supercurio said:

@Dosingpsychedelics uh-ho. Which wheel model?
I'm asking because some wheels are known to drain a bit or a lot when off, some don't.

Rs19 hs it’s suppose to arrive next Wednesday 

[supercurio]

Then I suppose it's supposed to arrive full or nearly full, and an owner should confirm but I don't expect it would drain by itself.

If you want to be excessively careful you could always disconnect the packs from the mainboard and each other, which will pretty much ensure nothing can happen with the packs. It will require opening up the wheel, which require plenty of caution however.

[Begodecrashtestdummy]

21 minutes ago, supercurio said:

Then I suppose it's supposed to arrive full or nearly full, and an owner should confirm but I don't expect it would drain by itself.

If you want to be excessively careful you could always disconnect the packs from the mainboard and each other, which will pretty much ensure nothing can happen with the packs. It will require opening up the wheel, which require plenty of caution however.

Ewheels  are sending me a reinforced shell.they said the new batch of Rs19’s are coming with exceedingly weak shells and that the screw receivers inside the shell are breaking off.they said they are planning on switching out the shells but are low on manpower to do it but that if I wanted to do it myself they would take 100$ off the wheel and send me the reinforced shell .so it’s all on its way and I’m gunna try to tackle it myself 👌🏻

[Tawpie]

@Dosingpsychedelics Tell your friend that even tearing a wheel down to the shell and replacing the shell shouldn't take 12-18 months, even with bad behavior. There may be some bad language involved, but choice words shouldn't extend the down time much. Couple days max, so no worries about the batteries. 

[mrelwood]

11 hours ago, lagger said:

Cell imbalance doesnt kill your battery pack. Overcharging, deep discharging or overheating cells is what does.

1) The BMS prevents overcharging from happening, so no worries there.

2) Deep discharging isn’t an issue, since even with badly balanced packs the individual cells won’t get much below 3.0V.

3) Overheating the cells hasn’t been referred to as being the cause or even a suspect in a single EUC preliminary battery failure.

4) Cell imbalance has clearly been demonstrated as being the cause for numerous EUC battery failures. I don’t understand how this data doesn’t get through your scrutiny.
 

11 hours ago, lagger said:

Its also more probable that you forget your wheel discharged for longer time -> self-discharge -> deep discharge.

Advice for storing the battery between 40-60% is rather common by now. And only a few EUC models have any notable self discharge. Li-ion batteries themselves don’t.

11 hours ago, lagger said:

The imbalance in a fit battery is developing slowly. If your battery pack has significant imbalance after a few cycles it might mean more that the battery is already at the end of life or defective.

That is possible. What we’ve seen many times though is that it’s not at all rare for the packs to have been balanced badly from the start. One of the S20 demo wheels for example had a variance of 0.3V between cell groups after just a few cycles. At that point the difference accelerates already pretty fast.

11 hours ago, lagger said:

But yeah, if you are usually utilizing ALL of your battery then its better to have well balanced cells.

Yes.

11 hours ago, lagger said:

So again, the question should be: is it beneficial and should you do it? Are there risks?

Further more, how beneficial is it, and whether it’s worth it in the first place.

11 hours ago, lagger said:

The question about benefit / losses is subjective and everyone can have different opinions and usage patterns on their wheels.

Absolutely! What makes this to be even worthy of a discussion is that there are many things that are “bad” for the battery. You can pick any of them and try to avoid it at any and all costs. What’s missing though is how bad is it, and if the badness is even relevant before you have ridden several tens of thousands of miles. And as you avoid the one bad thing, are you doing something that harms the battery even more. An order of magnitude more in the case of 80% charging.

11 hours ago, lagger said:

my usage pattern allows me to charge 90% and discharge to 40% on my daily work commuting.

As long as you balance the V10F pack often, your charging behavior helps you extend reaching the 80% SoH from roughly 25000km to around 30000km. That is great!

[lagger]

10 hours ago, Vanturion said:

Effect of Battery Temperature while Charging:

Nice data you put there but I see that some of it differs from what I have found. The basic results are the same as in other studies but the absolute values are sometimes a few fold different. Might be chemistry or methodology. Anyway I would like to say you normally wont even discharge the packs at 0 C. Even if its freezing the joule heating in batteries especially if you have some performance wheel and riding a little aggressively will be warming your battery.  Depends on your drive but yeah if its little freezing it doesnt directly mean that your battery is also freezing.

10 hours ago, Vanturion said:

Second, discharging to 2.5V is really not great for capacity fade. Luckily we don't have to worry about this at all.

But this is theoretical. The BMS cut off at 3.2 (worst case as you said) and thats like 5-10% energy left in the cell. Now depending on the self discharge of cells / BMS circuit / Control unit it can take weeks or months to discharge the cell to 0% and basically kill it. I dont have illusions, so I think not really all the riders will charge their wheels and also some of them will just forget but this might be expensive. Storing the battery for long time is recommended at some 40-60% capacity and around 5 C as you said.

5 hours ago, mrelwood said:

1) The BMS prevents overcharging from happening, so no worries there.

2) Deep discharging isn’t an issue, since even with badly balanced packs the individual cells won’t get much below 3.0V.

3) Overheating the cells hasn’t been referred to as being the cause or even a suspect in a single EUC preliminary battery failure.

4) Cell imbalance has clearly been demonstrated as being the cause for numerous EUC battery failures. I don’t understand how this data doesn’t get through your scrutiny.

You are very losely quoting me there because I explained most of the things in text. But if you say the imbalance directly kills the cells. What is the mechanics of killing those cells then?
I can think of that if the BMS is wrongly engineered it could discharge some of the cells more. Seen that already. In that case if you are not balancing frequently enough your pack would be dead. Wondering if someone have looked into that dead packs and saw some pattern of dead cells.

5 hours ago, mrelwood said:

 

That is possible. What we’ve seen many times though is that it’s not at all rare for the packs to have been balanced badly from the start. One of the S20 demo wheels for example had a variance of 0.3V between cell groups after just a few cycles. At that point the difference accelerates already pretty fast.

But then the question is if it just wasnt properly paired in manufacturing process or what is happening there because thats not something which is any normal. Maybe the pack was defective form beginning.

Also I am not promoting to not balance the cells by full charge. I am doing it like every 5-10 ride and I fully discharge my wheel only occasionally like once 2-3 months. Because I dont have need for that.

Edited April 23, 2022 by lagger

[mrelwood]

 

4 hours ago, lagger said:

You are very losely quoting me there because I explained most of the things in text.

I only quoted the text you had typed. Let me try and be more precise then:

4 hours ago, lagger said:

But if you say the imbalance directly kills the cells.

They do, but that’s not what I replied to. I replied to was your comment that said “Cell imbalance doesnt kill your battery pack”. A pack can be useless with fully healthy individually cells.

The balancing feature on the EUC BMS’s is quite ineffective, and it’s unable to balance the pack if the cell group voltages are vastly different. At that point the BMS shuts off the charging process well before the pack is anywhere near balanced, and the total voltage is less than it should be after a full charge.

4 hours ago, lagger said:

What is the mechanics of killing those cells then?
I can think of that if the BMS is wrongly engineered it could discharge some of the cells more. Seen that already. In that case if you are not balancing frequently enough your pack would be dead.

It doesn’t take a wrongly engineered BMS to discharge the more depleted cell groups more than the others. Both voltage sag and the voltage difference gets larger as the cells’ SoC is depleted. That too can be seen in the S20 reports, the 0.3V I mentioned was only 0.15V at 124V charge. The lowest cell groups gets stressed at a lower voltage, which lowers their SoC faster than the rest, increasing the voltage difference further.

 At this point the lowest cells themselves are just fine, they are just operated at a lower voltage than the rest. But once they get closer to 2.5V, they will wear out faster, and once below 2.5V, the cells themselves start to get directly damaged.

 Since the BMS’s high voltage cutoff is usually at 4.25V per cell, a 100.8V wheel can seem to be charging normally despite one cell group remaining at 2.25V. (23x2.25V=97.75, and 100.0V still shows as 100%). So an imbalance can keep damaging one cell group without the rider ever knowing about it until it’s way too late.

 And all that was if the owner tries to balance the pack at every charge. If they don’t balance at all, the imbalance will be further accelerated.

4 hours ago, lagger said:

Wondering if someone have looked into that dead packs and saw some pattern of dead cells.

I have, many times on my 16S, as was described in the video. Planning to do it again soon.

4 hours ago, lagger said:

But then the question is if it just wasnt properly paired in manufacturing process or what is happening there because thats not something which is any normal. Maybe the pack was defective form beginning.

Clearly the cells were badly paired from the factory. But how much pairing happens when building the packs is anyone’s guess. Or if it happens at all. I’ve seen many reports of a wheel not charging to full when new, but doing so just fine after a few full charges followed by a few hours of balancing. My guess is that the cells are supposed to come out from the battery cells’ production line at a certain voltage, and that it isn’t checked after that.

4 hours ago, lagger said:

Also I am not promoting to not balance the cells by full charge. I am doing it like every 5-10 ride

I hope for your sake that your cells are well enough paired that your plan is enough. For my 16S it wasn’t, three times over.

Edited April 23, 2022 by mrelwood

[lagger]

I Agree. But basically you are saying what I am with the mechanics. That the cell groups are being damaged when they get deeply discharged or overly charged when having some imbalance. Small imbalance alone is not a problem when the depth of discharge isnt close to full and if the BMS can handle some reasonable imbalance of cells when charging. With my standard discharge maximum to 30-40% SOC. I have normally not problems with this. Basically the pack is as strong as its weakest cells so if it would be that bad, the pack would be trash anyway.

Anyway, the range and voltage drop of my wheel is very similar to when it was new so the drop in performance at this moment is insignificant.

[mrelwood]

On 4/23/2022 at 9:58 PM, lagger said:

I Agree. But basically you are saying what I am with the mechanics.

Fundamentally yes, but without the “imbalance doesn’t kill cells/pack” part.

On 4/23/2022 at 9:58 PM, lagger said:

That the cell groups are being damaged when they get deeply discharged or overly charged when having some imbalance.

Again, I can’t see how overly charging is an issue when the BMS prevents pretty well that from happening.

On 4/23/2022 at 9:58 PM, lagger said:

Small imbalance alone is not a problem when …

Sure, and even a larger imbalance is ok for a while if you don’t go below 50% battery. But how much imbalance can the BMS balancing recover from, how quickly would it get worse, and can the wheel then ever be ridden below 50% charge without damaging the cells? These we don’t know without individual cell group monitoring.

On 4/23/2022 at 9:58 PM, lagger said:

Basically the pack is as strong as its weakest cells so if it would be that bad, the pack would be trash anyway.

And that is exactly what my enthusiasm towards balancing the cells is intended to prevent.

On 4/23/2022 at 9:58 PM, lagger said:

Anyway, the range and voltage drop of my wheel is very similar to when it was new so the drop in performance at this moment is insignificant.

As I explained, once the user notices an issue, it can easily be too late at that point. But here’s hoping that all is well!

[Boofheadeu]

I am not sure what is the best routine. then i watched this. What do the rest of u think? Advice would be appreciated.

 

[.....]

Mr.Elwood knows his sh*t. There's also a large thread about batteries and a plethora of random postings. Its not as complex as people make it out to be. Im beginning to think that asking about batteries for euc, is akin to asking what kind of oil to use in a motorcycle.....

I charge mine to 100% and balance regularly. Only time I short charge them is when I KNOW I'll be not using them for 90 days (doesnt happen). I charge once cooled after a ride, and ride once cooled after a charge. Mine sit at 100% ready for me at all times, with the headroom I paid for.

https://forum.electricunicycle.org/topic/18317-battery-faq/

 

Edited May 7, 2022 by ShanesPlanet

[OneLeg]

Nice video.

Here are a couple of alternatives:

When you build your own packs and you have a dumb BMS, it is best you solder balancing leads... and use an external balancer periodically.

If you are daring build the pack without the BMS, use your charger to charge it to 4.0V CC and the CV and then  then connect the balancer to the leads a balancer. The balancer won't care that the cells aren't fully charged like the BMS does.

Alternatively, If you have a 20S pack, find a 21S BMS,  and install an extra battery in each series. When you build the pack make sure the batteries are not fully charged, otherwise you will bat 88V and blow the controller. If you have the cells at 3.5V build the pack and then charge the full, do the balacing, the actual full/max voltage the cell is at is 4.0V.  I still have not had a chance to prove it but I think running 0.2V less volts will unlock more capacity at the lower voltages.

Edited May 16, 2022 by OneLeg

[Tawpie]

@mrelwood As my go to rider on proper care of battery packs, what is your opinion of the variable charge-stop capability of the S22? Good or bad or unknown-as-yet?

I did a little test the other day and used the app to set the charge stop at 80%. Because the S22 BMS allows viewing of the individual cell voltages, I wanted to see what happened. Yes, this is a brand new wheel with 2 total cycles on the packs so the batteries should be balanced and I didn't check their values before starting a BMS controlled charge to 80%, but within each pack they finished 18 mV apart, closer than they are at the end of a 100% charge cycle (24 mV max to min). The difference is almost certainly within measurement error, but on the face of it the cells were nicely balanced when the BMS shut down the 80% cycle.

Do you know of anyone that has looked at this 'feature' closely enough to give it thumbs up or down?

I'll probably migrate to a "if you don't need the range, charge to less than 100% and check the max/min cell voltages" routine for the S22 for a while but it that's a terrible idea...

[mrelwood]

2 hours ago, Tawpie said:

what is your opinion of the variable charge-stop capability of the S22?

There are so many ifs that I still don’t know what to think about it.

- It’s been shown that the voltage readings in the cell groups, battery packs, and controller don’t match, even close. So trusting any of the readings as an absolute value goes out the window. You can’t even reliably compare them to each other.

- The S22 chargers are often badly calibrated, and hence won’t often even let the BMS to calibrate at 100% charging.

- The details of the charge stop option are not known, so despite your test giving carefully positive vibes, we don’t know how it operates, and when and how it balances.

- The S22 is still new, and long term results of various charging behaviors don’t yet exist. So we can’t deduct the actual balancing behavior from them either.

I really don’t know. On one hand it might make sense to have the charge stop at something like 95% in case the charger doesn’t output full voltage, but then again, in the worst case scenario doing so would prevent balancing.

So yeah. I just don’t know. If I had an S22, I would’ve of course tested the functionality as thoroughly as I could. But I don’t so I won’t.

Edited November 24, 2022 by mrelwood

[Tawpie]

If the internal measurements are bogus, we’re blind and might as well have a dumb BMS. I wouldn’t put it last KS to have failed here too, but they also rely on those measurements so I’d hope they’re meaningful!

They might be good enough to spot a weak cell at a gross level, one that lags the others by several tenths of a volt.

As always I’ll keep an eye on the voltages, as they say: time will tell!

Edited November 24, 2022 by Tawpie

[mrelwood]

Yeah, the larger issue seem to be the total voltages of the packs. A few % out at 4V will still give sufficient information, but a few % at 126V can make or break a balancing feature as a whole.

[Chriull]

51 minutes ago, mrelwood said:

Yeah, the larger issue seem to be the total voltages of the packs. A few % out at 4V will still give sufficient information, but a few % at 126V can make or break a balancing feature as a whole.

As it is and was with all wheels and chargers till now.

Small deviations will make balancing less effective, but not fully break it.

Don't know exactly about bms firmware capabilities but this possibility could lead to many (necessary) improvements. They should have the possibility to calibrate the different bms and motherboard voltages! It's very confusing now to get three values for one voltage and the consequences resulting therefrom.

Once KS eventually fixed this they could make the bms firmware measure the no load charger voltage once the charger is plugged in and adopt balancing thresholds...

So real partial charges would be possible and misadjusted chargers nothing to worry about anymore...

[Tawpie]

7 hours ago, Chriull said:

As it is and was with all wheels and chargers till now.

Small deviations will make balancing less effective, but not fully break it.

Don't know exactly about bms firmware capabilities but this possibility could lead to many (necessary) improvements. They should have the possibility to calibrate the different bms and motherboard voltages! It's very confusing now to get three values for one voltage and the consequences resulting therefrom.

Once KS eventually fixed this they could make the bms firmware measure the no load charger voltage once the charger is plugged in and adopt balancing thresholds...

So real partial charges would be possible and misadjusted chargers nothing to worry about anymore...

It's the promise of "smart" BMS right? The info available from various measurements could be used to manage sub optimal conditions that a casual consumer is (rightfully) unaware of. By doing so, you might be able to materially reduce the chance that consumer behavior would raise the probability of cell damage. Even something stupid like delaying the green LED until the pack is properly balanced would be beneficial (but I know that the perception of 'faster charging' is a strong selling point)

I'd like to see the per-cell monitoring be used to identify when any individual cell is sagging significantly worse than its companions... avoiding under voltage damage to individual cells is the root of why we want to balance our packs.

Edited November 27, 2022 by Tawpie

[Cress]

 

@Tawpie -   'manage sub optimal conditions that a casual consumer is (rightfully) unaware of.'  Yup.  Agreed.  That's the point.  Automobile manufacturing is a good model:  1) Quality controls in EV battery manufacturing  2)  BMS actively manages EV charging and battery balancing  3) BMS takes pack off-line if voltage imbalance is detected  4)  BMS takes pack off-line if temperature rise is detected.  I'd be really unhappy if my electric car forced me to monitor battery voltages and temperature.  Car companies know how to sell cars.

[Chriull]

3 hours ago, Tawpie said:

It's the promise of "smart" BMS right?

It could be the first a bit smarter bms.

3 hours ago, Tawpie said:

Even something stupid like delaying the green LED until the pack is properly balanced would be beneficial (but I know that the perception of 'faster charging' is a strong selling point)

It's not really known at which thresholds the led turns green for the various wheel and their chargers, but from some tests it seems the green light is often a good time to stop charging.

There is not much happening after the green light, just some slight stressing of the cells.

3 hours ago, Tawpie said:

I'd like to see the per-cell monitoring be used to identify when any individual cell is sagging significantly worse than its companions... avoiding under voltage damage to individual cells is the root of why we want to balance our packs.

If one cell is significantly worse than the others passive balancing cannot help anymore. It is just able and designed to balance small deviations at every charge.

With the high burdens in EUCs no system can make the whole pack better rhan the weakest cell.

in low burden systems active balancing could help to average cell capacities.