Gotway MSX battery safety margin

[Rywokast]

36 minutes ago, WI_Hedgehog said:

The other item of concern is bearings...are the bearings in Wheel_X serviceable with moderate effort, like changing a tire's innertube?

no idea about that.. ive only ever heard of one person doing that before, not even sure why they did it should last tens of thousands of km before needing servicing.. i cant imagine it would be too difficult though the motor comes apart easily enough

[mrelwood]

 

Quote

people see that and think full charges but thats not actually a full cycle

Exactly. Even running an 18XL to 0% at rest isn’t yet a full li-ion charge cycle.

From Battery university: “there are no clearly defined standards of what constitutes a cycle”

But they do mention a range of 100% - 20% being generally announced for being available for use, so whether the end of the cycle is 2.7V or 3.0V, it’s still only 80% of a full cycle. And since one rarely rides an EUC past 20% at rest, an EUC charge cycle is only about 60% of a full li-ion charge cycle.

 With this math a full cycle on a 1600Wh battery gives usually a range north of 100km. 300 of those would be at least 30 000 km. And one would still have 80% of the original range. I’m at 11 000 km and getting a new wheel “already“.

Quote

So you can charge to 80% and get that 80% now, or charge to 100% until it gets degraded to 80% and get that 80% later.

This is a marvelous point! I’ll take the 100%. I could be at 93% already, but my riding has become faster and more aggressive as well, so I really can’t tell.

Edited May 5, 2020 by mrelwood

[.....]

10 hours ago, WI_Hedgehog said:

On one hand, for guys like myself who live in cold weather climates it's about battery longevity, so buying more battery than needed is not optimal.

On the other hand, commuters like yourself are indeed the ones who have to look at how many charges per day they're doing. One full charge might mean replacing the batteries yearly. Two half-charges (to 4.2V) might mean you get an additional year of use due to shallower depth-of-discharge, so bringing a charger to work (or leaving a second one there) may have excellent Return On Investment.

 

What voltage does the Battery Management System start to balance at? 84V, right? (4.20V x 20 cells = 84.00V) So you need more than 84.00V to keep pushing electrons into that system to end up at 84.00V. The LiPo chargers I've used turn off when current flow is lower than (let's say 100 milliamps, for simplicity). So if the BMS clamped voltage at 4.20V like it should, you could put 86V into it and the BMS would cut off charging on each cell branch as it reached 4.20V, when all branches were full only a trickle current would flow to power the BMS (under 100mA), and the charger goes into Standby mode (charge complete).

Disclaimer: I don't know your BMS and how it is functioning.

 

Damn, I was really hoping for a black and white concrete answer to this. It also occurred to me that if I only allow the charger to put out 84.0v, it MAY not have the voltage required to allow the bms to manage what it is intended to do? I know that when i set the charger to 84.0v and let it run for hours, I wind up with... 83.9-84v on the wheel. If i bump the charger up to 84.2v and let it run, I end up with a charge of 84.18-84.2 at the wheel battery. All i can deduce from this is, the bms is NOT limiting voltage to a level that is lower than 84.2.   Silly me, I had always thought that batteries get their final balance, AFTER they were near full. I wrongly assumed that you would want to put higher voltage into the circuit to charge, as the bms would limit it as needed. Maybe it was dangerous and wrong of me to think that once i battery was full, it would dispose of any over voltage as heat. With a trickle charge, I thought that any low cells would continue to charge and all full cells would shuck off energy as heat, therefore balancing the weaker cells.

Damn, I need to read up some more. Sadly, I think that reading too much bullshit is what has got me so damn confused. Its so easy to change from 84.0-84.2 in the charger, I just wish I knew for sure which. @Mike Sacristan makes an amazing point. Use the full battery now and take advantage as youll get to use 80% only, soon enough? Regardless, I cant even make an informed decision as my ignorance of all this undermines it. My typical gas station run takes my battery from 100%, down to <15%. Charging in route is an impossibility, so the battery discussion does kind of matter in my use. Running out of juice during the trip would be a DEFINITE safety issue.

Edited May 5, 2020 by ShanesPlanet

[Mike Sacristan]

52 minutes ago, mrelwood said:

I could be at 93% already, but my riding has become faster and more aggressive as well, so I really can’t tell.

I feel you! I am at about the same battery %. Although since the beginning I can't really say that it ever charged to 100% either.
It would usually charge to 95% unless I left it in the charger for ages... and even then it would dip fast down to 90% anyway.
It could be the charger that has always been charging a bit on the low side.
Like you say... we ride differently now than we did in the beginning. The winters probably take their tolls on the batteries as well.
So in the end... I try to forget all of this and just ride the wheel and get what range I get.
I'd rather take the full power and range when the wheel is young.

I guess as long as we aren't doing things that aren't directly damaging our batteries we should be fine.
So things like extreme temperatures, leaving the wheel in the charger for a day, etc... could be detrimental to the battery.

And like @ShanesPlanet says... there is way too much to read about this stuff.
My understanding on batteries and charging and balancing is the same as yours though Shane.

[.....]

Just to add more confusion. My Ks18L charger came with an output of 83.8V. I (like Mike) noticed that it would only go full (per side wheel lights) if i left in charge for many hours past when the charger light went to green. Then within a few 100 yards of operation, I would lose the last light and be quickly hitting low 90's in percent. If i unplugged the charger when it went green, the top most battery light on my wheel would NOT light up. I'm suspecting my charger was 'undercharging' by a little, if "FULL" Charge is 84.0+.  Sorry about the derail, as I know this is a GW thread. I am VERY tempted to bump it back up to 84.2V and call it a day. Hell, at my current rate, the odds are that something else catastrophic will likely happen, before A battery fail (I hope).

[Rywokast]

4 minutes ago, ShanesPlanet said:

Just to add more confusion. My Ks18L charger came with an output of 83.8V. I (like Mike) noticed that it would only go full (per side wheel lights) if i left in charge for many hours past when the charger light went to green. Then within a few 100 yards of operation, I would lose the last light and be quickly hitting low 90's in percent. If i unplugged the charger when it went green, the top most battery light on my wheel would NOT light up. I'm suspecting my charger was 'undercharging' by a little, if "FULL" Charge is 84.0+.  Sorry about the derail, as I know this is a GW thread. I am VERY tempted to bump it back up to 84.2V and call it a day. Hell, at my current rate, the odds are that something else catastrophic will likely happen, before A battery fail (I hope).

hmm.. thats a definite QC fail.. im not sure about the stock chargers since i havent taken any apart.. but i have taken apart multiple fast chargers and i would imagine it would be similar, where there is a potentiometer inside that you would just give a slight adjustment to, it will change the output voltage so you could increase it slightly to get the full charge

[.....]

6 minutes ago, Rywokast said:

hmm.. thats a definite QC fail.. im not sure about the stock chargers since i havent taken any apart.. but i have taken apart multiple fast chargers and i would imagine it would be similar, where there is a potentiometer inside that you would just give a slight adjustment to, it will change the output voltage so you could increase it slightly to get the full charge

No way, a lack of quality control in a chinese euc factory? Surely ye jest?! Mine has a potentiometer and is VERY easy to adjust. It came at 83.9 and then I bumped it up to 84.2(at 84.2, all the charge indicators on the wheel side would light up and i wouldnt lose the top one so quickly). THEN I read about how over voltage kills and lowered it back to 84.0. NOW, Im pondering if it SHOULD be at 84.2.   Maybe i should flip a coin, as 50/50 chance seems better odds than i can give the QC guys about getting it right after the  first 1000+ tries. Higher or lower voltage made ZERO change in board temp during charge as its a 1.5amp charger(95 degrees).

Edited May 5, 2020 by ShanesPlanet

[Justin Boivin]

Thanks to everybody who is participating in this thread, there is a lot of knowledge to gain here!

Edited May 5, 2020 by Justin Boivin

[Rywokast]

6 minutes ago, ShanesPlanet said:

No way, a lack of quality control in a chinese euc factory? Surely ye jest?! Mine has a potentiometer and is VERY easy to adjust. It came at 83.9 and then I bumped it up to 84.2. THEN I read about how over voltage kills and lowered it back to 84.0. NOW, Im pondering if it SHOULD be at 84.2.   Maybe i should flip a coin, as 50/50 chance seems better odds than i can give the QC guys about getting it right after the  first 1000+ tries.

lol, i guarantee they slap that stupid little sticker on there without even checking 1/10,000 of them... tbh im not entirely sure where the ".2" comes in.... 84V is the traditional full charge of 4.2V per cell but perhaps its for balancing or something... not too sure, but tbh i dont think it makes a difference anyways you could set it to 90 and the BMS would of course stop it at a full charge anyways, i dont even know how it happened perhaps i bumped it wrong or something lmao but one of my fast chargers one day magically changed from 84.2 to 84.4, however it is incapable of ever reaching the 84.4 because the bms wont allow it it just turns the charger off... i should check it on a gotway wheel since they have a live charging port to see the display but the highest ive ever seen it go was 84.1... perhaps they do allow the cells to reach 4.21V temporarily for balancing i dont honestly know... however it should almost immediately go down once you power it on

Edited May 5, 2020 by Rywokast

[.....]

12 minutes ago, Rywokast said:

lol, i guarantee they slap that stupid little sticker on there without even checking 1/10,000 of them... tbh im not entirely sure where the ".2" comes in.... 84V is the traditional full charge of 4.2V per cell but perhaps its for balancing or something... not too sure, but tbh i dont think it makes a difference anyways you could set it to 90 and the BMS would of course stop it at a full charge anyways, i dont even know how it happened perhaps i bumped it wrong or something lmao but one of my fast chargers one day magically changed from 84.2 to 84.4, however it is incapable of ever reaching the 84.4 because the bms wont allow it it just turns the charger off... i should check it on a gotway wheel since they have a live charging port to see the display but the highest ive ever seen it go was 84.1

The bms in mine, doesnt seem to care if i charge it at 84.2. After a charge of 84.2 and a let it rest, the battery packs themselves show me 84.18-84.2. Maybe 84 IS max voltage and Im seeing a slight overage that settles back to 84 overnight. Hearing how dangerous and finicky these cells are, 84.2 had me worried.  I do know that 83.9 wont charge up to 84v, no matter how many amps it sends at the voltage. Im not really willing to try pushing 90v to see if the bms reduces it internally. But if the bms DOES regulate voltage, wouldnt it make sense that it would be better to set the charger a little higher and just allow the bms to work? Pushing lower wouldnt help as the bms cant INCREASE voltage, afaik. Why cant this be simple enough for me to grasp? SUrely Im making this more difficult than need be? Prolly those 3 pesky letters : O.C.D.

*again I apologize, this has NOTHING to do with the original topic of safety margin.

Edited May 5, 2020 by ShanesPlanet

[WI_Hedgehog]

HAS A POSSIBLE CAUSE OF PREMATURE CELL FAILURE JUST BEEN DISCOVERED?

Charging to 80% and then balancing the cells, which requires balance leads, means a loss of 20% range per trip, but gain of 100% to 200% more battery charges.

Say the EUC can safely go 30 miles on a charge:

30mi x 100% range = 30 mile trips (or less)
30 miles x 300 charges = 9,000 mile battery lifespan

30mi x 80% range = 24 mile trips (or less)
24 miles x 600 charges = 14,400 mile battery lifespan

If you're fanatical and don't discharge below 20% battery capacity:

30mi x 60% range = 18 mile trips (or less)
18 miles x 1,200 charges = 21,600 mile battery lifespan

- THAT'S THE THEORY -

So @ShanesPlanet is was seeing 83.9V / 20 cells = 4.195V/cell => 4.20/4.195 = 0.12% or one-eighth of one percent low,
and after his adjustment is  seeing 4.21V/cell, 0.24% (one quarter of one-percent) difference.

It is interesting @ShanesPlanet Battery Management System isn't cutting off at exactly 84V, though 2% maximum error on components is realistic (84.00 +/-1.68V = 4.20V +/-0.08V max), perhaps 1% on LiPo circuits.

 

Perhaps that's why we're seeing premature battery failures.

Balancing should happen at 84.0V where the BMS clamps the maximum voltage at 84.0V and sinks the remaining current to ground, bypassing any additional charging for that cell and allowing other cells to creep up to 84.0V and balance.

If the charger is putting out 83.9V and the BMS isn't "balancing" until 84.2V (or more), the cells aren't being balanced [properly], leading to premature cell failure.

 

 

[.....]

@WI_Hedgehog    I think my OCD is the problem. With QC and tolerances the way they are, I am probably trying to split hairs with a butter knife. In the end, I have my charger outputting 84.1v 1.5amp. The charger turns to green somewhere around 83.8 (about where the adapter was set from factory). If unplugged as soon as it goes green, my batteries settle around 83.8 and tho my apps say 100% charge, it drops FAST with almost no use. The wheel light loses one VERY soon. IF i leave the charger on AFTER it goes green, the batteries will climb up to ALMOST the voltage set on the power brick. Atm I have my brick set at 84.1.  I reconnected the brick and let it sit with green light on for an hour. In that hour, the batteries have charged to 84.04 and settled there(during charging). So, even after the brick was green and app was claiming 100%, the batteries were still taking trickle charge. Leaving it on for even longer has done nothing past 84.04.  We are talking a .06v variance from charger output to battery output, but even MY ocd ass will chalk it up to excuses of resistance in testing points or the like. I could go further and take temp reading at various spots and wire in an ampmeter, but to hell with it. 9000 miles sounds good enough to me. I can almost positively say that by then, the old 18l will have been replaced for a myriad of other possible issues/reasons. Thank you so much for all your time in sharing the information with me. I can't say I know FOR SURE, much more than I did(not your fault). However, the bigger picture is that I'm satisfied with the fact that I can leave my charger on for a few hours AFTER it goes green, and I wont go past 84.04v. IF the bms is sending the .06v excess to ground, so be it. IF the batteries are dispelling .06v at trickle, Im hoping its also a non issue.  End result.. 84.04 fully charged and an owner that's fine with 9000mi lifespan.

Edited May 5, 2020 by ShanesPlanet

[mrelwood]

 

Quote

It came at 83.9 and then I bumped it up to 84.2

To pull the carpet from under your OCD, please check the datasheet of your multimeter for accuracy at 84VDC. It can be as bad as +-1%.

Quote

Say the EUC can safely go 30 miles on a charge:

30mi x 100% range = 30 mile trips (or less)
30 miles x 300 charges = 9,000 mile battery lifespan

Again, a li-ion “charge cycle” is very different from how far you get on a charge on an EUC.

And even when the capacity goes down to 80%, calling it the end of the battery’s lifespan is quite funny if the remedy is to use 80% (or less) of the capacity since day one!

 

Quote

Balancing should happen at 84.0V where the BMS clamps the maximum voltage at 84.0V and sinks the remaining current to ground

The way I’ve understood how the BMS balancing works is that the 100 ohm balancing resistors are connected in parallel with each battery cell, which doesn’t stop charging the cell but just slows it down so that the lower cells can catch up. There is no mechanism to completely separate the high cell from getting any charge, and the cell will still get many times the current that flows through the resistor.

I don’t even know whether the balancing resistor is switched on or if the transistor ramps up the action. If the latter, the balancing would start a bit before 4.20V, but it would just be less effective while the resistor is being brought in. Switching it in would cause less heat, so it might be the way it works.

[.....]

39 minutes ago, mrelwood said:

 

To pull the carpet from under your OCD, please check the datasheet of your multimeter for accuracy at 84VDC. It can be as bad as +-1%.

Again, a li-ion “charge cycle” is very different from how far you get on a charge on an EUC.

And even when the capacity goes down to 80%, calling it the end of the battery’s lifespan is quite funny if the remedy is to use 80% (or less) of the capacity since day one!

+- 0.1%  so my ocd is far worse than my multimeter's accuracy. I've spent too much time of mine and your on this already. Totally moot as it only takes 3 seconds of the wheel being powered on, to bring the voltage down to 83.9 or less anyhow. I'm gna go obsess about something else with no real answer (like tire pressure) and let the big boys carry on. As for me, I have nothing usefull to add and I think the OP would like to have his thread back. Thanks so much for the info and view points!

Edited May 5, 2020 by ShanesPlanet

[svenomous]

Wow, another fun battery thread. I've participated in a couple, and read through some older ones. I used to obsess more about this that I do now.  I think I've come down off the "baby the battery packs" thinking since getting my wheel last year. I (now) believe there's a little bit of over-thinking applied to the topic.  It's true that these cells will give absolute best performance if charge rate, discharge rate, maximum (charge) voltage, and minimum (discharge) voltage are kept within certain limits. Environmental temperature also plays a factor. Let's not forget storage charge level, and of course the balancing problem. On the other hand, as pointed out in this thread and others, a reasonably well operated battery pack (the most important thing being that the cells remain in balance and no dead cells/strings develop) will give tens of thousands of km of total usage, and then they don't just die instantly, they just settle down to about 80% of their when-new capacity. In fact, even that doesn't happen like a switch, it's a gradual process that will happen no matter what we do, and all we can do is affect the rate at which it does. Continue to run them well after the magic 300-cycle threshold, and they'll probably give another few tens of thousands of km before their ability to retain a charge decays to the point that the wheel's range is considered fatally compromised. If a 20% range reduction after a couple of years is a deal breaker due to daily commute requirements, I submit that the wheel's battery pack capacity is too small to begin with for the intended mission.

So, my remaining rules for myself are:

• Generally charge the pack(s) to 100% (let's say that's 84.2V for a 72V/20s pack) with a charger that taper-charges in CV mode, and ensure the 100% charge level is held down to a fractional amperage to give the BMS time to perform balancing. For each new charger used, cross-check that it really does reach the right voltage. If it does that, I can trust it does the tapering correctly as otherwise the voltage would drop or rise after that point.
• When I'm using the wheel less, I occasionally reduce charge to 80 or 90 percent, to give the pack a little "vacation" from being at 100% and create a warm fuzzy feeling that I've done something nice for the packs. Still, 100% charges should be done at least every 2nd or 3rd charge, especially as the pack gets older, since balancing becomes ever more important as some cells begin to develop higher internal resistances than others and drop lower than others during discharge.
• For off-season storage, charge to 80%, pat the wheel, and leave it in a cool dry place until the next riding season. Maybe check on it once a month to ensure charge level remains at 50% or higher.
• Try not to run the battery down below 30% or so too often. Don't sweat it if I have to do it, but don't make it the daily norm. Deep cycling is not good for any battery chemistry, and in any case I don't like the pucker factor of range anxiety, nor when the wheel gets all tilt-backy and beepy on me.
• Watch for unusual symptoms, like sudden decrease in range, or inability to charge past a certain voltage, as that can be a symptom of one ore more cells having gone "bad."

I say don't sweat it so much, don't obsess over it, just enjoy the wheel. Don't baby the battery pack, but also don't abuse it unnecessarily. Watch for any weird stuff happening. Don't be afraid of charging to 100%, as the good of that (balancing) usually outweighs the bad (some tiny incremental loss of capacity), except when the wheel is about to go into 3 months of storage. Why should managing the battery on the wheel consume more CPU cycles in my brain than enjoying it?

[svenomous]

By the way, I'm quite aware of the irony of claiming "I'm not over-obsessing about batteries anymore," just days after starting a thread about BMS battery balancing . In my defense, I reserve the right to be curious about how stuff works, and I wanted to know how the BMS balancing works in these wheels, and what that might mean in terms of charge cutoff voltage, tapering, etc.  So, I'm not obsessing, I'm just a curious guy (in more ways than one).

[Planemo]

2 minutes ago, svenomous said:

In my defense, I reserve the right to be curious about how stuff works, and I wanted to know how the BMS balancing works in these wheels, and what that might mean in terms of charge cutoff voltage, tapering, etc. 

You're not alone

I find these type of threads really interesting. No doubt I worry about it too much as well, constantly wondering whether my charger is outputting correctly (it hits 100.7v at the moment but with my meter tolerances...who really knows?) and I am VERY anal about storage charges, consider when the last 'full' charge was (I usually do this every 3 charges) etc etc.

But then my buddy has had a 1230wh 100v MSX for a couple of years, doesn't care too much about any of this, concentrates purely on riding only and his wheel rips and charges just as well as it did on Day 1! I think it has at least 5k miles on it now.

[WI_Hedgehog]

8 hours ago, mrelwood said:

To pull the carpet from under your OCD, please check the datasheet of your multimeter for accuracy at 84VDC. It can be as bad as +-1%.

And even when the capacity goes down to 80%, calling it the end of the battery’s lifespan is quite funny if the remedy is to use 80% (or less) of the capacity since day one!

 

The way I’ve understood how the BMS balancing works is that the 100 ohm balancing resistors are connected in parallel with each battery cell, which doesn’t stop charging the cell but just slows it down so that the lower cells can catch up. There is no mechanism to completely separate the high cell from getting any charge, and the cell will still get many times the current that flows through the resistor.

I don’t even know whether the balancing resistor is switched on or if the transistor ramps up the action. If the latter, the balancing would start a bit before 4.20V, but it would just be less effective while the resistor is being brought in. Switching it in would cause less heat, so it might be the way it works.

Great point, I have a calibrated FLUKE meter.

As the battery ages and hits the 80% capacity mark, one tends to use closer to 100% of its capacity, and it fails rapidly. It's like a cell phone battery, where every day you use 30% and it has 70% charge left at night when you plug it in. Then one week you notice it's in the low 60s, then 50s, then 30% charge left, and you know you'd better get a new phone.

If a 100-ohm resistor were in parallel w/the cell it would constantly draw an average of 0.0385A (0.15W), not a huge draw, but the batteries would lose 3.5Wh a day in parasitic load...not a great BMS strategy.

 

5 hours ago, Planemo said:

I find these type of threads really interesting. No doubt I worry about it too much as well, constantly wondering whether my charger is outputting correctly (it hits 100.7v at the moment but with my meter tolerances...who really knows?) and I am VERY anal about storage charges, consider when the last 'full' charge was (I usually do this every 3 charges) etc etc.

But then my buddy has had a 1230wh 100v MSX for a couple of years, doesn't care too much about any of this, concentrates purely on riding only and his wheel rips and charges just as well as it did on Day 1! I think it has at least 5k miles on it now.

Personally, I'm interested in understanding the technology, and if doing a few "simple" things extends the life of the batteries greatly, that's awesome--money saved! Though for many of us the batteries will probably die of old age (shelf life).

The bigger thing I'm looking at is when a cell dies because of a bad BMS or charger component, and takes a few parallel cells with it, those are replaced and the pack dies again without realizing it's the BMS. For example, if @ShanesPlanet could get 3x the battery life by tweaking a pot correctly, that's pretty awesome. We've seen an OEM damaged BMS resister, bad solder joints, bad components, bad batteries, bad plug connections, etc. It's not "everyday common," but I ordered safety gear to survive pavement surfing, and it makes sense to understand how the wheel works to try to avoid pavement surfing altogether.

Edited May 5, 2020 by WI_Hedgehog

[Chriull]

15 minutes ago, WI_Hedgehog said:

If a 100-ohm resistor were in parallel w/the cell it would constantly draw an average of 0.0385A (0.15W), not a huge draw, but the batteries would lose 3.5Wh a day in parasitic load...not a great BMS strategy.

Only while charging for cells over about 4.2V. So they get charged slower but never discharged!

[RagingGrandpa]

14 hours ago, Mike Sacristan said:

since the beginning I can't really say that it ever charged to 100% either.
It would usually charge to 95%

^ This is what not to do. Top-balancing is defeated because 4.2V/cell is never experienced.
 

11 hours ago, WI_Hedgehog said:

BMS clamps the maximum voltage at 84.0V and sinks the remaining current to ground

As Mr Elwood explained, this is incorrect.

The BMS has two actions it can take:
1) Engage the balancing resistor for each individual cell group.
2) Completely interrupt the charging (open-circuit).

(If your charger is working properly and provides only 84.2V +/- 200mV, and your pack is healthy, #2 should never occur.)
 

10 hours ago, ShanesPlanet said:

I'm satisfied with the fact that I can leave my charger on for a few hours AFTER it goes green

IF the batteries are dispelling .06v at trickle, Im hoping its also a non issue.

This is correct.
The 'trickle' is theoretically undesirable (minimally accelerated wear), but practically unimportant compared to the alternatives (e.g.: imbalance leading to total failure of cell groups).
 

7 hours ago, svenomous said:

the most important thing being that the cells remain in balance

 

1 hour ago, WI_Hedgehog said:

it makes sense to understand how the wheel works to try to avoid pavement surfing altogether

This thread was considering two failure modes
A )   Pack aging and reduction of capacity; and
B )   0V ('failed') cell groups resulting from discharge of an excessively imbalanced pack.

I don't think either of these failure modes typically results in actual "cut-out" of the EUC during riding. Both should cause more rapid voltage sag to the 3.3V/cell threshold (Gotway), at which point you get a rideable but beepy tilty wheel.
And possibly flames shortly after... but no such occurrence has been reported here?

Edited May 5, 2020 by RagingGrandpa

[mrelwood]

 

53 minutes ago, WI_Hedgehog said:

As the battery ages and hits the 80% capacity mark, one tends to use closer to 100% of its capacity, and it fails rapidly.

Can you show me an EUC rider that has had this happen to them? ‘Cos I can easily find half a dozen cases of failed battery packs at just a few thousand km due to 80% charging or other behavior that was intended to prolong the battery lifetime.

 Cell phones and laptops are different in the sense that they usually spend time in the charger for many times the required amount. Laptops being the worst, as it’s not uncommon to have something like one charge cycle after a year of active use. Of course a battery will fail then. But I’d love to hear of an EUC battery that has aged to the point of being useless before 20 000 km due to charging to 100%. If it can be an issue, I’m absolutely certain that someone somewhere in the world has had it happen.

53 minutes ago, WI_Hedgehog said:

If a 100-ohm resistor were in parallel w/the cell

The resistor is behind a transistor that connects them to the specific cells only when needed.

53 minutes ago, WI_Hedgehog said:

if doing a few "simple" things extends the life of the batteries greatly, that's awesome--money saved!

That’s all good. Just don’t end up at the “failed battery due to 80% charging” statistic. There is a whole lot I don’t understand about these batteries, but I read this forum a lot, so I have a reasonably good understanding of the issues that are common. And 80% charging and other behavior that may neglect cell balancing are the largest red flags for me at the moment. Cells aging to the point of being useless? Zero cases that I’ve seen so far.

[WI_Hedgehog]

44 minutes ago, Chriull said:

Only while charging for cells over about 4.2V. So they get charged slower but never discharged!

Then it's not a passive component like a resistor...maybe something more complex like a reverse-bias diode, or transistor, depending on how inexpensive a manufacturer wants to make it. My understanding is King Song started out as a BMS manufacturer, and therefore have solid BMS units.

Electric Bike has a good BMS article.

Example 4-cell balancer.

[WI_Hedgehog]

17 minutes ago, mrelwood said:

Can you show me an EUC rider that has had this happen to them? ‘Cos I can easily find half a dozen cases of failed battery packs at just a few thousand km due to 80% charging or other behavior that was intended to prolong the battery lifetime.

 Cell phones and laptops are different in the sense that they usually spend time in the charger for many times the required amount. Laptops being the worst, as it’s not uncommon to have something like one charge cycle after a year of active use. Of course a battery will fail then. But I’d love to hear of an EUC battery that has aged to the point of being useless before 20 000 km due to charging to 100%. If it can be an issue, I’m absolutely certain that someone somewhere in the world has had it happen.

The resistor is behind a transistor that connects them to the specific cells only when needed.

That’s all good. Just don’t end up at the “failed battery due to 80% charging” statistic. There is a whole lot I don’t understand about these batteries, but I read this forum a lot, so I have a reasonably good understanding of the issues that are common. And 80% charging and other behavior that may neglect cell balancing are the largest red flags for me at the moment. Cells aging to the point of being useless? Zero cases that I’ve seen so far.

The biggest problem seems to be lack of proper balancing (the 80% group largely neglecting balancing), however, there are plenty of reports of failed cell groups on the forum by 100% chargers. I want to learn how & why things work and maximize pack life, but that's just me. Maybe I'll build this out later, right now I need to finish the miniPRO Speed project.

EUCs haven't been out long enough to age the cells to death, but we should start seeing that in a year or so. The other thing is early EUC adopters that ride 10,000 miles seem to buy several EUCs, so don't tend to notice if the batteries wear or not since they are riding newer, faster technology. It looks like this could become an issue now that EUCs are running 30-40 MPH and reaching maturity (other than the suspension development). Luffy was great in its time, but the M10 is a total upgrade, so why worry about the Luggy's batteries? But if you own an M10, will there be an appreciable future improvement?

I'm looking to understand how things work and be ahead of the curve.

 

Edited May 5, 2020 by WI_Hedgehog

[RagingGrandpa]

16 minutes ago, WI_Hedgehog said:

it's not a passive component like a resistor

It's a resistor.

Let's assume for a moment that this behavior is real (as I believe it is) -

On 5/2/2020 at 6:28 AM, Chriull said:

At around 4.2V the bleeding resistor is put side n parallel to this cell group, so it gets charged slower.
At around 4.28V the BMS cuts off the charger.

Generic visual aid:

As an example:
If the pack is excessively imbalanced, while charging, let's say Cell S4 reaches 4.28V. (The S4 load resistor was already engaged, since the cell had exceeded the 4.20V balancing threshold.)
This 4.28V condition triggers Charge Stop for the entire pack - the BMS disconnects the charger.

My question:
In this example, does (GW) BMS keep the S4 load resistor engaged after Charge Stop? (I hope so...)
And, what condition resets the Charge Stop event?
(Has anyone demonstrated this reset behavior with an EUC pack? @Chriull?)

[Chriull]

1 hour ago, WI_Hedgehog said:

Then it's not a passive component like a resistor...

It is, with some switching element (mosfet) and "logic" (comparator) ..

1 hour ago, WI_Hedgehog said:

maybe something more complex like a reverse-bias diode, or transistor, depending on how inexpensive a manufacturer wants to make it. My understanding is King Song started out as a BMS manufacturer, and therefore have solid BMS units.

Electric Bike has a good BMS article.

Example 4-cell balancer.

In this example the for identical rows of components on the left side perform the balancing.

The two R101 resistors are the "bleeding resistors. The mosfets (Q1 - 4) switch them in parallel to the according cell (group). U1 to U4 are the comparators spupervising the according cell (group) voltage and switching the mosfet.