Begode 900Wh Battery Recall Options

[RagingGrandpa]

1 hour ago, sevin7 said:

The new Gotway Hero uses the same battery packs as the RS19

Let's not be presumptive... Hero has not yet been delivered, and GW/BG has multiple variants of the modular 24s2p pack now, such as:

MonPro: (new M50LT cell, plus output-shutoff BMS)

Nik+ 2700wh: (Samsung 48X, new 3-pin output connector)

EX.N 2700wh: (same as Nik+ above)

Always changing.

[YourAubsome]

32 minutes ago, RagingGrandpa said:

2 hours ago, sevin7 said:

The new Gotway Hero uses the same battery packs as the RS19

Let's not be presumptive... Hero has not yet been delivered, and GW/BG has multiple variants of the modular 24s2p pack now, such as:

I'm certain I saw that the cells in the Hero were LG M50T from somewhere (not trying to find it right now), which is also the reason that ewheels is only selling their version with modified packs (which is also why the ewheels version of the hero cost more)

[Paul A]

What was the reasoning for eWheels to contract a third party to assemble battery packs, and have other safety measures then?

What was the reasoning for the recall then?

[Paul A]

And cause of the service center fire?

[AtlasP]

As a reminder which seems relevant to the present discussion, here are the set of areas/improvements eWheels + LiTech chose to focus on with the contracted LiTech packs. Surely the culprit is connected to some subset of these--whether as faults with the original packs and/or limitations of the original packs in conjunction with Begode's firmware and the demands it places on the packs. (So for example perhaps some of these changes/improvements are only necessary to compensate for the faults of the Begode firmware but wouldn't be necessary otherwise.) It would be immensely interesting to see testing of the Begode packs in other brand wheels as against other EUC brand packs in the affected Begode wheels.

On 10/12/2021 at 12:06 PM, AtlasP said:

eWheels has officially launched its unique line of Begode wheels featuring eWheels-sourced battery packs, starting with the T3/Tesla V3, EX.N (HS & HT), Hero, and Commander (HS & HT). [Presumably all their new RS wheels as well, but which isn't documented on the eWheels site along with these others.] This marks the first new Begode wheels added by eWheels in a year(!) (since the RS in Oct of 2020). The eWheels-sourced packs are assembled by LiTech Power and feature the following improvements/capabilities:

• Imported Korean Samsung 50E cells (Versus the Begode packs which use LG M50LT made in China)
• Short detection/protection, transient threshold at 300-450A
• Back up inline 30A physical fuses
• Each pack fitted with 5x temperature probes & buzzers to alert at a 65°C level & shut-off when the temps exceed 75°C
• Voltage differential management for packs of unequal voltage, no balance lead required
• Improved cell-holders for better survivability in the event of crash
• Upgraded pure nickel strips that are both thicker & wider to withstand higher currents with less resistance—Begode had used ‘inferior plated’ strips on at some of their packs  
• Automated weld production methods for consistent quality
• Capacity & stress-testing processes done at the pack level after assembly
• Photos with further details, here:
  • https://ewheels.com/wp-content/uploads/2019/03/LiTech-Pack-Features-1-600x437.jpg
  • https://ewheels.com/wp-content/uploads/2019/03/LiTech-BMS-Features-600x437.jpg
  • https://ewheels.com/wp-content/uploads/2019/03/LiTech-Cell-Holders-Nickel-Strips-1-600x437.jpg
  • https://ewheels.com/wp-content/uploads/2019/03/LiTech-Cell-Holders-Nickel-Strips-2-600x437.jpg

[Source: https://forum.electricunicycle.org/topic/17309-euc-comparison-table-project-with-monthly-updatesnews-roundups/page/4/?tab=comments#comment-389532 ]

Edited November 11, 2021 by AtlasP

[Paul A]

Thank you AtlasP.

Seems to show eWheels taking responsibility for QC that is lacking.

[Patrick Robert]

If anything, Adam's video on his friend's fire with his Nik+ helped me realize one thing: the importance of the max discharge current of a cell to be used in an electric vehicule.

Previously, I had only considered the max charge current. There were people in my EUC squadron arguing that charging at 5A all the time is bad for the batteries in the long run. I was wondering myself, so I looked into it. 

Turns out that a 4-pack wheel like the Nik+ is fine to always charge at 5A, since each of the four parallel packs will take in a quarter of that 5A, and 1.25A is well below the max charge current of 1.455A for the LG M50T. I also charge my S18 at 5A all the time, even though 5A/3 = 1.66A slightly exceeds the max. But as others have said, the max is actually a table showing that if the cells are warmer than 25C the max becomes 3.395A instead, and that this table is probably a crude way to represent a curve to model how the cell chemicals behave. So I figure it's within tolerance, as long as the cells are warm enough. Not an electrician, but I can read and analyse.

But I never paid much attention to the max discharge rate. Regardless of what happened to Adam's friend's wheel, whether he drove it too hard and over the long run slowly damaged the cells, or perhaps he always had a few bad cells that over time grew from bad to worse... one thing is for certain: the LG M50T has a max discharge rate of 7.275A (per cell and parallel pack) over 25C (and just 2.425A below 10C). It's right here in the spec from LG: 
https://www.batteryspace.com/prod-specs/11514.pdf

I previously said the S20 used the LG M50T, which was wrong. It will have the LG M50LT, which has an updated lifespan of 1000 charge cycles, compared to 500 for the M50T. But they both have the same max discharge rate.
https://www.illumn.com/21700-lg-inr21700m50lt-5000mah-high-discharge-flat-top.html

First conclusion:, the LG is an inferior battery, compared to other models that can do 10A. Why choose anything else than a 10A cell, price? I would have paid extra to have more watt-hours in my S18 if there was an option, and I would have done the same to upgrade to top-level batteries to have better amperage for less chance of a cutout and/or cell damage. And the possibility of one or more bad batchs of these cells in circulation doesn't point to an award winner either, hence the upcoming recall and replacement of these packs by eWheels.

Second, I understand one more reason why the S18 is so anemic. Oh, so much fun (like I want to ride it even after my Sherman), but oh so sadisticly underpowered. Strictly a commute wheel for me. I thought it was because of the 3 packs. But on top of that, I now know that those 3 packs supply less power than if they had used another battery like the Samsung 50E, with a max discharge current of 10A. It could have had a third more juice, just by picking another cell.

Third, the KingSong S20 is now much less interesting to me because of its use of the lower-rated LGs, combined with 126V. Its 120 cells could have been 5 packs instead of 4 if they had chosen to go 100.8V. More parallel packs, more amperage, and amperage is better than volts. I initially got hooked on the idea of boosting up to 126V a torque motor in order for it to also have good top speed, along with all the other cool stuff about it, but now I can picture the possibility that it might just be as performance-challenged as the S18. History has been known to repeat itself.

Edited November 11, 2021 by Patrick Robert

[Tawpie]

2 hours ago, Patrick Robert said:

It could have had a third more juice, just by picking another cell.

and changing the firmware to allow more juice. Personally I think KS is playing on the very conservative end of the field, protecting the batteries and electronics by deliberately limiting the power pulled from the batteries. If you push further than they want you to go, the firmware will put you on your face… and your wheel will probably be fine! Without a firmware change, the high discharge rate capability would be a moot point.

Begode is on the other end… let the rider decide for themselves, even to the point where they risk life, limb, and apartment. Without tighter firmware limits, their packs would be much better served by high discharge rate cells—not home free, just ‘better’. But I believe Begode knows they usually must trade capacity for discharge rate, and last I checked high discharge rate high capacity cells were considerably more expensive than their more pedestrian models.

Regardless, using any component outside its intended operating envelope is asking for trouble. Li ion cells are not exempt.

Edited November 11, 2021 by Tawpie

[Chriull]

2 hours ago, Patrick Robert said:

Third, the KingSong S20 is now much less interesting to me because of its use of the lower-rated LGs, combined with 126V. Its 120 cells could have been 5 packs instead of 4 if they had chosen to go 100.8V. More parallel packs, more amperage, and amperage is better than volts.

No. With the same power current and voltage are designwise interchangeable - with different drawbacks/advantages.

The same power output can be reached either way. Or the same maximum speed and/or torque.

So whatever config one uses for the 120 cells - if 20s6p, 24s5p or 30s4p - the current draw per cell will be _similar_ for the same output power.

[yoos]

3 hours ago, Patrick Robert said:

the LG M50T has a max discharge rate of 7.275A (per cell and parallel pack) over 25C (and just 2.425A below 10C)

As you correctly noticed elsewhere, that datasheet is rather crude and the stepwise data is but a representation of an otherwise smooth curve. The datasheet says 14.5A at 10-25C and 7.2A at 25-55C. This just means that even at 55C you can still draw 7.2A. The sustainable current at 30C is much closer to 14.5A than to 7.2A but alas, a more accurate I_max(T) curve is nowhere to be found. Moreover, even if such detailed data were available, you would need temp sensors on the batteries to properly limit the discharge current at all times. Furthermore, as mentioned in this thread many times, those specs are for continuous draw. Pulse (i.e. short in time) draw might allow much higher instant current. Again, this data is not on the sheet. [The oversimplified nature of the datasheet is understandable. It's easier for LG to establish simple and blunt conservative usage limits rather than to formulate complex physical limitations which would be misunderstood somewhere down the road by a non-engineer/scientist person in charge of sourcing the right batteries on the vast chinese market]

You are praising Samsung 50E for 10A, yet there are also 35A and even 45A cells readily available on the market, with price of the same order (the S18 might become some 50$-200$ more expensive -- small price to pay for performance and safety). So the solution for fixing the S18 into a powerful and light short-range all-terrain EUC is hiding in plain sight.

[Paul A]

Difficult to have confidence and trust in the competency or safety.

 

[supercurio]

3 hours ago, yoos said:

You are praising Samsung 50E for 10A, yet there are also 35A and even 45A cells readily available on the market, with price of the same order (the S18 might become some 50$-200$ more expensive -- small price to pay for performance and safety). So the solution for fixing the S18 into a powerful and light short-range all-terrain EUC is hiding in plain sight.

Agree I wish the EUC industry would move away already from cells focusing strictly on high capacity vs high current.

First, when you measure the high capacity cells, their actual capacity doesn't match the name. A "3500" cell typically output 3300 mAh over the whole voltage range, a "50" cell 4800 mAh. And that's when tested at low currents: it goes down quickly at higher currents.

I've been super impressed by benchmarks of the Molicel P42A for instance, by lygte-info.dk or Mooch (Facebook link).
A cell like this would provide significantly lower sag during aggressive/high speed riding - when it matters most: meaning better safety margin as well.
I would love for someone to make both a low speed range test and high speed range test with these cells vs the LGs or Samsung.

Possibly the range with Molicel P42A would be shorter at low speed, and longer at high speed compared to LG/Samsung "50" cells.

Most importantly for this topic, the stress on these cells would be well below capacity, and they could also be fast-charged (for real) safely.
(Standard charging rate 1C, max 2C. Charging your EUC in 40 minutes anyone?)

At least that's on paper, I don't know much about historic safety characteristics and QC with this cell manufacturer.
I hope someone will make a full review of the Molicel P42A packs in when actually ridden in a EUC 

[mike_bike_kite]

4 hours ago, Tawpie said:

Begode is on the other end… let the rider decide for themselves, even to the point where they risk life, limb, and apartment. Without tighter firmware limits, their packs would be much better served by high discharge rate cells—not home free, just ‘better’. But I believe Begode knows they usually must trade capacity for discharge rate, and last I checked high discharge rate high capacity cells were considerably more expensive than their more pedestrian models.

Are you saying that Begode could release a firmware for their existing wheels that significantly reduces the fire risks? I'm quite a slow rider and would happily exchange performance (that I rarely use) for additional safety. Or do I get that safety by rarely pushing the wheel?

I tend to run the wheel down to about 40% then fully charge with an extra hour of (battery) balancing. I tend to ride at around 25mph. I don't have power pads and don't feel like I'm ever stressing the wheel though I do have a couple of steep hills that I need to get up. Saying all that, I did notice that I had a max speed of 34mph on my last trip to tennis  Maybe I need to upgrade my bicycle helmet and tracksuit?

[Tawpie]

4 hours ago, mike_bike_kite said:

Are you saying that Begode could release a firmware for their existing wheels that significantly reduces the fire risks

Actually, I'm talking out my a$$ because I don't know anything about the detailed design of Begode hardware or software.

But I am a perfectly good hardware head (gone soft), and even if the Begode hardware doesn't directly monitor current drawn from the batteries, that value can be inferred. It really should be very straightforward to have firmware limit the power delivered to the motor (which in turn limits the power drawn from the battery) and I accuse KS of doing this—there really isn't a reason that I can see for the 16X to crap out at 30 mph (without burning parts up) other than firmware.

The problem for Begode is that such a limit would reduce the performance from "golly that's a rush" to "gee, that's kinda ok". I ride KS because I literally don't like going crazy fast and am not disappointed that I can only go 28 mph. From what I've observed, those that buy Begode do not feel as I do and would be royally disappointed if their wheels were limited by something as stupid as ample design margin.

Edited November 11, 2021 by Tawpie

[/Dev/Null]

So for those of us who haven't (or couldn't) buy from ewheels.  How do we get our batteries replaced?  Anyone have any ideas?  I wouldn't mind paying out of pocket to get better batteries & more safety even if it was out of pocket.  

Are there any stats on fires vs wheel age, or fires based on manufacture date of the wheel?  My Nik+ is going on 2 years old (IIRC March 2022) and has about 1500 miles on it.  I typically don't ride over 25mph (maybe a hair over rarely), don't jump or crash the wheel nor ride in rain.  I balance the wheel typically a day before riding, and never leave it on the charger for an extended time past 2-3 hours of the green light.  What is the real risk?

Issues of service, recalls, and quality are what really keep people from signing on to the sport.  I've had people ask me who were interested "SO who replaces the tire or repairs this thing when I crash it?" I and I can't give them an answer.  

 

Edited November 11, 2021 by /Dev/Null

[supercurio]

24 minutes ago, /Dev/Null said:

So for those of us who haven't (or couldn't) buy from ewheels.  How do we get our batteries replaced?  Anyone have any ideas?  I wouldn't mind paying out of pocket to get better batteries & more safety even if it was out of pocket.  

Are there any stats on fires vs wheel age, or fires based on manufacture date of the wheel?  My Nik+ is going on 2 years old (IIRC March 2022) and has about 1500 miles on it.  I typically don't ride over 25mph (maybe a hair over rarely), don't jump or crash the wheel nor ride in rain.  I balance the wheel typically a day before riding, and never leave it on the charger for an extended time past 2-3 hours of the green light.  What is the real risk?

Issues of service, recalls, and quality are what really keep people from signing on to the sport.  I've had people ask me who were interested "SO who replaces the tire or repairs this thing when I crash it?" I and I can't give them an answer.  

 

I'm guessing the most sensible option will be to replace the Gotway packs by LiTech drop-in replacements, as they appear to be designed for this purpose.
I thought they were an eWheels-only thing but maybe not, since a local distributor here in Sweden advertised a LiTech pack with Molicel as a replacement, for 6999 SEK.
It's a seller with way over-inflated prices on everything tho so I wouldn't take the price as absolute reference.

If the volumes allow it, eWheels could sell a lot of these replacements to those not covered by any sort of recall. Hopefully will recoup a fraction of the recent losses added to the recall expenses.

Edited November 11, 2021 by supercurio

[mike_bike_kite]

31 minutes ago, /Dev/Null said:

Are there any stats on fires vs wheel age, or fires based on manufacture date of the wheel?  My Nik+ is going on 2 years old (IIRC March 2022) and has about 1500 miles on it.  I typically don't ride over 25mph (maybe a hair over rarely), don't jump or crash the wheel nor ride in rain.  What is the real risk?

This!

[Elwood]

52 minutes ago, /Dev/Null said:

So for those of us who haven't (or couldn't) buy from ewheels.  How do we get our batteries replaced?  Anyone have any ideas?  I wouldn't mind paying out of pocket to get better batteries & more safety even if it was out of pocket.  

Are there any stats on fires vs wheel age, or fires based on manufacture date of the wheel?  My Nik+ is going on 2 years old (IIRC March 2022) and has about 1500 miles on it.  I typically don't ride over 25mph (maybe a hair over rarely), don't jump or crash the wheel nor ride in rain.  I balance the wheel typically a day before riding, and never leave it on the charger for an extended time past 2-3 hours of the green light.  What is the real risk?

Issues of service, recalls, and quality are what really keep people from signing on to the sport.  I've had people ask me who were interested "SO who replaces the tire or repairs this thing when I crash it?" I and I can't give them an answer.  

 

I bought a used ewheel nik a few months ago, think it was a 2019 version. Only had 175 miles on it, charges fine to 100v, dream to ride, even came with one of the covers, never dropped, probably never ridden hard ( with only 175 miles, basically brand new tire and no body dings, so probably no board replacement). Hard to decide if it's problematic or not. I guess second owners will still be in if there's a recall, not sure how it would be sorted out or notified. For the "it was just sitting there and burned up" stories, it seems that's rare without some previous problems. That Swedish msp guy in the wrongway vid comments added that he smacked a building, blew the board, replaced it, then two weeks later " it just burned up". I used to build fpv drones and had a few of them burn up with shorts, but that was the soldering learning curve; if they lasted a few flights and survived a few ground bounces, things worked fine. There has to be 1,000's of niks out there of original stock ridden all kinds of ways not burning up. I'll just not let the bat go below 40%, charge to full each time, alarms at 25 mph and 25 amps so I know where I'm at. Sit by it when charging, store in the basement wrapped in a 2100 degree ceramic blanked and keep on trucking.

[Tasku]

1 hour ago, /Dev/Null said:

My Nik+ is going on 2 years old (IIRC March 2022) and has about 1500 miles on it.  I typically don't ride over 25mph (maybe a hair over rarely), don't jump or crash the wheel nor ride in rain.  I balance the wheel typically a day before riding, and never leave it on the charger for an extended time past 2-3 hours of the green light.  What is the real risk?

Short answer: I would say your changes are about the same as getting hit by lightning. I will knock wood on that!

Obviously someone answering to you from world wide web is not as reliable as someone from insurance company or some other investigation department for that matter. Or li-on battery expert for that matter.

Longer answer yet, in many ways simplified:

Based on some calculations how long a battery pack totalling around 1800wh, your battery could start to get dentrites that could start to be noticable about 24000-30000km. Some of these engineers in this forum can do their own evaluations but..(14900-18640miles) Even still you could use the device. I feel uncertain doubling those numbers given. But you would likely notice device not performing what you used to(when it is time to let go), like battery drains faster and you will not get as far. Should you notice considerable loss in performance, it is good time to call quits (fast draining would become more dangerous). The invidual batterys still could work for low stress enviroments like batterywalls. Continous torture and living on edge will affect how fast dentrites will form. If your battery pack gets hit it might not survive that long. So natural cycle of battery pack is not always so simple as conditions affect in various ways. Majority of people break something other than batterypacks, that is a good thing to know. The battery tends to live longer than the will of user to use it, in almost all/most cases.

Just to be clear li-on batteries will start forming dentrites at somepoint. These invidual battery has protection to shut the battery unit off. Yes, every battery in the pack should have its own protection to shut it off, and the whole pack will be affected notable way. If in anyway your max VOLTAGE of batterypack should fall, from what you are used to by several Volts, never use this pack but take it to professional. This would be the normal end state for many EUC. 

Instead of getting new batterypack people dispose the wheel and get new, latest and greatest. (atleast the battery, most battery firms take it from you) Makes sense, the new wheels have hopefully better smart BMS (invidual cell measuring), better waterproofing, better protection against shortcircuits and other less likely bad events, now taken better in account for. 

This community will prolly never know how exactly some fires happened, but this is hot potato. Mainly we have been trying to voice concern about this. Even insurance companies usually settle for the most likely reasons, not being the absolute truth of the occasion, just more likely for the scenario. Somethings are common in most cases but here are some guessing I can throw in in no particular order:

DIY fixes, battery pack extensions, battery damaged from crashes.

Water in the battery pack

Transit in car - battery pack nickel strip/plating gets loose and shortcircuits the pack - bumbs and vibrations, or device powers on during transit.

Overcurrent draw from batterypack and no design to stop continous damage to batterypacks. General design flaw

Any form of moving about is not 100% safe, yeah even planes crash (1 in 11million), yet winning big in lotto draws somewhat near (1 in 14million) 

---------------------

By the way just to be clear, it seems e-bikes have way more problems with this battery draining being too big, yet the fire incidents are less talked about. This talk about batteries and fear is good tho, hopefully something better comes out of it, safer products for one. More knowledge for riders, that be my 2 cents.

[RagingGrandpa]

2 hours ago, /Dev/Null said:

My Nik+ is going on 2 years old (IIRC March 2022) and has about 1500 miles on it. 
I balance the wheel typically a day before riding, and never leave it on the charger for an extended time past 2-3 hours of the green light.

Sounds perfect; keep riding!

And perhaps you can use this thread as an excuse to motivate yourself to do some fire safety planning. It's healthy for all EUC owners.

 

1 hour ago, Elwood said:

1,000's of niks out there of original stock ridden all kinds of ways not burning up.

 

On 5/20/2021 at 9:27 AM, RagingGrandpa said:

My take:

The safest EUC is the one you already own.

(Assuming you didn't modify it electrically; or soak it.)

Once the EUC has survived 200 miles of riding, it's proof that there were no big defects, and failures during normal use are much less prevalent.

All great motivation to appreciate the wheel you have, vs chasing the latest model

[supercurio]

A rider in my local chat group noticed a few days back that his MSP HT wasn't charging fully and had poor range.
We talked about risks involved and the necessity to neither charge or discharge any of the pack further and disconnect both packs.

Still today, he tried to charge the battery after unplugging/plugging things back to see if there would be any change.
I was  and did my best to convince him to stop immediately, including with material coming from this forum on previous fires, which he did soon after following the charger stopped charging - then proceeded to resume, at ~94V.

I learned that despite clear instructions, people are just not taking the safe steps and don't acknowledge the risks with the large packs in our wheels.
Hence an announcement, and maybe recall seems wise.

Later we got the whole story: one of this MSP HT packs got damaged earlier by being briefly shorted with pliers.
And that was enough to permanently damage at least 2 cell groups.

While a short is a lot more current than what a controller can draw, it was also an extremely short duration.

Has anyone measured:

• how much current a controller and its large capacitors can draw in peak?
  Like a high-time resolution measurement to truly measure peaks. Wondering if it's like many repeated shorts.
• how much current is drawn from the pack when you re-connect them to the board without an anti-spark in between?
  Definitely like a short each time.

Another example of damage by too much current being drawn I guess.

Edited November 11, 2021 by supercurio

[YourAubsome]

On 11/10/2021 at 10:43 AM, RagingGrandpa said:

As we scrutinize cells, it will be important to keep the product names straight. There are 3 LG cell models being conflated- they are not the same.

 

I assumed the king song site had a typo as it says it uses LG 50LT cells. There is no such cell. The question is, are they using LG M50LT or LG M50T? I bet its nearly impossible to get an answer.

 

 

[Paul A]

On 11/9/2021 at 3:37 AM, Jason McNeil said:

With the previous incidents, I believe the scope of the recall is going to have to be to recommend that Customers dispose of their existing 900Wh packs for the Nikola+, RS, MSP [the risks are too great] then await for a replacement packs with better preventative measures in place,

 

It seems eWheels is stating to immediately remove the batteries for the stated models and dispose them.

To not have the batteries in the home at all.

[Paul A]

On 11/9/2021 at 6:43 AM, Jason McNeil said:

all the cases have been with the LG M50LT cells

EWheels would likely to have informed KS about fire, and battery recall.

Would seem it is just a typo and the LG M50LT would not be used by KS.

 

 

[RagingGrandpa]

On 11/11/2021 at 7:17 PM, Paul A said:

 

On 11/8/2021 at 3:43 PM, Jason McNeil said:

all the cases have been with the LG M50LT cells, which are what Begode use exclusively in the 21700 format. 

EWheels would likely to have informed KS about fire, and battery recall.

Would seem it is just a typo and the LG M50LT would not be used by KS

That's my point about the 3 cell models under discussion... M50LT only recently entered production, and there were no EUC's sold with this cell before September July 2021.

The recent Gotway fires have been M50T; and earlier the Panasonic "Tesla Motors spec" 21700.

We have no basis for distrust of the new M50LT, because the EUC community has just now started to use it.

(And yes, the names are confusing...)

Edited November 15, 2021 by RagingGrandpa
NikAR 2700wh seems to be the first to ship with M50LT