[WARNING] Gotway Nikola 100v 1800WH Battery Fire

[Archee Jan Bloch]

7 hours ago, davinche said:

Note: The model that caught fire was the first batch of Nikola 100V 1800WH Panasonic NCR21700A. I have never crashed my Nikola nor have I mistreated it in any way. Over the span of 5 months, I've only used it to commute to and from work (~7.5 miles each way). I did have a fast charger from ewheels (the Nikola did not come with a charger), but I have never charged it with > 3A or to 100% (used the 90% setting on the fast charger).

Below is official communication from EWheels. Jason asked me to withhold posting on the forum until he had time to communicate with Gotway and develop a plan of action. Many people in the EUC community praise EWheels for genuinely caring about the his customers and the community. I am glad I purchased my wheel through Ewheels and the communication from the email (below) was the evidence I needed that I bought from a good distributor.

One minor correction I want to make to the official email is that it didn't catch on fire immediately after turning on. I walked it to a parking lot near my place (so it was already on). When I got on it, it immediately shutoff and threw me off the wheel. It was only then after mysteriously powering off did the wheel go into thermal runaway and the wheel caught on fire.

Anyways, below is the official communication. I am posting to raise awareness for those that might not have purchased through a distributor. Once again my model was the PANASONIC NCR21700A cells.

 

you mentioned that you never charged more than 3 amps and only to 90% could you be so kind and tell us how often do you charge it if at all,  ever to 100% 

 

[meepmeepmayer]

Wow, this easily could have been worse! Too bad it intimidated your friends.

• How was the air humidity and temperatures when this happened? Maybe some condensation got into the battery somehow?
• Firmware problem, and after you removed the charger, the "heating" started? 10 mins later, boom?
• Did you ever balance the battery (charge to full and keep the charger in for a longer while)? I didn't think even completely unbalanced cells would catch fire, but maybe they can?
  I am NOT trying to blame you here, this should not happen even without balancing!
• Maybe it's just a random coincidence, or a problem with the Panasonic (used, as per some rumors?) cells.

That's the explanations I can come up with. I don't expect one from Gotway other than "We are using different cells now, problem solved". Jason doing great work as usual.

[meepmeepmayer]

6 hours ago, Chaindrop said:

Question regarding Lithium-ion fires: In cases where a fire starts, can they explode even when the device is turned off? or does it usually just happen when you turn on the device when the batteries are faulty?

The electrolyte fluid in Li-Ion batteries simply ignites when it gets too hot (like 200°C or so) and can "pop" the battery before that, too. Usually that heat is from a short (e.g. from water intrusion into a battery pack) or something else (in theory [never happened as far as it is known]: firmware bug) causing an uncontrolled runaway in the battery and the cells heating up rapidly until they pop or start burning.

So fundamentally whether the wheel is on or off does not matter, and switching it off won't stop a fire. But of course things can be set in motion when you engage the electronics. Not sure if this happened here or was just coincidence.

If you have a fire (extremely extremely rare) that you would need to stop (probably it's hardly possible to move an already burning wheel), the important thing is to cool the battery down (e.g. submerge the wheel in water). Realistically, all you can probably do is let it burn and contain the damage.

-

As always with battery fires, these are extremely rare, and this example should be cause for caution and introspection, but not concern

[WI_Hedgehog]

6 hours ago, davinche said:

Hmm... so what is the correct way to charge then? Also, is it only gotway boards that do "top-balancing"? What about Kingsong or Inmotion bms?

My understanding from the members here is all* major EUC manufacturers use top-balancing circuits. This is smart, as it is cheaper, smaller, lighter, and allows the most power into the cells the fastest. 

*The exception may be Ninebot/Segway, though I don't know. Both miniPROs I have use active balancing circuits that are always running and monitor each cell, even when the batteries are removed from the units and just sitting around. Because of per-cell monitoring and reporting, it is easy to see how fast a cell can start going out of balance and by how much. The miniPRO charges to 4.00V (80%) and throttles at 30%, encouraging the owner to use only 50% of the batteries total capacity, and by doing so extending battery life from 500 charges to 2000 charges (assuming battery life means the cells can hold only 80% of their rated capacity).

Because it appears in general EUCs do not monitor individual cells, nor cell groups during discharge, best practice seems to be (and I belive @Chriull mentioned this in the battery charging threads):

• Let the battery cool before charging (reduces risk, extends life). 
• Do not charge until going riding. (The further the charge is above 50% = 3.85V the more the batteries wear out, even during periods of non-use).
• Charge fully (100% = 4.20V, engages the balance circuits),
• Check the voltage at full charge (there are devices made to do so if the EUC app doesn't report voltage). Voltage should be 100% of rated voltage = 4.20V x # cells in series. (4.2V x 20 = 84.0V, 4.2V x 24 = 100.8V). If the voltage starts to fall below normal, there's a problem developing--get the problem fixed immediately. 
• Plan your ride so the battery is not discharged below 30% = ~3.30V. (Gotway's 0% remaining is 3.3V, so riders are forced to observe this. King Song 0% is 3.0V, so the rider can self-manage their usage.) 

Wheels need a certain amount of power to move you. Power is:
Voltage (volts) x Current (amps) = Power (watts)
so as voltage drops current goes up, because the amount of power needed to do something doesn't change. Current causes heat, heat cooks the enamal on motor windings, MOFSETs, and battery cells, therefore the less heat that is generated the longer things last, meaning the more voltage a battery has the fewer amps that are needed to do the same amount of work, and the cooler things are. This is part of why using the last 30% of battery capacity is so hard on everything: not as much voltage and more current. This is also why speed is limited as battery capacity drops. 

Think of it this way: Voltage is like strength. Current is like the amount of motion. If you have three hours to move a large pile of gravel from the road where it was dropped off to the back of your yard, it is better to take a wheelbarrow full of gravel and use your strength, than it is to use a child's sand pail and run back and forth very fast--you will soon be hot and exhausted. You will need more time to recover. If your wife does not let you sleep (recharge) fully, and if you get a new load of gravel every day for a month, by the end of the month you may be worn out and not able to continue. However, if you used the wheelbarrow each day (strength) and were allowed to sleep (recharge) fully, you would be much healthier--not like new, but much better (maybe you would not like your wife for making such a large project, but you would still love her and continue on with her). : Your battery will work better and last longer if you charge it fully, and do not exhaust it completely. 

Several members have said it is important for the owner to be able to monitor cell voltage. I think it is not so important IF you follow the usage guidelines above and expect to buy a new wheel every two years or so, or at least replace the batteries when they start to age (which is dependant on how they are used, the quality of the cells, and time). That is an IF. As wheels are made with more batteries in parallel and owners hope to use them for more than two years (or sell them to someone who will), it does become more important to monitor the cells/cell groups more closely. @Chriull and @RagingGrandpa, @meepmeepmayer and others have very smartly detailed in many posts why Smart BMS are [becoming] important (their posts are very informative and well-thought). 

 

tldr: Charge the battery 100% using the charger that came with the wheel and you'll be fine. When the wheel starts to wear out and/or the green "100% charge complete" light no longer illuminates, buy a new wheel.

Edited May 30, 2020 by WI_Hedgehog

[alcatraz]

When healthy cells are overcharged they don't lose the charge. They stay overcharged.

If they are showing signs of old age their internal impedance rises and so they develop accelerated self discharge.

So unfortunately one can't be sure ones battery pack is fine without measuring individual voltages. 

If you have the bad luck that somehow one of 20 groups isn't being voltage monitored and that group is the healthiest and gets repeatedly overcharged, then your pack end voltage still appears normal and there is danger developing.

I have three eucs. Two V8 and one Tesla. The V8 are great for monitoring because the packs have an enclosure. It means if you do open them and take the shrinkwrap off, the enclosure still protects. It means anyone can just take the shrinkwrap off and measure the cell groups for peace of mind.

On other wheels where the shrinkwrap is the outside of the pack, then one can't just leave it off. The only option is 1. pray that the bms is doing its job or 2. remove shrinkwrap and attach balance leads, followed by new shrinkwrap. 

The wheel manufacturers really need to have individual voltages visible in their apps. It could make a big difference for failsearching. Corrosion, dud cells, impact damage, crash damage, 2nd hand pack evaluation, bms evaluation, etc...

[Aibolit66]

8 hours ago, Gadgetfann said:

There was mention of the Nikola’s battery pack appearing “swollen” in this post by ecodrift back in December.   Photos - Ecodrift disassembles the Nikola 100V 21700

«swollen» - it is written in jest and means that the batteries are similar to the 18650, but thicker.

[Chriull]

2 hours ago, WI_Hedgehog said:

Wheels need a certain amount of power to move you. Power is:
Voltage (volts) x Current (amps) = Power (watts)
so as voltage drops current goes up, because the amount of power needed to do something doesn't change. Current causes heat, heat cooks the enamal on motor windings, MOFSETs, and battery cells, therefore the less heat that is generated the longer things last, meaning the more voltage a battery has the fewer amps that are needed to do the same amount of work, and the cooler things are. 

+1, but with the small restriction this is only true for the battery!

Motor output power is "driven" by the motor current. So any burden/acceleration/load demanded from the motor needs motor current. Together with the motor voltage (== motor generated back emv plus coltage sag over the motor coils) this equals to the motor input power. The mosfets (together with the motor coil inductance) transform battery voltage to motor voltage (and the two currents vice versa).

So motor determines and always gets full current, mosfets motor current(1), too and batteries stay with lower current loads the higher their voltage is.

(1) this is a bit simplified, as they are "pwmel'ed" with some duty cycle. During the "on" stage they have the motor current, during the "off" stage they theoretically have no current. But the motor coils force the motor current to be flowing - if this xurret would be cut by the mosfets the voltage over the coils would rise until the current flows again (arcing, ..). So this "coils current" is kept upright over the mosfets body diodes, some external freewheeling diodes (afaik not used in eucs) or the mosfets are switched to perform "active freewheeling" (less dissipated power cmpared to using the body diodes). 

So sinplified one can say that the mosfets have to bear the motor current.

 

1 hour ago, alcatraz said:

When healthy cells are overcharged they don't lose the charge. They stay overcharged.

From the current BMS analysis done till now the BMS balancing resistors discharge all overcharged cells downto ~4.2V again!

 

Edited May 30, 2020 by Chriull

[Aibolit66]

Do you think that the KIA3510A mosfets on the BMS Gotway are rated for 100V, and the 24S charger is 100.8V, is that normal?

1811151428_KIA-Semicon-Tech-KIA3510A_C116512.pdf

[Chriull]

8 minutes ago, Aibolit66 said:

Do you think that the KIA3510A mosfets on the BMS Gotway are rated for 100V, and the 24S charger is 100.8V, is that normal?

1811151428_KIA-Semicon-Tech-KIA3510A_C116512.pdf 295.76 kB · 1 download

They "only" have to stand the difference of charger input voltage and battery voltage. Once all the cells in series of the battery are below 0.8V the mosfet can die in peace, too - the pack is already long dead...

[davinche]

6 hours ago, Chriull said:

@davinche - were there still enough remains so they could do some forensics?

Gives a really bad feeling to hear from such incitents - great nothing happened to you!

It's time that all manufacturers use smart BMS with single cell (group) monitoring and cell temperature monitoring! (...without draining the batteries... )

... and that GW uses short circuit protection again, if they not already do...

It was too charred to try to bring home. I linked a google drive containing more footage/photos of the wheel. https://drive.google.com/drive/folders/16OmH-pp96OO1PcYyTYf2m-vYbLFyCFhW?usp=sharing

[davinche]

4 hours ago, meepmeepmayer said:

Wow, this easily could have been worse! Too bad it intimidated your friends.

• How was the air humidity and temperatures when this happened? Maybe some condensation got into the battery somehow?
• Firmware problem, and after you removed the charger, the "heating" started? 10 mins later, boom?
• Did you ever balance the battery (charge to full and keep the charger in for a longer while)? I didn't think even completely unbalanced cells would catch fire, but maybe they can?
  I am NOT trying to blame you here, this should not happen even without balancing!
• Maybe it's just a random coincidence, or a problem with the Panasonic (used, as per some rumors?) cells.

That's the explanations I can come up with. I don't expect one from Gotway other than "We are using different cells now, problem solved". Jason doing great work as usual.

1. It wasn't humid. It was a very nice day out on Sunday (Los Angeles).
2. It wasn't after removing the charger. I had charged the day before. The day of, my friends arrived and we turned on the devices and walked it to the parking lot. It was only when I tried to get on the Nikola when it shut off on me. After shutting off, 20s later it started smoking. Which of course then led to the fire...

3. I've only charged it to 100% maybe once or twice. The first couple of times when I was tweaking the settings of the fast charger. I then read that you shouldn't charge to full to preserve the life of the battery which is why I then switched to the 90% setting and only charged with the 90% setting since then. Amp-wise, I've always charged only at 3A.

One last point I want to make: regarding usage, I've only ever used my nikola to go to and from work. So I have never come close to depleting the battery and always juiced up after coming home from work.

[Chriull]

21 minutes ago, davinche said:

3. I've only charged it to 100% maybe once or twice. The first couple of times when I was tweaking the settings of the fast charger. I then read that you shouldn't charge to full to preserve the life of the battery which is why I then switched to the 90% setting and only charged with the 90% setting since then. Amp-wise, I've always charged only at 3A.

One last point I want to make: regarding usage, I've only ever used my nikola to go to and from work. So I have never come close to depleting the battery and always juiced up after coming home from work.

So absolutely nothing done wrong from your side - especially considering the milage!

Discussion about balancing/battery degradation in this case is purely academic.

32 minutes ago, davinche said:

It was too charred to try to bring home. I linked a google drive containing more footage/photos of the wheel. https://drive.google.com/drive/folders/16OmH-pp96OO1PcYyTYf2m-vYbLFyCFhW?usp=sharing

Thanks for the info!

Presumably nothing really survived this...

Kudos to @Jason McNeil for his fast and unbureoucratic response with the exchange of all the first batch batteries!

Hopefully this was the reason - as GW stated probs with the non LG li ion cells.

Not to imagine if such an incident happens in a flat - if someone dies it's unforgivable. And (i know a very narcisstic point) EUC's likely to get banned...

So @Jason McNeil please keep on your great work with pushing the manufacturers to provide safer and more reliable products! Additionally pushing GW to implement short circuit protection in the BMS again (if they not already did this with their latest wheels) - imho a great security fault from their side.

And also some smart BMS with cell voltage monitoring - wheels get used longer, batteries degrade and risk for incidents rise...

[davinche]

Coincidence that shortly after watching Kujis video on the V11, I put in a preorder for it through ewheels? Then today inmotion releases a Q&A video about the V11 and the first thing they talk about is battery technology / BMS?

 

Edited May 30, 2020 by davinche

[Archee Jan Bloch]

4 hours ago, WI_Hedgehog said:

My understanding from the members here is all* major EUC manufacturers use top-balancing circuits. This is smart, as it is cheaper, smaller, lighter, and allows the most power into the cells the fastest. 

*The exception may be Ninebot/Segway, though I don't know. Both miniPROs I have use active balancing circuits that are always running and monitor each cell, even when the batteries are removed from the units and just sitting around. Because of per-cell monitoring and reporting, it is easy to see how fast a cell can start going out of balance and by how much. The miniPRO charges to 4.00V (80%) and throttles at 30%, encouraging the owner to use only 50% of the batteries total capacity, and by doing so extending battery life from 500 charges to 2000 charges (assuming battery life means the cells can hold only 80% of their rated capacity).

Because it appears in general EUCs do not monitor individual cells, nor cell groups during discharge, best practice seems to be (and I belive @Chriull mentioned this in the battery charging threads):

• Let the battery cool before charging (reduces risk, extends life). 
• Do not charge until going riding. (The further the charge is above 50% = 3.85V the more the batteries wear out, even during periods of non-use).
• Charge fully (100% = 4.20V, engages the balance circuits),
• Check the voltage at full charge (there are devices made to do so if the EUC app doesn't report voltage). Voltage should be 100% of rated voltage = 4.20V x # cells in series. (4.2V x 20 = 84.0V, 4.2V x 24 = 100.8V). If the voltage starts to fall below normal, there's a problem developing--get the problem fixed immediately. 
• Plan your ride so the battery is not discharged below 30% = ~3.30V. (Gotway's 0% remaining is 3.3V, so riders are forced to observe this. King Song 0% is 3.0V, so the rider can self-manage their usage.) 

Wheels need a certain amount of power to move you. Power is:
Voltage (volts) x Current (amps) = Power (watts)
so as voltage drops current goes up, because the amount of power needed to do something doesn't change. Current causes heat, heat cooks the enamal on motor windings, MOFSETs, and battery cells, therefore the less heat that is generated the longer things last, meaning the more voltage a battery has the fewer amps that are needed to do the same amount of work, and the cooler things are. This is part of why using the last 30% of battery capacity is so hard on everything: not as much voltage and more current. This is also why speed is limited as battery capacity drops. 

Think of it this way: Voltage is like strength. Current is like the amount of motion. If you have three hours to move a large pile of gravel from the road where it was dropped off to the back of your yard, it is better to take a wheelbarrow full of gravel and use your strength, than it is to use a child's sand pail and run back and forth very fast--you will soon be hot and exhausted. You will need more time to recover. If your wife does not let you sleep (recharge) fully, and if you get a new load of gravel every day for a month, by the end of the month you may be worn out and not able to continue. However, if you used the wheelbarrow each day (strength) and were allowed to sleep (recharge) fully, you would be much healthier--not like new, but much better (maybe you would not like your wife for making such a large project, but you would still love her and continue on with her). : Your battery will work better and last longer if you charge it fully, and do not exhaust it completely. 

Several members have said it is important for the owner to be able to monitor cell voltage. I think it is not so important IF you follow the usage guidelines above and expect to buy a new wheel every two years or so, or at least replace the batteries when they start to age (which is dependant on how they are used, the quality of the cells, and time). That is an IF. As wheels are made with more batteries in parallel and owners hope to use them for more than two years (or sell them to someone who will), it does become more important to monitor the cells/cell groups more closely. @Chriull and @RagingGrandpa, @meepmeepmayer and others have very smartly detailed in many posts why Smart BMS are [becoming] important (their posts are very informative and well-thought). 

 

tldr: Charge the battery 100% using the charger that came with the wheel and you'll be fine. When the wheel starts to wear out and/or the green "100% charge complete" light no longer illuminates, buy a new wheel.

I asked how many times he chargers to 100% early because if he didn't , I see fundamental problem here. got my BMS balances around 85% , it is not smart BMS but it does continue balancing even after you  disconnect from the charger, if you see it on the flir camera, you will notice it will stay hot very long. most of the boards have program med to stop at 3.3 volt on cell. Even though the app says 0 you still are around 35% of the battery. 10% on the app was in reality about 3.45 volt on a battery cell * 24

however if he never charged 100% The difference between the cells increased, and it's possible that on the individual cell level 1 or 2 S could go bellow it's possible charging state.  this is specially know for Panasonic cells. if the voltage on cell drops below 2.5 it will start reading 0, it will pass through current  but the actual one of the 24 S will not change , which would mean that the other cells getting that extra voltage. I have personally experienced this numerous times in 20700B , but I forget my skateboard on for longer period of time and when I park the charger in it would show as charging but it would also melt the ABS enclosure around the BMS. when I dismantle the pack in fact 1st 4s4p out of 12s pack were done ... when I try to recover the cells with small current they would super heat up which can obviously affect anything around it. Even though he didn't ride it any possibility of small Regen while bringing the wheel outside houses could create tremendous heat from the bad cells... using 90% is great if you have BMS can balance on charge and discharge at all times   or let's say start balancing at 50% of cells or 0.05 V DIFFERENCE between cells , use diodes instead mosfets etc.  Gotway uses only charge bms , no discharge , any high current goes back to battery directly on Regen.  so if the cell can't accept charge it will heat up and the other cells will get overcharged.  .. basically there is no protection because main board thinks u are at 90% of pack but you can as well be on 120% on individual cell level except 2 or 3  that are at 0. On regen the current goes from mainboard directly back to battery without any protection... the over current protection kicks in at 102 V.  so if we take scenario well this would happen for quite some time and let's say that two S are affected.  that's 4 cells out of 24s pack..if cells gets charging 4.5 x 22 you get 99 V but way beyond manufacturer recommend limit   if they are below point of charge, BMS shouldn't allow charging but since  You have 2 packs  the voltage will try to equalize on both packs and that goes through discharge cable at much higher current creating tremendous heat on the bad pack - packs are connected together on Discharge cables xt60, charge cables xt30 and HV communication cable from BMS JST plug  .  if you disconnect charging cable on one pack the battery pack will get charged via discharge cable and there is no protection

 

[Chriull]

4 minutes ago, Archee Jan Bloch said:

I asked how many times he chargers to 100% early because if he didn't , I see fundamental problem here. got my BMS balances around 85% , it is not smart BMS

That's how EUC BMS work, too. But in his case it does not matter - after about 10 charges without any serious low discharge a battery pack cannot be seriously misbalanced, if the cells were not gravely mismatched from the beginning.

4 minutes ago, Archee Jan Bloch said:

butit does continue balancing even after you  disconnect from the charger, if you see it on the flir camera, you will notice it will stay hot very long.

Balancing happens by resistors beeing paralleled to the cells with voltages above 4.2V. and they stay, no matter if charger is connected or not, until the cells are below 4.2V again. (Kudos to @RagingGrandpa)

4 minutes ago, Archee Jan Bloch said:

however if he never charged 100% The difference between the cells increased,

He did, as he stated lately some times.

But this is nothing that is allowed to get the wheel go up in fire.

It could mismatch the cells a bit more, but not seriously in his case.

4 minutes ago, Archee Jan Bloch said:

  if they are below point of charge, BMS shouldn't allow charging but since  You have 2 packs  the voltage will try to equalize on both packs and that goes through discharge cable at much higher current creating tremendous heat on the bad pack - packs are connected together on Discharge cables xt60, charge cables xt30 and HV communication cable from BMS JST plug  .  if you disconnect charging cable on one pack the battery pack will get charged via discharge cable and there is no protection

 

GW and KS (and the other manufacturers most probably too) have synchronisation wires to prevent this - if one BMS cuts off, all others cut off too.

If you like and find time, i'd highly regard your input to https://forum.electricunicycle.org/topic/18317-battery-faq/?do=findComment&comment=307636

Critics/discovered mistakes/additional info is always welcome!

[Archee Jan Bloch]

26 minutes ago, Chriull said:

That's how EUC BMS work, too. But in his case it does not matter - after about 10 charges without any serious low discharge a battery pack cannot be seriously misbalanced, if the cells were not gravely mismatched from the beginning.

Balancing happens by resistors beeing paralleled to the cells with voltages above 4.2V. and they stay, no matter if charger is connected or not, until the cells are below 4.2V again. (Kudos to @RagingGrandpa)

He did, as he stated lately some times.

But this is nothing that is allowed to get the wheel go up in fire.

It could mismatch the cells a bit more, but not seriously in his case.

GW and KS (and the other manufacturers most probably too) have synchronisation wires to prevent this - if one BMS cuts off, all others cut off too.

If you like and find time, i'd highly regard your input to https://forum.electricunicycle.org/topic/18317-battery-faq/?do=findComment&comment=307636

Critics/discovered mistakes/additional info is always welcome

however there is no protection on the discharge cables and if one cell is bad and gets heat up it can affect the other cells as well, and the communication between the BMS HV and LV works only at specific voltage it doesn't affect individual cells if you measure the voltage on 24S pack out of the JST plug it will give you only two voltage readings so if one of them is bad it will not communicate with the other

[Chriull]

3 minutes ago, Archee Jan Bloch said:

however there is no protection on the discharge cables and if one cell is bad and gets heat up it can affect the other cells as well,

Yes. With regen charging there is no protection and there should also be just warnings for rider safety.

But this does not regard this case in any way.

[Archee Jan Bloch]

1 minute ago, Chriull said:

Yes. With regen charging there is no protection and there should also be just warnings for rider safety.

But this does not regard this case in any way.

The manufacturer of the BMS is the same manufacturer that does that main board it is also the same manufacturer that will make the veteran and they make the packs for Gotway, when we asked them what the LV /HV stands for, they said it's a secret and that they don't want to tell and they said that Gotway never ask them for it either also thank you for the link but it's basically what we just said and what have been known regarding the batteries and charging, my personal take out of this is that one or two cells in his back and got bellow charging capacity, which could even possibly damage the BMS, and he could just charge the other pack that will try to equalize  wit the damage pack

[Chriull]

29 minutes ago, Archee Jan Bloch said:

thank you for the link but it's basically what we just said and what have been known regarding the batteries and charging

I know. That's why i asked you, if you'd like to add adfitional information missing/correct rumours or false information(1)

But back to the topic - i fear that in this case there is a high chance that it has nothing to do with the batteries.

Most likely low speed overburden killing the mosfets/causing molten insulations snd by this shorted wires (to the motor).

Maybe a BMS/contact fault heating up the batteries. Or any other QC issue causing heatup/shorting.

This in combination with the missing short circuit protection led to a thermal runaway.

Could be any other problem with bad cells from the beginning or whatnot, but above cases would be my first guess.

Definitely the described charging behaviour of @davinche would be my very last guess.

Edit: i started this topic in the hope that it evolves into some kind of knowledge pool for the community sometimes...

Edited May 30, 2020 by Chriull

[Gadgetfann]

11 hours ago, davinche said:

The last bit of the (translated) post says: "Gotway encountered some difficulties when using Panasonic elements, so LG is now used." What is this "difficulty" that Gotway encountered?

I wondered the same.  It’s interesting that they only used the Panasonic batteries in a small batch of wheels and switched rather quickly to LG.

[davinche]

1 hour ago, Chriull said:

I know. That's why i asked you, if you'd like to add adfitional information missing/correct rumours or false information(1)

But back to the topic - i fear that in this case there is a high chance that it has nothing to do with the batteries.

Most likely low speed overburden killing the mosfets/causing molten insulations snd by this shorted wires (to the motor).

Maybe a BMS/contact fault heating up the batteries. Or any other QC issue causing heatup/shorting.

This in combination with the missing short circuit protection led to a thermal runaway.

Could be any other problem with bad cells from the beginning or whatnot, but above cases would be my first guess.

Definitely the described charging behaviour of @davinche would be my very last guess.

Edit: i started this topic in the hope that it evolves into some kind of knowledge pool for the community sometimes...

Is it only Gotway that lacks shortcircuit protection in their wheels? Now that I've had this experience... I'm willing to sacrifice some speed for a wheel that won't randomly blow up.

[Chriull]

9 minutes ago, davinche said:

Is it only Gotway that lacks shortcircuit protection in their wheels? Now that I've had this experience... I'm willing to sacrifice some speed for a wheel that won't randomly blow up.

Yes. But as written before i have no idea how it is with their latest/actual wheels. Their firmware and hardware changes and it takes quite some time until reports arrive here. I have none of the actual GW wheels.

And as written this are my guesses - some remote diagnosis. Discuss this with @Jason McNeil - (from all the reports and discussions i had with him) i'm sure he's experienced and will give you great advice on how to proceed for having carefree fun again!

Keep on wheeling! 

[davinche]

Kingsong advertises a "smart bms" with short-circuit on their website...

https://www.mykingsong.com/king-song-16x-electric-unicycle

[houseofjob]

Sorry for your incident @davinche

No one seems to have noted this, but I was skeptical on where the heck Gotway was getting their supply of these Tesla-use-only Panasonic NCR21700A cells from the start.

Back when this Nikola 21700 was launched, and still today, you cannot buy these cells from the usual reputable online outlets, only sketchy Chinese direct pages and sources (for reference, all our favorite EUC cells like MJ1, Pano GA, LG 50T, are widely available through reputable dealers).

The only thing I can think is that these are bulk salvaged cells from Tesla car wreckages, which would concern me regarding buying a new wheel.

[Chriull]

26 minutes ago, davinche said:

Kingsong advertises a "smart bms" with short-circuit on their website...

https://www.mykingsong.com/king-song-16x-electric-unicycle

It has short circuit protection, but is not smart enough in the sense of cell voltage monitoring.

And as stated before my knowledge of actual GW BMS is limited - i just know whats reported here...