Genevamech
Active Member
Nov 2019. I've not yet received any official communication regarding the recall, but it comes up on a VIN search. Wondering if there's going to be a pattern of who gets fixed first based on date of manufacture?
Worldwide recall according to this 11 Oct news item.
- Thread starter KiwiME
- Start date
- Replies 194
- Views 37K
Nov 2019. I've not yet received any official communication regarding the recall, but it comes up on a VIN search. Wondering if there's going to be a pattern of who gets fixed first based on date of manufacture?
July 2019 for mine. Also no formal notice from Hyundai as yet. My dealer is about 2 hours away, so I will likely also wait for more detail and others experience before scheduling a visit.
Thomas Mackay
New Member
June 13th 2019. Ran my vin against the NHTSA website and it came back clean.
On the Hyundai recall website it comes back for Recall 196.
Now that's different, in Canada it's the opposite:
https://insideevsforum.com/communit...n-us-hyundai-recall-website.9755/#post-112283
Also, welcome to the forum.
I suspect these unfortunate circumstances will draw more new members into discussion more so than before.
Have you had the BMS update yet?
https://insideevsforum.com/communit...-kona-recall-in-form-of-software-update.8713/
Last edited:
KiwiME
Well-Known Member
I thought I'd posted this article from 5 days ago but couldn't find it. It takes a few reads to absorb but it's an interesting article once you work around the few translation errors.
https://tekdeeps.com/이슈분석kona-electric-car-fire-cause-battery-safety-margin-considered-light/
This study seems to conclude that the safety margin Hyundai has selected is statistically unsafe due to trying to manage a large number of cells placed both in parallel and series, and how closely they approach the maximum allowable Li-ion per-cell voltage at 100% SoC. It also contrasts the Kona with another problematic power storage product they refer to as ESS.
https://tekdeeps.com/이슈분석kona-electric-car-fire-cause-battery-safety-margin-considered-light/
This study seems to conclude that the safety margin Hyundai has selected is statistically unsafe due to trying to manage a large number of cells placed both in parallel and series, and how closely they approach the maximum allowable Li-ion per-cell voltage at 100% SoC. It also contrasts the Kona with another problematic power storage product they refer to as ESS.
Genevamech
Active Member
So the crux of the problem seems to be that aren't allowing enough headroom for cell-to-cell voltage deviation, potentially resulting in some cells being overcharged, which causes damage and eventually failure.
If that's the case then it's immediately clear why the BMS update tightened the cell voltage check by an order of magnitude: they want to be very certain that no cell is getting overcharged. If they detect that something is out of balance, then the only completely safe remedy is to rebuild the battery pack to remove any cells that may have experienced overvoltage.
This could be a significant problem for anyone who relies on 100% charge. If you only rarely charge to 100% then any potential damage is mitigated, since you would have extra buffer space for cells to be at a higher voltage than the others without going over the chemistry's limits. The BMS update would also be able to detect such issues before they became failures.
You might think that charging to 100% would help since it allows the cells to balance, but if the balancing algorithm isn't sensitive enough it can still result in damage... so I would expect part of the update to also improve cell balancing to tighter tolerances, unless the hardware isn't capable for some reason. I have to wonder now, if all the incidents were vehicles that were regularly charged to 100%... I wouldn't be surprised if that's the case.
The only proper long-term solution to this kind of problem is to increase the buffer space, so that means either a redesigned pack with larger actual capacity or reduced range for the same pack. It seems Hyundai is aiming for the third option: being really careful about not crossing that line...
ericy
Well-Known Member
Translation wasn't bad - I still wonder what "National Fruit Tree" actually is, and how it writes a report. I can read some Mandarin (and to a much lesser extent Japanese). I have never studied Korean, so I can't make anything out of the original.
It does raise interesting questions though. For each of the cars that caught on fire, most if not all were attached to a charger. And the question then comes up what the charging limits were set to for each of them.
It also talks about the packs in an e-Niro. Different chemistry, different characteristics, so any analysis of the Kona won't apply to e-Niro.
In the end, for those of us who regularly only charge to 80%, there ought to be little to no danger. But on the other hand, my understanding had been that the only way to balance the pack is to charge to 100%, and it seems like there need to be other ways to do this.
In my case, the one cell was 0.12V below the others, so in theory there is no danger of over-charging. I don't know why it would be that this one cell was slightly lower than the others. That's still a mystery. And then there is the question of whether attempting to balance the pack could bring it back into line with the others, and resolve the difference without needing to replace the battery.
Jimct
Active Member
I think you just drew some very reasonable conclusions and I do appreciate your posting them. It seemed to me, for the past couple of months, the sensible thing to do was to set the default charge to 90%, 100% only when I'll be driving a fair distance soon after. In my case that amounts to 100% about 1 out of 3 times. If what you say about balancing the pack is accurate, that should cover it. I agree that these fires seem to occur while charging, so that's done outside from now on.
I want to mention one thing, and it may be nothing, but I almost always use my level 2 charger and have had no problems. But 3 times I tried using the supplied 120V level 1 charger, and every time it tripped the GFCI outlets in my garage after about 1 hour - I used different outlets each time, same results. I thought it might be a faulty charger, but now I'm wondering...
Ed C
Active Member
The article states that 'extreme operating conditions' leads to 'poor progression' that eventually causes one of the cells to fail. When that cell fail, it can cause a short circuit which can cause an ignition. 'Extreme operating conditions' was not really described in detail. Is it always charging to a high SOC??? Since it is pointed out that the Kona EV has a lower 'safety margin' of capacity, compared to a Tesla or Bolt. Kona has available 62kW of its 64kW, compared to a Tesla that has a 75kW of its 80kW battery. The same for a Bolt which has available 55kW of its 60kW battery. So that's how Korea's Ministry of Land, etc calculated that the Kona EV has only a 3% 'safety margin' buffer.
As for 'poor progression', that is not defined either. Does it mean battery degradation from frequent charge/discharge cycles???
Finally, it is pointed out that the eNiro uses a NMC811 as opposed to a Kona EV NMC622 battery. Meaning that eNiro's battery is composed of 80% Nickel, 10% Cobalt, 10% Mangenese; as compared to Kona's 60% Nickel, 20% Cobalt, 20% Mangenese battery. Cobalt is the more volatile element that will tend to cause fires; but more Cobalt makes for a better battery (Teslas has 80% Cobalt). It does not necessarily mean that the eNiro will be safer; but will tend to be less fire-prone compared a Kona EV if it was to operate under the same 'extreme operating conditions'.
So, it may appear plausible that in order to increase the 'margin of safety' in a Kona EV, they may have to increase the capacity buffer above 3% in the existing battery. That means that you will have a decreased available battery capacity, meaning a slight loss of range. If they will increase the buffer capacity to industry standards like the Tesla or Bolt, then you will probably see that the Kona EV might get a 59kW capacity to the current 62kW. a loss of 3kW of current range.
As for 'poor progression', that is not defined either. Does it mean battery degradation from frequent charge/discharge cycles???
Finally, it is pointed out that the eNiro uses a NMC811 as opposed to a Kona EV NMC622 battery. Meaning that eNiro's battery is composed of 80% Nickel, 10% Cobalt, 10% Mangenese; as compared to Kona's 60% Nickel, 20% Cobalt, 20% Mangenese battery. Cobalt is the more volatile element that will tend to cause fires; but more Cobalt makes for a better battery (Teslas has 80% Cobalt). It does not necessarily mean that the eNiro will be safer; but will tend to be less fire-prone compared a Kona EV if it was to operate under the same 'extreme operating conditions'.
So, it may appear plausible that in order to increase the 'margin of safety' in a Kona EV, they may have to increase the capacity buffer above 3% in the existing battery. That means that you will have a decreased available battery capacity, meaning a slight loss of range. If they will increase the buffer capacity to industry standards like the Tesla or Bolt, then you will probably see that the Kona EV might get a 59kW capacity to the current 62kW. a loss of 3kW of current range.
Last edited:
Good article. thanks for that. Apparently we still get to keep our ~300 mile range EV, we just have to wait until degradation kicks in, about 3% should do it
ericy
Well-Known Member
I made a half-hearted attempt to find the original report, but Google doesn't work on Korean. Or perhaps it does, but you would need to enter Korean search terms.
Extreme operating conditions are probably going to be either battery temperature, state of charge, and for a high state of charge, time (how long the battery is at a high state of charge).
apu
Well-Known Member
I don't think this a problem with your car. GFCI devices are know to trip on occasion during EV charging, you might want to try it with a non protected circuit. There is some thought that there is high frequency feed back from vehicle inverters to the GFCI, inducing a leakage current on the ground line, and thereby tripping the GFCI. This can appear like a random effect, appearing to work most times and not others.
At the rate batteries seem to degrade or rather not degrade in these cars, that should only take ten years or so.....lolGood article. thanks for that. Apparently we still get to keep our ~300 mile range EV, we just have to wait until degradation kicks in, about 3% should do it
Ed C
Active Member
This section hits me:
"Accordingly, it is necessary to upgrade the battery management system (BMS) to alleviate battery operating conditions through a recall measure by the Ministry of Land, Infrastructure and Transport, and to accurately filter out any damage to the batteries of existing vehicles sold or operated. It is pointed out that these two methods must be used in parallel to prevent future accidents."
So from my interpretation, 'alleviating battery operating conditions' means increasing the margin of safety of the battery buffer capacity above the 3%.
So it may appear the BMS update in the recall might very well decrease the available battery capacity, resulting in a decreased in range.
In parallel, they will also check the battery of existing cars sold for battery damage.
"Accordingly, it is necessary to upgrade the battery management system (BMS) to alleviate battery operating conditions through a recall measure by the Ministry of Land, Infrastructure and Transport, and to accurately filter out any damage to the batteries of existing vehicles sold or operated. It is pointed out that these two methods must be used in parallel to prevent future accidents."
So from my interpretation, 'alleviating battery operating conditions' means increasing the margin of safety of the battery buffer capacity above the 3%.
So it may appear the BMS update in the recall might very well decrease the available battery capacity, resulting in a decreased in range.
In parallel, they will also check the battery of existing cars sold for battery damage.
ericy
Well-Known Member
It might, but they might only need to increase the buffer by 1-2%. How much would people complain if they lost 5 miles of range? I can't speak for others, of course - I would barely notice it myself.
They might also adjust BMS to watch for cell groups that approach the maximum voltage, even if the pack as a whole is not fully charged. Which is sort of related to watching for the maximum deviation in cell group voltages.
Checking existing batteries for damage is something I am confused about. What would they actually be doing to the things? Visual inspection is unlikely to reveal anything, so for the most part they would need to rely upon the BMS to read out the cell group voltages and the ground fault impedance. But the existing BMS update they already pushed out is already watching for cell group voltage differences - so what new thing would they be doing? I guess we will find out once there is a TSB for us to read.
Clamps
Active Member
I know it’s isn’t helpful but I sure wish I would have picked the Model 3 now. I’ll take a few misaligned body panels over igniting batteries.
Having said that, I remain hopeful. I really don’t want a loss in range but if it’s minimal, five miles as is suggested I could deal with that.
Having said that, I remain hopeful. I really don’t want a loss in range but if it’s minimal, five miles as is suggested I could deal with that.
Ed C
Active Member
Or you can have the roof fly off if you would like a Model Y....
https://electrek.co/2020/10/05/tesla-model-y-roof-fall-off/
Pick your poison....
Anyways, like I said, in order for Hyundai to be on par with its peers like Tesla and Bolt on the buffer capacity, you will lose 3kW of range. Depending how you drive, whether you usually get 3 miles/kWh or even 6 miles/kWh, you can be looking at 9-18 miles of range reduction.
In addition, since Hyundai deliberately lower their margin of safety of its buffer capacity compared to its peers, to maximize the Kona EVs range....you are looking at a lawsuit. It was not an accident, but an intentional act of lowering the car's safety in order to maximize its range for selling purposes. Oh boy....
https://electrek.co/2020/10/05/tesla-model-y-roof-fall-off/
Pick your poison....
Anyways, like I said, in order for Hyundai to be on par with its peers like Tesla and Bolt on the buffer capacity, you will lose 3kW of range. Depending how you drive, whether you usually get 3 miles/kWh or even 6 miles/kWh, you can be looking at 9-18 miles of range reduction.
In addition, since Hyundai deliberately lower their margin of safety of its buffer capacity compared to its peers, to maximize the Kona EVs range....you are looking at a lawsuit. It was not an accident, but an intentional act of lowering the car's safety in order to maximize its range for selling purposes. Oh boy....
Last edited:
navguy12
Well-Known Member
May I assume your figure of 80% Cobalt in the makeup of a curent Tesla battery is a misprint?
If not, here is a link to a story that lists a much lower (cobalt) figure:
https://www.mining.com/tesla-battery-day-a-possible-blow-to-cobalt-miners/
Ed C
Active Member
The 80% was from the Dr. Dalm or whatever-his-name video that was posted. It could be the old Tesla batteries which uses 80% cobalt and 20% ferric (?) oxide. Tesla obviously had changed its battery composition recently. It appears they are using Nickel-Cobalt-Aluminum for their batteries and are planning cobalt-free batteries.
Cobalt causes fires when exposed to oxygen. It is the most volatile, but it helps make a longer lasting battery. In addition, cobalt is very difficult to mine due to its scarcity and its toxicity during processing.