[coder]

Fascinating. If this is the root cause (dendrites) than charging to 90% only suddenly makes a lot more sense. It would be interesting to know the age of the batteries where a fire did occur. I may be wrong, but it seems we have seen a lot more cases in Korea than elsewhere. Maybe the age of these vehicles was a factor.
A few takeaways from the story: The software update does not solve the issue. Charging practices do not 100% solve the issue, although make the fire less likely.

The article asserts that dendrites were implicated in many LION battery fires, including laptop, cell phone and aircraft fires, and indicates that we can expect an announcement soon from Hyundai, to point a finger at LG Chem. The allegation in the article is that the internal separator in the cells is sub standard. It is supposed to be able to prevent dendrite formation that crosses the separator boundary, and it allegedly does not do this.

Dendrites are crystalline, conductive structures that can form in the battery electrolite, and, if they grow to connect the 2
electrodes inside the battery, a short, and possible fire will result. The article explains this fairly well, it is worth reading.

 

[coder]

Oh one more thing to tie this new info to what we know/guessed: Dendrite formation is promoted by overcharging. So that is what the software update, and the practice of
"under-charging" aims to prevent. Seems like we are dealing in probabilities at this point (practices and workarounds that are thought to minimize the probability of thermal runaway).

 

[Kiwi]

We have to credit the 196 BMS software update in that it allowed Hyundai to acquire suspect battery packs in S.K. that are still in one piece. A handful of global owners have reported their cars being affected on FB and other forums. It also seems to have a hair trigger as a few have been sidelined have been put back on the road after clearing the error while others are getting pack replacements. Several already have had replacement packs installed with the shortest wait being 3 weeks.

What's fascinated me about this is that it appears that there are limited means of detecting a separator fault early, really only cell group voltage (v.s. input energy) and temperature. The voltage drop due to a partial short in a defective cell, being in a parallel group of three, will be diluted as the other two good cells will supplement the charge loss. Compounding that is their (apparent) choice of a 0.02 V A to D resolution which might be perfectly adequate for balancing but might be poor at detecting the onset of a separator failure. The Bolt has three decimal places resolution with the same 3 x 60Ah cells in parallel.

We know from BM-2 data (aux battery voltage) that the car gets busy for short intervals every hour for several hours after a traction charge event, powered solely off the aux battery. I could imagine that it's checking the 98 cell group open-circuit voltages, not specifically for balance but for unexpected drops over some hours indicating charge loss across a separator.

I can't imagine there's any way out for Hyundai other than replacing all battery packs with entirely new ones in affected examples, or a buy-back. It seems unlikely that any existing cells or modules could be re-used.

 

[coder]

Kiwi, could you explain this "partial short" concept? Ae you proposing that a dendrite which is partially formed (perhaps, poking through the separator), but not yet connecting the 2 electrodes can cause detectable voltage drop and / or a temperature spike, but not a thermal runaway quite yet?

 

[La Trucha Loca]

I wanted to post an update about my mileage range loss after the Campaign 196 Software Update. I had the update done on November 23, 2020. Before that, I always charged my car to 100% battery capacity and it always averaged a mileage range of 320 miles or so (highs of 330-340)...a little over 300 miles with the heat on (first 2 images). After the software update was done, my mileage range dropped dramatically to an average of 250 miles or less (last 2 images). You can also see how my dashboard display changed after the software update. The dates of the range snapshots are on each of the picture files (August, October, November, & December, 2020).

In an earlier post, I raised my concern and some folks on here said, "don't worry - it may take some time, but your range will come back / readjust." Well, I've charged about a dozen times since my software update, and I've had no range return. I did Recall Campaign 199 and had the Service Manager check the car out, and he said that it was operating fine, with no issues that he could see whatsoever. No service or troubleshoot codes came up, also.

I've opened up a case with Hyundai Customer Care, and am asking if the Dealership can take the car for the day and charge it up to 100% on a Level 2 Charger (like I have at home) and monitor what happens when the car hits the 80% charge mark. I am wondering, in my case, if my car just stops charging - or if there is something in the software that prevents it from charging to 100% (but yet gives the artificial impression on the dashboard that the car is charged up to 100% - again, see the pictures above that show that).

Any advice on how to proceed? I am convinced in my gut that something is awry and the loss of range really bothers me. I drove this car for 14 months before the software update - through the New Hampshire winter last year and all 4 seasons, and never saw the range this much in the toilet. Something is definitely up, because the range loss is real and I'm having to charge my car more to go less miles. There is no denying something changed...the question is, what?

 

[Kiwi]

Well, I'm more suggesting that a loss of charge in one group of three parallel cells is about all that could be detected by the BMS, and I would assume the defects discussed in the article would lead to that loss. To attain thermal runaway would require that the defect worsened in a relatively short period of time with the increase in temperature.
I gather the purpose of the software upgrade was to carry out whatever diagnostics it could based on existing hardware sensors to catch the defect before thermal runaway took over.

Here, the AC charging finished at 04:20 (to 60% SoC) and momentary hourly checks follow over three hours. I can only assume it's looking at the 98 cell group voltages while the traction battery is disconnected and looking for any that drop unexpectedly. I've noticed this happening after every charge event. What happens during the delay at 82% is another question.

 

[coder]

Kiwi, what are we looking at on this chart? I reckon the x axis is time elapsed in hours. What is the y axis? Some voltage 0..15v. Aux battery voltage?
The square spikes being the aux battery being charged back up to nominal? The dips are the aux battery being drawn down, because it is supplying power to the BMS computer, which performs the actions you describe ( while powered by the Aux battery) ?

 

[coder]

Trucha, are you charging to 100% now? Do you have SoulEvSpy to read out your cell group voltages?

It is unlikely that Hyundai silently changed the charge percent so that they fake it out and report 100% while only charging to something less %.
I think people would notice. In any case, the cell voltages would indicate that.
OTOH never say never.. It is rumoured that Tesla actually has dropped their available range by 10% on the Model S in response to their battery fire issues.

Several of us have the software, and we can compare readings. If there has been any changes introduced by the software update, the (allegedly) 100% charged state would report a lower voltage for the cells than what it is supposed to be fully charged.. I think somebody posted the "fully charged" voltage here already.

There are so many things that impact your actual range, it is hard to compare "before and after". At this point, it is just a feeling that maybe your battery capacity or your efficiency got worse for some reason. Maybe it did, but to have an accurate comparison, you would have to keep the major variables the same. For example, the winter you are comparing to (before the update), may have been warmer on the average than the current winter. To say " well it was winter in both cases" is not accurate enough. Average outdoors temperature has a surprisingly huge impact on range. When I have seen the temp vs range graph, I was floored. I was like: "Oh.. if it gets cold enough that may actually cut my range in half! That sucks..."
.
In any case my expectations (like I mentioned before) have been lowered, I am happy if my car does not catch on fire on me. Never mind the range.

Joking aside. I think there are 2 elements to the range: The battery voltage achieved when you are done charging, and the average efficiency of the car (which it tracks overall and per resettable interval) The efficiency is expressed in "miles per KWH achieved" .

We discussed this here before, you cannot take the GOM (Guess-O-Meter ) predictions as "THE" range.
It is only a heuristic estimate. For all we know, the update may have changed the GOM algorithm.

 

[Kiwi]

Kiwi, what are we looking at on this chart?

Exactly as you said. It's the aux battery voltage on the Y-axis but you could also consider it the 12V system voltage. When the LDC powers up off the traction battery it takes that to a regulated 14.6 V. The aux battery will of course draw as much current as it needs to charge, typically starting at up to 50 A. Over the 20 minutes the current (and therefore energy) will normally trend towards zero as the battery takes up a charge.

At the left starting at 00:00 AC traction battery charging starts and the aux is charged for 30 minutes exactly, the normal time period. After that, and while the AC charging continues, the system voltage is regulated to 13.1 V which keeps the aux battery from being discharged.

 

[Kiwi]

News of pending pack replacements in S.K.

Hyundai reportedly set to replace Kona LG batteries in Korea - Electrek

Hyundai to announce a recall of all Kona, Ionic and Electic to replace the batteries due to fire concern after repeated fires across their lineup.

electrek.co

 

[coder]

Thanks for posting this. A few key points from the article:

• There have been now 5 reported fires in SK that involve cars with the updated BM software.
• Stay tuned for an announcement on the 19th, of Hyundai replacing all batteries in Konas, Ioniq EV-s and Ioniq Hybrids (in SK) .
• LG who was claiming the battery problems were not their fault, is picking up at least half of the cost of the recall (because, hey, it had nothing to do with them...)
• Hyundai is threatening to switch battery suppliers.. except the next logical supplier, SK is banned from the US for the next 10 years.

On top of this bad news for Hyundai, comes the recent news that the Biden administration reinstates the tax credit for Tesla EV-s. So suddenly it makes a lot more sense to buy a Model 3 than a Hyundai. This battery issue IMO cost Hyundai a fair amount of goodwill and credibility. If I was looking for an EV today, it would be hands down a Tesla model 3, and I would not touch a Hyundai EV with a 10 foot pole. One of my major differentiators for picking the Kona vs a Tesla was the existence of the federal tax credit (and the Massachsussets $2K EV rebate).

 

[Kiwi]

I wasn't aware of the renewal for the Tesla credit. I'm a US citizen and am liable for US taxes despite not living there, but I could not claim the federal (or CA) credit on my Kona.

I don't see any reason why LG Chem can't make suitable replacement batteries provided the fault has been positively identified and corrected. I think we all would agree that Hyundai now cannot avoid replacing all LG Chem Kona batteries worldwide. For the inconvenience of avoiding the top 10% SoC for the year or so it will take, I think a brand new battery at 100% health is a fair exchange.

I have complete confidence in Tesla long term but until the production quality matches the rest of the car market I'd be reluctant to buy one. As an engineer, I'd notice and be irritated by minor defects even if they are inconsequential in practice. Here the Model 3 SR+ is cheaper than the base Kona, but there's no Tesla dealer or service center within 300 km of me and I'm unsure how efficiently warranty repairs would be completed, never mind any collision repairs. Hyundai are just a 10 min walk down the street and my small dealer has proven to be reliable and honest. There's another one a 20 min drive away as well.

 

[coder]

Kiwi,

I get your point about the availability of servicing in your locality. I would probably come to the same conclusion.

Are you saying that the Tesla model 3 has more, or comparable problems, (small or large) than the rest of the EV market (for example comparing it to the Kona EV?)
I would be curious to hear about some of those "minor defects" which would annoy you and make a Model 3 worse for you than the Kona EV.
I have not researched the topic, so I go by superficial information I randomly come across here and there. My impression was (could be wrong) that the Tesla Model 3 currently is a more mature product than the Kona. My impression is that Tesla aggressively solves any problem that they find. Case in point, their redesigned battery cover (3 layers, with a titanium layer, which sounds like a "take no prisoners" radical fix) . This contrasts rather unfavorably with Hyundai's dragging their feet on the tackling the battery problem head-on.
I think Hyundai is yet to ponder the problems that road debris can cause to the propulsion battery, and has no similar protection in place.
I also have the impression that Tesla's self-driving capability is significantly ahead of Hyundai-s. The Hyundai design in many respect strikes me as "me too".
This may change with the advent of new Hyundai EV-s coming out, but clearly the Tesla vehicles were designed as pure electric platforms. Vs The Kona currently is just a converted ICE platform, and it shows. The Kona ends up as "a curiously cramped, small suv" vs the model 3 is "a surprisingly roomy sedan".
Just my impressions. I am kind of down on the Kona at the moment, can you tell?

 

[Kiwi]

Tesla's problems are mostly about fit and finish according to JD Power. A quote from CNBC: "Owners of 3-year-old Teslas reported 176 problems per 100 vehicles, compared with the industry average in the U.S. of 121 problems per 100 vehicles. Tesla owners reported more problems with their exterior and interior than with other systems like propulsion, battery or infotainment and navigation."
I noticed this (photo below) in a video about the Model Y 7-seat version. It may be just a matter of pushing it back under the cushion in this brand-new car but sloppiness like this would annoy me. There are many other videos on YouTube pointing out Tesla paint quality issues and delays in getting repairs done. Collision repair costs are also another source of complaints. Car insurance here is very expensive because of these costs, twice what the Kona is.

I haven't seen any EV-specific industry quality stats. With Hyundai/Kia EVs the motor/reducer noise has anecdotally led to the 2nd most complaints from what I've seen on other forums, dead 12V batteries are the first, but incidents seem to still be relatively-rare considering the large numbers on the road. And thanks to LG Chem it's now the main battery, which identically affects the Bolt and is not really Hyundai's fault and they appear to be acting to rectify this, as fast as a large company can. Tesla is now learning the downside of placing most controls into one device, the model S/X screen life issue, where, disappointedly govt intervention was required.

Tesla are absolutely responsive to design issues which of course is good. They also place vehicle performance (including charging) above what I would label serviceable conservative design, somewhat like BMW has done for decades. E.g., all Tesla battery cells are glued together in large slabs to maximise heat transfer to the cooling media, but those are not serviceable. Hyundai/Kia use a more conventional design (check out John Kelly's teartown video on the similar Bolt pack). There are also Sandy Munro's videos on Tesla's design and production engineering (but sometimes I think he's a bit star-struck by Musk.) As I said, I'm all over Tesla and have a significant investment in the company but it's more about the potential than what they can deliver to me as a consumer today.

As for the the Kona's size, I think that's up to the customer to determine what suits them before purchase. I'm personally fine with it and the model 3 would be too wide for my inner-city driveway, which was designed for 1920s cars. I'd struggle to believe that Hyundai did not intend an EV when designing the Kona's lineup. The independent rear suspension is nicely compacted to the rear and the EV-version body accommodates the extra clearance for the battery. The detailed design under the hood doesn't indicate to me any compromises. The heat pump integration with the HVAC is really good (I know US version didn't get that). Everything is well laid out and appears serviceable, as I'd expect.

My Kona has been flawless however my vehicle requirements have diminished and I was thinking of moving on to an even smaller ICE car just to have one on-hand when I need to get around. The battery thing has complicated that as the resale is not great at the moment. But my 3-year warranty is up Oct this year and it's always easier to move a car on while it's still covered. Battery warranty goes to 2025 I think.

The self-driving thing does nothing for me as a consumer but I can't ignore Tesla's accomplishments from a investment perspective. I've always wondered why they complicated their EV rollout with such costly technology. But they're leaders now and are doing things with cameras that I would not have thought possible without LIDAR.

EDIT: an owner over at Speak EV has an short comparison after momentarily changing to a Model 3.

 

[R P]

Kiwi,

I get your point about the availability of servicing in your locality. I would probably come to the same conclusion.

Are you saying that the Tesla model 3 has more, or comparable problems, (small or large) than the rest of the EV market (for example comparing it to the Kona EV?)
I would be curious to hear about some of those "minor defects" which would annoy you and make a Model 3 worse for you than the Kona EV.
I have not researched the topic, so I go by superficial information I randomly come across here and there. My impression was (could be wrong) that the Tesla Model 3 currently is a more mature product than the Kona. My impression is that Tesla aggressively solves any problem that they find. Case in point, their redesigned battery cover (3 layers, with a titanium layer, which sounds like a "take no prisoners" radical fix) . This contrasts rather unfavorably with Hyundai's dragging their feet on the tackling the battery problem head-on.
I think Hyundai is yet to ponder the problems that road debris can cause to the propulsion battery, and has no similar protection in place.
I also have the impression that Tesla's self-driving capability is significantly ahead of Hyundai-s. The Hyundai design in many respect strikes me as "me too".
This may change with the advent of new Hyundai EV-s coming out, but clearly the Tesla vehicles were designed as pure electric platforms. Vs The Kona currently is just a converted ICE platform, and it shows. The Kona ends up as "a curiously cramped, small suv" vs the model 3 is "a surprisingly roomy sedan".
Just my impressions. I am kind of down on the Kona at the moment, can you tell?

Latest J D Powers survey just came out and Tesla is #30 out of 33. Last year they were dead last.

Tesla Fares Poorly In The 2021 J.D. Power Dependability Study

JD Power released its annual dependability survey and ranked Tesla for the first time. It wasn't good.

insideevs.com

My son has a Tesla M3 which is now out of warranty, and he still has a few outstanding problems which he is just living with for now. But over the last 3 years (120,000 kms now), he has it in the shop for probably 10-15 times for repairs and recalls, more that all my cars I ever owned put together.

I much, much prefer our Kona EV over my son's M3. Besides reliability, there are many other reasons that make it a far inferior car than the Kona. This center monitor is just stupid. When I drive with him, 5 minutes doesn't go by with him looking and poking at it, taking his attention off the road. I consider it dangerous, actually. He got the top end interior, and I can tell you that a top trim Corolla looks far more luxurious inside. That fake wood paper strip across the front dash especially looks hideous.

 

[coder]

OK. Thanks for the Tesla info. Like I said, my opinion was not based on any research, just an apparently incorrect superficial impression.
I stick with the Kona for now. Let's see where the battery problem goes. I have not seen any further announcements. Did you guys see any further news?

Does anybody know if the Kona EV needs any maintenance? (Other than a list of "check this and inspect that" BS)
I am coming up on 5K miles.

 

[R P]

I think we need to wait until LG Chem and Hyundai work something out. I have no doubt that Hyundai will not hesitate to spend billions if necessary to make things right with their customers and preserve their reputations. I would say this, for not just the SK corps, but also the Japanese corps. I have been to both SK and Japan on business, and can tell you they are a very, very proud people and take these kind of failures very, very seriously. BTW, I would not say the same about our domestic corps.

If you think of how these battery failures/fires happened, and the actions that Hyundai took (software updates to catch them before they totally failed), I think they did the right thing. No doubt the SK govt was all over them, too, as they should. But I think we are now seeing some very good progress towards a final fix.

And this is coming from a guy (me) who has had both his reduction drive and motor replaced, and now facing a possible battery replacement. However, I knew this was a new model, and they all have new technology risks. Yeah, maybe some companies (like Toyota) are better than others with their quality control. But for me, I am still hopeful that in the end all the problems will be fixed, and we will enjoy the car that we thought we bought, for a long time to come.

 

[Kiwi]

Does anybody know if the Kona EV needs any maintenance? (Other than a list of "check this and inspect that" BS)
I am coming up on 5K miles.

Presumably you have the maintenance schedule, on the web perhaps, although I'm learning the US one may be different than RoW. At 5,000 miles a tire rotation F-R is really all that's needed. There may be a cabin filter change called out yearly but I ask the dealer to ignore that due to my low kms and they do - but still sign off the book. Saves me $90 a year. At two years there's a brake fluid change and I had that done, I think it's important on the Kona due to the very complex hydraulics. Inspections are the usual things, coolant level, brake condition, and the underside of battery housing for dents.

 

[R P]

At two years there's a brake fluid change and I had that done, I think it's important on the Kona due to the very complex hydraulics. Inspections are the usual things, coolant level, brake condition, and the underside of battery housing for dents.

Is there anything about the EV Kona that is different than the ICE Kona, that may require a more frequent brake fluid change?

 

[coder]

I do my own TR, I have a couple of those rapid racing jacks, compressor and air wrench so it is pretty easy. I can bleed the brakes as well, have the equipment for it.
Do you guys know what brake fluid the Kona EV takes? Something special (unlikely) , or jus a high quality dot 3/ dot 4 (maybe low viscosity) ?
This is probably in the owner's manual, I should really go find it and look.

It is recommended practice to replace the brake fluid every couple of years on many modern cars for optimal braking performance. Brake fluid absorbs moisture from the air over time, so this is strictly a time thing, not a miles thing. Givent moisture present in the fluid, when braking aggressively, steam bubbles form, making the fluid mildly compressible. This makes braking less effective, but perhaps not enough to make a noticeable difference in daily driving. Also keep in mind, that a top shelf brake fluid is less prone to absorb moisture from air, so tends to last longer. Choosing the fluid wisely makes a difference.
There are many cars on the road running 5-10 years with the original factory fluid fill. But those older cars often have bigger problems interfering with braking efficiency.
Sticky sliding pins, hung up pads, uneven pad wear, beat-up rotors, those effects of a poorly maintained brake system is probably worse than the old fluid.

One way to do brake fluid replacement on a rational basis is checking the moisture content, and base it on that. There is a tool for that.

I imagine most of us do not use brakes aggressively enough to warrant bleeding the brakes every 2 years, but, if you are keen on maintaining peak efficiency in an emergency braking,
that maybe makes this worth while...