Hyundai Kona Forum banner

101 - 120 of 160 Posts

·
Registered
Joined
·
187 Posts
This leaves us with an interesting question...How long do people have to wait, if the software update reveals a problem? Are they safe during that time, with just the software update?
 

·
Registered
9/18 EU base model
Joined
·
101 Posts
Well, given that those are notes under the applicability list I'd expect that they apply to the work being done.
I'm currently in a discussion with someone who has an appointment (no idea what country) and was told by the service manager without a hint of irony to charge up to 100%. He thinks they are confusing it with the 12V battery.
 

·
Registered
Joined
·
187 Posts
Yes reading the bulletin that would make sense. I had a vague uneasy feeling about the instructions.
I have my doubts about the average service tech, when he has to follow a fairly difficult to parse text. Many of us completed the 960 campaign, so items c and d will be up to date already. Ok.

I wonder how many cars going to get hosed because they do not follow the guidance to "turn on the car for 20 minutes" so that the 12v aux battery can charge itself. (Resulting in update failure or controler damage".), or despite of the red lettered warnings, perform all updates in automatic mode, where the 1st 2 updates won't take...

Whenever I have something done at the dealer, there is always an inevitable follow-up visit to make to "fix the fix". They replaced the tranny on on of our sonatas, and forgot to reconnect some wire harness. Then they did some work and "forgot" to put the skid plate back. When I yelled at them, they just grabbed some random old beat up skid plate, off of another car and slapped it on, with only half of the mounting screws installed. So based on my experience of the sloppy, careless, inattentive work I get at the dealer, I fear how this is going to go. This is the very reason why I do all my maintenance on all of out cars.
 

·
Registered
9/18 EU base model
Joined
·
101 Posts
... This is the very reason why I do all my maintenance on all of out cars.
Years back I took my new 2003 VW Golf TDI into a San Diego VW-Audi dealer to have a minor issue looked at. An hour later I felt in my gut something had gone wrong and sure enough got a call to say they had an "accident", bent rear wheel and torsion bar, body smashed in on one corner. Of course they wouldn't fix it themselves and the owner even verbally threatened me. A month later my car came back with a half-baked repair from a nearby body shop. It never drove the same, the back would hop to one side on bumps.
To top it off, the original problem turned out to be that someone had sabotaged my car by pushing a plastic glove down the filler tube. I think it was an employee from the dealer I had purchased it from because I had asked the salesman to double-ensure he didn't accidently put gas in it, a not uncommon occurrence with these rare diesels at the time.
 

·
Registered
Joined
·
187 Posts
So finally got my Konnwei dongle. The "free" version of SoulEvSpy does not work too well. (Kept disconnecting and reconnecting, and would not display the battery cell map, nor the BMS info) The $15 version actually works. I had my car sitting outside for a couple of days, at about 40% or 50% charge. All the cells started out at a uniform 3.62v. Bank temps started out at 15,14,14,16 celsius. Brought it into the garage ad plugged it into my chargepoint. I am watching it as it is charging, to see if there is any variation at any point. What I see, as I am charging, periodically some cells are lagging behind the others by 0.02V but they catch up after a while. I am Charging at a steady 7KW rate.

Thinking about this: I think the "limit charge to 80% is sound advice. It eliminates one possible failure mode, the accidental overcharge of a cell because of cell voltage variation. If you limit the charge, you limit the max cell voltage and even if you have cell voltage variation, you are still avoiding overcharging any cells. The lower you set your limit, the safer you will be, at least with regard to this one particular issue. The charger is not charging individual cells, it charges banks, which contain multiple cells. Hence if the cells are not of the same voltage, it is possible to overcharge those that are at a higher level. Kiwi, correct me if I am getting this wrong :)

Now if you have some other problem, like Hyundai says "cell separator physical damage" or some other physical defect, I think that would manifest first as weird cell voltage levels.
I am in my living room and connected to the dongle via blue tooth, so I can keep an eye on what is happening as I am charging
I notice the aux battery is getting charged from the getgo.
 

·
Registered
9/18 EU base model
Joined
·
101 Posts
Every ‘bank’ (we’ve been calling it a group) of 3 cells in parallel is by definition at the same voltage. But although those 98 groups in series all see exactly the same current (again by definition) very slight variations of internal resistance mean that some groups could receive or provide a slightly different amount of energy.
In some or perhaps most BMS implementations the highest voltage cell group would determine the termination of charging, for safety but at the downside of loosing some total capacity. So, the system will try to draw down higher groups by bypassing current locally via a resistor.
It’s thought that this happens at 98% while charging, so if you rarely get there it rarely balances. Also if you’re DC charging the bleed current will be too small to matter.
It’s possible that the SW was not clever enough to cover all these situations but that’s just speculation at this point.
 

·
Registered
Joined
·
187 Posts
What I observe: SoulEvSpy reports uniform cell voltage for some time. Say 5-10 minutes. Meanwhile the vehicle reports a steady charge, (say 7kw) and an increasing charge percentage. for example at 64% .all cells report 3.86V. As soon as 68% was reached, started seeing some cells light up showing 3.88V. Then more and more cell light up as 3.88V, until all of them do.
So now all cells remain at 3.88V until the whole process starts again. It seems to progress on 0.02V increments. This may be an artifact of SoulEvSpy, maybe it can only display 0.02 increments.
The battery bank temperature goes up steadily as charging progresses. Started at 15,14,14,16 and now at 18,17,18,19
 

·
Registered
9/18 EU base model
Joined
·
101 Posts
It would not surprise me if the analog to digital conversion only has a resolution of 0.02 V.
 

·
Registered
Joined
·
187 Posts
So: I am done charging to 80%. . At this point, no more charge flowing tino the battery. Most of the cells show 3.94V, except from time to time , a seemingly random set of cells light up as 3.96V. It is a shifting pattern of 1 to 10 cells that light up for a few seconds, then a different set of cells light up showing 3.96V. So some cells seem to be oscillating between 3.94 and 3.96. I do not get why a random set of cells cells go up and down in voltage. The cells that do this are no necessarily the same ones. Finally the cells settle down. Just one cell is 3.96V, the others are all 3.94V

What's with that?
 

·
Registered
Joined
·
176 Posts
So: I am done charging to 80%. . At this point, no more charge flowing tino the battery. Most of the cells show 3.94V, except from time to time , a seemingly random set of cells light up as 3.96V. It is a shifting pattern of 1 to 10 cells that light up for a few seconds, then a different set of cells light up showing 3.96V. So some cells seem to be oscillating between 3.94 and 3.96. I do not get why a random set of cells cells go up and down in voltage. The cells that do this are no necessarily the same ones. Finally the cells settle down. Just one cell is 3.96V, the others are all 3.94V

What's with that?
What you are seeing is very normal. Good news, is that your battery looks good and does not show any bad cell groups. Re the variances, here is what the current SoulEVspy developer, Henrik Scheel, told me on email:

"The precision of the voltage measurement is. 02 volt, so the difference in cell voltages can be anywhere between 0.0000001 Volts and 0.0399999999 Volts. During charging, all cells do not read the same value when changing. A difference of. 02 is neglible, and likely to not be the same cells, after your next drive or charge."
 

·
Registered
Joined
·
187 Posts
Cool, thanks.
I am guessing he rounds the actual value, possibly up or down, so if a cell happens to be just a tiny bit shy of the rounding limit, a minuscule voltage change may show up as a 0.02V oscillation. So the rounding may either amplify or minimize the displayed value vs. the actual voltage. Do we know if it is the actual measurement that has this resolution limit, or it is SoulEvSpy-s interpretation?
I tend to think it is the latter, because It tracks the lowes N and the highest K ( where N and K are counts) batteries, and these change even when all cell voltages are shown the same.

Am I right assuming that the Software update and "inspection" is strictly a software based self-diagnostic,
and no visual inspection of the battery will be performed?
Sounds like the new software comes with some new diagnostic routines.

I think there are 2 possible culprits have been proposed so far: 1) Cell overcharge damage 2) Battery cell Separator manufacturing defect.
Both of these would be diagnosed by software, right?
 

·
Registered
Joined
·
176 Posts
Yes, .02V is a pretty small variation, and he clearly says that is within the margin of error. It might help if there were 3 decimal places. Maybe Torque Pro has that?

As for exactly how Hyundai determines whether the battery is defective or not, we can only speculate. But it seems pretty reasonable to assume that if the software finds large cell group variances, they would deem the battery defective. As I have posted previously, I do know of one instance where an owner found a major variance with SoulEVspy. And when he took the car to the dealer, they confirmed it and ordered a battery replacement. This was even before the latest BMS software update. Here again is that example showing one bad cell group:
4804
 

·
Registered
9/18 EU base model
Joined
·
101 Posts
... and he clearly says that is within the margin of error. It might help if there were 3 decimal places. Maybe Torque Pro has that?
Well, rather than a "margin of error" he stated that the "precision of the voltage measurement" is 0.02 V. We don't know the overall accuracy of the analog voltage measurement. Adding more than 2 decimal places in the TP display doesn't help, which strongly implies that's a limitation of what's available over the CAN bus.
I am guessing he rounds the actual value, possibly up or down, so if a cell happens to be just a tiny bit shy of the rounding limit, a minuscule voltage change may show up as a 0.02V oscillation. So the rounding may either amplify or minimize the displayed value vs. the actual voltage. Do we know if it is the actual measurement that has this resolution limit, or it is SoulEvSpy-s interpretation?...
...
1) Cell overcharge damage 2) Battery cell Separator manufacturing defect.
Both of these would be diagnosed by software, right?
I wondered this myself but realised that if you only have data in steps of 0.02 you can't assume anything that would allow rounding. However, you could just add 0.01 to all values to evenly distribute the error.
As best as I can understand (2) would be detected by a persistent above-average loss of charge in any cell group showing up as an imbalance. Historic data would be useful if that's collected at a group level but frankly I doubt that's the case.
As noted above in the image, the variation that would cause concern is that over 0.10 V.
 

·
Registered
Joined
·
55 Posts
My recall was completed today, was told the battery is in good condition and I can keep it indoors and charge to 100%. It will be on charge tonight but might not get over the threshold to pause.I will plug it in each night during the week to see it hit the pause thresholds. I will be able to see the usage on my inverter to see how long it pauses for.
Various settings were reset, android auto wasn't installed. That also happened last time I had BMS and firmware updates earlier this year.
Firmware screenshot for what its worth
4807
 

·
Registered
Joined
·
187 Posts
Congrats Mr P!

Were you able to get any more/new info out of the dealer or the techs?
Mostly Interested if the DIY cell monitoring and charge limit workaround some of us are practicing makes
sense>
 

·
Registered
Joined
·
55 Posts
Congrats Mr P!

Were you able to get any more/new info out of the dealer or the techs?
Mostly Interested if the DIY cell monitoring and charge limit workaround some of us are practicing makes
sense>
Not really, was an outward facing service guy so not the EV mechanic. We did discuss how bad it would be to change the battery, sounded like a lot of logistics they have done before. Took it on a 200km commute this morning, went fine. Will charge it up again the next two nights and hit the 75 / 80 thresholds to make sue it does pause as expected.

One thing I did here this morning was the chook wheel / wheel of fortune sound in my transmission again. It already had the 80 deg rotation so now I'll be monitoring that closely as well.That 80 deg rotation solution sounded suspect to me.
Hopefully Gen2 of the Kona has all this stuff sorted. I'd also be super happy if they did a SantaFe BEV with 500km range.
 

·
Registered
Joined
·
187 Posts
Mr P.

Sorry I must have missed an earlier post of yours in this thread or another. Your EV tranny is making a strange sound? I thought the EV had no tranny.
What is the 80 deg rotation? Thanks for explaining...
 

·
Registered
Joined
·
187 Posts
So to summarize:

If this is all they need to do (a software upgrade, and a small number of battery swaps) and indeed it takes care of the problem,
it may end up a minor bump on the road for them. I still wish they were more up front and open kimono about this whole thing.



.
 

·
Registered
Joined
·
55 Posts
Mr P.

Sorry I must have missed an earlier post of yours in this thread or another. Your EV tranny is making a strange sound? I thought the EV had no tranny.
What is the 80 deg rotation? Thanks for explaining...
Yeah check it out here. Hyundai Kona EV powertrain clicking sound
Basically sounds like the wheel of fortune clacking, fair bit of explanation in that service report of what they adjusted.
 

·
Registered
9/18 EU base model
Joined
·
101 Posts
I believe they meant (and carried out) 180 degrees rather than 80, just a typo. Not that it matters if they failed to click their heels together afterwards.
 
101 - 120 of 160 Posts
Top