RECALL) KONA EV - BMS UPDATE and 20% Reduce cell capacity

  • Thread starter Thread starter kennykim
  • Start date Start date
  • Replies Replies 177
  • Views Views 40K
KiwiME said:
I would assume if a cell voltage is well below the average the "check" light is triggered.

I can imagine that an example that has been DC charged frequently misses out on cell balancing and could suffer damage when eventually charged to 100%. But that doesn't explain the several examples that apparently failed when nearly new.
Click to expand...
Why would Fast charging affect cell balance?? I DC Fast charge about once per week and my cells are all in balance... within 0.02 volts.. Cell group 39 is the only one that is always the last one to get to the next 0.02 higher voltage but it gets to it before the first cells get to the next 0.02 higher Voltage.. I consistently monitor my cell balance almost every time that I charge the car..
 
BlueKonaEV said:
Why would Fast charging affect cell balance?? I DC Fast charge about once per week and my cells are all in balance... within 0.02 volts.. Cell group 39 is the only one that is always the last one to get to the next 0.02 higher voltage but it gets to it before the first cells get to the next 0.02 higher Voltage.. I consistently monitor my cell balance almost every time that I charge the car..
Click to expand...

The BMS had the ability to add load to individual cells that are reading high. It can only compensate so much though. It is the reason most charge algorithms taper off at the end.


Sent from my iPhone using Tapatalk
 
KiwiME said:
I think that situation will remain and replacing the entire pack is still the only practical option. Checking balance occasionally using TP or SoulEVspy could give an early warning. I've had one that is always slightly down (~0.02) so I'm checking it frequently.

I would assume if a cell voltage is well below the average the "check" light is triggered.

I can imagine that an example that has been DC charged frequently misses out on cell balancing and could suffer damage when eventually charged to 100%. But that doesn't explain the several examples that apparently failed when nearly new.
I don't see how it's possible to measure cell voltages while charging. Charging must stop entirely and cells allowed to settle.
Click to expand...

$30 hobby chargers do can and do, do it.


Sent from my iPhone using Tapatalk
 
KiwiME said:
I don't know. Why would they get out of balance at all? Production variations obviously.
Click to expand...

Normal cell variation. Damage is cumulative. Over charge a cell a few times by .x volt is no big deal. Do it a few hundred times and things change.


Sent from my iPhone using Tapatalk
 
Is it clear/provable that the Kona BMS/CMU setup does any balancing at all, passive or otherwise?
I'm dubious that watching a jump in declared SOC near the top of a charge really says anything.

My usual high/low cellblocks are 1 and 84, with occasional departures. I figured that was an
artifact of the order in which the taps are scanned, vs. the rate at which my OBD2 queries come
in. To make the scanning process faster I sometimes put the high/low together on a running
graph as the *only* two things I'm asking for, and they basically stay right together under various
forms of abuse, never more than the .02V measurement granularity apart.

At some point I expect i'll go to charge to 100% and watch the final 4 or 5 % closely to see if any
cells are flying a little high at the end. I've never seen any individual cellblock voltage over 4.1,
and generally a tad lower than that even after finishing a 100% charge at level 2 rates. I wonder
if I'd see anything different by changing to a lower current ... although full 7.x kW is only 20A,
a 0.1C-ish rate.

_H*
 
BlueKonaEV said:
Why would Fast charging affect cell balance?? I DC Fast charge about once per week and my cells are all in balance... within 0.02 volts.. Cell group 39 is the only one that is always the last one to get to the next 0.02 higher voltage but it gets to it before the first cells get to the next 0.02 higher Voltage.. I consistently monitor my cell balance almost every time that I charge the car..
Click to expand...
The faster you dump energy into cells, the more the internal resistance becomes a factor. This varies (slightly) across cells and isn't a constant -- it changes with temperature, age, and also SoC. So yes, I would expect differences at the pack fills up, especially using faster charging.

This actually reminds me of driving a Model S loaner (long story) several years ago. The one pedal driving was completely different when the pack was near 100% -- it wouldn't regen very much and in fact one pedal driving was not really possible (this behavior may have changed over the years, but this was definitely the case ~2016). As the pack discharged a little bit, you had to re-train your "lift-off" response as the regen became much more aggressive.
 
hobbit said:
Is it clear/provable that the Kona BMS/CMU setup does any balancing at all, passive or otherwise?
I'm dubious that watching a jump in declared SOC near the top of a charge really says anything.

My usual high/low cellblocks are 1 and 84, with occasional departures. I figured that was an
artifact of the order in which the taps are scanned, vs. the rate at which my OBD2 queries come
in. To make the scanning process faster I sometimes put the high/low together on a running
graph as the *only* two things I'm asking for, and they basically stay right together under various
forms of abuse, never more than the .02V measurement granularity apart.

At some point I expect i'll go to charge to 100% and watch the final 4 or 5 % closely to see if any
cells are flying a little high at the end. I've never seen any individual cellblock voltage over 4.1,
and generally a tad lower than that even after finishing a 100% charge at level 2 rates. I wonder
if I'd see anything different by changing to a lower current ... although full 7.x kW is only 20A,
a 0.1C-ish rate.

_H*
Click to expand...
Set your voltages on Torque Pro to 2 decimals to get more accurate voltages.. My voltages at 100% show 4.16 volts each BUT my total voltage shows 409 Volts which translates to 4.173 volts per cell group. The 2nd digit updates in 0.02 intervals, so it will go from 4.16 to 4.18 once it reaches 4.18 volts.. it will still show 4.16 if the voltage is 4.17
 
BlueKonaEV said:
Set your voltages on Torque Pro to 2 decimals to get more accurate voltages.. My voltages at 100% show 4.16 volts each BUT my total voltage shows 409 Volts which translates to 4.173 volts per cell group. The 2nd digit updates in 0.02 intervals, so it will go from 4.16 to 4.18 once it reaches 4.18 volts.. it will still show 4.16 if the voltage is 4.17
Click to expand...

It's a PITA to change 98 pid's one at a time from default (1 decimal place) to 2 in Torque Pro.

John.

Edit changed from:-

3 decimal places surly? but it's a PITA to change 98 pid's one at a time from default (2 decimal places) to 3.
 
Last edited:
Anaglypta said:
It's a PITA to change 98 pid's one at a time from default (1 decimal place) to 2 in Torque Pro.

John.

Edit changed from:-

3 decimal places surly? but it's a PITA to change 98 pid's one at a time from default (2 decimal places) to 3.
Click to expand...
I did it and it was a pain in the butt.. took me a while..
 
I *do* display 2 decimal places, that's why I show high/low that lets me see the 0.02 [and never more so far]
deviation. Rather than dork with 98 PIDs, I let the BMS do the work and sniff out the extremes, and then just
ask for 4 PIDs for a pretty good picture of the pack's health. [FD: I'm running Obdlink's companion app to my
bluetooth widget, not TP, but it does custom PIDs and the effect is the same.]

_H*
 
hobbit said:
I *do* display 2 decimal places, that's why I show high/low that lets me see the 0.02 [and never more so far]
deviation. Rather than dork with 98 PIDs, I let the BMS do the work and sniff out the extremes, and then just
ask for 4 PIDs for a pretty good picture of the pack's health. [FD: I'm running Obdlink's companion app to my
bluetooth widget, not TP, but it does custom PIDs and the effect is the same.]

_H*
Click to expand...
ah, ok... I got confused as you said ". I've never seen any individual cellblock voltage over 4.1"
I would have expected 4.10 in that case.. I assumed only 1 decimal place for that reason..
 
I'm just curious to know

How is the 2021 any different than the 19-20 in regards to this recall ?

If the 2021 has the issue resolved, wouldn't it imply they were aware of an issue and actually do have a fix ?

I have a 2021 and it doesn't make me feel any better about all this, I honestly think buyers no matter 2019 or 2021 wont touch our cars if we try to sell them
 
persianrider said:
I honestly think buyers no matter 2019 or 2021 wont touch our cars if we try to sell them
Click to expand...

Eh, buyers have short memories. Once Hyundai identifies a fix and gets it rolled out, that will be that. Worst case, you might get the best price possible from CarMax. I’m looking at selling my 2019 Kona next Spring/Summer because I have an order on a Mach-E. That decision wasn’t based on the battery issues/recall though. The Mach-E is a very attractive car!
 
TheLight75 said:
Eh, buyers have short memories. Once Hyundai identifies a fix and gets it rolled out, that will be that. Worst case, you might get the best price possible from CarMax. I’m looking at selling my 2019 Kona next Spring/Summer because I have an order on a Mach-E. That decision wasn’t based on the battery issues/recall though. The Mach-E is a very attractive car!
Click to expand...
Well if you want to trade in on a Tesla be prepared for a shock. Just for fun I asked for an online trade quote on my 10 month old 2020 Kona EV, they were willing to take the car of my hands for a measly 36% haircut off of MSRP, lol. That was way below Canadian wholesale blackbook or wholesale auction comparables.
 
persianrider said:
I'm just curious to know

How is the 2021 any different than the 19-20 in regards to this recall ?

If the 2021 has the issue resolved, wouldn't it imply they were aware of an issue and actually do have a fix ?

I have a 2021 and it doesn't make me feel any better about all this, I honestly think buyers no matter 2019 or 2021 wont touch our cars if we try to sell them
Click to expand...
In part I suspect the BMS ROM ID version, check the manufacture date (will probably be different than the vehicle manufacture date on the door jam) and ID # on the label at the back of the battery pack drivers side behind the suspension,
I will be starting a new thread on this after some more of this information is disclosed from 2021 Kona EV owners and some more LG production detective work.
 
apu said:
Well if you want to trade in on a Tesla be prepared for a shock. Just for fun I asked for an online trade quote on my 10 month old 2020 Kona EV, they were willing to take the car of my hands for a measly 36% haircut off of MSRP, lol. That was way below Canadian wholesale blackbook or wholesale auction comparables.
Click to expand...
From what I have read online, this is not unusual.

Tesla is not in the used car business (per say) and would rather low ball you to get you to dispose of your current vehicle via some other fashion.

As always, YMMV.
 
You must log in or register to reply here.
Back
Top