[thejudge]

So far in over 2 years (10.5 km so really low mileage) ive had 1 dead 12 battery which was after sitting at 19% for 4 days , which 20% is the threshold i believe. Kind of sux coz now the only way to calibrate the BMS is to actually drive the I5 to below 5%.

Uniq + relax AWD bought in Germany.

Ps I have my I5 connected to home assistant through the kia_uvo app and i have an automation which will send a message to our main alexa and sends a message to all connected phones if the 12v battery drops below 50% , which it never has , lowest it seems to drop to is 75%.

 

[CoSolar]

There are many arguments around the theme of it not being the chemistry of the 12v battery that's the issue but the incorrect selection of the type of lead/acid unit. As has been mentioned, the last thing that an EV needs is a huge cranking ability. But such a battery is usually installed. They are also very intolerant to being exposed to a low state of charge for long periods and will fail rapidly.

However, the type of lead/acid battery that is used for leisure purposes such as in boats, caravans and motorhomes will happily endure a deep cycle to low charge rates and work best when charged and discharged across the full 100% available power. There is a good case to be made to replace with such a battery except that they tend to be physically too large to fit.

My experience with 12 volt AGM lead acid battery systems used in camp trailers is that discharging the batteries below 50% SOC should be avoided. On the other hand they can be kept fully charged without a problem. This is based upon recommendations from people with far more experience than I and it is advice I've followed. Lithium batteries for recreational vehicle use have become more popular because they handle low SOC levels well thereby allowing access to the battery's full capacity. They do require different charging regime settings on a charge controller when charging from solar panels.

 

[HKtech]

So far in over 2 years (10.5 km so really low mileage) ive had 1 dead 12 battery which was after sitting at 19% for 4 days , which 20% is the threshold i believe. Kind of sux coz now the only way to calibrate the BMS is to actually drive the I5 to below 5%.

Uniq + relax AWD bought in Germany.

Ps I have my I5 connected to home assistant through the kia_uvo app and i have an automation which will send a message to our main alexa and sends a message to all connected phones if the 12v battery drops below 50% , which it never has , lowest it seems to drop to is 75%.

When EV is powered on it will keep maintaining 12V battery all the way down to limp mode.
HV DC pack SOC is considered only for stationary purposes when 12v battery requires maintenance.
Latest TSB have changed E-GMP to 10% SOC on HV DC.

 

[HKtech]

I'm speaking as a mechanical hardware, & software, engineer of many years experience (now retired), designing machine-control systems since the days of the earliest microprocessors. I created several patents, both in the mechanical engineering arena, as well as IT.


This is unacceptably poor design. Everyone is familiar with aftermarket add-ons to car systems, e.g. fancy sound-systems with huge sub-woofers, lighting bars, dashcams etc. Any/all of these may add noticeable extra loads to the 12V battery at any time, and the car should cope with them. I'll explain why it doesn't below, and what the correct design should have been.


You're describing a polled system. This is the simplest way to code up a control system that needs to respond to external events (in this case "low battery"). As any competent software engineer who has worked on microprocessor real-time control systems knows, this method is prone to failiure when the rate of events exceeds the rate of polling. The correct design is to use an interrupt driven system.

What is meant by "exceeding calibration values" ? The way the Ioniq 5 won't charge the 12V battery in this polled way at all, if the main traction battery is below 20% SOC or whatever that threshold is? Fair enough, but a bit cautious perhaps.
Edit 2 Jan 2024: Post elsewhere here says ICCU update now continues to topup 12V battery down to 10% SOC - a step in the right direction IMHO.

My Ioniq '38 has this inadequate polling system, and originally it polled once per day. Here's a trace of how it attempted (not!) to deal with an unexpected discharge scenario. The cause of this remains unknown, wasn't damp in the door switches, wasn't bluelink.
View attachment 53488

As you can see, the voltage dropped to well below the stage where a topup was needed.

My battery was damaged by this deep discharge, and another, similar one previously. It was the previous failure that made me design & build my own built-in battery-booster based on 22 Ah motorcycle battery + electronics to ensure my backup cannot be drained by these "vampire" attacks, and install a BM2 Battery Monitor that captured the 2nd such vampire attack shown here. The quick recovery from the attack shown is thanks to this backup working. I flip 2 switches and the 12V battery is changed over.

Finally, late 2022 or early 2023, long after Hyundai has changed to 6 topups/day on the Kona, H decided to apply the same update to the Ioniq! Similar cars & electronics. (Why the Ioniq didn't get this update MUCH earlier baffles me).

So here's a shot of how the 6 topups/day algorithm works with a damaged-but-reconditioned Delkor OEM battery in it.
View attachment 53492

A few things have changed with the recent BMS update to 6/day topups. My damaged battery is managing to sit around 12.7V thanks to the regular topups, but immediately after a 5 hour AC charge, where the car holds the 12V battery at 13.0V steadily, the 12V battery drops to 12.5V. So this use of 13.0V during charging is insufficient to maintain a level state of charge in the 12V battery. Fortunately H have added a 12V battery test at the end of an AC charge session, and this test decides that 12.5V is too low, and sets off a series of 5 quick topups each about 20 minutes, and a re-test after.
Having done 4 such tests, the 12V is still only at 12.69V afterwards, but perhaps this is close enough to the pass/fail threshold which seems to be at 12.70V judging by the previous regular topups. At some stage in this re-test process it appears the timer for Regular topups gets reset to zero, as the regular topups restart at about 21:15 on 26 Dec.


Incorrect.
Onboard systems aren't going to fix a bad battery, but they should prevent it going bad in the first place. As they clearly have failed to do in my example, and many others in this forum dating back several years.

The correct system design is one where any instance of the 12V battery dropping below a predetermined safety-level (let's say 12.70V as that seems to the the threshold for the regular topups) triggers an interrupt that wakes up a subsystem dedicated to monitoring the 12V battery state. The hardware required for this is minimal, a voltage comparator plus a voltage reference are all that is needed. It's quite possible this setup already exists in the battery-status-monitoring device already connected to the +ve terminals of the battery which reports to the car's computer.

Once such an interrupt is triggered, the code should first determine if it's a momentary spike due to noise or some other transient that can be safely ignored. Things like the spike above at 13:00 in Dec 25 where the wing mirrors operated momentarily. This kind of transient can be filtered out by waiting a few seconds then retesting the voltage to see of the transient passed. This is typically how things like mechanical switch bounces are filtered out, you do a few consecutive retests to see if the condition stabilises, and when it does, only then do you act on it.
Given an interrupt-driven topup mechanism, it would be perfectly possible to start providing topups at a rate of say 3 per hour, or more if neccessary. But you now have the precise timing information of when these topups are needed, and this gives you the ability to take sensible preventative action and log the actual rate for historical analysis.

A pattern of the number of topups per day increasing gradually from say 4/day, to 5/day, to 6/day over a period of weeks/months indicates a battery that's wearing out and needs replacement/reconditioning soon.

A pattern of suddenly needing 2 topups/hour out of the blue, indicates an anomaly of some kind. At this stage you keep applying regular topups, as well as provide other alerts to the owner. Such as a bluelink alert suggesting a door was left open, or something else draining the 12V. If bluelink not in use, a momentary blast on the horn every 5 minutes might be in order. Or flash the headlights. Or something.

A polled system such as Hyundai have implemented cannot cope with the scenario where the power consumed in the 12V devices powered up exceeds the power replaced by the regular topup approach. And the worst scenario happens when something keeps a relatively high-power device active, such as the ?car's computer? consuming maybe 240W, or 20A at 12V. This kind of power draw is enough to drain my 40 Ah battery in a couple of hours, which is less than the interval between regular topups. This kind of sudden high-current drain illustrates perfectly why a polled approach is unable to cope with real-world "extra-ordinary" events. And this is precisely why an interrupt-driven approach is far better.


Agreed. Mine's been down to 8.4V before the drain posted above. Fortunately it's still limping on, thanks to 6 topups/day, and regular use of a de-sulphating + reconditioning cycle on my Ctek smart charger. But I can see my battery is struggling!
If H had implemented the correct algorithm, both of my vampire attacks would have been spotted, and I would have been warned, before the voltage dropped to anywhere near 12.0V, let alone the 8.4V I discovered many hours later.
Neither of these damaging, sulphating discharges should have happened.


People have every right to expect a decent amount of 12V battery capacity. I see that Hyundai decided to fit a remarkably small battery to my Ioniq 38, something more suited to a Toyota Aygo sized car. But there's no high-current starter motor, so, hey-ho, let's fit a tiny one now!

When there are more and more "clever" things happening, like bluelink, OTA updates someday, remote-fob-sensing circuitry etc placing demands on the 12V system overnight when the car's asleep, it seems rather short-sighted to go downsizing the 12V battery so much! Unless you have a proper charging support system in place, which Hyundai clearly have not.

@HKtech , I'd like to thank you for your participation and help here. I really wish there were more around like you, as it's almost impossible to make any kind of technical connection with large corporations like Hyundai, which tend to hide the technical people behind armies of press departments & management etc. I worked for one such large corporation.

If there's anything you can do to contact the designers of the 12V subsystem, please do, and please pass on my opinions and charts. Thank you again.

I will try to detangle this out with you, as soon i get enough time to gather information on latest ICCU update logic.
Here are abilities that ICCU can read about 12V battery
Battery voltage
Battery temperature at the terminal post
Battery internall resistance
Battery SOC read by shunt
SOH degradation ( this is another subject we can discuss later)
At what voltage saver+ start is dependent on temperature.
Here are abilities of ICCU DC-DC converter strategies.
There are 8 modes of operation and they include:
Battery Sulfation Mode
Normal Mode
Fuel Economy Mode
Headlamp Mode
Voltage Reduction Mode
Battery Maintenance Mode
Factory floor mode
Transport mode
There is in the works to support another different types of 12V battery chemistries ( I'm not allowed to speak publicly if/ and when they will available)

 

[HandyAndy]

Li-ion is an obvious possibility for a new 12V battery type, Hyundai already use this in the Hybrid I believe, so maybe that's acting as a test-bed platform to see get a few years real-world experience.

Advanced Lead-Carbon looks a promising development of Lead-Acid, acc to BU-202: New Lead Acid Systems .

Claimed advantages are: cheaper than Li-ion, better operation in sub-zero temperature than Li-ion, and can operate between 30 & 70% SOC without suffering sulphation, a problem that's hitting us when we get these "vampire" drains down to as low as 8.4V that I & others have seen, or which can happen if the owner puts a large load on it w/o using Utility mode or other hack to protect the 12V (insert V2L adapter, ...).

Disadvantage is voltage drops off like a super-capacitor as the battery drains. I guess you really don't want the 14V at full-charge (my guess) dropping down to 10V if it's only at 71% SOC (I'assuming linear drop), but if the car had an interrupt-driven ICCU then it would be simple to keep the minimum level at 12.0V when car is idle, and boost it when car's out on the road.

 

[rbdavis808]

Latest TSB have changed E-GMP to 10% SOC on HV DC.

@HKtech, can you clarify exactly which TSB (or TSBs) dropped this number from 20% to 10%?

I watched an IoniqGuy video 2 or 3 weeks ago that mentioned this but it seemed a bit unclear as to whether this change to 10% only happened in a very recent TSB, or was perhaps also bundled into a somewhat older TSB (like the regen-triggers-brake-lights TSB, which I already have).

 

[HKtech]

@HKtech, can you clarify exactly which TSB (or TSBs) dropped this number from 20% to 10%?

I watched an IoniqGuy video 2 or 3 weeks ago that mentioned this but it seemed a bit unclear as to whether this change to 10% only happened in a very recent TSB, or was perhaps also bundled into a somewhat older TSB (like the regen-triggers-brake-lights TSB, which I already have).

Here is screen shot.

 

[rbdavis808]

Here is screen shot...

Thanks, @HKtech !

Here is a screenshot of part of the preceding TSB 23-01-067H which was applied to my car in August 2023:

As you can see from the items listed in the Description section, it is virtually the same as the description shown in your screenshot of the October 067H-1, except that the August TSB says nothing about the 12v top-up threshold being lowered to 10%.

To me this looks like my car, and every other car that got the brake light logic update in August, needs to go back to the dealer to get the 067H-1 update applied if they have not already done so and they want to be sure their 12v will top up all the way down to 10% SOC on the HV battery. Can you please confirm or deny?

 

[kapinski22]

I'm with you.
I don't really understand the the details like many but it is clearly a design problem and, again not understanding the details, it seems unacceptablethat this is not fixed going into the 4th year.
Illustrating my ignorance, it blows my mind that you're sitting on this high tech HUGE battery with an incredibly high tech electron system and the system fails because an old fashion 12v battery can't hold or be recharged as well, as a 1940's car.

AGREED!!! When are they going to get this all straightened out? I spent $60,000 for my 2022 ionic five deluxe model and I don’t have a warm and cosy feeling I can trust it all the time.

 

[kapinski22]

Thank you.

You have just described what currently happens. As many of us already know.

But what you haven't done is explain why the OEMs appear to be incapable of designing a system that takes ongoing note of unexpected power drains - that may be outside their initial calibration estimates - and react accordingly to cope with that unexpected condition.

Why do they just abrogate all responsibility at that stage and allow the DC/DC converter to go to 'preserve' conditions, cease power transfer, and allow the 12v battery to be destroyed by constant low voltages?

That is the question that remains unanswered. Because to us laymen that appears to be a really simple amendment to the initial program that calls for DC/DC action. But for some reason, they do not do that and to make matters worse deliberately withdraw sessions of DC/DC transfer so that failure is inevitable.

“…deliberately withdraw sessions of DC/DC transfer so that failure is inevitable….” Excellent point!

 

[kapinski22]

2022 I5 SE AWD. Bought new May 2022. 22K miles. No issues with 12V battery until last night. To borrow a line from "The Princess Bride" it was not dead, just mostly dead. Car seemed to start up fine, but then dash displayed "12V battery voltage low. Stop safely". But car would not shift into gear. I had noticed once or twice in the past few weeks I was prompted to touch the fob to the starter button to start the car (but assumed I had to fob in a bag that was blocking the signal). Anyone experience anything similar ?

Unreal coincidence… This exact event just happened to me. I’ve got a 2022 I5. I’ve had a few bad battery experiences going back about 10 months ago, but the ICC whatever update at the dealership in August seem to fix everything. I’ve had no problems with the car until literally just today … I come home and I read this post of yours. Stopped at a friends house for an hour or two went back out to my car and there was that message.”12V battery voltage low. Stop safely"… never seen it before. Open the hood to take out my lithium battery pack which has worked great in the past - the 3 or 4 times I’ve had to use it but no go this time. I to then got the weird message on the screen to push my fob against the start button… What the heck is that all about?! I actually gave it a try but no go. Hyundai emergency sent AAA over and his battery pack did the trick … I drove the 5 miles home but I’m now noticing on my battery monitor that I’ve got low power… Only 54% and 12.5 V . figuring the battery is permanently damaged from all these experiences so, time to get a new battery / demand a new battery from Hyundai?? Troubling to see the post re that $500 lithium battery from OHMMU… thought that might be the solution

 

[Hitstirrer]

................but I’m now noticing on my battery monitor that I’ve got low power… Only 54% and 12.5 V . figuring the battery is permanently damaged from all these experiences so, time to get a new battery / demand a new battery from Hyundai??

You are probably right. Standard lead/acid batteries are often killed after being exposed to a very low charge for some time. Sometimes they can be resurrected by using a modern smart charger that has a desulphide programme. Such chargers first analyse the condition and then go into a repair program before moving on to a lengthy pulse charge sequence. That can take many hours but there are reports that a troublesome 12v battery can be brought back to life.

But even if that doesn't work and a new battery is fitted it is still a good plan to attach a smart charger once a month during the tyre kicking and screenwash filling routine. And do that whether it seems to need it or not. Such regular preventative maintenance has worked for some years for me and can be recommended. I know that it should not be needed but life's too short to stress over such things.

 

[kapinski22]

You are probably right. Standard lead/acid batteries are often killed after being exposed to a very low charge for some time. Sometimes they can be resurrected by using a modern smart charger that has a desulphide programme. Such chargers first analyse the condition and then go into a repair program before moving on to a lengthy pulse charge sequence. That can take many hours but there are reports that a troublesome 12v battery can be brought back to life.

But even if that doesn't work and a new battery is fitted it is still a good plan to attach a smart charger once a month during the tyre kicking and screenwash filling routine. And do that whether it seems to need it or not. Such regular preventative maintenance has worked for some years for me and can be recommended. I know that it should not be needed but life's too short to stress over such things.

many thanks Hitstirrer! Off to do battle with the Hyundai dealer… I’m demanding they install a new battery, gratis. It will probably be as crappy as the original one, but I’ll treat it with the loving care you suggest. ( Lucky guy, you have a TR3A, 1959… I had one of those. Sweet.)

 

[msilvers50]

HKtech do you have advice on replacement 12v batteries we IONIQ5 owners should consider? Any lithium batteries that are compatible to ICCU control? Any AGM batteries vs. lead acid ones? There are frequent posts about lithium and occasional ones about AGM type.

 

[IoniqShark]

HKtech do you have advice on replacement 12v batteries we IONIQ5 owners should consider? Any lithium batteries that are compatible to ICCU control? Any AGM batteries vs. lead acid ones? There are frequent posts about lithium and occasional ones about AGM type.

Search a bit in this thread. He responds to the use of lithium alternatives.

 

[HKtech]

HKtech do you have advice on replacement 12v batteries we IONIQ5 owners should consider? Any lithium batteries that are compatible to ICCU control? Any AGM batteries vs. lead acid ones? There are frequent posts about lithium and occasional ones about AGM type.

AGM would be good substitute, but to keep warranty on the 12 battery you will have to stay with factory 12V battery replacement.
Stay away from LFP chemistry or lithium ion chemistry or any other aftermarket solutions that has dedicated BMS to manage this chemistry.

 

[Hitstirrer]

AGM would be good substitute, but to keep warranty on the 12 battery you will have to stay with factory 12V battery replacement.
Stay away from LFP chemistry or lithium ion chemistry or any other aftermarket solutions that has dedicated BMS to manage this chemistry.

There has been much debate over the years about fitting a deep-cycle lead/acid 'leisure' battery instead of a standard car battery. The logic being that EVs don't need the huge cranking ability of standard batteries, together with the idea that leisure batteries can cope better with low voltages and can survive the odd incident without needing replacement.

Apart from the obvious issues over such a battery actually fitting the tray, do you have a view on this type of replacement?

 

[HKtech]

There has been much debate over the years about fitting a deep-cycle lead/acid 'leisure' battery instead of a standard car battery. The logic being that EVs don't need the huge cranking ability of standard batteries, together with the idea that leisure batteries can cope better with low voltages and can survive the odd incident without needing replacement.

Apart from the obvious issues over such a battery actually fitting the tray, do you have a view on this type of replacement?

One you mentioned can be used.
But it will need to be maintenance free.
Deep cycle lead acid has 50% DoD
AGM battery has 80% DoD
Both if drop too low will affect SOH degradation.
AGM are usually maintenance free
Deep cycle lead acid usually requires maintenance ( this is a reason why automotive industry has moved away, and for ICE engine it usually needs hig cranking amps, what this chemistry dosent like)
Also AGM battery is quite close to the stock battery regarding internall resistance ( what ICCU is checking also).
Another substitute would be
EFB and EFB+ chemistry ( not many on the market and usually carry higher prices for no reason, even though that AGM battery will have longer-term life expectancy in my opinion)

 

[HandyAndy]

Afaik the Ioniq 5 uses Calcium-enhanced Lead-acid, as they do in the '28 & '38, so different to AGM.
I tried to find a Leisure battery to suit my 38 (40 Ah), but they were all far too massive.

So my preference woild be Yuasa YBX series as these are also Calcium-enhanced similar to the Delkor in my '38.
Lots of people fitting these to the '38s successfully.

I don't know what the Ioniq 5 battery physical size is, or the post diameters, but I'd be tempted to check out the Yuasa YBX5075 which is 60 Ah, and looks to have similar terminal layout. Yuasa are available in USA.

 

[Goluchowski]

One you mentioned can be used.
But it will need to be maintenance free.
Deep cycle lead acid has 50% DoD
AGM battery has 80% DoD
Both if drop too low will affect SOH degradation.
AGM are usually maintenance free
Deep cycle lead acid usually requires maintenance ( this is a reason why automotive industry has moved away, and for ICE engine it usually needs hig cranking amps, what this chemistry dosent like)
Also AGM battery is quite close to the stock battery regarding internall resistance ( what ICCU is checking also).
Another substitute would be
EFB and EFB+ chemistry ( not many on the market and usually carry higher prices for no reason, even though that AGM battery will have longer-term life expectancy in my opinion)

Does the Hyundai service department have access to historic 12 volt battery voltage to see the discharge rate like we get when a BM2 is hooked up?