[Blucid]

I took a quick look at BM2 today on the lab bench. Some interesting observations. I got the QUICKLYNKS Bluetooth4.0 Battery Monitor II (Amazon).

First note is the spec of 1.5 mA load current, no big deal, but my average load current is 2.3 mA, still no issue for harmlessly living on the battery. Just an aside, there are fast spikes to 8-10 mA, however the average is much lower at about 2.3 mA.

Next, using a power supply (a settable voltage reference), I re-connected the BM2 to reset it for each different voltage on the chart below for the BM2 standard battery. Not a huge issue for our general use, however I believe the standard 12V battery BM2 calibration of 0% soc at 12.0V to 100% at 12.6V, to be off from our standard 2022 aux battery. According to the Hyundai tech notes, the 2022 I5 aux battery is a flooded cell type which should be 0% soc at 11.6V, not 12.0V.

Now the reason I had to re-connect the BM2 for every voltage above is that on boot, BM2 just uses its standard 12V battery BM2 chart (albeit not our chart as explained above). I noticed that if I changed the voltage while the BM2 was running, the BM2 reported voltage changes correspondingly within a few seconds and the fast 5 minute curve follows within seconds as well. However, it is many minutes later before the soc starts to change up or down. So, the BM2 reported soc is based on voltage and elapsed time, probably just some very very incredibly slow averaging routine.

Below the voltage chart is another chart showing a quick example of the very slow change in BM2 reported soc for a step change in battery voltage. I booted the BM2 at 12.51V, then changed the voltage to 12.55V. The BM2 reported the new voltage and the fast graph also changed correspondingly within a few seconds. However, the soc did not increase from the initial 85% to 91% for over an hour! Best understanding to date is that ignition off, our I5 charges the aux battery from 85% soc to 92% in up to 20 minutes. My guess is the rough BM2 voltage numbers are pretty close, but the reported soc is probably lagging behind actual soc by tens of minutes. [oh, the BM2 soc symbol changes to "charging" above some voltage, I bet in that algorithm mode, the BM2 wants to see some higher voltage, e.g. 13.8V for a period of time to move the soc higher at a faster rate.]

Update: Just had a quick look at the custom battery page (see last picture below). At first glance, (under bluetooth device in the setup) changing from "intelligent power algorithm" to "Voltage corresponding to power" seems to make for an instant look up table response of voltage to aux battery soc. Not sure why they call it "power" instead of soc, probably a minor technical error, not a big deal. Also, it looks like one could manually enter the table below for the I5 flooded leaded acid battery. The BM2 will still display the correct voltage without any of these changes, which all apply to the BM2 displayed soc number.

UPDATE: as shown below by @Zuinige Rijder, the battery may not be a flooded cell battery as described in the Hyundai Tech note.

Later: I entered the proper flooded battery table, however the instant soc does not seem to reflect the custom table, maybe I did something wrong. Did not exit the last setup page with the blue check mark, see the post below (exiting by the "X" does not save the custom settings).

 

[Blucid]

These should be the correct custom BM2 App settings for the IONIQ 5 2022 OEM aux battery. After changing all of the values (hold down "-" or "+" to increment faster), on the last page, push the blue check mark, not the small X in the upper right corner.

 

[KiwiME]

Was your BM2 voltage accurate against the bench supply? Certainly I use mine (on both EV and ICE) only for voltage logging and care little about the developer's idea of SoC or any of the other mostly ICE-related features. There's a trimpot inside for voltage calibration as I'm sure you've seen.

 

[Zuinige Rijder]

@Blucid I use another battery chart, because my IONIQ 5 battery is sealed, according to what I know:

This also matches the (default) values of my BM2 monitor.
Or will BM2 report other voltages, when you configure this how you did describe it?

37110-3X100 M60 CMF60L-DIN(12v) 60Ah 20HR RC 92min CCA 550A

 

[Blucid]

@Blucid I use another battery chart, because my IONIQ 5 battery is sealed, according to what I know:

This also matches the (default) values of my BM2 monitor.
Or will BM2 report other voltages, when you configure this how you did describe it?

37110-3X100 M60 CMF60L-DIN(12v) 60Ah 20HR RC 92min CCA 550A

Hyundai says the battery is a flooded cell type battery. See the attached TSB which, although from 2018 this TSB seems to be linked as relevant to the Ioniq 5 OEM 37110-3X100 battery. Also, BM2 voltage graphs on this forum show the aux battery charging at around 12.5V, which is consistent with the 85% value shown the Hyundai diagram.

So, I think the standard chart is not correct for the I5 battery. The chart my be important where recharging events are described in Hyundai documents and diagrams by aux battery soc, and we need some way to at least roughly correlate aux bat soc to voltage as a tentative check on what is happening.

The BM2 voltage graph is not changed by the BM2 battery voltage chart (standard or custom), only the BM2 soc values use the chart.

 

[Blucid]

Was your BM2 voltage accurate against the bench supply? Certainly I use mine (on both EV and ICE) only for voltage logging and care little about the developer's idea of SoC or any of the other mostly ICE-related features. There's a trimpot inside for voltage calibration as I'm sure you've seen.

My BM2 is surprisingly close in absolute calibration, it reads about 20 mV low (voltage reference set to 12.72V, BM2 reports 12.70V).

I agree with your note on soc, however some users may rely on the reported soc value, so I was hoping to make some sense of it. If the recently re-published Hyundai diagram that says ignition off charging begins at aux battery soc 85% is correct, the BM2 fast 5 minute graphs seem consistent where around 12.5V, systems including the LDC turn on, which drop the voltage under momentarily high load to 12.2V or 12.1V, then with LDC on, the voltage goes up to whatever the LDC charging voltage is for current conditions.

The longer term 1 day graphs can be confusing, without the detail the 5 minute view. Is there a way to zoom the 1 day graph, or to recover the raw BM2 data?

 

[Zuinige Rijder]

@Blucid Is there a difference in 12 volt Battery between Europe and USA? Because I see 12.89 Volt in my BM2 readings, so that would charge the 12 volt battery to 120%?

 

[Blucid]

@Blucid Is there a difference in 12 volt Battery between Europe and USA? Because I see 12.89 Volt in my BM2 readings, so that would charge the 12 volt battery to 120%?

Could it be the LDC is on, or the battery voltage is settling from the LDC having been on? AFAIK, the batteries are the same. The LDC (DC/DC converter in the ICCU) can run very near, but always at least just above battery voltage, LDC voltage changes according to conditions.

 

[Zuinige Rijder]

@Blucid No, the LDC is then not on, I see it charging to around 12.89 Volt and it will start charging around 12.5 Volt.
12.5 Volt is in my chart around 70%. AFAIK there is a sealed flooded 12 volt battery and non-sealed flooded 12 volt battery, both have different characteristics.

Here my BM2 readings:

 

[Blucid]

Certainly interesting, thank you for the video. I agree there are at least two types of batteries. Your curves do seem to show the 12V sealed battery case (almost AGM like). Soon enough, I should be able to say what is in my 2023 I5. If the Hyundai links are incorrect as to flooded cell, it will be good to know that.

On Nov 29, Is the I5 parked (plugged in L2 (or L1), or not?) to about 5, drive to about 7:30, parked (plugged in?), 14:27 is I5 automatic battery maintainer at 12.5V?, status 16:30?, then parked (plugged in?) through midnight?

I am starting to wonder how many changes, if any, happened between the drawing that shows the ignition off 85%/92% soc aux battery charge logic, and the production models. Also, I hope to be able to decode LIN battery sensor data output recorded by a LIN logger (at least the reported aux battery soc and current).

 

[Zuinige Rijder]

Here is my tripinfo for those 3 days, generated with my tool:



31-11-2022 I did make no trips. All the 3 days NOT on a charger, just standing still, except for the trips. And the key in a box, so the car would not be awakened by walking by the car.

2022-11-29      Regen        Consumption    Engine   Climate   Electr.  BatteryCare
   3.2kWh        1.3kWh        5.6km/kWh     2.3kWh    0.2kWh    0.7kWh     0.0kWh
    18km         39.5%       18.0kWh/100km    70%       7.7%     22.3%       0.0%
                  Trip                      Distance  Avg speed Max speed  Idle time
  (0.7kWh)    16:25-16:35     (9.7km/kWh)     6km      45km/h   108km/h     0 min
  (2.1kWh)    14:24-15:08     (5.9km/kWh)     12km     34km/h    73km/h     20 min

This matches the comment by the video:

nov 29/2022:  
6:10 forcerefresh by tool hyundai_kia_connect_monitor (github) 
6:10-7:27 automatic 12 volt battery recharge (probbly triggered by bluelink refresh)
14:25-15:15 trip 
16:30-16:40 trip

And for 30 november:

2022-11-30      Regen        Consumption    Engine   Climate   Electr.  BatteryCare
   1.0kWh        0.5kWh        4.0km/kWh     0.6kWh    0.1kWh    0.3kWh     0.0kWh
    4km          46.2%       25.2kWh/100km    61%       9.2%     29.8%       0.0%
                  Trip                      Distance  Avg speed Max speed  Idle time
  (0.7kWh)    12:39-12:45     (3.9km/kWh)     2km      31km/h    67km/h     0 min
              12:21-12:26       (2.5km)       2km      32km/h    61km/h     0 min

This also matches the comment by the video:

nov 30/2022: 
1:30-2:13  automatic 12 volt battery recharge 
6:10 forcerefresh by tool hyundai_kia_connect_monitor (github) 
11:25 drop of 12 voltage, nothing done 
12:21-12:40 two small trips 
16:30-17:19 automatic 12 volt battery recharge

And the comment by the video of 31 November:

disabled forcerefresh by tool hyundai_kia_connect_monitor (only cacheupdate from now on)
0:02 12.86 Volt 
9:45 12.78 Volt

 

[Blucid]

Excellent, nice work! I might try to set it up on my spare Raspberry Pi.

Just looking at two cases of aux battery saver, Nov. 30 12.68V (1:31, 12.68V, 85% std chart); Nov. 30 12.55V (16:21, 12.55V, 75% std chart). That is where I am wondering what the LIN battery sensor was reporting before and when the aux battery saver comes on. The battery sensor probably reports 24/7, although possibly at different rates when ignition is on and when ignition is off, I5 idle for some time. The decision to start the aux battery saver may be based on battery sensor reported soc, which may or may not match the BM2 voltage to soc chart (TBD).

Does your owner's manual have this comment about the aux battery saver? (somewhere around page 1-51 or so, seems to vary by version): The Aux. Battery Saver+ activates maximum of 20 minutes. However, your battery saver seems to run for about 45 minutes +/- both times, which is certainly more desirable if needed. Maybe an update since the manuals were written?

 

[Zuinige Rijder]

@Blucid

I also see that sometimes the 12 volt battery is charged on higher or lower voltage. I also see that sometimes there is a sort of 2 hours battery drain. But then I should make a longer video with all data in combination with the trip data.

No, in Europe there is no comment about the aux battery saver. Not in Dutch manual, English manual or German manual.

But "battery saver" is mentioned in the UK manual on page 5-64, but probably this is not what you mean:

Battery saver function
The purpose of this feature is to prevent
the battery from being discharged.
The system automatically turns off the
position lamp when the driver turns the
vehicle off and opens the driver-side
door.
With this feature, the position lamps will
turn off automatically if the driver parks
on the side of road at night.
However, the position lamps stay ON
even when the driver-side door is
opened if the light headlamp switch is
turned to the position lamp or AUTO (if
equipped) position after the engine is
turned off.
If necessary, to keep the lamps on
turn the position lamps OFF and ON
again using the headlamp switch on
the steering column after the engine is
turned off.

 

[KiwiME]

The longer term 1 day graphs can be confusing, without the detail the 5 minute view. Is there a way to zoom the 1 day graph, or to recover the raw BM2 data?

The BM2 sampling time when 'sleeping' is 2 minutes and I see that temporal resolution discernible in the 24-hour log graphs. I doubt the raw data has anything better.

As far as battery types go these EVs use the common lead-calcium "maintenance free" (CMF) type which I believe is described best by the 100% = 12.89 V chart.

On my early telematics-free Kona I see mostly this (below) while parked. The termination of either traction charging or driving both invite hours of traction battery "maintenance" which does pull it down quite a bit, but generally not under 12.5 V. The dealer has managed to get it down to about 11.5V while doing BMS updates where the LDC must be off.

My main criteria of battery health is that it always fully recovers from any normal discharge after a 20-minute charge at 14.65 V. It always has and that's why I think it has lasted so long.

 

[Zuinige Rijder]

I asked ChatGPT about the sealed and non-sealed lead-acid batteries. Answers should be taken with a grain of salt. 🧂

q! What is the difference between a sealed and non-sealed common lead-calcium "maintenance free" (CMF) 12 volt battery type

A sealed lead-calcium "maintenance free" (CMF) 12 volt battery is a type of lead-acid battery that is sealed and does not require regular maintenance, such as adding water to the cells. A non-sealed lead-calcium "maintenance free" (CMF) 12 volt battery, on the other hand, is a type of lead-acid battery that is not sealed and does require regular maintenance, such as adding water to the cells.
The main difference between the two is that sealed batteries are sealed and don't require regular maintenance, while non-sealed batteries are not sealed and do require regular maintenance.
Sealed batteries have a sealed design that eliminates the need for adding water to the cells, reducing the risk of leakage and spillage. They also have a longer service life and are less susceptible to damage from vibration and impact.
Non-sealed batteries, on the other hand, require regular maintenance, such as adding water to the cells, to keep them in good condition. They also have a shorter service life and are more susceptible to damage from vibration and impact.
It's important to note that sealed batteries are generally considered to be more reliable and convenient, but they may also be more expensive than non-sealed batteries.

I was of course interested in the voltage chart.

q! what are the battery voltage chart for those 2 types? Are they different?

Both sealed and non-sealed lead-calcium "maintenance free" (CMF) 12 volt batteries are lead-acid batteries, and as such, they have similar voltage characteristics. The voltage of a lead-acid battery is dependent on its state of charge, with the voltage increasing as the state of charge increases and decreasing as the state of charge decreases.
Here is a general voltage chart for a lead-acid battery:

• 100% charged: 12.7-12.9 volts
• 75% charged: 12.4-12.6 volts
• 50% charged: 12.2-12.3 volts
• 25% charged: 12.0-12.1 volts
• Discharged: Below 12.0 volts

It's important to note that the specific voltage range may vary depending on the manufacturer and the specific battery model, and this chart should be used as a general guide.
In terms of voltage, there is no big difference between sealed and non-sealed lead-calcium "maintenance free" (CMF) 12 volt batteries, but as mentioned before, sealed batteries are generally considered to be more reliable and convenient, but they may also be more expensive than non-sealed batteries.

q! How long does a lmaintenance free CMF lead-acid battery last?

The lifespan of a maintenance-free lead-acid battery, also known as sealed lead-acid (SLA) or sealed lead-calcium (SLC) battery, can vary depending on several factors such as usage, charging habits, and temperature. However, on average, these batteries can last between 3 to 5 years with proper care and maintenance. It is important to note that the lifespan of a maintenance-free lead-acid battery can be shortened if it is frequently deep-discharged (discharged below 20% of its capacity) or overcharged (charged above 100% of its capacity).
It's also important to note that as a lead-acid battery ages, its performance and capacity may decrease. A good indicator of a battery's health is its open circuit voltage (OCV) measured with a multimeter when it is not under load. A healthy battery should have a OCV of 12.6V or higher, and a battery that has a OCV lower than 12.4V is considered weak and may need to be replaced.
It's always recommended to use a charger that is specifically designed for sealed lead-acid batteries and to follow the manufacturer's recommendations for charging and storage to prolong the battery's life.

 

[Zuinige Rijder]

I made a long video with all my BM2 data and also in the description the trip information, so you can look up if something special occurred.

IONIQ 5 BM2 12 volt battery readings from 28 November 2022 till 20 January 2023.