BQ34Z100-G1: Output look strange after convresion from I2C to HDQ

I could not read the calibration values in data memory, but when I push it back into I2C mode everything looks fine.
It seems like something goes wrong in the conversion to HDQ very consistently...

Reagrds,

Martin

I can see in Table 7-16 update status definitions, that the Golden image should be created at status 0x02.
I created the Golden image at status = 0x06

Could that have any effect?

The battery pack on which I performed the learning and created the golden image from, that works fine in HDQ mode. This board I converted back to I2C today to check the data. When I converted to HDQ using the button I had the same result with "black" battery in the dash board showing 157mV. I managed to get it back into I2C mode and when I sent 0x29E7 it converted correctly to HDQ showing propper status and carried over all the correct values in the data memory.

I hope any if this helps you narrow down the issue.

Best regards,
Martin

Hi Martin,

The "black" battery is indicating SOC%, and incorrect SOC reading is not surprising if a learning cycle has not been performed. Getting accurate SOC is a separate issue that we can tackle next.

What I am most concerned about is the voltage and current reading drastically changing when the device is sent into HDQ mode. These reading are controlled by the calibration parameters. If these values did not change when switching to HDQ mode, this issue is probably related to the reading the HDQ data. 

Please navigate to window -> preferences-> advanced mode. then turn off scan on all registers except Voltage, temperature and current. 

Please also disable auto-refresh and the bit registers. Check if the voltage current and temperature readings are now correct.

Regards,

Diego

Hi Diego,

I am not sure if I do it correctly, but I logged the Voltage, Current and Temperature using the watch window in respectively I2C mode and HDQ mode.

In I2C mode I get:

Data Log

TimeStamp,Voltage,Current,Temperature,
2025-06-04 08:13:04,7818,0,22.0,
2025-06-04 08:13:06,7818,0,22.0,
2025-06-04 08:13:08,7818,0,21.9,
2025-06-04 08:13:10,7820,0,21.9,

In HDQ moce I get:

Data Log

TimeStamp,Voltage,Current,Temperature,
2025-06-04 08:31:34,157,-22273,-273.2,
2025-06-04 08:31:36,157,-22273,-273.2,
2025-06-04 08:31:38,157,-22273,-273.2,

Clearly something goes wrong, but I have no clue what.

Best regards,
Martin

One additional observation: When I activate my switch to show LED capacity in HDQ mode, I have 3 LEDs on! This correlates well with remaining capacity/voltage shown in I2C mode.

Hi again Diego,

Another observation: I restarted BQStudio and it started in I2C mode, so maybe the 3 LEDs (now 2) was somehow in I2C mode even thoug the dashboard said HDQ?!

Also, I always receive this error message when start BQStudio with the EV2400 module:

The HW engineer who made the initial design and work instruction made a note to just click OK, but I am wondering if there is a HW/SW/FW discrepancy?

We have a production stop due to this issue, so I am getting kind of desperate to finding solutions

Thanks,
Martin

Hi Diego,

More observations:

I have seen this on multi fresh boards I have flashed now with the golden image. I have seen it both in I2C mode (in data memory it clears when I "read all data" 3-4 times... then all values populate, but it starts with this error.

Best regards,
Martin

Diego,

You may feel that I am getting desperate.
This is the normal issue I see when trying to convert from I2C to HDQ:

When I power cycle it starts again in I2C mode

Cheers
Martin

Diego,

Last question for tonight (I hope you have something good for me when I wake up tomorrow):
Can I do anything fancy in this mode:

;.)
Martin

ThomasCosby

ThomasCosby This relates to this thread where you gave good advice on: 
https://e2e.ti.com/support/power-management-group/power-management/f/power-management-forum/539017/bq34z100evm-no-display-of-state-of-charge
I found out that the Chem ID was incorrect + multiple settings. That all fixes and I have a board that cosistenly shows accurate remaining capacity levels with four LEDS.
It just seems, that something goes wrong when converting to HDQ.

I hope you and Diego can help.

Thanks,
Martin

Hi Martin,

Try reducing your log/scan interval.

Martin Lastrup Hviid said:

Also, I always receive this error message when start BQStudio with the EV2400 module:

your device is not autodetecting, BQstudio should automatically open, you are most likely missing the correct bqz, which could be the reason for the HDQ error.

Martin Lastrup Hviid said:

More observations:

This could be related to your voltage on the bat pin or flash update ok voltage.

Martin Lastrup Hviid said:

This is the normal issue I see when trying to convert from I2C to HDQ:

looking at the screen shot your device is not communicating with BQstudio.

I also recommend upgrading to latest test version of BQstudio.

Regards,

Diego

I have upgraded to latest version. Still get the same errors.

Can I manually download the latest bqz. somewhere?

Please let me know if there is anything else I should upgrade or do?

Best regards,
Martin

For the record: My golden unit performs as expected. The LEDs show remaining capacity with a very tight margin of 25% for each LED. I just ran a new discharge and recharge cycle, and the LEDs turn on/off at the correct levels, so my big challenge is generating a working golden image.