BQ34Z100-G1: Unable to complete Learning Cycle

Hello Hicham,

Can you send me the data sheet for the battery you are using.

Regards,

Nick Richards

Here's the datasheet, for reference I've been using chemID 0710 during my testing.
7.2V 26A Huahui Battery.pdf

Hello, 

Please expect a delayed response as Nick is currently traveling for work. 

Regards, 

Robert. 

Hey Nick,
Thanks for the suggestion, I'll give it a try and let you know how it goes.

Hi Hicham,

Sounds good.

Regards,

Nick Richards

Hey Nick,

I tried running a new Learning Cycle and during the first discharging step I got an error on Battery Monitor Studio about losing comms to the BQ IC. 
Now when I try reopen Battery Monitor Studio, the BQ34Z100 is initially not Autodetected but then when I select a target and fully open the program, the BQ chip gets recognized for about a second before it disappears.
I've checked that the MSP430 software has been updated to the newest v0.28 build and that Battery Monitor Studio is running v1.3.127 but the BQ chip is still unrecognizable.

My circuit involves a BQ chip on a PCB with the I2C pins routed through an Arduino Relay board which controls charging and discharging flow. I've also tried connecting the I2C connections straight from the MSP430 into the BQ chip and sill can't connect to the device.
I've used the setup with the Arduino relay board for my previous Training Cycle attempts where I've been able to communicate with he BQ chip for weeks at a time without any comms issues until now. I've also tried using an entirely new board with a new BQ chip which has yielded the same comms issues.

I've attached the logs below, any help is appreciated.

Thank you,
Hicham

 FourthAttemptErrorLogs.zip

I also wanted to add that I am connecting port 2 on the EV2400 to my BQ24Z chip to enable I2C. I tried running both v0.18 and v0.28.

Hello Hicham,

Please allow me some time to look into this issue.

Regards,

Nick Richards

Hey Adrian, any updates available on this?

Hello Hicham,

Please ensure that the gauge is being properly powered. Also, can you check to see if the gauge is in ROM mode? You can do a read word of 0x0D and if the results is 0x03 0x70 then the gauge is in ROM mode.

Regards,

Adrian

Hey Adrian,

I'm currently unable to connect to the BQ chip at all through the EV2400 so unless there's another way I'm not aware of I don't think I can read any memory/registers to check the mode.

Hello Hicham,

Sorry I should have been more specific. Force open the bq34z100.bqz file by opening bqstudio clicking "Gauge" and opening the corresponding bqz file (this is completely fine, the EV2400 does not need to autodetect the gauge). Then go to the advanced comms tab and do a 2 byte read from register 0D. Please share with me what the results are from trying to perform this read.

Regards,

Adrian

Hello,

Thanks for clarifying, attached below are the results from attempting the 2 byte read:

Hello Hicham,

Thank you for verifying this. Can you verify that the SCL and SDA lines are pulled high?

Regards,

Adrian

Hey Adrian,

Both SDA and SCL are measured at 2.2V.

Hey Hicham,

Can you get a digital logic capture of trying to send communications to the device? I just want to make sure that communication is trying to be sent to the device. After this, I feel the next best step is to do an ABA swap of the physical bq34z100 IC.

Regards,

Adrian

Also, can you check the TS pin on an oscilloscope and see if there are periodic pulses on the plot.

Regards,

Adrian

Hey Adrian,

Thanks for the suggestion, I already started a new Learning Cycle with a fresh BQ IC but if the same error occurs I'll follow your recommendations.

Regards,
Hicham

Hello Hicham,

Sounds good, let me know the results of the new learning cycle test.

Regards,

Adrian

Hey Adrian,

Still running a Learning Cycle in the background but I just wanted to ask is there a risk to resting too long in between charge/dscharge cycles? Can't find much about that in any documentation but I just wanted to confirm so I can avoid any future learning cycle failures.

Hello Hicham,

There is no risk to resting too long. It is better to rest for too long than to not rest long enough.

Regards,

Adrian

Hey Adrian,

i just finished a learning cycle and it was successful. I think the main issue was using a low discharge current.

Thanks for all the help,
Hicham