BQ27427: The accuracy of temerparue measured function as external NTC on BQ27427

Hello Johnny,

Could you re-send the schematic, the link you attached goes to a Microsoft bing page.

Thank you,
Alan

hi ,Alan Frias

     The schematic diagram has been resubmitted. thanks

hello,Alan Frias

      the schematic has been resubmiited.

and the 2nd question: can explain  why get the different result by changing value of registor R2713 or cancel.

thanks

Hello Jonny,

It seems like the issue can be coming from VDD being connected to BIN and could also be from the ADC_NTC being used as well. You would only pull it to VDD with a 20k resistor if you are planning on not using a thermistor.

Thank you,
Alan 

hello, Alan Frias

I have already connected the NTC resistor externally through the BIN, as shown in the previous circuit. My question is that the circuit has already been pulled up with a 10k ohm resistor. Keeping the circuit unchanged, how can I modify the software to make the external NTC detection correct? thanks       

Hello Johnny,

If you would like to setup the thermistor software wise than you can calibrate the thermistor in BQStudio within the calibration tab or you can also go within the the Data Memory >> Calibration >> Data >> Ext Temp Offset and update the offset error. However in terms of hardware I would validate your offset error on a EVM.

Thank you,
Alan

hello, Alan Frias

      I have tried these methods before, but the error values detected are still significant.  do you have other ways ?thanks

hello，Alan Frias

    the  2nd question: Can you explain the reason for the large difference in test results when external pull-up resistors with different resistance values are connected?

Hello Johnny,

Would it be possible to grab a EVM program the same golden image (.srec) and see if it's a hardware issue or not for confirmation.

Thank you,
Alan

Hello Johnny,

After reviewing your schematics I would remove the connection you have to VDD and have a direct connection to the thermistor that connects to ground.

Thank you,
Alan

hello，Alan Frias

       how to grab a EVM program the same golden image (.srec) from our design pcb-board? have you better ways?

Hello Johnny, 

You would export the .srec from the golden image tab and than with the new EVM import that onto the programming tab with the .srec.

Thank you,
Alan

hello, Alan Frias

        Our design board cannot access images through EVM, but I print some data memory through software, including subclass ID=82/89/64/104/105/107, as shown in the picture. 

Hello Johnny:

I would recommend using BQstudio since it would be only be needed to be used during development period. I am not to sure what method/software you are using but we don't support software's outside of bqstuido.

Thank you,
Alan

hello,Alan Frias,

how about this test result from you ? we have get this result by this way.

My question is that the circuit has already been pulled up with a 10k ohm resistor. Keeping the circuit unchanged, how can I modify the software to make the external NTC detection correct?

thanks.

Hello Johnny,

The test result you sent doesn't have any meaning with my current knowledge. However if you don't want to remove R2713 the pullup resistor and R2720 than the only other suggest I can make is going into the calibration tab and calibrating the temperature portion.

Calibration E2E to reference:
https://e2e.ti.com/support/power-management-group/power-management/f/power-management-forum/1390346/bq27427-bq27427-calibrating-fuel-gauge-for-given-cell

Thank you,
Alan

hello, Alan Frias,

      The reference case is an introduction to voltage, current, and battery calibration,But according to the current circuit design, how do we want to solve the problem of correct temperature detection.

      Additionally, the data provided to you is the data memory in BQ7427-0, which includes numerical values for several subclassID

hello, Alan Frias,

       I get some value about data flash include subclassid-82,89,64,104,105,107. Refer to attach file.

Hello Johnny,

The way we would solve the problem is by removing R2713 the pullup resistor and R2720, thermistor don't need a pull-up resistor since the thermistor themselves create a voltage. Also I would recommend using BQstudio since we have a lot of documentation on the product page with this tool and use this tool to send files of different events to debug the users issues.

BQstudio: 
https://www.ti.com/tool/BQSTUDIO

Thank you,
Alan