BQ27541: BQ27541-xx - Calibration steps (not using TI bqEVSW)

Sime Darby

Intellectual 470 points

Part Number: BQ27541
Other Parts Discussed in Thread: BQSTUDIO

Need help, can somebody share information on the calibration steps for BQ27541-xxx

I am currently working on a project with BQ27541 and need to write the code for Calibration to automate the process.

I cannot find any document regarding calibration steps and formula for BQ27541-xx.

over 7 years ago

0 Onyx Ahiakwo over 7 years ago

TI__Guru* 76230 points

Hi Sime,
See app note below:

www.ti.com/.../slua640b.pdf

thanks
Onyx

0 Sime Darby over 7 years ago in reply to Onyx Ahiakwo

Intellectual 470 points

Thank you.

I will give it a try.

0 Sime Darby over 7 years ago in reply to Onyx Ahiakwo

Intellectual 470 points

Hi Onyx,
Can you help me clarify the Checksum formula:
unsigned char newChecksum = calculate_New_Checksum_1B(13, (char) vOffset);
printf("newChk is %x\n", newChecksum);
send_extendedCommand_1B((0x40 + (13 % 32)), offset);
send_extendedCommand_1B(0x60, newChecksum);
_delay_cycles(100000);
send_extendedCommand(0X3E, 0x00, 0x68);

With this, I have to replace the offset13 with the new PackVOffset. What's the formula to calculate for Checksum?

0 Onyx Ahiakwo over 7 years ago in reply to Sime Darby

TI__Guru* 76230 points

hi Sime
see similar posts

e2e.ti.com/.../209823
e2e.ti.com/.../703893
e2e.ti.com/.../714815
e2e.ti.com/.../678295

thanks
Onyx

0 Sime Darby over 7 years ago in reply to Onyx Ahiakwo

Intellectual 470 points

Hi Onyx,
Thank you for the links above... Checksum is working now following these steps:
Here is the procedure to calculate the checksum.
1. Add all 32 bytes in a data row together.
2. Do an "integer divide" by 256 on the sum to get the most significant byte of the result.
3. Subtract out the most significant byte from the total sum, leaving only the least significant byte
4. Invert the result in #3, which can be done by subtracting it from 255.

I have another question, I cannot calibrate the voltage and temperature based on the tolerance that i wanted. I haven't tried the current yet.
So for Voltage it cannot calibrate within 10mV with respect to the reference voltage?
it just jump back and forth +253mV to -98mV difference with respect to the known applied voltage.

Same on temperature, I tried to calibrate it to 25degC, it just jump from 38degC to 13degC.

Any clue?

By the way, the actual partnumber of this FG is SN27541DRZT-H200.

0 Onyx Ahiakwo over 7 years ago in reply to Sime Darby

TI__Guru* 76230 points

Hi Sime
Have you looked at a different chip for the voltage and temp calibration to know if it is not just associated with that chip in question.?
This device is a custom device and ought not to be discussed on a public forum. i suggest getting in touch with your local FAE if you have one to forward questions to us directly
thanks
Onyx

0 Sime Darby over 7 years ago in reply to Onyx Ahiakwo

Intellectual 470 points

I been able to calibrate the Temperature to the accuracy of less than 1degC. But still having issue on Voltage Calibration, i cannot get a better accuracy closer to 10mV with respect to the reference voltage.

I tried the steps in indicated on SLUA640B, but still it cannot get the desired accuracy.
I tried to calibrate using TI BQ Evaluation SW, it works fine on accuracy.

0 Sime Darby over 7 years ago in reply to Sime Darby

Intellectual 470 points

Hi Onyx:
on SLUA640B, the VPACK Offset should be stored on SubCLass ID 104, Offset 13.
VPack Offset is simply the difference between the know voltage (analog measurement) and the voltage measurement from BQ.

Please correct me if im wrong on how to calculate the VPack Offset.

0 Sime Darby over 7 years ago in reply to Sime Darby

Intellectual 470 points

I tried to compare also the SLUSA11B document and the SLUA640B, in SLUSA11B to enter in Cal Mode is 0x0040 but in SLUA640B it is 0x0081.
There's no command also in SLUSA11B to exit from Cal Mode.

Please clarify.

0 Onyx Ahiakwo over 7 years ago in reply to Sime Darby

TI__Guru* 76230 points

correct

0 Sime Darby over 7 years ago in reply to Onyx Ahiakwo

Intellectual 470 points

Onyx Ahiakwo said:
correct

So, what command should i use, should i follow the commands from SLUA640B which is 0x2D, 0x81? or follow SLUSA11B with 0x0040 (cal mode)?

But if we will use 0x0040, how are we going to exit from calibration mode?

0 Sime Darby over 7 years ago in reply to Sime Darby

Intellectual 470 points

Hi Onyx,
I followed document SLUA640B and was able to write calibration values and checksum. I verified also the calibration value in the BQManagement studio and it shows the same result.
The only issue left is the calibration formula to get the desired accuracy.

With the formula based on SLUA640B, it seems its not correct because the accuracy is far from what it should be. We cannot calibrate within 10mV window with respect to reference voltage.

Hope you can help me clarify on what formula to use to get the Pack V Offset for Voltage calibration.

0 Onyx Ahiakwo over 7 years ago in reply to Sime Darby

TI__Guru* 76230 points

hi Sime,
What i would recommend at this point will be to connect your device to bqstudio and then calibrate voltage. Snoop the bus using an i2c analyser to see what you are doing wrong.
thanks
Onyx

Power management

Power management forum

BQ27541: BQ27541-xx - Calibration steps (not using TI bqEVSW)