• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 58 subscribers
  • Views 174 views
  • Users 0 members are here
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

BQ25887: ADC characteristic

Ivan Suvorau
Prodigy 90 points
Part Number: BQ25887

Hello,

We are using BQ25887 in our design. We don't need cell balancing feature in it, so, bit 6 in REG2A set to 0. Also, the CBSET pin left floating, and MID pin are connected to the mid point of two 33k resistor divider. The values of resistor are the same, I have double checked it. The problem is that when I check ADC readings in REG26 and REG1F, the difference between them are 100-200mV during charging(for example, 3.8780 and 4.1000). I checked that with voltmeter on the PCB, the readings the same, ADC works fine.

Can you please describe, what are the characteristics of MID pin and ADC input? In datasheet there is only one lane, about bias current. 

Also, can you mention, which values of the resistor should I choose to prevent this problem? We have battery powered device, so, I can't choose too low resistance.

Thanks!

  • 0
    TI__Genius 10385 points

    Ivan,

    So the cell balancing action has two pins as you are aware.

    The MID pin is purely a sense pin and is where the ADC senses. This pin benefits from the inclusion of 300 ohm series resistor in the event of a reverse battery polarity event.

    CBSET is the cell balancing power pin. If cell balancing is not being used, then it is okay to leave this pin floating as you have done.

    Could you show me exactly how MID is connected?

    Best Regards,

    Ricardo

  • 0 in reply to Ricardo
    Prodigy 90 points

    Ricardo,

    Here are the connection of these two pins. All other pins of BQ25887 are connected as they should from datasheet.

  • 0 in reply to Ricardo
    Prodigy 90 points

    25887log.txt 
    * Created: Wed Mar 03 11:32:04 MSK 2021
*
* Format: Register Name  tab Character,\t  Register Address  tab Character,\t  Hexadecimal register value.
* BQZ Container: Charger_1_00-bq25887.bqz
*
REG00	00	A0
REG01	01	5E
REG02	02	00
REG03	03	39
REG04	04	71
REG05	05	8D
REG06	06	7B
REG07	07	22
REG08	08	0A
REG09	09	F6
REG0A	0A	01
REG0B	0B	05
REG0C	0C	B2
REG0D	0D	00
REG0E	0E	00
REG0F	0F	00
REG10	10	00
REG11	11	00
REG12	12	70
REG13	13	00
REG14	14	00
REG15	15	B0
REG16	16	00
REG17	17	B5
REG19	19	40
REG1B	1B	13A8
REG1D	1D	1F93
REG1F	1F	103C
REG21	21	261
REG23	23	38
REG25	25	29
REG26	26	F4B
REG28	28	2A
REG29	29	F4
REG2A	2A	90
REG2B	2B	00
REG2C	2C	00

    Also, added register config for you convenience

  • 0 in reply to Ricardo
    Prodigy 90 points

    Ricardo,

    I just find what causes these bug. This bug only appear, when bit 6 of REG 2A is set to 0. When I set it to 1, the bug disappear, and all the readings from ADC stays correct(the difference between cells are 10-20 mV). When I set this bit to 0, the difference goes to 100-300mV.

    What should I do with that? Is that ok, that this bit will be set to 1 without connection of CBSET pin to real mid-point of battery?

    Thanks,

    Ivan

  • 0 in reply to Ivan Suvorau
    TI__Genius 10385 points

    Ivan,

    It is not a bug. My apologies I thought you already had set this bit. This bit will need to be set 1 whenever the charger PORs if you will not be using the cell balancing.

    Best Regards,

    Ricardo

  • 0 in reply to Ricardo
    Prodigy 90 points

    Ok, so let me sum it up:

    I need to disable cell balance feature, what should I do after POR?

    1. If I set 1 bit 6 of REG2A (Enable auto cell balancing), ADC works fine, but what consequents may happen with the connection I showed you?

    2. If I set 0 bit 6 of REG2A(Disable auto cell balancing), ADC works with error 100-200mV and I see that battery didn't get enough charge(Because of the difference, BQ25887 gets to CV earlier). So I can't use BQ25887 with disabled cell balancing?

    Thanks!

  • +1 in reply to Ivan Suvorau
    TI__Genius 10385 points

    Ivan,

    Thank you for the summary!

    I see what is happening now. 

    At POR:

    CB_AUTO_EN = 1 (it should already be 1 by default)

    CB_CHG_DIS = 0 (not a default setting)

    As you have biased MID to always be VBAT/2, the cell balancing algorithm will never kick on to force a current through CBSET. CBSET can be NC.

    By disabling CB_CHG_DIS, your charging cycle will not be interrupted for the cell balancing measurements.

    Best Regards,

    Ricardo