• State TI Thinks Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 3 answers
  • Subscribers 63 subscribers
  • Views 531 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

BQ25895: Register value not retained when changing the /CE (pin 9) pin

FFabio
Expert 1455 points
Part Number: BQ25895
Other Parts Discussed in Thread: BQSTUDIO, EV2400,

Dear Team,

I'm working with the BQ25895EVM-464 board and the EV2400 interface board (firmware v. 0.18), BQSTUDIO v1.3.101 on a Win10 64 bit machine.

Communication with the evaluation board seems ok but I can not understand why the parameters on the registry is changed when the /CE pin is switched.

Actually I'm working with a little LiPo battery (420 mAh but able to give in discharge up to 15C) so I've set a fast charging current equal to 384 mA (REG04), precharge current and end-charge current are set to 64 mA (REG05), the minimum settings available inside BQSTUDIO and from BQ25895 datasheet.

The Input Current Limit is set by the registry to 3.25A (REG00) and the R19 trimmer on the evaluation board was tuned to achieve 0 ohm, so the only effective resistace on the ILIM pin is given practically only by R1=130R so the value of the maximum input current set by the limiting resistor is equal to IINMAX=390/130=3 A (suppose KLIM_max=390).

The idea is keep the /CE pin high, write the parameters on the board and the force a low level on the /CE pin to enable the charging of the battery avoiding then a too high charging current that is set from the default register value...

But when I perform the registry write and then I manually read back the registers content the value seems not updated and equal to the default values though the Device ACK into the BQSTUDIO is OK hence I think there are no communication errors with the device (the EV2400 I2C Clock Speed is set to Low Speed 100 kHz).

I've attached my registry settings for your reference.

Thanks and best regards.

F.

Register settings

Fullscreen 5353.bq25895-reg-test-1.txt Download 
* Created: Mon Dec 28 14:28:13 CET 2020
*
* Format: Register Name  tab Character,\t  Hexadecimal register value.
* Device: bq25890
* BQZ Container: Charger_1_00-bq25890_895.bqz
*
Input Current Limit, HIZ and ILIM pin	7F
OTG Temperature thresholds, and Input Voltage DPM Offset	05
ADC controls, OTG Frequency, ICO, DCP, and Input Detection	7D
Battery Load, WDT Reset periods, OTG and Charge Enable, and MINSYS setting	3A
Fast Charge Current Limit settings and MTK Pump enable	06
Pre-charge and Termination Current Limit settings	00
Charge Voltage Regulation Limit, Battery Low Voltage and Recharge threshold	5E
Termination Current Enable, STAT pin, WDT, Safety Timer, and JEITA Charge Current	8D
IR Compensation and Thermal Regulation thresholds	03
ICO, Extended Safety Timer, BATFET control, MTK Pump Voltage, and JEITA Charge Voltage	44
OTG Regulation Voltage and Current Limit settings	93
Charger VBUS, Charging Mode, PG, USB Input, and System Voltage Regulation status	A6
WDT, OTG/VBUS OVP, Thermal Shutdown, Safety Timer, and TS pin faults	80
Input Voltage DPM settings	13
Thermal Regulation Charge Status, and ADC Battery Voltage conversion	5C
ADC System Voltage conversion	5F
ADC TS pin voltage as percentage of REGN conversion	52
VBUS status, and ADC VBUS Voltage conversion	98
ADC Charge Current conversion	00
DPM status, and Input Current Limit setting	3F
Registers reset, Input Current Optimizer status, Device ID, and Temperature Profile	39

  • 0
    Expert 1455 points

    UPDATE #1

    I've shorted DP with DN on the BQ25895EVM-464 board, to do that I've removed the jumper from JP3 and put it on JP5.

    Now the registry setting value are retained.

    UPDATE #2

    Another question is about the Status window (Field View), when the VBUS is applied I can see inside the Input Current Limit textbox 500 mA, when the VBUS is removed the text change into 3.25A but with the selected settings the Input Current Limit should be always fixed to 3.25 A? Why this value change?

    To solve the Input Current Limit I've removed the check on the Enable Auto-Input Detection checkbox, now the VBUS report aways the string OTG, voltage on PMIS is 5V and the boost mode is activated 30 ms after the VBUS is removed as stated into the datasheet.

    ISSUE ABOUT VBUS label on BQSTUDIO

    In my application I'm not using the PMID output so I've disabled this feature by pulling down the OTG pin through the JP6 jumper, now when the VBUS is not applied at PMID I've the battery voltage, instead the 5V from the boost (and cyclically I can see a little drop long 75 ms on the PMID voltage, measured with the battery in fast charging stage), but in BQSTUDIO into the VBUS textbox I can read "No input" everytime even if the VBUS is applied.

    In order to understand when the VBUS is present or not I can see that inside the "VBUS Good" text box I've "VBUS attached" or "No VBUS attached" so I can use the bit 7 of the REG11 to understand when the VBUS is present or not, but the question is if the "No Input" text field is a BQSTUDIO issue or I've to perform some more checking or change other options.

    I've attached two screenshots when VBUS is attached and also removed.

    Thanks and best regards.

    F.

  • 0 in reply to FFabio
    TI__Guru 57481 points

    F,

    What is REG02 value when VBUS shows "No input" and "VBUS attached"?

    Thanks,

    Ning.

  • 0 in reply to Ning T
    Expert 1455 points

    Dear Ning here below the requested data:

    - VBUS not present

    REG02=7C / FC (I've enabled the ADC reading so the bit 7 CONV_START can be 0 or 1) => rewritten in binary form: X1111100 (where X can be 0 or 1)

    - VBUS present

    REG02=7C / FC (I've enabled the ADC reading so the bit 7 CONV_START can be 0 or 1) => rewritten in binary form: X1111100 (where X can be 0 or 1)

    hence the REG02 do not change respect the VBUS.

    Here below the main BQSTUDIO screenshot when VBUS is not applied (first) and when is applied (last).

    Thanks and best regards.

    F.

  • 0 in reply to FFabio
    Expert 1455 points

    QUESTION RELATED TO THE BEHAVIOR WHEN THE BATTERY IS GOING TO VBATT MIN AND VBUS IS NOT PRESENT

    I've did a fully battery discharging test starting from a charged battery, hence I've removed the VBUS and then the load was properly powered by the battery.

    My load, attached to the J3 connector of the evaluation board, is made by a parallel of 5 block made each one of a 165R resistor in series with a red led.

    At the start of the discharge testing the battery voltage was equal to 4,125V with a overall load current (SYS pin) equal to 69.16 mA.

    After approx 6h 23' my 3.7V 420 mAh single cell LiPo battery was fully discharged, when the voltage was going below the VBAT(DPL) threshold all the leds have started to blinking for some seconds at a rate of approximately of 0.5Hz - 1Hz then after a few seconds the LEDs have stopped flashing. I immediately measured the voltage on the terminals of connector J4 (the battery connector) and the reading was equal only to few mV so I've checked again and the voltage was really at so low level.

    My question then is if the charger perform a true battery disconnection in some way when the battery is depleted and also the VBUS is not available, in such situation the BATFET is really open or not? About this last point is possible to have a little more information about the BATFET structure?

    Thanks and best regards.

    F.

  • 0 in reply to FFabio
    TI__Guru 57481 points

    F.

    When AUTO_DPDM_EN (REG02[0]) is disabled, the Input Source Type Detection is bypassed. The Input Current Limit (IINLIM) register, VBUS_STAT, and SPD_STAT bits are unchanged from previous values.

    Referring to 8.2.2 Device Power Up from Battery without Input Source, If only battery is present and the voltage is above depletion threshold (VBAT_DPLZ), the BATFET turns on and connects battery to system. Referring to 8.2.6.3 Supplement Mode, BATFET turns off to exit Supplement Mode when the battery is below battery depletion threshold.

    Thanks,

    Ning.

  • 0 in reply to Ning T
    Expert 1455 points

    Dear Ning,

    thanks for your support!

    About the first point, when the Input Source Type Detection is bypassed, hence the only way to know if the VBUS is present or not is to check the bit 7 of the REG11, I'm right?

    About the battery disconnection when the VBUS is not present I need some clarification.

    From BQ25895 datasheet, point 8.2.6.3 page 20 the following sentence:

    "When the system voltage falls below the battery voltage, the BATFET turns on and the BATFET gate is
    regulated the gate drive of BATFET..."

    when VBUS is removed the system voltage should drop to the V(BUSMIN) voltage (page 7, value 3.8V) then the SYS output voltage should be temporarily regulated by the buck stage to a value between VSYS_MIN and 4.848V (in my situation OTG mode is disabled). Is correct?

    Just as a note from BQSTUDIO the VBUS voltage is equal to 2.6V because the ADC have a minimum offset equal to 2.6V and no voltage is present at VBUS input.

    Now, when the VBUS is not present the battery is the only element able to supply the load, so the SYS voltage should be decrease until it reach the battery voltage and at this point the BATFET should be turned fully on to connect directly the battery to the SYS output (the buck stage then is not working). The battery then continues to gradually discharge over time until the battery voltage reach the V(SYS_MIN) that in my case is set to the default value equal to 3.5 V. What I can see is that without the VBUS clearly the SYS voltage follow the battery voltage so the BATFET should be always fully on because there is no way to perform any regulation and I think that the system is always working in Supplement Mode. If my thinking is correct the BATFET should remain in this state until the battery voltage falls below the minimum threshold given by VBAT(DLP)=2.15V ÷ 2.5V (page 7) then the BATFET should be opened and then the battery voltage should be stay at the VBAT(DLP) value.

    Could you please let me know if my interpretation is correct?

    Thanks and best regards.

    F.

  • 0 in reply to FFabio
    TI__Guru 57481 points

    F,

    It is better to use VBUS_GD bit (REG11[7]) to check if valid VBUS is present.

    8.2.6.1 Narrow VDC Architecture on the d/s applies when valid VBUS is present.

    If only battery is present and the voltage is above depletion threshold (VBAT_DPLZ), the BATFET turns on and connects battery to system. VSYS is VBAT minus BATFET voltage drop. Please refer to 8.2.6.3 Supplement Mode Figure 12 on the d/s for the details.

    It is highly recommended to get familiar with the charger operation by evaluating the EVM.

    Thanks,

    Ning

  • 0 in reply to Ning T
    Expert 1455 points

    Dear Ning,

    thanks for your info.

    About the behavior when only the battry is present I'm just testing the system through the EVM, I'm also investigating the battery behavior around the cutoff voltage that, for the battery I'm using, is around 3V so I think that the reason about the 0V reading is due to the over discharge battery protection that is inside the battery, upon this point I'm just waiting some info from the battery manufacturer, then I'll update the test results in that sense.

    Happy new year!

    Thanks and best regards.

    F.

    UPDATE - 04/01/2021

    From the battery manufacturer the battery have a cut-off voltage equal to 3.0V±0.1V so the protection inside the battery will open the battery connection to prevent the over-discharge and this should be the reason of the strange behavior.

  • 0 in reply to FFabio
    TI__Guru 57481 points

    F.,

    Thank you very much for your update!

    Happy New Year,

    Ning.