• State Resolved
  • Locked Locked
  • Replies 9 replies
  • Subscribers 62 subscribers
  • Views 400 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.

BQ25155: 4-8 mA current in shipping mode

user6067694
Prodigy 80 points
Part Number: BQ25155


Tool/software:

Hi,
My design uses a BQ25155 to charge a single Li-Ion-battery and power a circuit with a microcontroller. This microcontroller will set the BQ to shipping mode via I2C once it needs to shut down. This mostly works, but in a few circuits I see a rapid loss of charge when it is switched off. In the most egregious case I measure 4.8 mA current from the battery even though most lines show zero voltage. The only parts where I measure any voltage (which is Vbat in all those cases) are the line to MR and VBAT on the BQ25155.Three other BQs in other circuits of the same design (out of 308 so far) also draw some current, but only between 0.22 and 0.5 mA.

Current starts to be drawn as soon as voltage is applied and the current rises with voltage. With 4.08 Volts at the battery, I measure 4.8 mA. When switched on, the circuit behaves as designed (with the almost 5 mA on top of the regular consumption, that is). The BQ communicates over I2C and delivers the 3.0 V requested by the controller. Battery charging also works properly. But when the circuit is switched off by sending the BQ via I2C the command to switch to shipping mode, the battery continues to be drained. I replaced the battery already and see the same behavior with the new battery, too.

Somewhere there must be an effective resistance of 833 Ω to GND. None of the resistors in the circuit has anything close to that. Also, an IR photo of the circuit (see below) shows a hot spot right where the BQ25155 sits.

The pins on the BQ are as follows:

MR and VBAT:  4.08 V from the almost fully charged battery

VIO, LS/LDO, LP  are all interconnected and show 0 V

SCL, SDA are connected to VIO via their pullups. I can only measure them indirectly (at the pullups) where I see 0 V.

ADCIN, TS, GND  are connected to ground, directly or via appropriate resistors

Has anyone seen this before? Is there a way to improve my circuit which makes the BQ stop to consume so much power in shipping mode? Again, three others out of a total volume of 308 circuits also draw some current in shipping mode, but less than 1 mA. The others are all doing fine, drawing a current below the range of my equipment. Are there known cases of shorts between single balls of the BGA which would explain this current?

  • 0
    TI__Mastermind 18425 points

    Hi,

    Ship mode quiescent current is one of the many parameters that are tested on these devices before they are shipped out, any devices not matching the specification aren't made available to customers. So quiescent current consumption greater than 150nA from these devices typically indicative of damage or soldering issues.

    Can you attempt an ABA swap with these high current ICs? The objective would be to check if another IC on the same board has the same current consumption, and if this IC on another board causes the same current consumption.

    Best Regards,

    Juan Ospina

  • 0 in reply to Juan Ospina
    Prodigy 80 points

    Hi Juan,
    thank you for the answer! It's good to know that you test shipping mode, and I also suspect a soldering issue. The next steps will be to X-ray the PCBs and swap the IC for a new one. However, I was wondering whether the amount of current gives already a hint which contacts might be shortened.
    Best Regards,
    Peter

  • 0 in reply to user6067694
    TI__Mastermind 18425 points

    Hi Peter,

    Unfortunately not. 4-8mA does exceed any shipmode or active battery quiescent current so they imply some kind of load current. Since the device is in ship mode then the BATFET should be disabled (confirmed by 0V on PMID) which means that any additional current must be connected directly to the battery pin. 

    If your schematic indicates any other devices connected to BAT those would be items to look at, otherwise it may be a soldering issue or device damage. Since you mention this only happens on some units and with varying magnitudes, I don't expect it would be a schematic design caused behavior.

    Best Regards,

    Juan Ospina

  • 0 in reply to Juan Ospina
    Prodigy 80 points

    Hi Juan,
    I think the thermal imaging precludes any other part to be the culprit. Also, all my voltage testing only showed any difference from GND at the LED (which could be expected, as any LED is a tiny photo element). Now I have the X-ray result of the 4.8 mA circuit, and it does not show any faults in the soldering:


    Best Regards,
    Peter

  • 0 in reply to user6067694
    Prodigy 80 points

    Sorry, I guess I could do better:

  • 0 in reply to user6067694
    TI__Mastermind 18425 points

    Hi Peter,

    Without any soldering issue then this points to the most likely culprit being IC damage of some sort. Can you try swapping out the BQ25155 with another IC to confirm that the board does not show the error with a working unit?


    Best Regards,

    Juan Ospina

  • 0 in reply to Juan Ospina
    Prodigy 80 points

    Hi Juan,


    yup, this will be the next step. Please allow me one week to report results.
    Kind Regards,

    Peter

  • 0 in reply to user6067694
    TI__Mastermind 18425 points

    Hi Peter,

    Looking forward to your results.

    Best Regards,

    Juan Ospina

  • 0 in reply to Juan Ospina
    Prodigy 80 points

    Hi Juan,

    now the three PCBs got their BQ25155s replaced and all show zero leak current. Now it should be proven beyond doubt that the parts were defective (3 out of a run of 1000). Two of them will be sent back to me - are you interested in them?

    Kind Regards,

    Peter