Post Go back to editing

LTC3554 - 100uA draw when battery is below threshold voltage

Category: Hardware
Product Number: LTC3554

Hi,

I am using LTC3554 in my project as PMIC for battery charging and running the all the other circuitry. When I simulate the discharged battery with an external power supply and drop the voltage below ~2.5V the 2 switching regulators stop working as expected, but the current draw on the battery is still around 120uA. I tried in both standby and non-standby modes and there is no difference. This seems quite a lot and I cannot find any such figure in the datasheet. Is this expected for normal operation mode?

Any idea what might be causing this. I also couldn't find much detail on how the chip behaves when battery is empty in the datasheet. It is weird to me that standby mode doesn't make a difference, also if I disable one of the switching regulators the draw is the same ~120uA.

Notes on the schematic:

- pbstat & pgood are connected to each other here - I cut the trace on the circuit board so there is no contact - no difference

- tried replacing 0ohm jumpers with 100k resistors for PWR_ON1 & PWR_ON2 - no difference

The ideal working pattern would be:

 - LTC3554 works in standby mode when the battery is charged enough

 - LTC3554 goes to POFF when battery is empty - less than 3.0V - I am assuming another voltage detector IC will be required here.

 - MCU can bring LTC3554 out of standby mode if necessary

Thanks,

Matej

  • Hi Matej,

    Thank you for the information. I do not see anything obvious with the schematic. 

    How is the current from the battery supply being measured?

    I do not expect a difference but is it possible to disconnect the /ON pin and use the internal pull up?

    Is it possible to disconnect BVIN just as an experiment to see where the current might be going?

    Is there anything else on VOUT that is not on the schematic?

    Thanks

  • Hi Marty,

    The current is measured with a normal multimeter between the PSU that simulates the battery voltage and the circuit.

    I also disconnected the ON from pin VOUT by cutting the trace and left the pin floating - does not make the difference. 

    It would be really hard to disconnect the BVIN from VOUT on the board, I would need to cut the trace right next to the chip. I am not sure what I could measure though.

    At the moment nothing is connected to VOUT, I rerouted the led driver from 3.3V buck to VOUT which works fine if voltage is high enough but doesn't make a difference in UVLO mode. 

    If the voltage is high enough the quiescent current is somewhere in a couple uA range which seems the correct.

    Also noticed that the current is just about the same (~120uA) when powering on the device. If the voltage is high enough this period will be second or even less. If the voltage is low (doesn't have to be below UVLO threshold) the device would be stuck in the 120uA draw mode until I increase the voltage to 3V. This makes me think that the chip might be in PUP2 state instead of POFF when UVLO is below the threshold. 

    I don't quite get the UVLO, I thought it was to protect the battery from over discharging the but it looks like this is not the case here. For now, I updated the design with the voltage detector that drives the ON pin and should shut down the whole chip in the under voltage case.