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.

BQ25015: BQ25015 EN input handling, abs max rating

Part Number: BQ25015
Other Parts Discussed in Thread: CC85XXDK-HEADSET-RD

Hi,

I'm using the BQ25015 in a simple charge/supply application, where it is powering a small MCU (3.3V nominal) and charging a 250mAh LiPo pouch cell.

I'm wondering about how to handle the EN input. Obviously it must be set HIGH for the DC/DC to start up, I've looked at the CC85XXDK-HEADSET-RD and here EN is driven high both from VBAT but also from VBUS if connected. The reference design has a small circuit that inhibits EN in precharge (through a 74AHC1G32, STAT1 ORd with STAT2 output). The OR gate is supplied from VBUS, so EN would be pulled high to +5V (or thereabouts).

However, the absolute maximum input voltage for EN is Vout + 0.3V. If I'm not mistaken, Vout is the charger output voltage, which is 4.2V at charge termination and can be much lower depending on the charge state of the battery. So, would not connecting EN to VBUS violate this constraint?

Please help shedding some light on this. What am I missing?

  • Hi,

    We are currently testing this issue in the lab. We think that the VCC from the OR gate may be VBAT which would not violate any abs max conditions but we will get back to you in a few days to confirm.

    Thanks for your patience,

    Jackson

  • Hi Jackson,

    thanks for the reply. However, if the battery is deep discharged and the charger circuit needs to do a precharge cycle, there might not be enough voltage to supply the OR gate. Also, if you supply the OR gate from VBAT, with the STAT1 and STAT2 outputs being pulled up to VBUS, abs max constraints of the gate inputs might become violated.

  • Hey Matthias,

    I've been digging into this today and I also share your same concerns. I have put in an order for an EVM for testing and reached out to the design team for their input here so as soon as that goes in, ill be able to get back to you with some more definitive answers. I would expect sometime early next week as the EVM won't get here for 3-5 days.

    As soon as I get results, I will let you know.

    Thanks again for your patience,

    Jackson

  • Hi Matthias,

    For your scenario a good workaround to avoid the issue with this OR gate is to monitor the stat pins with an MCU and pull down the EN pin as necessary. Tie EN to VOUT as well so that the EN pin remains within spec. Essentially, I would just follow what is on the actual BQ25015 EVM instead of the reference design.

    Thanks and let me know if there's any further info needed,

    Jackson

  • Hi Jackson,

    for some applications that might be a viable solution. It creates a bootstrapping problem for me though: I only have one MCU in the application and since the DC/DC has to be off in case of precharging, there is nothing powering the MCU and obviously it cannot monitor the STAT outputs in that case. I'd have to add an additional power path and a selection mechanism to have the MCU always available. As my application does not require that, I don't really want to put additional lines on the BOM and allocate precious PCB area.

    The EVM circuit doesn't seem to provide a solution either, because it only breaks out the EN input to a header and pulls it up to Vout.

    A simple solution could be using an OR gate with open drain output. But it seems that doesn't exist.

    Anyway: For me it's enough if you confirm that the circuit in the headset reference design is indeed wrong and should not be used.

  • Hi Jackson, would you confirm that the following circuit is a viable solution? The +BATT net is connected to BAT/OUT.

  • Matthias,

    I can confirm you should not use the circuit in the headset reference design. 

    The more recent circuit you sent looks good to me at first glance but I can't confirm for certain so I would do testing on your own to verify.

    Thanks,

    Jackson

  • Excellent. I think there's still a problem with my circuit if the +5V rail is floating, but it's enough to know that the documentation regarding abs max for EN is correct and must be obeyed. It'd be nice if the headset reference design could be corrected, though. The error has a good chance of releasing a lot of magic smoke.