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.

BQ24171: External source instead of Vref

Part Number: BQ24171

Hello,

IC: BQ24171RGYT

Battery: 3P2S pack of 18650s

Thermistor: the pack has a 10k NTC thermistor in it which I have connected to TS as described in the BQ24171RGYT datasheet.

Ext power: 12V wall plug to power the charger

Extra TS connection: TS is also connected to ADC of an MCU so MCU can read battery temperature and turn on a fan as needed.

Problem: when 12V charger cable not plugged in, Vref = 0V.  When Vref = 0V, TS circuit is off and MCU can't read thermistor value.

Proposed solution: replace Vref with an external 3.3V supply that can be controlled by MCU.

Question 1: does proposed solution sound reasonable?

Question 2: how stable does 3.3V supply need to be?  Is 3.3V from main system buck regulator reasonable or should I add a dedicated LDO?

Question 3: should I replace all Vref with the external 3.3V supply or only TS.  I only need it for TS.

Question 4: do you have any other solutions to consider?

  • Hi KXCO,

    As you can see from the datasheet block diagram, the TS function uses a set of comparators to compare V(TS) to reference voltages derived from VREF.  If you provide a separate VREF when the charger is charging, the TS trip points will no longer vary relative the VREF voltage and will be out of spec.  However, if you can only provide the voltage when no input power is available or provide a slightly lower voltage than 3.3V, using a diode from your MCU's voltage source, then in theory it should work.

    Regards,
    Jeff    

  • Hi Jeff,

    Thank you for the reply.  Yes I do see the set of internal comparators.  I guess the problem with disconnecting Vref and using external 3.3V supply (aka 3V3_ext) is if Vref and 3V3_ext aren't the exact same then there will be extra error.  Is that the only concern with my original proposed solution?

    Does this seem like a reasonable worst case?  Vref is 3.267V and 3V3_ext is 3.333V.  That is an error of 0.066V which I think corresponds to an error of about 3.5°C.  That is an acceptable amount of error but I would not want to add too much more error than that.

    I am hesitant to add diode due to:

    1 - risk of mistakes when having a separate but similar equation for MCU temperature readings vs charger temperature readings.

    2 - error created by variance in forward voltage drop of diode.

    If I disable 3V3_ext when Vref is enabled and vica versa I worry about:

    3 - If 3V3_ext is on and connected to Vref which is disabled, will current flow into output of Vref LDO?  Will that damage Vref LDO?  

    4 - If there is a brief time when both 3V3_ext and Vref are on, for example when user plugs in 12V, Vref turns on and MCU takes a brief time to turn off 3V3_ext, can that cause damage?

  • Hi KXCO,

    You are correct in your assessment and concerns.  The only additional comment I have is that it is not only the DC regulation VREF that will be different but also the transient VREF.  For example, if the charger AVCC that powers the VREF linear regulator has a line transient, VREF will see part of that line transient too.  If both the external resistors and thermistor plus the internal divider have the same voltage, there is less risk of an inaccurate "glitch" in TS reporting than when the 3.3V comes from an external source.

    If you are ok with the potential DC inaccuracy and transient glitches, then I do not see an issue with using an external 3.3V for TS pin components.

    Just to be clear, the charger's TS functionality does not function without valid AVCC.

    Regards,

    Jeff

  • Thank you for the help.  I think I'll add a 2nd thermistor on a separate circuit for the MCU so I can leave this circuit alone.