Forum Rule: Always post complete source code & details to reproduce any issue!
Results 1 to 8 of 8

Thread: choosing FET for fuel gauge

  1. #1

    choosing FET for fuel gauge


    I need to monitor the battery level on one of my project but I would like to move away from the simple voltage divider solution since it unnecessarily drains some current. I only need to monitor the battery level every couple of minutes so a FET sounded like the ideal solution to quickly switch the fuel gauge on and off. The gate will be connected to a digital output and the source to an analog input of the Teensy.

    I found this very nice blog here and here is the interesting part:

    "But there’s a catch: we need to be able to turn the N-channel MOSFET on and off, which means we need to be able to apply a voltage to its gate which is a few volts above the drain pin (the bottom one, attached to AIO). With a resistive divider of 10 + 10 kΩ on a 6V PWR line, that voltage will immediately rise to 3V, and there’s no way the DIO pin can keep the MOSFET on (it can only go up to logic “1”, i.e. 3.3V). The solution is to use a different divider ratio: say 50 + 10 kΩ. Then, a 6V PWR level leads to a 1V level on the AIO pin, i.e. on the drain of the MOSFET. With DIO set to “1”, that means the MOSFETs gate will be 2.3V above the drain – enough to keep it turned on."

    I am a bit confused, I though the drain pin of an n-channel MOSFET was connected to PWR and the source pin to GND... Is it a typo ? did he mean "source pin, the bottom one attached to AIO" ? I contacted him and I am waiting for his reply.

    I am having really a hard time choosing the right FET. I will use the 50k+10k divider as advised and I expect my battery voltage to go from 4.2v to 2.5v (absolute max and min). When I turn it on, the gate will see the 3.3v logic from the digital output, and the top pin sees at most 0.7V when the battery is fully charged (divider ratio 6). Does that mean that I need a FET with Vgs < 2.6V ? There aren't that many... most of them are rated 5V to 60V ! I must have gotten that wrong.
    My understanding is that Vgs is the most critical characteristic when it comes to choosing a FET. I also want a very small Rds(ON) and ideally a small power dissipation if I want to go for a SMD package. Am I right ?

    Thanks !

  2. #2
    Senior Member
    Join Date
    May 2017
    Consider using a NPN transistor in that circuit. Choose the resistor divider such that the voltage across the lower resistor is less than ( 3.3 - 0.6) volts. Then you will not need to find a FET with a very low turn on voltage.

  3. #3
    Interesting thought. Thanks

  4. #4
    I had a look at the NPN transistors but I kind of have the same problem: choosing the right component.

    Click image for larger version. 

Name:	Screenshot 2019-03-28 at 16.15.43.png 
Views:	15 
Size:	30.4 KB 
ID:	16276

    I have an extremely low collector current (that's the point I don't want to drain the battery) from 0.21mA (battery fully charged @ 4.2V) to 0.07mA (flat battery, 2.5V) and I selected a hFE=10. I used 3 different online calculators to find Rb but I ended up with different results...




    The NPN solution is interesting but it does not solve the problem (the main problem being I don't have a clue about electronics...). I have been reading a lot about transistors lately but what I found was either too simplistic or too detailed and nothing that I can relate to my own application. Any help ?


    I may have found a work around: digital transistors that incorporate a couple of resistors, the data sheet is easier for me to understand. This one seems ok to me but I'd like confirmation pls.
    Last edited by MattH; 03-28-2019 at 05:41 PM.

  5. #5
    Senior Member
    Join Date
    Feb 2013
    Not sure if this will be any help, but this is what I have used in a project to monitor a lipo battery, and use this library to do all the calculations on battery life etc
    I use an FDV302 p-ch mosfet along with a BC846 NPN
    I am no expert and maybe this is wrong, but seems to work very well and does what I want.
    For the battery sense library I also changed it to use the ADC library to improve performance and use 1.2V reference.
    R16 is actually 25.5k 0.1% resistor and R17 is 10.2k 0.1%. The image has the wrong values.

    Click image for larger version. 

Name:	battery monitor.png 
Views:	21 
Size:	11.2 KB 
ID:	16277

  6. #6
    Senior Member
    Join Date
    May 2017
    turtle9er's circuit looks good. For the other, try hfe of 100 in your calculators and maybe a 5k in the emitter side. In the single transistor approach you should keep in mind that the base current will also produce a voltage across the emitter resistor, so you will get a small voltage reading when reading a zero voltage. So the result will be like a linear equation of the slope intercept kind where zero x has a slight positive y intercept.

  7. #7
    Turtle9er, your schematic looks very neat but maybe a bit overkill for a fuel gauge. The presence of C7 for instance makes me think this was designed for a very accurate high sampling rate application. Definitely good practice but not required here. But at least it works, so thanks. From the reference you gave me however, I was able to find a few other good candidates including the FDV303N, thanks for pointing in the right direction. I'll buy a few different ones and test them directly.

    rcarr you made a very good point. I'll do a calibration round to get the exact slope (resistor tolerance) and intercept (contaminating base current). What about the digital transistor I found ? I would have no external resistors to worry about. The specs looks fine to me but I'm not an expert...

    Cheers guys
    Last edited by MattH; 03-29-2019 at 09:37 AM.

  8. #8
    Senior Member
    Join Date
    May 2017
    About the transistors in the link, in my opinion the base resistor(10k) is too small for your circuit. Going back to your circuit with a 10k in the emitter lead and considering that the base current also produces a voltage across this resistor, I would try a 100k resistor for Rb. This would produce an error voltage 1/10th of the voltage under test. The circuit could be built on a protoboard and values adjusted if needed. A 2n3904 NPN should work fine.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts