Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Mosfet Load switch + weird load

  1. Jun 17, 2013 #1
    I have a N-ch Mosfet load switch that connects to a WiFi module on a circuit board.
    When the switch is turned OFF, I still see a voltage of 1.4V at the load.
    Ckt attached.
    I decreased the bleed resistor from 1k to 100 ohms. The off voltage decreased to 1.1v.

    This doesn't happen with other loads like cellular modules, only with a particular WiFi module (WPER-172GN).
    I don't have other wifi modules at the moment.

    The N-Ch Mosfet is DMN1019
    The WiFi module is WPER-172GN. The IC is Ralink RT5390. Cannot find Datasheet online!
    The cellular module is EM820W and others. None of them gave me problems.

    I can't figure out what kind of load would make the circuit behave like this.
     

    Attached Files:

  2. jcsd
  3. Jun 17, 2013 #2
    What's your voltage across the mosfet when it's Off? Is it actually 3.3? Are you sure?

    I will preface this by saying I'm not an engineer I'm only a student, but as I understand it it is inadvisable to drive ground connected loads with an n channel.
     
  4. Jun 17, 2013 #3
    Across the Mosfet, I get (3.3 - the voltage at the wifi module). If the voltage at the load is 1.4v (when switch is off), the voltage across Mosfet is 1.9v

    Why is it inadvisable to drive ground connected loads with N ch?
    The ckt works for every load except the wifi module.
     
  5. Jun 17, 2013 #4

    rbelli1

    User Avatar
    Gold Member

    Check the CE voltage of the 2222. You may find that you are not driving it to saturation. If this is the case then you could be using the FET in its linear region depending on the load current. This is not what you want. Try replacing the 2222 with a FET. You will need a very small gate resistance or use a logic level FET such as BSS138 where you shouldn't need a gate resistor at all. Reducing the 2222 base resistor to bring it to saturation will work as well.

    Do you have any signals connected to the 3.3V module back to the 5V section? You have to be careful not to drive any signal connected to the 3.3V module outside of power supply pins. When you have the load module off it has all of it circuitry at 3.3V. If any of the pins connected to the active parts are driven to anything other than 3.3V then you are in danger of biasing the protection diodes or parasitic BJT structures. That is generally not a good idea (you can lose the magic smoke!).

    BoB
     
  6. Jun 18, 2013 #5
    Sorry i didn't get back to you. I was thinking along the same lines as rbelli above me though, that somehow you're not saturating the fet. Maybe the chip floats a small voltage for some reason? Shouldn't but you never know...
     
  7. Jun 18, 2013 #6
    I checked the CE voltage, it's about 50-100mV. I even grounded the Mosfet gate. But I still see 1.4V at the load.
    I also connected Mosfet gate directly to LAB power supply and manually varied the voltage from 0 to 5V. But no change.

    I double checked this too. No signals connected. The mini PCIe connector has provision for 5V, but the corresponding pins on the module are NC. Just to eliminate any possibility of this 5V getting into module, I lifted the connector pins off the board. No use though.
     
    Last edited: Jun 18, 2013
  8. Jun 18, 2013 #7

    rbelli1

    User Avatar
    Gold Member

    Maybe you fried the FET? They can be very sensitive to static electricity. Try switching back to a module that worked in the past.

    BoB

    PS: Regarding off state voltage levels. They will be at 0 not 3.3 like I said (too late at night!). Driving them outside of the rails has the same consequences in either case.
     
  9. Jun 20, 2013 #8
    The FET is working fine. No damage.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Mosfet Load switch + weird load
  1. Weird Science - VIDEOS (Replies: 1)

  2. Weird 555 pathology (Replies: 6)

  3. Weird PCB stuff (Replies: 23)

Loading...