N-Channel Mosfet

Discussion in 'General Electronics Chat' started by Hk606, Sep 8, 2015.

  1. Hk606

    Thread Starter New Member

    Oct 19, 2014
    17
    1
    I have this N-Channel mosfet I'm checking using a DVM. Mosfet is a FQP50N06 When I charge the gate with 9VDC [spec. is +- 25V] Then check from the drain to source the ohms reading is 5 ohms for only 3 milliseconds. then goes back up to 28meg. indicating the gate is opening. My question is: Shouldn't the gate stay closed until the gate is shorted to the drain or grounded? Internal Diode checks ok.
     
  2. MikeML

    AAC Fanatic!

    Oct 2, 2009
    5,450
    1,066
    Tie an external capacitor of ~1uF from gate to source, and repeat the test. The gate capacitance is quite small, so discharges fast due to leakage...
     
  3. OBW0549

    Well-Known Member

    Mar 2, 2015
    1,308
    884
    If your MOSFET is staying on literally for only 3 milliseconds after you let the gate float then I'd say you've got some significant gate-source leakage there, more than I've seen in any of my FETs. Several seconds to several minutes would be more typical. What happens when you take a resistance reading between gate and source?
     
  4. alfacliff

    Well-Known Member

    Dec 13, 2013
    2,449
    428
    I have been checking some mosfets using my fluke meter on diode test. charging up the gate to source to turn on the mosfet, then checking the source to drain. a good one will charge up the source to gate and stay turned on for quite a while, sometimes for days. the source to drain is turned on both polarities at the same time with a very low resistance. a simple short from gate to source will turn it off, even just touching both leads at the same time. ther is quite a bit of capacitance between source and gate, and a good low leakage mosfet will stay charged up quite a while.
     
    OBW0549 likes this.
  5. Papabravo

    Expert

    Feb 24, 2006
    10,137
    1,786
    Using a multimeter on a FET is a good way to damage it in my experience. We always made a test jig with a socket for this kind of testing.
     
  6. alfacliff

    Well-Known Member

    Dec 13, 2013
    2,449
    428
    an irf840 will not be damaged by a fluke dmm. only very low signal mosfets will have any damage from normal test equipment. an fqp50n06 is not as touchy as a 2n7000. the milisecond "stay on" is probably the gate junction capacitance charging up.
     
  7. GopherT

    AAC Fanatic!

    Nov 23, 2012
    6,004
    3,757
    Is it in-circuit or an I mounted component?
     
  8. OBW0549

    Well-Known Member

    Mar 2, 2015
    1,308
    884
    Assuming the user takes commonsense precautions against ESD, how does a multimeter cause FET damage?
     
  9. alfacliff

    Well-Known Member

    Dec 13, 2013
    2,449
    428
    maybe some of the real old ones that used 22.5 volt batteries in their ohmmeter.
     
  10. Papabravo

    Expert

    Feb 24, 2006
    10,137
    1,786
    A 9-volt battery can do more than enough damage. You pays yer money and takes yer chances.
     
  11. Hk606

    Thread Starter New Member

    Oct 19, 2014
    17
    1
    no it's not. I removed it from the board using a rework station as not to damage the mosfet.
     
  12. Hk606

    Thread Starter New Member

    Oct 19, 2014
    17
    1
    The test I did was hold the negative probe on the drain put positive probe on gate for a second then moved the positive probe to the source.
    When I put the positive probe on the source meter read 5 ohms them immediately went to high 2.8 meg. I miss stated when I first posted as saying 28 meg. anyway this test if I did it right shows the gate closing then immediately opening. Like major leakage! This is a fet on a board for instrument cluster light dimming in a car. [factory board]
     
  13. GopherT

    AAC Fanatic!

    Nov 23, 2012
    6,004
    3,757
    Was it working or malfunctioning?

    I had the same transistor model (logic level version) fail recently. Behaved as though it was switching very slowly-running much hotter than expected. Also had low megOhm resistance when no no gate voltage applied. Should be infinite (on my meter anyhow).

    I purchased those as part of a kit on eBay (don't know the true source of the component). Replaced with units from Digikey and no troubles since.
     
  14. alfacliff

    Well-Known Member

    Dec 13, 2013
    2,449
    428
    when doing the leakage test, short the gate and source, then test, any charge on the gate might turn on the fet. why try to turn on the mosfet between the drain and gate? isnt the bias usually between the gate and source? short the gate and cource, then measure the leakage from source to drain, remembering that there is a diode in there also. - on source, + on drain, should be open, then move + to gate for a little bit. then bck to drain, should read very low resistance till the gate discharges.
     
  15. OBW0549

    Well-Known Member

    Mar 2, 2015
    1,308
    884
    Since this is an N-channel FET, you've got your polarities reversed: both the drain and the gate should be positive with respect to the source.
     
    alfacliff likes this.
  16. Hk606

    Thread Starter New Member

    Oct 19, 2014
    17
    1
    I have no instrumentation lighting in the car. The feed is from the main light switch to this board for dimming. The board is quite simple,
    dial is a 0-10K smd pot, Op-amp 2904D Dip8, Mosfet FQP50H06 QFET 4 caps. and a LED
     
  17. Hk606

    Thread Starter New Member

    Oct 19, 2014
    17
    1
    Thanks I guess online info is wrong.
     
  18. Hk606

    Thread Starter New Member

    Oct 19, 2014
    17
    1
    You are correct. I retested results indicate the Fet is good. Thanks
     
    OBW0549 likes this.
  19. OBW0549

    Well-Known Member

    Mar 2, 2015
    1,308
    884
    Excellent!
     
Loading...