IRF510 won't turn off fully

Discussion in 'General Electronics Chat' started by aziz.s.lok, Mar 11, 2016.

  1. aziz.s.lok

    Thread Starter New Member

    Mar 11, 2016
    5
    0
    Hi, I just got a few mosfets the irf510's I tried to rig up a test circuit to see how things will work .. I have a 12V rgb led strip so I connected the 12v + to led stip the - of any colour to the drain of the mosfet and the source to ground with nothing connected to gate I expected the mosfet to be off but sadly it still conducts the leds are dimmed but they light up .. If I try to connect the gate and source(ground) to the a battery with 3.7v I get the leds to light up fully..
    I even tried shorting the gate to ground but no luck the LEDs still light up ...

    Is it something wrong I'm doing ?

    Aziz
     
  2. ScottWang

    Moderator

    Aug 23, 2012
    4,853
    767
    Attach your circuit.
    To make sure you connect the right pins, it seems you reverse the D,S pins.
     
  3. wayneh

    Expert

    Sep 9, 2010
    12,094
    3,033
    That's my guess too, that you are passing current thru the body diode from source to drain. The LEDs might brighten if you raise the gate voltage, since the voltage drop of the diode will then be bypassed.
     
  4. aziz.s.lok

    Thread Starter New Member

    Mar 11, 2016
    5
    0
    I verified the pins from the data sheet and also tried inverting the pins ;(
     
  5. ScottWang

    Moderator

    Aug 23, 2012
    4,853
    767
    You need to in series a 560 Ω resistor with led, it make Vds of mosfet getting into the lock status.
    If you don't have 560 Ω then using 560 Ω ~1K.
     
    aziz.s.lok likes this.
  6. GopherT

    AAC Fanatic!

    Nov 23, 2012
    5,997
    3,754

    Note that a mosfet's gate must be connected to a 10v or 0v to turn it on and off. You cannot leave it disconnected.

    Also, the pinout from left to right is 1 (Gate), 2 (Source), 3 (Drain).

    image.jpg
     
    aziz.s.lok likes this.
  7. aziz.s.lok

    Thread Starter New Member

    Mar 11, 2016
    5
    0
    @ScottWang
    Ok I'll try out the series resistor .. Thanx guys...
    @GopherT yeah I know.. I wanted to test the circuit so I just left it hanging earlier but later connected it to ground all well to turn off the mosfet..

    Hey also can you guys explain what is lock status?

    Regards
    Aziz
     
  8. wayneh

    Expert

    Sep 9, 2010
    12,094
    3,033
    So with either the gate and the source connected to ground, or with the gate and drain both connected to ground, you still got current thru the LED? Something's wrong. One configuration should work and the other not. Do all of your 510s behave this way?

    Also, are you sure your LED is a normal LED or is it meant for 12V operation? If it's a normal LED, I'm surprised it wasn't already destroyed by excess current.
     
  9. aziz.s.lok

    Thread Starter New Member

    Mar 11, 2016
    5
    0
    Yeah I did try two new ones.. Both dud the same thing..

    Yes the LEDs are 12v it's a rgb led strip;)

    Aziz
     
  10. wayneh

    Expert

    Sep 9, 2010
    12,094
    3,033
    Ah, OK, well that explains why the LED is OK.

    Can you post a photo of your circuit? Other than something wrong with your wiring, about the only thing I can think of is counterfeit parts. But let's take a look first. It could be something simple overlooked. Very easy to do.
     
  11. ScottWang

    Moderator

    Aug 23, 2012
    4,853
    767
    That is to describe the Vds get into the saturation state too deep, and the mosfet still not destroyed.


    [​IMG]


    [​IMG]

    ScottWang's blog
     
  12. wayneh

    Expert

    Sep 9, 2010
    12,094
    3,033
    I don't know what that means. Ground the gate, it turns off. Raise the gate to +12V, it turns on. Are you suggesting there could be some other result?
     
  13. ScottWang

    Moderator

    Aug 23, 2012
    4,853
    767
    Because the TS didn't mention that the led is a 12v rgb led strip, I think if he just connected the led(2V, 3V) to the 12V and gnd, and the power didn't destroy the led, that is quite strange, and when he can't used the Vg to disable the mosfet, so I assumed that it could be some reasons to caused the Vds get into the deep saturation as a locked status.
     
  14. BobTPH

    Active Member

    Jun 5, 2013
    782
    114
    Re-read the original post.

    Bob
     
    ScottWang likes this.
  15. aziz.s.lok

    Thread Starter New Member

    Mar 11, 2016
    5
    0
    Guys I tried the resistor before the drain... I tried 10k between gate and ground I tried shorting ground directly... But this mosfet just won't turn off

    As an experiment I connected a 12v relay instead of the LEDs.. Well when I first connect power the relay is Off which is a good sign.. Giving a 5v pulse to the gate turns on the mosfet and the relay but disconnecting or grounding the gate doesn't do anything .. It stays On for some time n then clicks back Off!!

    You guys think this mosfet is not suitable for 12v application ?
     
  16. ScottWang

    Moderator

    Aug 23, 2012
    4,853
    767
    The IRF510 is designed for higher voltage Vds max = 100V and low current Ids max=4A, Rds(on) max = 0.6 Ω.
    But in most normal testing, it still should be works as 12V power supply.

    Make sure you connected the pins are right.
    Which circuit did tested?
    Did you test the first circuit that I attached on #11?
     
  17. ian field

    Distinguished Member

    Oct 27, 2012
    4,413
    782
    A static charge on the capacitance of an unconnected gate can bias it anywhere between fully on and fully off.

    If you touch a loose gate, it will pick up mains hum just like the input terminals of an audio amplifier - what charge you leave on the gate capacitance, depends where the hum waveform was at when you let go.

    You could shunt the gate to source with about 1k, unless you need the high input impedance. Or use a shunt transistor and pull up resistor.

    If the LEDs still light with the gate shunted - you may have the wrong polarity MOSFET or supply. The MOSFET has a "body diode" intrinsic to the die structure, normally its reverse biased so you don't know its there.

    Getting back to static charges - they can blow the gate so it'll never turn off again.
     
    Last edited: Mar 13, 2016
Loading...