Electronic – Why does the 2N7000 show a gate current of more than 4mA

leakage-currentmosfet

I have the following circuit:

schematic

simulate this circuit – Schematic created using CircuitLab

The D1 is in fact a character LCD backlight (with 4.2V input voltage and a maximum current of about 300mA). R2 is a peculiarity of my test setup, where I control the gate voltage using a potentiometer.

When measuring the current into the gate, I find that my multimeter shows 4mA to 5mA. The current drawn by the LCD with R1(*) in place like that is about 90 mA.

I wonder why there is such a large gate current. The datasheet of the 2N7000 suggests (but I might well be reading it wrong) that there should be a maximum of 10nA of gate current.

I tried the setup with two different voltage sources: A lab supply current-limited at 500mA and, after making sure that the setup doesn’t draw more than 100mA in total, 5V from my computers USB port. Both have shown the same behaviour.

(*): Don’t nail me on the value of R1. R1 is constructed of four 10Ω resistors in parallel, each with a tolerance of 10%. I just now measured 4Ω (8Ω before changing batteries on the multimeter) for that construct.

Best Answer

Gate of a MOSFET cannot drain 5mA current. If you said 5pA I could understand, but 5mA is practically and theoretically impossible.

Either;

  • There is a leakage somewhere across the gate and source pins of the MOSFET.
  • Your multimeter has a bias.
  • The MOSFET is broken.
  • There is a wrongly done connection around the gate.

Suggestions:

  • Connect a gate resistor of 10\$\Omega\$ or 100\$\Omega\$. Send 1kHz, 0-5V pulses to the gate. If you can, amplify the voltage on the resistor by using an opamp difference amplifier. Observe the voltage drop on the resistor by using an oscilloscope (not with that multimeter).
  • If you constructed your circuit on a breadboard, reconstruct it in a different region.

The circuit is indeed constructed on a breadboard. Which effects would show this behaviour?

Dirt inside the breadboard structure my cause leakage current.

I’m still open for blaming the multimeter though

You should. That's probably the only culprit.

The FET seems to do what it is supposed to. Is there a failure mode which introduces such a large leak current but makes it still work correctly?

The system will still work if there is a ghost resistance between gate and source pins which will draw 5mA current. The system will still work even if that resister were very small in value as long as the power supply can supply enough current.

Related Topic