1
\$\begingroup\$

I have a project with LEDs which are switched on/off with MOSFETs. But there is this "off mode" slowly reducing voltage period which takes 12ms to go truly to 0 V. I have this Vishay Intertech SI4477DY-T1-GE3 P-Channel MOSFET. Is it possible to reduce this off period to go down faster? Is this MOSFET too slow to do that? This is my shematic, and the processor is an Arduino Nano.graph showing voltage decay period

schematic

simulate this circuit – Schematic created using CircuitLab

\$\endgroup\$
8
\$\begingroup\$

Ignoring the poor attempt at a circuit, you are actually switching off current when you turn your MOSFET off but, because the LED will go high impedance you are left with a circuit node with parasitic capacitance that is charged to nearly the same voltage at which the LED operated at.

Without any "load" resistance this voltage will dwindle away quite slowly. The parasitic capacitance will be several hundred pF and is due to the MOSFET drain-source capacitance and the LED parasitic capacitance.

\$\endgroup\$
  • \$\begingroup\$ So where should I put load resistor? \$\endgroup\$ – user45189 May 9 '18 at 9:57
  • 1
    \$\begingroup\$ Or make the gpio pin sink the current it previously sourced. By pulling the pin LOW instead of making it high impedance. \$\endgroup\$ – ratchet freak May 9 '18 at 10:03
  • 2
    \$\begingroup\$ @user45189 Before I recommend anything, you need to have a workable diagram. I criticized your first diagram but the redrawn diagram also has problems. Your circuit (as now shown) will barely work - are you intending that the LED current is somewhat less than 1 mA when the MOSFET you are using can deliver over 20 amps? I still have doubts that what you show is not actually what you have. Also, having R1 in the source doesn't make much sense either. So, clarification is needed of the exact circuit. \$\endgroup\$ – Andy aka May 9 '18 at 10:20
  • 1
    \$\begingroup\$ If we put both resistor and led in drain. And if we put a strong pull down resistor between ground n gate, will it improve switching speed ? But I think arduino's digital pin is already a low source resistance to ground. \$\endgroup\$ – Meenie Leis May 9 '18 at 14:12
-1
\$\begingroup\$

I think this is because of charge left on the gate of your mosfet. The arduino perhaps doesn't pull off the charge in the off position. Try adding a resistor from the gate to ground. (maybe 1 k ohm.)

\$\endgroup\$
  • 2
    \$\begingroup\$ Arduino does normally pull to ground when driving a pin low, and 1k is practically infinite resistance compared to an MCU pin actively pulling low. \$\endgroup\$ – Dmitry Grigoryev May 9 '18 at 15:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.