0
\$\begingroup\$

i've built the following circuit to short the PS_ON to GND and turn on a PSU.

enter image description here

The problem is that, despite the pull-low resistor or the state of the switch, the MOSFET will always be on.

With this circuit i've also tried to turn on an LED and it was always on, just a little 'less bright when the MOSFET should have been turned off and I thought of a broken part, but the same happens with other IRF540 i have...

Any idea? I can't understand what I'm missing or wrong

Improve this question
\$\endgroup\$
4
  • \$\begingroup\$ 1 Meg is a very weak pull down, how did you choose that value? Could anything be being coupled into it? \$\endgroup\$ – Colin Apr 18 '18 at 14:24
  • \$\begingroup\$ PS_OK is an output from the ATX PSU. The signal you want to drive low to turn it on is PS_ON# \$\endgroup\$ – Finbarr Apr 18 '18 at 14:24
  • \$\begingroup\$ @Colin__s i'm quite noob and i've designed this circuit folliwing some tutorials to create a switch with mosfets i've found online, what do you think it's a correct value? \$\endgroup\$ – Sempri92 Apr 18 '18 at 14:30
  • \$\begingroup\$ @Finbarr yeah...now it's correct \$\endgroup\$ – Sempri92 Apr 18 '18 at 14:31
0
\$\begingroup\$

I've never tried a mosfet to do this, but I've used a 2n3904, no emitter or collector resistor, with just a 1.1K resistor on the base between the transistor and switch.

but in those, I used a latching flip-flop before so I could use a momentary switch.

And if you are hooking up a standard toggle switch you only have to ground the atx pwrOn lead (green wire usually)

Improve this answer
\$\endgroup\$
1
  • \$\begingroup\$ i've tried with a 2n3904 and it works well, but still i can't understand why the mosfet isn't working as it should \$\endgroup\$ – Sempri92 Apr 20 '18 at 10:46
1
\$\begingroup\$

Is it possible you have misinterpreted the pinning for your IRF540. The behavior you are seeing is similar to what I would expect if you has the source and drain connections interchanged. In that case the body diode forward biases and your load looks always on.

Improve this answer
\$\endgroup\$
3
  • \$\begingroup\$ double checked the wirings: ps_on coming from the psu and going to drain, source going to gnd....is a classic TO220 package, pinout checked on datasheet...btw i've tried to invert the wirings with the same result \$\endgroup\$ – Sempri92 Apr 18 '18 at 14:54
  • \$\begingroup\$ an ifr540 is not really suitable for this application, because it has too much parasitic resistances due to the substrate type. An enhancement mode mosfet like a tn0702n3-g would work. \$\endgroup\$ – drtechno Apr 22 '18 at 10:54
  • \$\begingroup\$ thanks you @drtechno i've solved this using an optocoupler...btw i was about to buy IRL540, what do you think about it? can it suite the needs for a similar application? \$\endgroup\$ – Sempri92 Apr 30 '18 at 14:16

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.