2
\$\begingroup\$

I am adding reverse polarity protection to my circuit. I can simply use a diode/schottky diode to do that; however, my main concern with a diode is its voltage drop. Therefore, I am considering the P-channel mosfet. I didn't add a zener diode with a resistor to protect its gate because my power will never exceed the voltage rating of the gate.

I want to simulate the voltage drop of YJL2301F in LTspice. I don't know how to work with directive to change the parameter of the FET to get the estimated voltage drop in LTspice, so I use Rds_on of the Fet. Below is the Rds_ON from the datasheet, 94mOhm at Vgs = -4.5V.

enter image description here

Below is the equivalent circuit of the FET when it's conducting, which is simply represented by the RdsOn. I use a current source at the load side, varying from 0.01A to 0.5A.

enter image description here

The graph in blue shows the voltage at the load side. The voltage drop is shown to be around 0.005V - 0.05 V. Is this a good way to estimate the voltage drop of the P-mosfet?

\$\endgroup\$
1
  • \$\begingroup\$ I would flag the current source as a load when doing this so it can never get into a condition where it generates power. Rightclick->Advanced->Parasitic Properties->"This is an active load" checkbox \$\endgroup\$ Commented May 19, 2023 at 4:37

2 Answers 2

Reset to default
3
\$\begingroup\$

It may be a good way to estimate the voltage drop of the MOSFET (or you could just use Ohm's law and a calculator and save yourself some time). Good under 25°C die temperature conditions only. The maximum Rds(on) might be 50% higher if the die is very hot. See figure 6.

You should also consider transient and (perhaps pathological) conditions such as brownout and charging whatever capacitors you might have on the other side of the MOSFET. See figure 7, for example. Typical 1A+ diodes are very forgiving and can withstand large surges in relation to their steady-state capability. They don't burn out or unsolder themselves if your input voltage is low and the load perhaps starts drawing more current than normal. They don't generally die suddenly if there is a transient on the input that exceeds 10V.

A MOSFET is a good way to go if those considerations are covered and voltage drop is a concern.

\$\endgroup\$
1
  • 1
    \$\begingroup\$ Note- this is a '2301' type of MOSFET and there are literally dozens (maybe scores) of manufacturers with very similar products and part numbers. I suspect many of the datasheets are copies of copies. \$\endgroup\$ Commented May 18, 2023 at 19:30
3
\$\begingroup\$

I'd be a tad concerned about contradictions in the data sheet.

For instance, on this typical \$I_D\$ vs \$V_{DS}\$ vs \$V_{GS}\$ graph, I see a typical value of about 0.094 Ω when the gate voltage is -4.5 volts: -

enter image description here

So, 1 volt ÷ 10.7 amps is 0.094 Ω and, remember this is a typical value and not a maximum value so, either the data sheet is lying or they got their graphs a bit screwed up. Either way, I wouldn't be using this device because I can't trust it.

\$\endgroup\$
3
  • \$\begingroup\$ Good catch ! what if we ignore that, is using a Rds_on to represent the Fet when it's conducting a good way to estimate the voltage drop? \$\endgroup\$ Commented May 18, 2023 at 18:35
  • 1
    \$\begingroup\$ @Alex I don't ignore things like that; from my analysis, the MOSFET looks suspicious and I'd never use it for anything. I always use the typical graph and apply a 1.5x weighting factor to get a good feel for RDS(on) but, if the numbers in the data sheet appear to contradict things then I don't use it. \$\endgroup\$ Commented May 18, 2023 at 18:39
  • \$\begingroup\$ I got you! Thanks ! \$\endgroup\$ Commented May 18, 2023 at 18:50

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.