# PMOS High Side Switch

I have been working on a project and one of the ideas was to control power supply of LCD 16x2 module and IR sensor via I/O pin of microcontroller, basically to have them on/off when i need them to be. I know that that can be done using MOSFET and BJT transistors (aka. High/Low Side Switches). I am using pic18f4550 for this project.

In worst case those two modules consume around 30mA on 5V. I have done the math and made circuit that provides me output of 33.3mA on 5V.. Here is what i have done so far:

simulate this circuit – Schematic created using CircuitLab

I have googled for high/low side switches but so far i haven't been satisfied with answers I found. Is this going to work or do i need something else to make this work? Thanks in advance.

• Basic idea is great. Assuming the load is designed to work on 5V, then R3 is not needed. R1 can be much larger. Like 470k. M1 should be a type of PMOS that has Rds(on) specified when Vgs is 5V or less. The IRF9530 may work, but would not be my first choice. It may not reliably turn on fully with only 5V. – mkeith Sep 9 '18 at 20:05
• Low-voltage load switches are a solved problem. For example have a look at: ti.com/power-management/power-switches/load-switches/… just a single component is needed. – filo Sep 9 '18 at 20:17
• @filo If a suitable PMOS is chosen, this circuit may conly cost US\$0.04. It is simple and reliable. You can even do it in a single SC70 package if you get a PMOS/NMOS dual transistor (using NMOS instead of NPN). There is nothing wrong with doing it this way. Sometimes the extra features of a load switch make it worthwhile. Sometimes not. – mkeith Sep 9 '18 at 20:38
• So in other words this will do the work, it'l be good enough to drive my modules? – A.F. Sep 9 '18 at 20:49
• R3 troubles me. Unless you want Vout to be zero when 33mA flows through R3, then I think you made a mistake there. And I would change M1 as Matt Young suggested. Then it is good enough. There is no point in using 4.7k for R1, but it will work fine. – mkeith Sep 9 '18 at 20:57

In principle, your circuit is fine. I do not like that the IRF9530's $R_{ds on}$ is only spec'd at 10V. You would be better served to choose something like a BSS84.
This is a higher $R_{ds on}$ part, but the curve is much better suited for operating at 5V.
• AO3401 is a nice jellybean part if you need a lower Rds(on). 60m$\Omega$ max with 4.5V drive. – Spehro Pefhany Sep 9 '18 at 20:31