Is a 3.3 V level converter required for interfacing a 5V PIC

arunb · Joined: 08 Sep 2003 Posts: 492 Location: India

Hi,

I am trying to develop a project using a 18F4550 and a MMC (this is the first time i am using a MMC and the 18F4550). I need to interface them.

The MMC works on a 3.3 V system whereas the PIC works on 5 V system, do I need to convert the I/O levels of the MMC to 5 V or will the PIC detect the 3.3V level as a high.

Also is it OK to use a potential divider to power down the 5 V level to 3.3V or should I use a device like 74VHC08.

Please bear in mind the project is a datalogger for use in an industrial environment.

Your comments would be most helpful....

thank you
arunb

newguy · Joined: 24 Jun 2004 Posts: 1909

I think you answered your own question. Since this is for a commercial unit meant for industrial applications, why not use a dedicated level converter IC/ICs?

asmallri · Joined: 12 Aug 2004 Posts: 1635 Location: Perth, Australia

libor · Joined: 14 Dec 2004 Posts: 288 Location: Hungary

I found this Microchip booklet very useful:
3V Tips 'n Tricks

ckielstra · Joined: 18 Mar 2004 Posts: 3680 Location: The Netherlands

Why not use an PIC18LF4550 at 3.3V?
Makes things a lot easier and you will save some power too (cheaper power supply). Only limitation is a maximum clock frequency of 18.7MHz at 3.3V.

newguy · Joined: 24 Jun 2004 Posts: 1909

If you're set on using a 5V PIC, I can recommend the MAX3371 level shifter IC.

kender · Joined: 09 Aug 2004 Posts: 768 Location: Silicon Valley

Here's a selection tool for TI level shifting chips http://focus.ti.com/logic/docs/translationselection.tsp?templateId=5985&navigationId=11413
_________________
Read the label, before opening a can of worms.

libor · Joined: 14 Dec 2004 Posts: 288 Location: Hungary

As a makeshift solution you can also improve the voltage levels of your normal PIC (i.e. if you don't have the LF version) to lower threshold levels by lowering the supply voltage to 4.2 Volts (which still fits the specification according to the datasheet), this will give you much better aquisition margin to properly interpret the 3.3 V signals usually without any level shifting in the upward direction.

arunb · Joined: 08 Sep 2003 Posts: 492 Location: India

Thank you all for the help.

I think the level shifter idea is better.

arunb

foodwatch · Joined: 18 Apr 2006 Posts: 66

A simple level shift method is to take the 5v data out, go through the cathode of a diode, tie the anode to 3v through a 10k resistor and bring the anode to the desired 3v input of the connected line. When the source goes high to 5v it is blocked by the diode and the 10k resistor acts as a 3v pull up. When the source goes low, it pulls the anode of the diode low and provides a safe low for your 3v source. This is cheaper and easier that a level shifter unless you have many i/o pins to deal with.

rnielsen · Joined: 23 Sep 2003 Posts: 852 Location: Utah

This idea may work when you are going from the output of a 5v device to the input of a 3.3v device but, without a shifter, how do you get the 3.2v coming out of your 3.3v device to reach the 4.0v required by the input of the 5v device?

Ronald

foodwatch · Joined: 18 Apr 2006 Posts: 66

This can be done as follows: Take the 3v output line and connect it thru a 10k resistor to the base of an NPN transistor (ie 2n2222). Tie the emitter to ground. Take the collector through another 10k resistor to 5v supply. Take the output to your 5v device from the junction of the the 10k and the collector. You will need to invert your logic... ie a 1 at 3v = 0 at 5v, or if you like lotsa parts add another transistor configured the same way to get the logic back to 1 in = 1 out.