It is a correct method for translating voltage levels from 5V to 3.3V and the converse? / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

It is a correct method for translating voltage levels from 5V to 3.3V and the converse?

1 2 next

Move Thread

LAN_403

TheMerovingian

Thu Dec 03 2009, 01:24PM

Registered Member #14 Joined: Thu Feb 02 2006, 01:04PM
Location: Prato/italy
Posts: 383

For a wind datalogger project i am interfacing an SD card via SPI and a PIC18f2550 microcontroller (working at 5V TTL/CMOS).
I need to understand if this method is correct (in simulator it works):

Level conversion from 5V to 3.3V , using a 74LS125 quad buffer powered with 3.3V (hope it will work at this low voltage) with 5V tolerant inputs

Level conversion from 3.3V to 5V , using a 74HCT125 quad buffer powered with 5V , the 74LS125 is TTL compatible with a threshold of 2V

Is this correct?

Another proposition is to use a 74HC125 powered at 3.3V with inline resistors on the input (letting the internal diodes clamp to 3.3V) but i am afraid it will deteriorate the signal due to capacitance/inductance of the resistors

MinorityCarrier

Thu Dec 03 2009, 04:59PM

Registered Member #2123 Joined: Sat May 16 2009, 03:10AM
Location: Bend, Oregon
Posts: 312

LS-TTL should be run at 4.5-5.0V to guarantee internal stability. You can use a simple transistor level translator to interface to 3.3V.

HC and HCT devices can run at lower voltages, and your approach for this may work, but you may have a problem with noise. I'd recommend you Google level tranlation design for info on what works with HC logic.

Z28Fistergod

Thu Dec 03 2009, 06:14PM

Registered Member #2040 Joined: Fri Mar 20 2009, 10:13PM
Location: Fairfax VA
Posts: 180

You mean nobody makes a purpose built chip?

TI has capacitive isolation chips that uses two independent supplies, the ISO724*. They are compatible with 3.3V and 5V, and the two supplies can be run at different levels to achieve the translation you want. They have four channels per chip, and various configurations from all 4 channels going in one direction to an even split of 2/2. The only catch is that they are SMD, if that matters to you.

Bjørn

Thu Dec 03 2009, 06:21PM

Registered Member #27 Joined: Fri Feb 03 2006, 02:20AM
Location: Hyperborea
Posts: 2058

The best method is usually to run the microcontroller at 3.3 V. Series resitors always work fine with frequencies as low as this. If you are very worried about noise, buffer with 74HC14 instead.

WaveRider

Thu Dec 03 2009, 08:24PM

Registered Member #29 Joined: Fri Feb 03 2006, 09:00AM
Location: Hasselt, Belgium
Posts: 500

I always use pics at 3.3V with 3V logic (74LVXX series works well...and is fast as well!) 5V logic brings back memories of the computer stone age!!

TheMerovingian

Thu Dec 03 2009, 10:18PM

Registered Member #14 Joined: Thu Feb 02 2006, 01:04PM
Location: Prato/italy
Posts: 383

cannot run the microcontroller on 3.3V because i also need to interface the micro with a DS1307 RTC (although it can be run at 5V and work fine with 3.3V since the VIH is 2.2V) and with an LCD 16x4 display that wants to be driven at 5V.

For the dedicated chips i want to buy the components on Futurlec to avoid paying abnormous shipping charges (as on RS, mouser and digikey)

For the 3.3V->5V the 74HCT125 will work just fine (VIH = 2.0V , powered with 5.5V)

for the 74HC125 to do the 5V->3.3V conversion why we have problems with noise?

EDIT:
The LCD seems TTL compatible (VIH=2.0V) so when powered by 5V should work fine with a 3.3V powered microcontroller but i'm worried that the pic18f2550 would not work properly at 20MHz with 3.3V

Another possibility:

74HC4050 buffer For the 5V->3.3V conversion it can work at 3.3V and tolerate up to 15V on inputs

the 3.3V->5V conversion using 74HCT125

Found interesting article:

Bjørn

Thu Dec 03 2009, 10:47PM

Registered Member #27 Joined: Fri Feb 03 2006, 02:20AM
Location: Hyperborea
Posts: 2058

I always use the LCD displays with 3.3 V microcontrollers, use a 10 kOhm pull-up to 5 V on the enable pin (EN) to be sure of correct operation.

PIC18F2550 need a minimum of 4.2 V, PIC18LF2550 works down to 2 V. According to the formula and graph in the datasheet PIC18LF2550 should be able to do 25 MHz at 3.3 V.

TheMerovingian

Thu Dec 03 2009, 10:52PM

Registered Member #14 Joined: Thu Feb 02 2006, 01:04PM
Location: Prato/italy
Posts: 383

I have only access to pic18f2550 or pic18f2580 or pic18f2680 microcontrollers with enough ram and program space for FAT16 handling

MinorityCarrier

Thu Dec 03 2009, 11:10PM

Registered Member #2123 Joined: Sat May 16 2009, 03:10AM
Location: Bend, Oregon
Posts: 312

Switching noise transients on the DC return can get superimposed on the logic outputs. For low voltage logic devices the transients can become significant, and can move the logic signal output from a True High or True Low State, to a Metastable State. Your idea of running a device at 3.3V from a 5V rail with other 5V logic, may put that 3.3V device output at risk for metastable conditions.

Those TI level translator chips look like what you might want to use.

WaveRider

Fri Dec 04 2009, 08:47AM

Registered Member #29 Joined: Fri Feb 03 2006, 09:00AM
Location: Hasselt, Belgium
Posts: 500

Open collector with pull-up resistors is often a convenient way to do level shifting . Just be aware of the increase in rise-time.

Going from 5 to 3.3V can be done by using a 100 ohm resistor in series with the 5V output and a reversed-biased low-barrier potential schottky diode tied between the 3.3V input and the 3.3V supply. Another reversed biased schottky diode can be tied between the input and ground to clip off undershoots...

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob