SN65HVD1473: Can be used this IC for RS-422 purpose too?

Hi Oscar,

First off I wanted to make sure you are wanting to use the full-duplex, because the schematic you are showing is a half-duplex?

Also I was wondering what your reasoning was for the FET on the TXD line? As I see it now you will disable the drive whenever you send a high on TXD line, therefore your output state on Y and Z is undetermined so would need some sort of failsafe biasing. And using the device in that way is outside the 485/422 spec for the part.

As for your question the SN65HVD1473 is RS-422 compatible.

Regards,

Daniel Noel 

What is the purpose of R3?

When the receiver is disabled, you need a pull-up resistor on RXD to get a defined level.

If you have failsafe biasing on the bus, or if all receivers have the idle fail-safe feature, then this scheme might work (for half duplex). However, by never actively driving high bits, you lose of the noise margin of the differential signalling of RS-422.

There are other circuits that use a monostable to enable the transmitter for a specified time after the start bit. (But this requires knowing the baud rate.)

The SN65HVD1473 is not full duplex; you'd need separate TX/RX lines for that. Oops, it actually is.

Hi Clemens,

This schematic was copied from the Sparkfun website because they sell a rs-485 shield for Raspberry Pi. You can see in this link the complete schematic: www.sparkfun.com/.../13706

My target was to have a reference design to use in the RS-485.

I think the R3 resistor is not necessary.

This is going to be, more or less, the schematic I want to design:

Best Regards.

Hi Clemens,

I made an error. I attached you the new one with everyting fixed:

Best regards.

Moving the switches to the bottom makes it hard to understand.

Anyway, I see no obvious errors.

Hi Oscar,

1. You do not necessarily need to deactivate the RX line to transmit data in half-duplex mode; your mcu will just be able to hear what it is transmitting, which might be helpful for debugging. As I see it now for JP1 for HD mode you are leaving RE and DE open and that will cause an undefined state on the bus A/B per the function table 1 and table 2, although R will be pulled high because you have a pull up on the line.

Also I believe your FET is only being tie to R_RE and R_DE when JP1.1/4 are on, and it is not actually controlling RE with your current schematic. 

2. For your Full-Duplex implementation you have the RE and DE states flipped with the current jumper configuration you have RE high and DE low. RE is active low so you need to tie RE to ground and DE is active high you need to tie DE to VCC unless you are using  a control lines like this:

So you might want to re-look at your JP1 settings and/or function table

The way the sparkfun is using the FET to control RE/DE is outside of the normal operation of RS-485/422. Some common ways of configuring RS-485/422 (besides the one above) are like this:

You might want to look into some of these. For you application I would think the Slave Enable Control config might be the most straight forward one to implement.

Regards,

Daniel Noel