Searching \ for 'protecting the RX and TX serial port lines from ex' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/io/serials.htm?key=serial
Search entire site for: 'protecting the RX and TX serial port lines from ex'.

Truncated match.
PICList Thread
'protecting the RX and TX serial port lines from ex'
1998\10\26@194556 by Ry Lato

picon face
I am using a PIC16C77 are part of a consumer product. The RX and TX
lines of
the PIC are routed external from this product for serial communications
with
another PIC based device. The SCI port is configured for asynchronous
operation at either 2400 or 4800 baud (speed decision still needs to be
made). The TX, RX and a GND line are coupled to the other device via a
3-conductor headphone cord with a 1/8" headphone plug on each end. Since
this
product is for the automotive environment which will be installed by the
consumer, one of the 1/8" headphone jacks may come in contact with
12VDC.
This would apply 12V directly to the RX or TX line causing immediate
destruction to the I/O of the PIC. Although a potentially rare
occurance,
this type of scenario did happen in one of our prototypes. The headphone
wire
was rested on top of the car battery. Possible solutions I have
considered:

I can use 2 10K series resistors between the RX and TX lines to the
outside world. This would limit current to the port in the event of
directly
applied 12VDC. This limited current should be safely dissipated by the
input
diodes on each pin of the PIC. But the 10K series resistance coupled
with the 10-15ft of cabling will create an RC network that will severely
attenuate/degrade the serial port signal.

I could add a 5.1v zener diode in paralell with the port pin and GND in
series with a 50-ohm resistor. But this is ineffective if the PIC's VDD
rail
drops below 5V for any reason, such as a brownout or if the unit is
powered down.

I could add a .6V silicon diode between the PIC pin and VDD, with a
50-ohm series resitance. But that would dump the external DC voltage
directly into the 5V VDD rail
through the diode.

A polyfuse, which is an auto-reset current limited would not work, since
they don't have values that go into the low mA range, nor are fast
enough to limit current before damage occurs.

I have considered using an IC solution for this. I have seen several
RS232 ICs from
Maxim. But RS232 has many more signals than a simple TX and RX. I guess
what I am looking for is a simple 2-port transceiver buffer with
transient and DC voltage protection, in a through hole package. A device
that has low standby current (<500uA) would be ideal.

A solution that doesn't require adding another IC would be my
preference. So far it looks like my only choice. I would appreciate any
advice.






______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com

1998\10\27@082133 by paulb

flavicon
face
Ry Lato wrote:

> Possible solutions I have considered:
...
> I could add a .6V silicon diode between the PIC pin and VDD, with a
> 50-ohm series resitance.

 But the internal PIC diode is already just that; a silicon diode, with
much the same voltage drop, so you can't guarantee the current will flow
preferentially through the external diode.  If, however, you use a
Schottky rectifier, this makes a lot of sense.  Don't forget the one to
ground also.

> But that would dump the external DC voltage directly into the 5V VDD
> rail through the diode.

 I don't see that as a problem at all.  Put a 5.1V zener or equivalent
function across VDD to absorb it.  The regulator may (or may not;
requires research) do this for you.

>  I have considered using an IC solution for this. I have seen several
> RS232 ICs from Maxim. But RS232 has many more signals than a simple TX
> and RX.

 No it doesn't!  You said it has TX and RX only, so that's what it has.

> I guess what I am looking for is a simple 2-port transceiver buffer
> with transient and DC voltage protection, in a through hole package.

 Maxim does that.  The popular MAX-232 or in fact (IIRC) MAX-231 does
this nicely.  You actually get two inputs and two outputs, but close
enough.

> A device that has low standby current (<500uA) would be ideal.

 Requires research in the Maxim manual - you need a "shutdown" control
line.

> A solution that doesn't require adding another IC would be my
> preference.

 You don't half want much!
--
 Cheers,
       Paul B.

More... (looser matching)
- Last day of these posts
- In 1998 , 1999 only
- Today
- New search...