I am currently working on a design project that involves communication
between a PIC 16C76 and a PC, and need a low power solution for serial
communications. The device will be battery powered and will not normally
use the serial connection. Thus the serial component should have the
capability of sleeping while not being used to conserve battery life, and
have the ability to wake up when it senses communication from the PC. I
have looked at the Maxim MAX223 which seems to be what I want, but I am not
sure if this is the best way to go. Any Suggestions?
>I
>have looked at the Maxim MAX223 which seems to be what I want, but I am not
>sure if this is the best way to go. Any Suggestions?
I'm using an rs422/485 transceiver (Maxim MAX1482, but there are plenty
of others out there) to talk to an rs232 machine with great success.
rs422 is lower power (and lower noise!) than the usual 232 transceivers,
since it's +-5V and doesn't require charge pumps. The trick is to attach
the serial gnd to Tx+, NOT to the circuit ground, and leave Rx+
unconnected. This effectively gives you rs423, which is compatible with
rs232 over short distances. Works for me just fine. (Incidentally,
Macintosh serial ports are rs422, and they use this same trick to speak
to rs232 devices. That's where I learned about it.)
|I am currently working on a design project that involves communication
|between a PIC 16C76 and a PC, and need a low power solution for serial
|communications. The device will be battery powered and will not normally
|use the serial connection. Thus the serial component should have the
|capability of sleeping while not being used to conserve battery life, and
|have the ability to wake up when it senses communication from the PC. I
|have looked at the Maxim MAX223 which seems to be what I want, but I am not
|sure if this is the best way to go. Any Suggestions?
----+----1----+----2----+----1----+----2----+----1----+----2----+----1----+----2
If you are in fact using a PC and have some control over its serial ports (e.g.
the ability to have it keep DTR asserted and CTS deasserted) you may do well to
use some fairly simple opto-based circuitry (the PC side of things is powered by
DTR and CTS). This has the advantages that:
[1] When the PIC side isn't transmitting or receiving, the power consumption on
its end is--literally--zero.
[2] No matter what happens to your PIC, your PC will be safe (or vice versa);
otherwise you run the risk of ground loops.
[3] The opto interface I built works up to 19.2K; to ensure reliability I use it
at
9600 baud to give something of a safety margin.
Of course, there are probably even better ways of designing such devices, but th
ere
is a lot to be said for simple solutions that work.
>
> |I am currently working on a design project that involves
> communication
> > |between a PIC 16C76 and a PC, and need a low power solution for
> > serial
> > |communications. The device will be battery powered and will not
> > normally
> > |use the serial connection. Thus the serial component should have the
> > |capability of sleeping while not being used to conserve battery life,
> > and
> > |have the ability to wake up when it senses communication from the PC.
> > I
> > |have looked at the Maxim MAX223 which seems to be what I want, but I
> > am not
> > |sure if this is the best way to go. Any Suggestions?
> >
> > ----+----1----+----2----+----1----+----2----+----1----+----2----+----1
> > ----+----2
> > If you are in fact using a PC and have some control over its serial
> > ports (e.g.
> > the ability to have it keep DTR asserted and CTS deasserted) you may
> > do well to
> > use some fairly simple opto-based circuitry (the PC side of things is
> > powered by
> > DTR and CTS). This has the advantages that:
> >
> > [1] When the PIC side isn't transmitting or receiving, the power
> > consumption on
> > its end is--literally--zero.
> >
> > [2] No matter what happens to your PIC, your PC will be safe (or vice
> > versa);
> > otherwise you run the risk of ground loops.
> >
> > [3] The opto interface I built works up to 19.2K; to ensure
> > reliability I use it at
> > 9600 baud to give something of a safety margin.
> >
> > Of course, there are probably even better ways of designing such
> > devices, but there
> > is a lot to be said for simple solutions that work.
You not necessarily need a max-232.
How about the "good-old" transistor solution. While I4m not
realy sure if this works with full-duplex, it will do it with
half-duplex.
The negative supply for transmitting from the pic 2 PC is derived from
the normal negative PC TX Signal and feed through a diode into a
capacitor to "store" some energie. This negativ voltage is your negative
supply for TXing PIC -> PC, positive voltage is normal PIC supply.
The tx-signal PIC -> PC then cover the specs ( between approx -15 .. +5
)
Receiving is also easy by just hooking another transistor
between PC-TX and PIC-RX
I can Mail/Post the circuit if it is of interest. The way I4ve seen and
used this circuit is for a simple two-wire (RX-TX) connection
without any (hardware)handshake.
I didn4t follow the whole thread, so this idea need not to be the
thing you where looking for.
> John Payson wrote:
> >
> > |I am currently working on a design project that involves
> > communication
> > > |between a PIC 16C76 and a PC, and need a low power solution for
> > > serial
> > > |communications. The device will be battery powered and will not
> > > normally
> > > |use the serial connection. Thus the serial component should have the
> > > |capability of sleeping while not being used to conserve battery life,
> > > and
> > > |have the ability to wake up when it senses communication from the PC.
> > > I
> > > |have looked at the Maxim MAX223 which seems to be what I want, but I
> > > am not
> > > |sure if this is the best way to go. Any Suggestions?
> > >
"Paul B. Webster VK2BZC" wrote:
>
> Hello Stefan.
>
> Was it mentioned? The diode between the collector load of the PNP
> transistor and ground was the wrong way up. Written as "A" instead of
> "V". You are probably aware there is a MAXIM IC which implements this
> circuit.
Paul, you are right, I made a mistake while copying the source to the
E-Mail.
Yes, I4m aware of the exsiting of various Driver IC, e.g. MAX 232,
LTC...
MC1488/1498 but the qestion was for a "Low-Power"-solution - and you
must agree -
the circuit below needs no charge-pump at the PIC-side.
erinet.com
