Searching \ for '[PIC] RS232 vs RS485' 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/microchip/ios.htm?key=rs485
Search entire site for: 'RS232 vs RS485'.

Exact match. Not showing close matches.
PICList Thread
'[PIC] RS232 vs RS485'
2005\02\12@071602 by Lawrence Lile

picon face


I am needing to talk to a device that speaks RS485.  Can a PIC UART talk
RS485 without a protocol converter of some sort?

--Lawrence Lile

_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today - it's FREE!
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/

2005\02\12@075424 by Bob Axtell

face picon face
Lawrence Lile wrote:

> I am needing to talk to a device that speaks RS485.  Can a PIC UART
> talk RS485 without a protocol converter of some sort?
>
> --Lawrence Lile
>
> _________________________________________________________________
> Express yourself instantly with MSN Messenger! Download today - it's
> FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
>
Like RS232, an interface chip is required. An RS485 network is just as
environmentally dangerous as RS232, so can't be driven by the PIC directly..

--Bob

--
Note: To protect our network,
attachments must be sent to
spam_OUTattachTakeThisOuTspamengineer.cotse.net .
1-866-263-5745 USA/Canada
http://beam.to/azengineer

2005\02\12@083256 by Russell McMahon

face
flavicon
face
> I am needing to talk to a device that speaks RS485.  Can a PIC UART
> talk RS485 without a protocol converter of some sort?

Yes.
RS485 is a physical layer specification. It has push pull voltage
levels on a balanced line. Actual data can be async serial.
Someone will now complain that I have over simplified things AND done
violence to the standard. :-)


   RM

2005\02\12@085009 by Howard Winter

face
flavicon
picon face
Russell,

On Sun, 13 Feb 2005 02:21:47 +1300, Russell McMahon wrote:

>...<
> RS485 is a physical layer specification. It has push pull voltage
> levels on a balanced line. Actual data can be async serial.
> Someone will now complain that I have over simplified things AND done
> violence to the standard. :-)

Or I'll make it worse by saying:  what you need is the  SN75176BP  transceiver chip to implement it!  :-)

Glitchbuster has them for US$0.68 so they're cheaper than MAX232s...

Cheers,


Howard Winter
St.Albans, England


2005\02\12@092632 by Byron A Jeff

face picon face
On Sat, Feb 12, 2005 at 06:15:45AM -0600, Lawrence Lile wrote:
>
>
> I am needing to talk to a device that speaks RS485.  Can a PIC UART talk
> RS485 without a protocol converter of some sort?

Lawrence,

Short answer is NO. You'll need a RS-485 transceiver between the PIC and the
link.

The are several issues involved with RS-485.

1) It's only an electrical specification. No higher level data-link protocols
are specified.

2) It's half duplex.

3) It's multi-drop. So there can be a bunch of potential transmitters.

In a lot of ways RS-485 behaves more like thin-wire ethernet that RS-232.
You'll have to resolve issues of turning around the transceiver and packet
collision management.

Jan Axelson of LakeView Research has written books and articles on the
subject. She really knows her stuff. Here's a pointer to an overview article
she wrote for Circuit Cellar:

http://www.embeddedsys.com/subpages/resources/images/documents/microsys_art_RS485.pdf

By the way, don't let any of this scare you off. The transceivers are just
about as cheap as a MAX232, are generally only 8 pins, and allows you to
crank up both the distance and the bit rate.

Hope this gets you started.

BAJ

2005\02\12@093301 by Byron A Jeff

face picon face
On Sun, Feb 13, 2005 at 02:21:47AM +1300, Russell McMahon wrote:
> >I am needing to talk to a device that speaks RS485.  Can a PIC UART
> >talk RS485 without a protocol converter of some sort?
>
> Yes.
> RS485 is a physical layer specification. It has push pull voltage
> levels on a balanced line. Actual data can be async serial.
> Someone will now complain that I have over simplified things AND done
> violence to the standard. :-)

I guess I'll be the first to step up.

Here's how I read Lawrence's question:

"Can a PIC UART talk RS-485 (RS-232) without a protocol converter of some
sort (like a MAX232)?"

If that's the correct interpretation, then the answer is NO. A
RS-485 transceiver is required between the PIC and the RS-485 network.

BAJ

2005\02\12@102350 by Jan-Erik Soderholm

face picon face
Byron A Jeff wrote :

> Here's how I read Lawrence's question:
>
> "Can a PIC UART talk RS-485 (RS-232) without a protocol
> converter of some sort (like a MAX232)?"

Is a MAX232 a "protocol converter" ?
I thought it was a *level* converter/shifter between
TTL/CMOS and RS232 voltage levels. That has nothing
to do with any protocols, if I understand correctly what
a "protocol" is.

IMHO, the answer to Lawrence is "it depends".
If there's only one RS232 and one RS485 "node", you could
probably live with just a level converter. But if it's a full
RS485 "network" with multiple "nodes", you must probably
add some "protocol" to cope with bus conflicts and collisions.

So, we (or rather Lawrence) need to know more about the
RS485 node(s) at hand. What does the actual signal/protocol
used look like ? Is it a single RS485 node on "the line" ?

Jan-Erik.



2005\02\12@112916 by Kenneth Lumia

picon face
Lawrence,

In terms of the physical interface it is rather trivial and I
could supply a schematic.  In terms of an actual protocol, RS485 is
similar to RS232 as both don't really care what is being sent; they
are only electrical specifications. The protocol can be anything the
designer wants. A PICs uart may or may not work depending on
how strangely the protocol was designed. For example, if they used
12 data bits, a parity bit and an oddball
data rate, the PIC uart isn't up to the task and you will have to
bit-bang a pin instead.  If the protocol is a more standard one,
say 9600 8n1 then no problem.

Ken
.....klumiaKILLspamspam@spam@adelphia.net

{Original Message removed}

2005\02\12@124416 by D. Jay Newman

flavicon
face
> bit-bang a pin instead.  If the protocol is a more standard one,
> say 9600 8n1 then no problem.

I've run PICs with rs485 up to the higher standard baud rates. At higher
baud rates the problems are the same as with RS232: you need the proper
crystal to get the exact baud rates you need.

You do need a tranceiver chip, but you need one with RS232 also. The chips
cost about the same.
--
D. Jay Newman           ! Polititions and civilations come and
jayspamKILLspamsprucegrove.com     ! go but the engineers and machinists
http://enerd.ws/robots/ ! make progress

2005\02\12@194922 by olin_piclist

face picon face
Lawrence Lile wrote:
> I am needing to talk to a device that speaks RS485.  Can a PIC UART
> talk RS485 without a protocol converter of some sort?

I don't know what you mean by "protocol converter", but there are no UART
lines on a PIC that can be directly connected to RS-485 lines.  RS-485 is a
single signal on a differential pair, and it is only driven in one direction
and passively pulled in the other.  No damage or high current results from
two or more devices driving the RS-485 bus, although the data will probably
be corrupted.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

2005\02\12@204007 by Bob Ammerman

picon face
All you need is the RS485 equivalent of a MAX232.

Bob Ammerman
RAm Systems

----- Original Message -----
From: "Lawrence Lile" <.....lawrencelileKILLspamspam.....hotmail.com>
To: <EraseMEpiclistspam_OUTspamTakeThisOuTmit.edu>
Sent: Saturday, February 12, 2005 7:15 AM
Subject: Re: [PIC] RS232 vs RS485


{Quote hidden}

> --

2005\02\13@005730 by Brent Brown

picon face
> > I am needing to talk to a device that speaks RS485.  Can a PIC UART
> > talk RS485 without a protocol converter of some sort?
>
> I don't know what you mean by "protocol converter", but there are no UART
> lines on a PIC that can be directly connected to RS-485 lines.  RS-485 is a
> single signal on a differential pair, and it is only driven in one direction
> and passively pulled in the other.  No damage or high current results from
> two or more devices driving the RS-485 bus, although the data will probably
> be corrupted.

Are you sure about that, or maybe you're getting your wires crossed with
CAN bus? My understanding of RS485 (correct me if I'm wrong) is that when
transmitting a '1', one line is actively driven high (5V) and the other is actively
driven low (0V), and the reverse for sending a '0'. Nothing passive about it.
There will be higher current draw with two or more devices attempting to
drive the bus at the same instant. Transceivers such as the MAX485 are
current limited and thermally protected.

--
Brent Brown, Electronic Design Solutions
16 English Street, Hamilton, New Zealand
Ph/fax: +64 7 849 0069
Mobile/txt: 025 334 069
eMail:  brent.brownspamspam_OUTclear.net.nz


2005\02\13@074013 by Jan-Erik Soderholm

face picon face
Brent Brown wrote :

> > RS-485 is a single signal on a differential pair, and it
> > is only driven in one direction and passively pulled
> > in the other...

> Are you sure about that... ?
> My understanding of RS485 (correct me if I'm wrong)
> is that when transmitting a '1', one line is actively driven
> high (5V) and the other is actively driven low (0V), and
> the reverse for sending a '0'. Nothing passive about it.

But then, how do you connect the 32 "nodes" on a single
RS485 "network" ? (32 according to the original spec, newer
hardware can handle a higher number of concurrent nodes
due to higher "off" impedance in CMOS then in the original
bipolar devices when the spec was written...)

Jan-Erik.



2005\02\13@081713 by Bob Axtell

face picon face
About 8 years ago, I designed some twisted-pair casino networks.

Most designs used two sets of twisted-pair wires, so that one pair is
always a listener pair (driven by the PC Master). Each unit on the
network listened on that pair, then responded on a second  pair,
being switched in only when addressed.  A 100-ohm terminating
resistor was at the far end  of each  pair.  Up to 127 casino slots
were attached to one network..

At 19.2Kb the network operated flawlessly at 1200m, using CAT3,
because some building were so old that only CAT3 business pairs
could be used.
.
It was driven by the most advanced PIC available at the time, a,
PIC16C73. It was an early "player card" system, could store some
player card locally. My first heavy application for Ramtron I2C
8K FRAM's, too. None ever failed.

It worked fairly well, but in those days, static discharge occasionally
"took out" network driver chips, not very well-designed then. We
had to install sockets for them for easy replacement.

--Bob

Jan-Erik Soderholm wrote:

{Quote hidden}

--
Note: To protect our network,
attachments must be sent to
@spam@attachKILLspamspamengineer.cotse.net .
1-866-263-5745 USA/Canada
http://beam.to/azengineer

2005\02\13@091622 by Jan-Erik Soderholm

face picon face
Bob Axtell wrote :

> About 8 years ago, I designed some twisted-pair casino networks.

[snip...]

Interesting, but I'm not sure how it relates to the
RS485 specifications ? Did it use RS485 ?

Jan-Erik.



2005\02\13@111524 by Matthew Miller

flavicon
face
On Sun, Feb 13, 2005 at 01:40:13PM +0100, Jan-Erik Soderholm wrote:
> But then, how do you connect the 32 "nodes" on a single
> RS485 "network" ? (32 according to the original spec, newer
> hardware can handle a higher number of concurrent nodes
> due to higher "off" impedance in CMOS then in the original
> bipolar devices when the spec was written...)

Well, those nodes just connect to the same twisted pair ;) The trouble is
deciding when you can transmit. I like the LTC1482 because it has a carrier
detect output which makes it great for single pair, half-duplex networks.

Matthew

--
The moon is covered with the results of astronomical odds.

2005\02\13@112035 by olin_piclist

face picon face
Brent Brown wrote:
> Are you sure about that, or maybe you're getting your wires crossed
> with
> CAN bus? My understanding of RS485 (correct me if I'm wrong) is that
> when transmitting a '1', one line is actively driven high (5V) and
> the other is actively driven low (0V), and the reverse for sending a
> '0'. Nothing passive about it. There will be higher current draw with
> two or more devices attempting to
> drive the bus at the same instant. Transceivers such as the MAX485 are
> current limited and thermally protected.

Yes, you may be right about that.  However the main point is that RS-485 is
a single signal transmitted by a differential pair, and can not be "just
hooked up" to a PIC UART.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

2005\02\13@115508 by Jan-Erik Soderholm

face picon face
Matthew Miller wrote :

> On Sun, Feb 13, 2005 at 01:40:13PM +0100, Jan-Erik Soderholm wrote:

> > But then, how do you connect the 32 "nodes" on a single
> > RS485 "network" ? (32 according to the original spec, newer
> > hardware can handle a higher number of concurrent nodes
> > due to higher "off" impedance in CMOS then in the original
> > bipolar devices when the spec was written...)
>
> Well, those nodes just connect to the same twisted pair ;)
> The trouble is deciding when you can transmit.

Now, how do you transmit *at all* if the other 31 nodes
are activily holding the line low (or high or whatever) ?

Never mind...

Jan-Erik.



2005\02\13@123539 by Ake Hedman

flavicon
face
Jan-Erik Soderholm wrote:

{Quote hidden}

To clear this up....

The RS-485 transceiver drives a differential pair. When active it put out a 0/1 on the pairs or a 1/0  for a zero or a one. When the tranceiver is inactive it tristate and output floats. The receiving part of the transceiver can of course still be active.


A variant is to send on one pair and receive on one pair. Same electrical specs. but this is RS-422.

Regards
/Ake

--  ---
Ake Hedman (YAP - Yet Another Programmer)
eurosource, Brattbergavägen 17, 820 50 LOS, Sweden
Phone: (46) 657 413430 Cellular: (46) 73 84 84 102
Company home: http://www.eurosource.se      Kryddor/Te/Kaffe: http://www.brattberg.com
Personal homepage: http://www.eurosource.se/akhe
Automated home: http://www.vscp.org

2005\02\13@124709 by olin_piclist

face picon face
Jan-Erik Soderholm wrote:
> Now, how do you transmit *at all* if the other 31 nodes
> are activily holding the line low (or high or whatever) ?

You don't.  The upper protocol layers have to ensure that no more than a
single node is transmitting at a time.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

2005\02\13@141354 by Jan-Erik Soderholm

face picon face
Ake Hedman wrote :

> Jan-Erik Soderholm wrote:
>
> >Now, how do you transmit *at all* if the other 31 nodes
> >are activily holding the line low (or high or whatever) ?
> >
> To clear this up....

It sure did :-) :-)
Thanks.
Jan-Erik.

{Quote hidden}

2005\02\13@150402 by Brent Brown

picon face
Jan-Erik Soderholm wrote:

> > > RS-485 is a single signal on a differential pair, and it
> > > is only driven in one direction and passively pulled
> > > in the other...
>
> > Are you sure about that... ?
> > My understanding of RS485 (correct me if I'm wrong)
> > is that when transmitting a '1', one line is actively driven
> > high (5V) and the other is actively driven low (0V), and
> > the reverse for sending a '0'. Nothing passive about it.
>
> But then, how do you connect the 32 "nodes" on a single
> RS485 "network" ? (32 according to the original spec, newer
> hardware can handle a higher number of concurrent nodes
> due to higher "off" impedance in CMOS then in the original
> bipolar devices when the spec was written...)

Yes, as others have said, you have to come up with a way of ensuring that
only one device is transmitting at a time and all other devices are listening.
The RS485 specification does not define how this is done, it just gives you
the hardware to do it on. It's up to you the designer to apply a suitable
protocol, but it's not real hard to do.

One fairly simple implementation is to make one of the devices on the bus a
"Master" and all other devices "Slaves". Only the Master is permitted to
transmit on the bus whenever it wants to. Otherwise refferred to a
command/response method. A Slave can not transmit on the bus unless
requested to do so by the Master. Timing must be well defined. Often the
Master will be a PC connected to a network of Slave devices.

Standard RS485 devices all for up to 32 nodes (eg. MAX485 chip). There
are also "quarter load" devices (eg MAX487 chip) that allow up to 128 nodes.

Just out of interest, be aware of one of the common misconceptions about
RS485. Although many refer to it as a "2 wire" this is misleading. The data
signals are relative to GND, and all nodes must share a common GND. It is
not (in it's standard configuration) an isolated bus. But RS485 does allow for
a degree of GND potential difference (-7V to +12V for MAX485) between
nodes. You can, for example, have an RS485 network between two buildings
and use just 2 wires for data, using the electricity supply GND. Because the
signal is differential, small GND voltages are eliminated by common mode
rejection (but beware of the limits, and pay special attention to fault
conditions that may arise in this kind of application).

I have actually seen an RS485 network with all devices isolated from GND
and using just a single twisted pair. By some strange quirk it appeared to
"mostly work", "most" of the time, although it suffered many 485 chip failures.

I recommend the Jan Axelson/Bob Perrin articles on RS485 in Jun/July 199
issues of Circuit Cellar.

--
Brent Brown, Electronic Design Solutions
16 English Street, Hamilton, New Zealand
Ph/fax: +64 7 849 0069
Mobile/txt: 025 334 069
eMail:  KILLspambrent.brownKILLspamspamclear.net.nz


2005\02\13@192437 by Russell McMahon

face
flavicon
face
> Just out of interest, be aware of one of the common misconceptions
> about
> RS485. Although many refer to it as a "2 wire" this is misleading.
> The data
> signals are relative to GND, and all nodes must share a common GND.
> It is
> not (in it's standard configuration) an isolated bus.

You may cvery well be correct BUT I can';t see why gnd is needed for
signalling purposes. As long as wires signal differentially at the
standard levels ground potential would not matter for signalling
purposes. Certainly if devices had limited common mode voltage
capability and if ICs were used which did'nt like eg 230 VAC to
ground, then you could get problems if you applied 230VAC common mode
to both leads ;-)

> I have actually seen an RS485 network with all devices isolated from
> GND
> and using just a single twisted pair. By some strange quirk it
> appeared to
> "mostly work", "most" of the time, although it suffered many 485
> chip failures.

Makes sense. If you had a floating bearer it could assume any voltage
relative to ground. If it did and a device connected cared about
voltage to ground then failures could occur.



       RM

2005\02\13@200842 by Brent Brown

picon face
{Quote hidden}

OK, well said. But the common mode voltage on the MAX485 (a very typical
RS485 transceiver chip) is limited, -7 to +12V. Not sure what the actual
RS485 spec requires, but suspect it is something similar. Not a big range to
work with. For two wire "floating" network (where there is no common GND
at all) you would IDEALLY have some biasing resistors to the local power
supplies on each node to keep the signal within this range - right? Without
that, common mode voltage could drift anywhere - possibly outside of the
specs of the transceiver chips thereby creating potential for communications
errors (at least) and/or chip damage.

--
Brent Brown, Electronic Design Solutions
16 English Street, Hamilton, New Zealand
Ph/fax: +64 7 849 0069
Mobile/txt: 025 334 069
eMail:  RemoveMEbrent.brownTakeThisOuTspamclear.net.nz


2005\02\14@015307 by Roland

flavicon
face
At 01:15 PM 14/02/2005 +1300, you wrote:
>> Just out of interest, be aware of one of the common misconceptions
>> about
>> RS485. Although many refer to it as a "2 wire" this is misleading.
>> The data
>> signals are relative to GND, and all nodes must share a common GND.
>> It is
>> not (in it's standard configuration) an isolated bus.
>
>You may cvery well be correct BUT I can';t see why gnd is needed for
>signalling purposes. As long as wires signal differentially at the
>standard levels ground potential would not matter for signalling
>purposes. Certainly if devices had limited common mode voltage
>capability and if ICs were used which did'nt like eg 230 VAC to
>ground, then you could get problems if you applied 230VAC common mode
>to both leads ;-)
.....

although the OUPUT signals are differential, they are BOTH + relative to
local ground, which means there must be a return path.

It's a bit like saying that if the I2C bus transmitted SDA and a not_SDA,
then ground would not be required

the RS485 input allows a +/- 7V shift between signal in and local ground,
which generates the confusion.


>From the most informative doc I found; (sorry, no pics from the doc
attached, but the app note should be quite findable)

______________________________________________>>


RS-485 Interfaces
National Semiconductor
Application Note 1057
John Goldie
October 1996
Ten Ways to Bulletproof RS-485 Interfaces AN-1057

..............
Data sheets for RS-485 drivers usually do not include VOL or
VOH specifications. The driver’s VOLis typically around 1V.
Even for CMOS devices, VOH is slighly above 3V, because
both the source and sink paths of the output structure include
a series-connected diode, which provides the
common-mode tolerance for an Off driver. Because VOL is
usually greater than 0.8V, an RS-485 driver is not
TTL-compatible.

VOS represents the driver’s offset voltage measured from the
center point of the load with respect to the driver’s ground
reference. VOS is also called “VOC” for output common-mode
voltage. This parameter is related to VCM.
VCM represents the common-mode voltage for which
RS-485 is famous. The limit is -7V to +12V. Common-mode
voltage is defined as the algebraic mean of the two
local-ground-referenced voltages applied to the referenced
terminals (receiver input pins, for example). The
common-mode voltage represents the sum of three voltage
sources. The first is the active driver’s offset voltage. The
second is coupled noise that shows up as common mode on
both signal lines. The third is the ground-potential difference
between the node and the active driver on the bus. Mathematically,
VCM = VOS + VNOISE + VGPD.
VGPD represents the ground-potential difference that can
exist between nodes in the system. RS-485 allows for a 7V
shift in grounds. A shift of 7V below the negative (0V) power
rail yields the -7V common-mode limit, whereas 7V above
the 5V positive power rail yields the other common-mode
limit of 12V. Understanding these parameters enables improved
component selection, because some devices trade
off certain parameters to gain others.
........


.......

Grounding and Shielding

Although the potential difference between the data-pair conductors
determines the signal without officially involving
ground, the bus needs a ground wire to provide a return path
for induced common-mode noise and currents, such as the
receivers’ input current. A typical mistake is to connect two
nodes with only two wires. If you do this, the system may
radiate high levels of EMI, because the common-mode return
current finds its way back to the source, regardless of
where the loop takes it. An intentional ground provides a
low-impedance path in a known location, thus reducing
emissions.

Electromagnetic-compatibility and application requirements
determine whether you need a shield. A shield both prevents
the coupling of external noise to the bus and limits emissions
from the bus. Generally, a shield connects to a solid ground
(normally, the metal frame around the system or subsystem)
with a low impedance at one end and a series RC network at
the other. This arrangement prevents the flow of DC
ground-loop currents in the shield.

Contention Protection

Because RS-485 allows for connecting multiple drivers to
the bus, the standard addresses the topic of contention.
When two or more drivers are in contention, the signal state
on the bus is not guaranteed. If two drivers are on at the
same time and if they are driving the same state, the bus
state is valid. However, if the drivers are in opposite states,
........




Regards
Roland Jollivet

2005\02\14@031215 by Russell McMahon

face
flavicon
face
>>> The data signals are relative to GND, and all nodes must share a
>>> common GND.

>>You may cvery well be correct BUT I can';t see why gnd is needed for
>>signalling purposes.

> although the OUPUT signals are differential, they are BOTH +
> relative to
> local ground, which means there must be a return path.

I disagree. If you present me with a wire pair with differential
signalling on it it is possible to extract the digital signal with no
other reference. There can be some issues over "DC restoration"  but
at every moment the relative polarity of the wires, and thus the
intended signalling state, is clearly measurable. (Crossovers of
course occur between symbols.)

An example of such a system is a telephone where the speech signal
(here analog) is not essentially ground referenced.



       RM

2005\02\14@045053 by steve

flavicon
face
> >>> The data signals are relative to GND, and all nodes must share a
> >>> common GND.

> I disagree. If you present me with a wire pair with differential
> signalling on it it is possible to extract the digital signal with no
> other reference.

Yes, it's possible, but it isn't RS485. RS485 is differential signalling
within a defined common mode voltage. That way it can operate without
transformers & opto-isolation being required.

> An example of such a system is a telephone where the speech signal
> (here analog) is not essentially ground referenced.

An example of another totally unrelated system is a fibre-optic link. No
doubt there are others.

Steve.
==========================================
Steve Baldwin                          Electronic Product Design
TLA Microsystems Ltd             Microcontroller Specialists
PO Box 15-680, New Lynn                http://www.tla.co.nz
Auckland, New Zealand                     ph  +64 9 820-2221
email: spamBeGonestevespamBeGonespamtla.co.nz                      fax +64 9 820-1929
=========================================



2005\02\14@064050 by Russell McMahon

face
flavicon
face
>> >>> The data signals are relative to GND, and all nodes must share
>> >>> a
>> >>> common GND.

>> I disagree. If you present me with a wire pair with differential
>> signalling on it it is possible to extract the digital signal with
>> no
>> other reference.

> Yes, it's possible, but it isn't RS485. RS485 is differential
> signalling
> within a defined common mode voltage. That way it can operate
> without
> transformers & opto-isolation being required.

I agree with you. But ...

The omissions from my quoting fail to clearly how we got to where we
were. I originally noted that TECHNICALLY it wasn't necessary for the
differential signals to be ground referenced for the purpose of
conveying data - that their values relative to each other were
entirely adequate to convey the signals. I said that because there was
a suggestion that the ground reference was an essential part of the
signalling. I also discussed common mode and isolation limits and so
on.

My point was to try and separate other factors from the technical
aspects of sognal transmission per se as their was some important EE
concepts involved.

>> An example of such a system is a telephone where the speech signal
>> (here analog) is not essentially ground referenced.

> An example of another totally unrelated system is a fibre-optic
> link. No
> doubt there are others.

Here I disagree :-)
I gave the telephone example because I consider that it is a very
close one. I could have used ethernet transmission on twisted pair. In
each case a differential signal is used to convey information without
ground reference per se. The telephone wires are DC ground referenced
for connection control and power feeding purposes (usually with legs
at ground and -50 volts for power purposes). I think this is a good
example. The fibre optic example is indeed much less related as there
is no differential signal involved and no ability to have a DC
component.

Not wishing to argue (especially much :-) ) but I was trying to make a
technical point for the edification of others and i still think it's
valid. I'm sure we don't actually disagree on the technical aspects
once we get the background cleared up.



       Russell McMahon











{Quote hidden}

> --

2005\02\14@090722 by olin_piclist

face picon face
Russell McMahon wrote:
> I disagree. If you present me with a wire pair with differential
> signalling on it it is possible to extract the digital signal with no
> other reference. There can be some issues over "DC restoration"  but
> at every moment the relative polarity of the wires, and thus the
> intended signalling state, is clearly measurable. (Crossovers of
> course occur between symbols.)

Yes, but this doesn't work so well for transmitting RS485 since the common
mode range of a complient receiver need not be that large.

> An example of such a system is a telephone where the speech signal
> (here analog) is not essentially ground referenced.

And 10base-T ethernet where the both ends are transformer coupled and the
common mode voltage of the twisted pair is irrelevant over a wide range.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

2005\02\14@111517 by Lawrence Lile

picon face
Yes, more precisely what I am looking for is a level shifter.  My PIC will be a master and the devices on the RS485 side are all slave devices, i.e. they shaddap unless they are told to talk, so there is no problem with collissions.  There is a well known protocol/language/command set what have you that talks to these devices that I need not bore you guys with there.  If you are curious, they are ADAM 4015 temperature transmitters reading RTD's. (yes I know I could probably read RTD's with a PIC but I don't want to reinvent the wheel)

You have already probably anticipated my next question - is there a simple chip that accomplishes this level shift ala' the venerable MAX232?

--Lawrence



{Quote hidden}

2005\02\14@113308 by Josh Koffman

face picon face
As others have mentioned, take a look at the SN75176B and the MAX485.
Both are simple interfaces that do TTL <-> RS485.

To be honest, I'm not sure "level shifter" is the right term for these
though. Usually they're listed as line drivers or something similar.

Enjoy!

Josh
--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
       -Douglas Adams

On Mon, 14 Feb 2005 10:12:16 -0600, Lawrence Lile
<RemoveMElawrencelileEraseMEspamEraseMEhotmail.com> wrote:
{Quote hidden}

2005\02\14@114015 by David Challis

face picon face
How about the MAX485 series from maxim.  If you'd like galvanic isolation,
check out the burr-brown (TI) ISO485.  Also there is the TI SN75ALS176A.

Nat Semi appnotes AN-409, AN-979, and AN-1057 are helpful, along with with
these app notes from TI:
www-s.ti.com/sc/psheets/slla036b/slla036b.pdf
www-s.ti.com/sc/psheets/slla112a/slla112a.pdf
http://www-s.ti.com/sc/psheets/slla070c/slla070c.pdf

Dave Challis

{Original Message removed}

2005\02\14@114159 by Ake Hedman

flavicon
face
Lawrence,

If price is a concern and you can live with 32 slaves the SN75176 is fine If you want more protection, less buss load and want to spend more money on it Maxim has plenty of transceivers. Look at the end of http://www.maxim-ic.com/appnotes.cfm/appnote_number/367

Regards
/Ake

Lawrence Lile wrote:

{Quote hidden}

>> --

2005\02\14@115627 by Mark Scoville

flavicon
face
Yes, there are lots of RS485 "level shifter" chips. I have used the
Dallas/Maxim DS3695 on several RS485 ModBus projects along with the internal
PIC UART. There are other chips too. Check out National Semi (75176 maybe?),
TI, Dallas/Maxim etc. I hope this is some help.

-Mark

> You have already probably anticipated my next question - is there
> a simple
> chip that accomplishes this level shift ala' the venerable MAX232?
>
> --Lawrence



2005\02\14@151743 by steve

flavicon
face
> > Yes, it's possible, but it isn't RS485. RS485 is differential
> > signalling
> > within a defined common mode voltage. That way it can operate
> > without
> > transformers & opto-isolation being required.
>
> I agree with you. But ...

But ..... It appears, you miss the point completely.

It is _possible_ to convey information without a ground reference and
ethernet and telephone are fine examples of that.

However, the subject is RS-485. RS-485 is an interface where
differential signalling is used and the ground reference is _required_
and specified.

It's an important aspect of the spec as it makes it more robust than
RS232, but less so than a fully isolated differential bus.

It needs to be emphasized as it is one of the most common RS485
misconceptions and is probably the number one reason why the
interface chips are socketed.

> > An example of another totally unrelated system is a fibre-optic
> > link. No doubt there are others.
> Here I disagree :-)

Why ? Just as your example, it fits into the group "interfaces that aren't
RS485"

> Not wishing to argue (especially much :-) ) but I was trying to make a
> technical point for the edification of others and i still think it's
> valid. I'm sure we don't actually disagree on the technical aspects
> once we get the background cleared up.

Your point is perfectly valid and I have no technical argument. Perhaps
you should change the subject to the one I have suggested above.

Steve.



2005\02\14@161638 by Peter L. Peres

picon face


On Mon, 14 Feb 2005, Brent Brown wrote:

> OK, well said. But the common mode voltage on the MAX485 (a very typical
> RS485 transceiver chip) is limited, -7 to +12V. Not sure what the actual
> RS485 spec requires, but suspect it is something similar. Not a big range to
> work with. For two wire "floating" network (where there is no common GND
> at all) you would IDEALLY have some biasing resistors to the local power
> supplies on each node to keep the signal within this range - right? Without
> that, common mode voltage could drift anywhere - possibly outside of the
> specs of the transceiver chips thereby creating potential for communications
> errors (at least) and/or chip damage.

None of the dc coupled protocols can take too much common mode voltage.
POTS is one of the more robust (you can boost RS232 to POTS voltage
swing levels and go in excess of 10km at limited baud rate, but
electrical interference will still bother you a lot, i.e. storms, nearby
power lines, factory interference etc. The far end of such a loop is
usually insulated optically or otherwise. Remember how nervous the POTS
operators are about users connecting *anything* to their POTS system.
The initial 52V of the POTS system can drop to 1V at the far end!

You can send RS422 and RS485 through pulse transformers is you use bit
stuffing to prevent dc bias. That fixes the common mode problem.

hope this helps,
Peter

2005\02\14@173427 by Bob Ammerman

picon face


{Quote hidden}

Russell, I agree with you, but with the proviso that receiving circuit have
no reference to ground itself. Probably best done via optoisolation.

Bob Ammerman
RAm Systems


2005\02\15@014046 by Russell McMahon

face
flavicon
face
> But ..... It appears, you miss the point completely.

At the risk of beating a dead horse.
No.

My comments need to be read in the context they were made to make
sense of them.
The email quote/quote/quoting ... system very quickly chews up
context.
The comments were relevant enough under the subject line in use (IMHO
of course) - they were addressing a technical comment about the
NECESSITY of ground for signalling purposes WITHIN the RS485 system,
rather than its necessity within the system for other reaosns.

> It is _possible_ to convey information without a ground reference
> and
> ethernet and telephone are fine examples of that.
>
> However, the subject is RS-485. RS-485 is an interface where
> differential signalling is used and the ground reference is
> _required_
> and specified.

Indeed. But not directly for the purposes of determining which symbol
is being sent. Which was the assertion made by someone. It seemed
worth addressing that point so a misaprehension was not maintained. No
doubt i introduced a fewother misaprehensions in the process :-).

> It's an important aspect of the spec as it makes it more robust than
> RS232, but less so than a fully isolated differential bus.

I think that the aspect that makes it more robust is its balanced
differential natire (like telephone) which RS232 of course has not
got. Common mode line noise is rejected while with RS232 it's not. (I
realise that you know that.) I don't think the ground referencing adds
immunity. Both it and 232 are ground referenced after a fashion.

>> > An example of another totally unrelated system is a fibre-optic
>> > link. No doubt there are others.
>> Here I disagree :-)
>
> Why ? Just as your example, it fits into the group "interfaces that
> aren't
> RS485"

The whole world of NOT(RS485 or RS422) fit into that category. But the
telephone system has many more relevant common points that make it a
pertinent example. eg twisted pair with DC path, ground reference of
sorts, "balanced" differential data transfer, nicely reject
longitudinal common noise on the dc bearer. And more.

{Quote hidden}

As above. The comment was within an RS485 context as it addressed a
"... xxx is done within RS485 because yyy ..."

Horse looks even deader now :-)


       RM

2005\02\15@105636 by Joe McCauley

picon face
I use a MAX3082. Connect to the TXD & RXD pins of a PIC and use one more PIC
output line to tell the device whether to transmit or receive. The PIC uses
this line to keep the device in transmit mode, switching it to receive mode
when the PIC is expecting a reply from the bus.

Joe

{Original Message removed}

More... (looser matching)
- Last day of these posts
- In 2005 , 2006 only
- Today
- New search...