CANBus with rs485 chips - what's different?
Robert Rozman
at kvaser pages I saw schematic how to connect to canbus with rs485
driver chip.
Is this solution true equivalent of let's say using 82PCA251 ?
What's the difference between RS 485 and CANbus on physical level ?
Can dominant philosophy (wired and) be used also on RS 485 network ?
Any help or pointer to more info would be appreciated.
Thanks in advance,
Robert Rozman
Mark A. Odell
news:3A2BC97C...@fri.uni-lj.si...
> What's the difference between RS 485 and CANbus on physical level ?
> Can dominant philosophy (wired and) be used also on RS 485 network ?
None. CAN does not dictate the physical layer, thus you can run CAN over
RS-485. Just like you can run TCP/IP over RS-232.
Sinikka ja Seppo Sauna-aho
> at kvaser pages I saw schematic how to connect to canbus with rs485
> driver chip.
> Is this solution true equivalent of let's say using 82PCA251 ?
>
> Can dominant philosophy (wired and) be used also on RS 485 network ?
'Wired and' is possible with RS 485. The transmitter sends only
'zeros'. During the 'ones' the transmitter is in tri-state mode and a
passive or active bias circuit will make the 'ones'.
Seppo
Peter Hansen
>
> "Robert Rozman" <roz...@fri.uni-lj.si> wrote in message
> news:3A2BC97C...@fri.uni-lj.si...
>
> > Can dominant philosophy (wired and) be used also on RS 485 network ?
>
> RS-485. Just like you can run TCP/IP over RS-232.
Just a clarification: CAN *does* dictate that the physical layer support
the concept of a dominant and a recessive state, where the dominant one
has priority and stomps all over anyone attempting to transmit the
recessive state. So the above is correct if it ensures RS-485 is
configured in this way.
Stef
>Hello,
>
>at kvaser pages I saw schematic how to connect to canbus with rs485
>driver chip.
>Is this solution true equivalent of let's say using 82PCA251 ?
>
>Can dominant philosophy (wired and) be used also on RS 485 network ?
>
not equavalent (or compatible) to using a 82PCA251, which is an ISO 11898
tranceiver. So you'll have to use RS485 or 11898 for your entire network
(segment), not a combination of the two.
Stef
Mark A. Odell
news:3A2C91FD...@engcorp.com...
> > What's the difference between RS 485 and CANbus on physical level ?
> > > Can dominant philosophy (wired and) be used also on RS 485 network ?
> >
> > None. CAN does not dictate the physical layer, thus you can run CAN over
> > RS-485. Just like you can run TCP/IP over RS-232.
>
> Just a clarification: CAN *does* dictate that the physical layer support
> the concept of a dominant and a recessive state, where the dominant one
> has priority and stomps all over anyone attempting to transmit the
> recessive state. So the above is correct if it ensures RS-485 is
> configured in this way.
Yes, good point. Dominance is a rather critical part of the physical layer,
whatever it might be.
--
- Mark -> "It's in the Organization"
--
Lars Wictorsson
> >at kvaser pages I saw schematic how to connect to canbus with rs485
> >driver chip.
> >Is this solution true equivalent of let's say using 82PCA251 ?
> >
> >What's the difference between RS 485 and CANbus on physical level ?
> >Can dominant philosophy (wired and) be used also on RS 485 network ?
> >
> It is possible to run CAN with RS485 chips if wired correctly. It is however
> not equavalent (or compatible) to using a 82PCA251, which is an ISO 11898
> tranceiver. So you'll have to use RS485 or 11898 for your entire network
> (segment), not a combination of the two.
I have an application (100+ machines) that uses in the same network the
an RS485 chip and the 82C250 (which is also ISO11898), so it is possible
to use a mixture of them for lower speeds. The reason why I use the MAX485
is that when this was design in 1993 there was not so many options of
tranceivers.
Here is a diagram showing my CAN tranceiver as using MAX485 (opto-isolated)
http://www.lawicel.com/can/can485.gif,
but I suggest you use a dedicated CAN tranceiver instead, like the new
TJA1050 from Philips which replaces both 82C250 and 82C251 in one new better
design (EMC/EMI) and also 3V tolerant on the TX/RX pins.
However I have heard that CAN tranceievers are not approved for space
applications yet, so therefor they use RS485 tranceivers, but I do not
know if it is still like this. It was a year ago (or so).
/Lars
------------------------------------------------------------
LAWICEL / SWEDEN Phone : +46 (0)451 - 598 77
Lars Wictorsson Fax : +46 (0)451 - 598 78
E-mail: la...@lawicel.com WWW : http://www.lawicel.com
------------------------------------------------------------
Embedded soft-/hardware together with 8051/HC12/C16x/AVR and
smart distributed I/O with CAN (Controller Area Network).
Need Rapid CAN development? See http://www.candip.com
------------------------------------------------------------
Grant Edwards
>"Robert Rozman" <roz...@fri.uni-lj.si> wrote in message
>news:3A2BC97C...@fri.uni-lj.si...
>
>> Can dominant philosophy (wired and) be used also on RS 485 network ?
>
>RS-485. Just like you can run TCP/IP over RS-232.
I believe CAN does specify some sort of wire-AND or wire-OR
capability so that a bit of one state can trump a bit of
another state (I don't know the details, but I presume this is
used for some sort of collision detection or arbitration).
If this is true, then you can't use RS-485, since it doesn't
provide this capability.
--
Grant Edwards grante Yow! .. Now KEN and BARBIE
at are PERMANENTLY ADDICTED to
visi.com MIND-ALTERING DRUGS...
Peter Dennett
> applications yet, so therefor they use RS485 tranceivers, but I do not
> know if it is still like this. It was a year ago (or so).
There were some rad tested and flown on STS-91....
--
Peter Dennett Email: pden...@padsoft.com
61 Harbor Lane Web: www.padsoft.com
Kemah, TX 77565 Web: www.boatbrains.com
Voice: 281 334 3800 Cell: 713 899 6100 Fax: 281 521 1032
Peter Hansen
>
> > However I have heard that CAN tranceievers are not approved for space
> > applications yet, so therefor they use RS485 tranceivers, but I do not
> > know if it is still like this. It was a year ago (or so).
>
> There were some rad tested and flown on STS-91....
Do you happen to have either a reference for that statement, or at least
the part numbers that flew?
Vlastimil Pohnetal
See http://www.mite.cz/emserver
with both the optocoupled RS422/485 and the optcoupled CAN bus.
Regards
V. Pohnetal
---------------------
Robert Rozman wrote:
>
> Hello,
>
> at kvaser pages I saw schematic how to connect to canbus with rs485
> driver chip.
> Is this solution true equivalent of let's say using 82PCA251 ?
>
> What's the difference between RS 485 and CANbus on physical level ?
> Can dominant philosophy (wired and) be used also on RS 485 network ?
>
Peter Dennett
get the info. These parts flew on AMS on STS 91:
>> Intel AN82527 44-lead PLCC CHMOS III 5V were use as CAN controller.
>>
>> Philips PCA82C250T SO8 were used as interface between CAN controller
>> and the physical bus.
>>
>> Philips 82C200 CAN controller was used by some subsystems.
>> We had a problem with this device. I would strong recommend do not
use
>> it any more.
--
Peter Dennett Email: pden...@padsoft.com
61 Harbor Lane Web: www.padsoft.com
Kemah, TX 77565 Web: www.boatbrains.com
Voice: 281 334 3800 Cell: 713 899 6100 Fax: 281 521 1032
"Peter Hansen" <pe...@engcorp.com> wrote in message
news:3A2DCDF4...@engcorp.com...
William Nace
Can you tell us what the problems with the 82C200 were?
Thanks,
Bill Nace
Lars Wictorsson
> Peter,
>
> Can you tell us what the problems with the 82C200 were?
Strange that they use this outdated CAN controller that isn't
manufactured anymore. When was this flight made? Just curious
to know why they didn't used the newer SJA1000 instead (but
if the flight was before 1997 I understand that, since then
the SJA1000 wasn't avalible).
/Lars
Peter Hansen
>
> Strange that they use this outdated CAN controller that isn't
> manufactured anymore. When was this flight made? Just curious
> to know why they didn't used the newer SJA1000 instead (but
> if the flight was before 1997 I understand that, since then
> the SJA1000 wasn't avalible).
>
> > >
> > > Sorry this took a while, I had to root thru some old email to
> > > get the info. These parts flew on AMS on STS 91:
Not so strange, perhaps. I did a quick google search for you and it
appears STS 91 launched on June 2, 1998 6:06:24 p.m. EDT. The reason
they used that chip probably is that they had to complete the design
long before the launch date... it seems likely they selected that chip
long before 1998, and almost certainly also long before 1997...
Lars Wictorsson
Thanks for the info!
/Lars