Re: [foxboro] Redundant Cabling for Modbus Integrator 30s

Mark,

    Perhaps our usage could help:
    We have redundant modicon integrator 30's with redundant selector switches
and redundant cables (A & B) communicating using modbus to Honeywell PLC's (in
modicon emulation).  The RS232 from the A & B selector switches is converted
(blackbox) on each cable to RS485, to allow for longer distances and for
multidrop.  Each PLC has two serial input cards, each with the same modbus
address, and with the A cable connected to one and the B cable to the other.  My
understanding is that the modbus address is only used for the modbus
communication by the GW30.  On the PLC side, each card is addressed separately
by a nodal address, so the PLC does not get confused as to which card to reply
to.

    The GW30 sends/reads data on one cable, and sends out periodic heartbeats on
the other.  If the first cable fails to a given PLC, it switches over to polling
on the other cable to that PLC and sends heartbeats on the first cable.  When
the first cable communication is restored, it goes back to polling on it.

Rich Bakke
Longview Fibre Co.


Mark Dumond wrote:

> Hey List,
>
> I have a question about Foxboro Modbus Integrator 30s........
>
> Equipment:
> 1. Redundant Modbus Integrator 30's with redundant Modbus cabling (not
> Modbus+).
> 2. One Modbus Slave device, with one serial input (configurable as either
> RS232 or RS485)
>
> Problem:
> 1. If the Modbus slave device you are communicating to only has one Modbus
> serial connection, how do you "merge" the two redundant cables from the
> Integrator 30 to this device?
> 2. Can you convert both of the RS232 serial lines from the Integrator to
> RS485 so you can electrically tie them together?
> 3. Are there times where both cables from the Integrator communicate
> simultaneously, which would cause double requests for data from the Slave,
> and also cause confusion for the return path of the Slave's reply?
> 4. If you can make the Slave device redundant (for example, use two Prosoft
> Modbus modules, and both would be configured to use the same slave address
> and have the same data), would there be any issues with connecting each
> cable from the Int 30 to each slave device individually?  That is, connect
> one cable to the first slave device, and connect the other cable to the
> "backup" slave device.
>
> The Foxboro documentation dances around a solution for this problem (doc
> B0193RL, Figure 2-15).  The document merely shows redundant wiring as two
> independent inputs to the Modbus device; it assumes the Modbus slave device
> has two serial inputs that work as one slave address.  Since Foxboro MDSCAN
> and ECBs do not allow for a backup/alternate slave addresses, the two Modbus
> serial inputs would both have to be the same Modbus slave address, which is
> not a common feature for Modbus devices to have.
>
> Any help or insight would be greatly appreciated.
>
> Thanks in advance,
> Mark Dumond
> FeedForward, Inc
>
>
>
> _______________________________________________________________________
> This mailing list is neither sponsored nor endorsed by Invensys Process
> Systems (formerly The Foxboro Company). Use the info you obtain here at
> your own risks. Read http://www.thecassandraproject.org/disclaimer.html
>
> foxboro mailing list:             http://www.freelists.org/list/foxboro
> to subscribe:         mailto:foxboro-request@xxxxxxxxxxxxx?subject=join
> to unsubscribe:      mailto:foxboro-request@xxxxxxxxxxxxx?subject=leave
>

 
 
_______________________________________________________________________
This mailing list is neither sponsored nor endorsed by Invensys Process
Systems (formerly The Foxboro Company). Use the info you obtain here at
your own risks. Read http://www.thecassandraproject.org/disclaimer.html
 
foxboro mailing list:             http://www.freelists.org/list/foxboro
to subscribe:         mailto:foxboro-request@xxxxxxxxxxxxx?subject=join
to unsubscribe:      mailto:foxboro-request@xxxxxxxxxxxxx?subject=leave

Other related posts: