Re: [foxboro] Foxboro and Allen Bradley SLC 504


Hmmm...I've never experienced the master wantonly writing over the slave's
data unless there was a configuration problem, on either I/A or brand H
systems.  Once you understand that there are two possible sources of
changes to a given register in the slave (the master and the slave's own
programming), that the slave's programming typically runs many times faster
than the master's, and that the master typically reads back the actual
slave register on writes to update displays etc., you can make it work.

Having said that, we have "lost" data with Integrator 30s as well.  We
found out, after a day spent with an Invensys application engineer, that an
Integrator 30 does not have a lot of horsepower, and that certain
configurations carry a lot of overhead that can bury it.  We hope to get an
opportunity to optimize our configuration and fix these problems on one of
our I30s.  If this does not work, we will probably look at the AW70 w/OPC
client.  We don't use 70-series boxes, but since this will be a
single-function machine, we are amenable to trying it.  It's capacity is
vastly greater than an I30's; its only real disadvantage is that it is not
redundant, but that will not be a huge drawback in our application.

I'd love to see an Integrator 60, or better yet a RISC- or Pentium
3/4-powered Integrator(and CP :) as well; that's ultimately the most robust
solution.  But I'd think you could do bidirectional communication with an
AW70 just fine, if redundancy was not critical.

Corey Clingo
Sr. Engineer
BASF Corp.



|---------+---------------------------->
|         |           Glen Bounds      |
|         |           <gbounds@bellsout|
|         |           h.net>           |
|         |           Sent by:         |
|         |           foxboro-bounce@fr|
|         |           eelists.org      |
|         |                            |
|         |                            |
|         |           08/15/2002 01:54 |
|         |           PM               |
|         |           Please respond to|
|         |           foxboro          |
|         |                            |
|---------+---------------------------->
  
>-------------------------------------------------------------------------------------------------------------------------------|
  |                                                                             
                                                  |
  |               To:  foxboro                                                  
                                                  |
  |        cc:                                                                  
                                                  |
  |        Subject: Re: [foxboro] Foxboro and Allen Bradley SLC 504             
                                                  |
  
>-------------------------------------------------------------------------------------------------------------------------------|





The whole issue of Allen Bradley "communicating" with I/A has been
a dark spot on Foxboro since the inception of I/A.  And OPC does
nothing to brighten this tarnish.  By definition, OPC requires a Server
and a client, Master and Slave, or whatever the designers want to call
it.  But to communicate with a PLC, you must have a seamless, bi-
directional communication.  You can get fairly close with OPC, but the
possibility of loosing data that was sent from the "slave" is great.  The
Master will write over the slave's data as if it had never changed.
Whether you make the I/A the server, or the PLC, you WILL loose
data if you need to have bi-directional communications using OPC.
The I/A gateway (Integrator 30, etc) is the only way to go if you have
the need for bi-directional communication.  Period.

Life is too short not to have an Integrator 60!
Glen Bounds
Corn Products International






 
 
_______________________________________________________________________
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: