Re: [foxboro] Sequence code (MON, DEP, IND) opinions

  • From: "Moore, Kenneth, Celanese/US" <ken.e.moore@xxxxxxxxxxxx>
  • To: <foxboro@xxxxxxxxxxxxx>
  • Date: Mon, 14 May 2007 15:04:20 -0500

 I work at batch chemical facility and we use a lot of sequence blocks.
Our system is modeled on the ISA S88 standard, so we have units, phases,
etc....

Our CP loading is quite heavy at times, depends on the number of phases
running at any given moment. We typically have to use more CP's than a
continuous process would require.
But...one of the keys to reducing loading is only run the blocks when
they are needed. By using a EXC block, the only thing that runs
continuously is the monitor, it then calls the EXC and the other
sequence blocks are executed as needed.

Sequence blocks are very powerful, and are ideal for sequenced control.=20
I've taken a look at the new Allen Bradley control Logix processors,
specifically the structured text. The AB offering is missing one
critical instruction the: "WAIT UNTIL".


Regards,

Ken Moore

Process Controls Specialist
Celanese Emulsions
Enoree, SC 29335


-----Original Message-----
From: foxboro-bounce@xxxxxxxxxxxxx [mailto:foxboro-bounce@xxxxxxxxxxxxx]
On Behalf Of Corey R Clingo
Sent: Monday, May 14, 2007 2:40 PM
To: foxboro@xxxxxxxxxxxxx
Subject: [foxboro] Sequence code (MON, DEP, IND) opinions

I see a significant amount of discussion on here regarding use of these
blocks and sequence code.  Not wanting to ignore a potential tool in my
toolbox....OK, really I'm just lazy* and do not want to pass up an
opportunity to get a computer to do something instead of me.

Some background: my plants are continuous processes.  I inherited
systems in both plants.  Neither system had any sequence code in it -
nada.  I had heard all the horror stories about sequence code loading
CPs, issues with memory fragmentation, etc. etc., so I never took a big
interest in learning it.


I then had an application, non-control, where a large number of complex
calculations were being performed.  I felt it was a good application for
sequence code, so I built an IND block and got after it.  It worked
fine, but at the default BPCSTM of 100, it caused a CP30A's idle time to
drop by over 30% (it was about 50 lines of code, with a WAIT loop back
to the beginning).  I set BPCSTM to about 10, and the idle time problem
went away, but I only got away with it because this application did not
need anywhere near real-time performance.  I sgain shied away for using
it do do anything that did have such requirements.


Which is a shame, really.  I used the sequence equivalent fairly
frequently in my Brand H days, partly because they did not really have a
CALC* equivalent (you had logic, or math, but not both in one block),
partly because some flavors of their sequence-equivalent had wide system
access that I needed.  I don't think I'd go crazy with it on I/A, but
there are 2 or 3 applications where it would be nice to have if it
didn't have the issues I saw.


So how do others use sequence on I/A?  Batch only?  How do you get
around the CP loading issues?  What other issues do you have?  How do
you get good performance and reesponsive interaction with a human
operator?  Vi or emacs? :)


Thanks,
Corey Clingo
BASF Corporation




=20
=20
_______________________________________________________________________
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
=20
foxboro mailing list:             //www.freelists.org/list/foxboro
to subscribe:         =
mailto:foxboro-request@xxxxxxxxxxxxx?subject=3Djoin
to unsubscribe:      =
mailto:foxboro-request@xxxxxxxxxxxxx?subject=3Dleave
=20
 
 
_______________________________________________________________________
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:             //www.freelists.org/list/foxboro
to subscribe:         mailto:foxboro-request@xxxxxxxxxxxxx?subject=join
to unsubscribe:      mailto:foxboro-request@xxxxxxxxxxxxx?subject=leave

Other related posts:

  1. Home
  2. foxboro
  3. Archive
  4. 05-2007