Emerson Exchange 365

• Not Answered

Skid Mounted Controller - Best Practice

Hi,

I have a controller mounted on a skid. This skid/controller will be powered down intermittently for anything from a few minutes to a few months. I am wondering what is the best practice for handling these situations. Should the controller be decommissioned prior to each power down or only for extended duration power downs (if so after how long does decommissioning become necessary) or is decommissioning necessary at all?

Thanks afor any advice on this subject

  

pobrien

• 17 Aug 2015 8:47 PM
• Cancel

4 Replies

• Patrick Truesdale

  

  • 18 Aug 2015 10:43 PM

  If you could please share a bit more details about what type, including model, will see what may be required.
  
  Regards
  Patrick Truesdale

  • Cancel
• pobrien

  

  • 20 Aug 2015 8:17 AM

  In reply to Patrick Truesdale:

  Hi Patrick,
  Thanks for the response. We are using DeltaV version 12.3.1. We have two examples of this setup, one using an MX controller, the other an SQ controller with redundancy.
  Regards

  • Cancel
• Rachelle McWright

  

  • 20 Aug 2015 3:13 PM

  Excellent question, pobrien. I think others could benefit from this knowledge. Can I ask you to post this in the DeltaV (Group)? I can do so, but by you doing so, you will get responses to your question in your inbox (which most folks find super handy). Visit the Process Control:DeltaV group HERE>>>

  Best Regards,

  Rachelle McWright: Business Development Manager, Dynamic Simulation: U.S. Gulf Coast

  • Cancel
• Youssef.El-Bahtimy

  

  • 21 Aug 2015 3:26 AM

  You have two competing concerns:
  1. If you just power down the controller, the peer-to-peer diagnostics of every other node on the DeltaV network will register the loss of communication with this peer every three minutes, filling the event chronicle with a lot of extra unnecessary information. More importantly, the network diagnostic status in the Operate screen of every HMI will show an abnormal status for those time periods.
  There is no way to tell DeltaV that a node is commissioned but out of service, so if you power it down, you will have to contend with this abnormal state condition unless you decommission the controller (the same goes for any local HMI workstation on the skid as well, except you can't decommission a worktstation, you must delete it)
  If you wish to leave the controller (and workstation HMI) running while the skid process equipment is shutdown (if your electrical isolation will allow for this), you'll probably still have to suffer with I/O errors from things like open circuits on control outputs.
  
  2. If you decommission the controller / delete the workstation, there's always that nagging risk of 'what if it doesn't come back' or what if there was an undocumented online change that did not get incorporated into the database.
  
  This has been a concern for mobile skid systems where communications via serial or VIMs is interrupted for a majority of time (these systems may only be in service a few hours a day or less and are moved out of the way when not in use), and unfortunately there's no good way to 'turn off' the I/O, controller, or hmi, as far as the DCS is concerned.
  
  It seems that a procedure-controlled decommissioning process is your best bet to keep from having a system with constant errors.

  • Cancel