Re: [foxboro] Check against entering ridiculous values

Adam,

There was a discussion of this topic in April of 2002 which is probably
captured in the archives.  The subject line was "Preventing bad setpoint
changes".

There were quite a few suggestions including mine which is repeated here...

--------------------------------------------------------------------------
Another possibility that I haven't seen mentioned is to use the SPROPT,
SPTARG, SPRATE, & SPRAMP options of the PIDA along with a supporting CALCA
block.  The SPTARG value doesn't go to the setpoint until SPRAMP is
transitioned from 0->1.  You could use the CALCA block to see if the
setpoint jump is OK (i.e. ABS SPTARG - SPT <= some limit) and drive SPRAMP
as needed.  If the setpoint jump is too high, you could use the CALCA to
enable a field on your custom PIDA overlay to allow the operator to enable
the new setpoint anyhow (use RCL so that the override is only good one
time).  The CALCA logic should be pretty simple, all required PIDA
parameters are standard, the mechanism only affects the Local setpoint, and
you get a setpoint ramping feature thrown in as well.  (If you don't want
setpoint ramping, just make SPRATE high enough to jump the setpoint in one
cycle).
--------------------------------------------------------------------------
Regards,

Doug Lloyd
DGL Controls


-----Original Message-----
From: foxboro-bounce@xxxxxxxxxxxxx [mailto:foxboro-bounce@xxxxxxxxxxxxx] On
Behalf Of Adam.Pemberton@xxxxxxxxxxxx
Sent: Monday, October 16, 2006 5:51 PM
To: foxboro@xxxxxxxxxxxxx
Subject: Re: [foxboro] Check against entering ridiculous values

Thanks Rocco
   I wasn't aware of the SPHLIM and SPLLIM parameters and so thank you.
A bit embarrassing really. I thoroughly checked out the PID but not the
PIDA. 

However in this case these limits won't be enough on there own since the
issue with the autoclaves is that the required setpoint range is quite
high, but the short term change must be limited.

   I like your two possible options, that of 1) limiting the setpoint
based on prevailing conditions and 2) using a special overlay. 

I can see how to limit the setpoint but don't know how to implement this
special overlay. Could you please elaborate? I guess I can't see where
the checking code would go.

 

Regards

Adam

 

-----Original Message-----
From: foxboro-bounce@xxxxxxxxxxxxx [mailto:foxboro-bounce@xxxxxxxxxxxxx]
On Behalf Of Penndorf, Rocco (RP)
Sent: Monday, 16 October 2006 4:05 PM
To: 'foxboro@xxxxxxxxxxxxx'
Subject: Re: [foxboro] Check against entering ridiculous values

 

 

Adam,

 

there are certainly some options how to accomplish this. The most simple
one might be to clamp the range (using SPHLIM, SPLLIM parameters of
PIDA) where you can enter a setpoint. But this approach does not provide
much flexebility in terms of operation of this controller. On some
occurances you might want to be able to enter a very low setpoint, what
is then not possible. 

The second option takes care of this. You also can make this clamp
depending on operating conditions, but this would require an additional
CALC block, what changes the SPHLIM and SPLLIM parameters.

In our plant we are using a special PIDA overlay, what checks if the new
entered setpoint deviates more than 10% from the old setpoint and
notifies the operator.

 

 

Best Regards

Rocco 

 

-----Original Message-----

From: foxboro-bounce@xxxxxxxxxxxxx [mailto:foxboro-bounce@xxxxxxxxxxxxx]
On Behalf Of Adam.Pemberton@xxxxxxxxxxxx

Sent: Monday, October 16, 2006 2:04 AM

To: foxboro@xxxxxxxxxxxxx

Subject: [foxboro] Check against entering ridiculous values

 

 

We had an incident on Saturday where an operator accidently entered a
very low setpoint (5kPa when it should be >2200 kPa) on the pressure
control for one of our Autoclaves. This caused the vent valve to open to
100% and the clave to rapidly depressurize. This is a potentially
catastrophic incident since high venting in an Oxygen rich environment
can cause a major fire.

 

 

We are at the final stages of a migration from Display Manager to
FoxView. In order to help these guys with the transition, we have
provided them with one Exceed XTerm running Display Manager off our
engineering development AW.

 

 

 

This incident illustrated a number of issues :

 

   - The operator claimed that the FoxView workstation he was working on
(Blade 150) locked up and he had to move to Display Manager on the Xterm
to correct the erroneous value.

 

   - We have disabled the arrow keys on FoxView while we wait for 9.0.2.
Using the arrows on some parameters gives very large changes.

 

   - We need a good way to prevent ridiculous values being entered in
situations such as this.

 

 

 

My question from all this is what have people in this group done to
resolve the issue of preventing operators from entering dangerous
values? One suggestion was to have a popup appear if the change is
greater than 200kPa (for example). Is there an easy way of doing this?

 

All our workstations are Sun Blade 150's or PC's running Exceed (usually
V7). I/A is version 7.1.1. FoxView is version 9.0.1

 

 

 

Adam Pemberton

Site Electrical & Control Systems Engineer Lihir Gold Limited

 

Ph: +675 9865 655

Fax: +675 9865 666

Mob: Nogat

Postal:

   Australia: GPO Box 905, Brisbane, QLD 4001

   PNG: PO Box 789, Port Moresby, NCD

 

 

 

 

 

 

_______________________________________________________________________

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

 

 

Re: [foxboro] Check against entering ridiculous values


 
 
_______________________________________________________________________
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: