from the Forum department...
Process Oscillations
 Posted by engineer on 10 August, 2012 - 5:28 am
hi,
consider the sitation:
You see a badly tuned loop, its (pv)oscillating
(1). now, you do a step change, it again oscillates(the response)
(2). now, is there any correlation between 1 & 2, whatsoever?

needless to say, its all in auto.

 Posted by Rob on 10 August, 2012 - 7:51 am
Put it in manual. Does it stop oscillating?

There are lots of reasons why a loop will oscillate. While bad tuning is a likely cause its not the only one. Worn valves can and will do the same thing, or maybe its oscillating because something else in the plant is driving it.

Rob
www[.]lymac.co.nz

 Posted by Bruno on 10 August, 2012 - 8:19 am
Hi
what are you waiting for?

no indication of process, of control type (L, T, P, F or other), of regulator type (P, PI, PID or other), of value parameter.

 Posted by engineer on 10 August, 2012 - 11:49 am

consider the situation,
You see a loop badly tuned, its (pv)oscillating(oscillation no.1). Now, you do a step change; it again oscillates(the response)(oscillation no.2). Now, is there any correlation between 1 & 2, whatsoever?

Needless to say, its all in auto.

 Posted by engineer on 10 August, 2012 - 12:27 pm
thankyou rob and bruno,

yes, i know :)
but its a slightly different question,

The first oscillation relates to load/disturbance while the second is for load/disturbance+ set point. The interest is in the similarity in the characteristics of the oscillations.

 Posted by Rob on 10 August, 2012 - 5:15 pm
>but its a slightly different question

No, its not. Put the loop in manual. Does it oscillate?

> The interest is in the similarity in the characteristics of the oscillations.

When you add 2 sine wave together, you don't get another sine wave - you get a more complex interaction.

So if you make a step change in the setpoint and you still get the same sine wave in the PV then .....

Anyway, talk about sine waves is all theory. In reality, you have an oscillating loop in your plant. It most likely has one of three possible causes - bad tuning, bad actuator or another bad loop somewhere else. You need to track down the root cause. And putting the loop in manual is the first step to doing that.

 Posted by engineer on 12 August, 2012 - 12:48 pm
thank you again rob,

>So if you make a step change in the setpoint and you still get the "same" sine wave in the PV then...<

that was never implied!

 Posted by engineer on 13 August, 2012 - 12:21 pm
Rob, please see if I am correct (otherwise I might be directed to this very thread in future :), as we discussed, I put it in manual,

If it stops oscillating- Its bad tuning, if error(or better, cv checked negative in the field)- cv problems, if up/downs,oscillations- something else is driving. And, the driving loop may not be performing as bad as the driven loop.

 Posted by Rob on 16 August, 2012 - 6:49 am
If it stops oscillating - the problem is probably (almost certainly) in this loop. You cant say its a bad valve or bad tuning yet. You need to do some bump tests to determine the Hysteresis (backlash) of the valve, the stiction (stick slip) of the valve and some appropriate gains and time constants for the loop. Of course, you may also have two (or more) tightly coupled and interacting loops and to solve this you need more process knowledge and a bit of practical theory.

If you know what to look for in the combination of output moves and PV responses while the loop is in Auto you can also sometimes diagnose valve and actuator problems directly.

Forgive me, but I'm going to put the hard sell on you here ... You're spending a lot of time looking at the loop and trying to diagnose the problems. Meanwhile your plant isn't getting fixed. How much money are you loosing while you say you cant afford software & training to fix these issues ?

 Posted by engineer on 17 August, 2012 - 7:28 am
thank you rob,

i think you got me wrong. most of the questions that i have put in this forum arose during commissioning of new plants/units and overhauling/ modifications. its not about a single particular plant/unit. the control system also varied from case to case(from relays to dcs, all the way through). its usual to find more problems during this stage than once the system is well set.
if all these problems would have occured in a single system then...

talking specifically about the present problem, let me admit, there is no such a bad loop in front of me right now, just think about the prospects of an engineer who when faced with such a situation, posts this question in the forum(and puts lots of smileys in the messages too) and waits to get a suitable reply while in the meantime, the loop keeps on oscillating!

though not right now, i have faced these sort of problems over and over again(more so at times, when there is no previous history of the loop), the question was meant to develop newer strategies for the time, i am at it again.

Your use of this site is subject to the terms and conditions set forth under Legal Notices and the Privacy Policy. Please read those terms and conditions carefully. Subject to the rights expressly reserved to others under Legal Notices, the content of this site and the compilation thereof is © 1999-2013 Nerds in Control, LLC. All rights reserved.

Users of this site are benefiting from open source technologies, including PHP, MySQL and Apache. Be happy.

Fortune
"I am ready to meet my Maker. Whether my Maker is prepared for the
great ordeal of meeting me is another matter."
-- Winston Churchill
Help keep our servers running...