Jump to content



Featured Articles

Check out the latest featured articles.

File Library

Check out the latest downloads available in the File Library.

New Article

Product Viscosity vs. Shear

Featured File

Vertical Tank Selection

New Blog Entry

Low Flow in Pipes- posted in Ankur's blog

Alarm Settings For Flow


This topic has been archived. This means that you cannot reply to this topic.
6 replies to this topic
Share this topic:
| More

#1 Padmakar Katre

Padmakar Katre

    Gold Member

  • ChE Plus Subscriber
  • 992 posts

Posted 11 September 2008 - 09:43 AM

Dear All,
I have confusion about the alarm settings values to be given in the flow instruments. I am currently working on a project which is to be designed for overdesign of 120% and the turn down of 40 %. While deciding the flow alarms what values I should select for lo alarm (FAL) and lo-lo Alarm (FALL)?
As the turndown is 40 %. So I have given 45% of the normal flow as FAL and 35% of the normal flow as FALL. For FAL 45% setting can be justified but there is confusion regarding FALL setting. Any comment on this issue will be appreciated. Thanks in advance.

#2

  • guestGuests
  • 0 posts

Posted 11 September 2008 - 10:10 AM

Padmakar,

As one who spent the last 10 years as an operations engineer and have only recently gotten into Process Design, I definitely appreciate the design engineer giving thought to alarm points.

To answer your question, a little clarification on terminology may be required.

You say the design is 40% to 120%. Is this the expected range of "normal" operation? Or is this the flow basis for process equipment?

If the nominal design flow is say 100 gpm, but the process is expected to see operating rates under normal conditions at 40 gpm, I would not set FAL at 45%. This will lead to alarm trips when nothing is wrong and become a nuiscance alarm that will be quickly disregarded by operators and potentially distract them from responding to other actual alarm conditions.

However, if other process equipment, say a feed pump, is designed for a minimum 40 gpm before cavitating or causing other equipment damage then your choice of 45% and 35% are probably good choices. This gives a window (between 45% and 35%) where the operators have an opportunity to respond before software (interlocks) take effect and allow them to avoid process interuptions while preventing equipment damage. Obviously this will depend on the severity of the effect of low flow, if you are really worried about damaging equipment you could also make FAL 50% and FALL 40%.

Hope this helps.

#3 Padmakar Katre

Padmakar Katre

    Gold Member

  • ChE Plus Subscriber
  • 992 posts

Posted 11 September 2008 - 10:38 AM

QUOTE (Merrill @ Sep 11 2008, 08:40 PM) <{POST_SNAPBACK}>
Padmakar,

As one who spent the last 10 years as an operations engineer and have only recently gotten into Process Design, I definitely appreciate the design engineer giving thought to alarm points.

To answer your question, a little clarification on terminology may be required.

You say the design is 40% to 120%. Is this the expected range of "normal" operation? Or is this the flow basis for process equipment?

If the nominal design flow is say 100 gpm, but the process is expected to see operating rates under normal conditions at 40 gpm, I would not set FAL at 45%. This will lead to alarm trips when nothing is wrong and become a nuiscance alarm that will be quickly disregarded by operators and potentially distract them from responding to other actual alarm conditions.

However, if other process equipment, say a feed pump, is designed for a minimum 40 gpm before cavitating or causing other equipment damage then your choice of 45% and 35% are probably good choices. This gives a window (between 45% and 35%) where the operators have an opportunity to respond before software (interlocks) take effect and allow them to avoid process interuptions while preventing equipment damage. Obviously this will depend on the severity of the effect of low flow, if you are really worried about damaging equipment you could also make FAL 50% and FALL 40%.

Hope this helps.


Dear,
First of all thanks for your immediate reply. Let me clear you that the plant is to be designed to operate at 120 % on higher side and 40% on lower side of its capacity. Now as you say that the FAL if I will set at 45% it will lead to trip as this is not the case. My FALL will lead to the trip or will activate the interlock. Now if my all equipments are desigend to run within the span of 40 to 120 % of the normal flow/capacity then I will set my FAL just above the 40% flow it could be either 5-10 % more flow value say 45% to 50% as FAL. As my all equipments are capable of running at more than 40% load so I need not worry about the offset like off spec products or anything else.
But once my flowa will start falling 40% of the normal value then I could face issues with equipments like Pumps and Columns. So I will set the FALL just below the allowable limit i.e. 40%. But I am really confused in selecting the the FALL setting what it should be either 40% or less than 40%?
Thanks.

#4

  • guestGuests
  • 0 posts

Posted 11 September 2008 - 02:00 PM

In my opinion, from an operations standpoint:

FAL - should indicate an abnormal flow condition that if left unchecked will lead to negative consequences. FAL should require a human (operator) response to remedy the situation

FALL - shoud indicate imminent issue and as you say trigger an interlock

My point was that if the process could be expected to operate at 40% of flow under a possible "normal" operating scenario, setting the FAL at 45% would cause the alarm to come in when nothing is abnormal and the operators will simply acknowledge and clear the alarm. Doing this over and over eventually conditions them to ignore the alarm as a nuisance alarm.

Set the FAL at or slightly below what you would consider the lowest possible operating rate. Set the FALL somewhat below that to give operators a chance to respond before the software takes over.

#5 Qalander (Chem)

Qalander (Chem)

    Gold Member

  • ChE Plus Subscriber
  • 829 posts

Posted 11 September 2008 - 11:08 PM

QUOTE (Merrill @ Sep 11 2008, 02:00 PM) <{POST_SNAPBACK}>
In my opinion, from an operations standpoint:

FAL - should indicate an abnormal flow condition that if left unchecked will lead to negative consequences. FAL should require a human (operator) response to remedy the situation

FALL - shoud indicate imminent issue and as you say trigger an interlock

My point was that if the process could be expected to operate at 40% of flow under a possible "normal" operating scenario, setting the FAL at 45% would cause the alarm to come in when nothing is abnormal and the operators will simply acknowledge and clear the alarm. Doing this over and over eventually conditions them to ignore the alarm as a nuisance alarm.

Set the FAL at or slightly below what you would consider the lowest possible operating rate. Set the FALL somewhat below that to give operators a chance to respond before the software takes over.


Dears,
From operating personnel's viewpoit/responses
    especially in rotating shifts 'Merrill's referred nuisance'is true and At times they tend to over ride the alarms thus jeopardizing the safety proctection
      This scenario was partly instrumental in the recent (March 2005 Fatal Catastrophe of BP Ameria Texas refinery icident refer CSB and Baker's Reports)
      Hope this helps
      Regards
      Qalander

      #6 Zauberberg

      Zauberberg

        Gold Member

      • ChE Plus Subscriber
      • 2,728 posts

      Posted 12 September 2008 - 01:45 AM

      Hello Padmakar,

      Just to put few more thoughts from my side, based on my experience as Operations/ Technical Service engineer:

      1. In general, setting FAL and FALL values is dependent on the system itself, and it should be calculated by taking into account available buffering time of each component in a given system. It means you have to go through the entire equipment train and examine the consequences of running equipment below its turndown capacity - in conjuction with available operator response time, which is different for each peace of equipment.

      2. Speaking in that way, you are not actually interested in % of flow, but rather in time required for an operator to respond before you are running into unstable and potentially dangerous regime, and definitely before safety interlocks/PSDs are activated. For example, setting the FAL on the pump suction vessel at 45% is unreasonably high, if it results in more than 1-3 minutes available response time before corrective action. On the other hand, setting FAL at 60% is insufficient if vessel volume and LT span/position are such that the difference between 40% and 60% is only 45 seconds.

      3. Similar approach should be adopted when looking at the difference between turndown capacity and FALL (PSD) values: criticality and consequence severity of each component operating below its minimum rated capacity should be thoroughly evaluated, and available operator response time extended as much as possible - without jeopardizing plant safety - before the protection system is activated.

      I believe most of the Operator companies should have something which is called "Alarm management" (in order to avoid "blind alarm acknowledgment", as explained by Merrill and Qalander), and similar documents are also developed by DCS providers. I know for sure that Honeywell has one such manual developed for TDC2000/3000 systems.

      #7 Padmakar Katre

      Padmakar Katre

        Gold Member

      • ChE Plus Subscriber
      • 992 posts

      Posted 12 September 2008 - 07:44 AM

      Dear All,
      Thanks a lot for your valuable comments.




      Similar Topics