Instrumentation Problems

[Concepts12]

We have a system in which the High High and Low Low pressure alarms on a pipeline were disabled. The process is such that when the level in a tank upstream of the pocess goes low, the valve closes and the pressure in the downstream line decreases. This causes the Low Low pressure alarm to trip incorrectly. Thus it has been disabled from the system.

Is there a way that the Low Low alarm can be put back in the system without incorrectly tripping in the situation described above.

Any help will be greatly appreciated.

 

[Ashereng]

I am guessing that the HH and LL are for line over pressure and line break.

If so, one solution would be to interlock the LL alarm to the pump. If the pump isn't running, then the LL alarm is disabled. I am also assuming you are pumping.

If you don't have a pump, then interlock the LL alarm to the control valve such that if the control valve is closed, then the LL alarm is disabled.

"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[JimCasey]

>>when the level in a tank upstream of the pocess goes low, the valve closes and the pressure in the downstream line decreases. This causes the Low Low pressure alarm to trip incorrectly.<<

Put a limit switch on the valve. Wire the limit switch in with the pressure alarm. So if the valve is open AND the LoLo pressure alarm occurs then you trip the unit. If the valve is closed you have demonstrated that the LoLo alarm occurs, but that it is not a cause for concern.

 

[Concepts12]

Thanks for the help guys, I included your suggestions but encountered the following problems.

We have decided to include the interlock and applied the following conditions:
Valve Open, Pump running, Pressure below a certain value and Flow Rate above a certain value - Low Low Trip occurs since these conditions indicate a rupture.
However in applying these conditions to the data obtained from previous reading there are instances where all these conditions are fullfilled but no pipeline rupture occurs causing a Low Low trip.

Therefore we need some way of detecting a rupture without Low Low tripping incorrectly.

Any help will be appreciated.

 

[Ashereng]

What is the LL trip set at? Perhaps it is too high?

Is your LL pressure switch downstream of your valve?

Here is another thought - maybe you have too many inputs to your interlock logic? If the valve is closed, then the LL should be disabled. If the valve is not closed, then the LL should be active.

"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[ADMIRALKEN]

I would guess your false trips are at start up or shut down.
Add a small time delay to your logic. If all the conditions are met for say 5-10 seconds, then trip.

 

[Concepts12]

Thanks guys for all the help. The suggestions you all made were very helpful but when applied to the data at hand didn't work.
The thing is that we have data with readings that would suggest a ruptured pipeline occurred.
Therefore any of these suggestions we implement would reduce the number of times a Low Low occurs incorrectly but a Low Low trip will still occur.

Any suggestions on how we can put things in place to detct a pipeline rupture without causing a Low Low trip will be greatly appreciated

 

[Ashereng]

Any suggestions on how we can put things in place to detct a pipeline rupture without causing a Low Low trip will be greatly appreciated

Maybe a more complete description of the various scenarios where this is happening?

"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[danw2]

Won't a rupture produce a high negative rate of change pressure alarm?

 

[don01]

hi all
It is implied that you have analog inputs for both flow & pressure . If not get it.
From this point the spread of pressure / flow rate is quite calculatable.
If you have a pipe rupture the pressure will fall below a deviation alarm and alert your operations. If the pipe is blocked or crushed by an alien space craft (or similiar event) then the pressure /flow will go high.

Mask the pump stop condition / valve closed and you should cover most events. (this is all assuming measurement at the supply end. If it as the delivery end then the inverse is the case.) Using a deviation alarm will give you "tunability".

If your customer (internal or external) comes up with another couple scenarios for you to solve then I suggest you walk away and contemplate trees falling in the forest .

but above all enjoy life & the experience
Don

 

[djack77494]

If your low-low pressure events are short term transients, you can dampen them out.

 

[controlnovice]

The process is such that when the level in a tank upstream of the pocess goes low, the valve closes and the pressure in the downstream line decreases. This causes the Low Low pressure alarm to trip incorrectly.

Add the following condition to the Low Low Pressure alarm:

If valve is closed, then disable alarm.
Do not use this if you can get a rupture with the valve closed.

Your statement conditions below seems 'backwards':

We have decided to include the interlock and applied the following conditions:
Valve Open, Pump running, Pressure below a certain value and Flow Rate above a certain value - Low Low Trip occurs since these conditions indicate a rupture.

______________________________________________________________________________
This is normally the space where people post something insightful.