3 replies to this topic

[Ghasem.Bashiri]

Dear all
Is there any general guideline to fill alarm and set point list document.
As you may know we should give data about high pressure, high high pressure, high temperature, high high temperature. But in process information we usually have normal pressure and temperature and design pressure and temperature.
Rehards
A.Bashiri
Ghasem.Bashiri@gmail.com

[Padmakar Katre]

Dear,
This is not a general comment which I am going to say here.The process package what we issue to the client.It consists one document i.e. Opertaing guidelines where we cover starting with the plant description to the main following points we cover are (as a bare minimum requirement and not limited to)
1.Precommissioning Activities
2.Start-up and Stand-by Conditions
3.Normal Plant Operating Guidelines.
4.Normal Shutdown
5.Emergency Shutdown (In which we consider all the probable reasons which can lead to plant upsets like cooling water failure,power failure and so on...)
In above point No. 4 where we give the Set point values of the normal running plant parameters.These are nothing but the Set Points of your controllers and the normal values of all the parameters (Open loops as well as closed loops) Along with this we give the operating margins and based on this, the alarm values.
So this is quite important as client may or may not know the technology in this case you are the one who knows the entire information so it is always a good practice to give the alarm and trip point settings for all the parameters along with the normal operating values.
If it doesn't convince to you please ignore my comment.

[JoeWong]

No straight-forward answer...all subject to your system design, hardware limitation, operation limitation, available buffer,...

Set point for start-up may needs to be readjusted once system already in operation...

Arbitrary...
HIGH setpoint is at 50% gap between Normal to Design
HIGH HIGH setpoint is at 50% gap between HIGH setpoint to Design

JoeWong

[Adriaan]

Any alerts need to be useful and significant. If a system keeps on giving the same alert under normal or near normal conditions the users will ignore that alert and it will therefore be worse than useless. Ideally and limits should be user changeable within limits, those fixed limits representing safety limits (with a reasonable margin).

Setpoints will always need updating, even more than alerts.

Having said this... with alerts - for obvious reasons - the tendency is for users to WIDEN the limits; it is adviseable to regularly tighten alert limits (if possible) or use frequent alerts as an indication for the need to change something in the process (like the need for a controlvalve with a somewhat higher capacity for example).