4 replies to this topic

[AymanAyadi]

In order to reduce the risk level in one of our satellite platform, we intend to adopt a total blowdown philosophy instead to the partial one.

i.e: Process area is divided in tow fire zone: GL and Production. In case of occurrence of fire in one zone only this later is depressurized and the second one remain under gas.

We run the depressurization module in aspen hysys for three credible scenarios: fire case, adiabatic & minimum temperature.

 

Results show:

 

With the actual configuration and without restriction orifices on blowdown lines:

-       Depressurization time in each separator is less than 3 min!!

-       Total blowdown cannot be adopted: peak flow of one separator exceed the HP flare tip capacity!!

-       Fire case gives us the highest peak flow.

 

With restriction orifice designed to get depressurization according API RP 521 criteria:

-       Fire case gives the highest peak flow.

-       Blowdown load (sum of peak flow from each vessel) is less than the HP flare capacity.

 

My questions are:

-       There is other credible scenarios that shall be studied in our case? (note: only separators are installed in the platform and the HP flare tip was designed for the relief case “blocked outlet scenario in the main production separator”)

-       Can we assume the blowdown load as the sum of peak flows?

-       Shall we run simulation on ASPEN FLARE ANALYZER to confirm backpressure is not exceeding MABP during blowdown? 

Best regards

[shan]

Check Mach Number of your flare header line for the peak flow.

[Zauberberg]

- No one here can confirm if you have covered all relevant relief scenarios. I think you should invite your Instrument engineer and sit together with Operations personnel, and go through the scenario checklist which you can find in API 521. The checklist and the experience of Operations are crucial in developing the relief load list. Not more, not less.

 

- This again depends on activation of blowdown valves - see if there is any delay between opening of the relevant BDV's in the high/maximum load scenarios. If there is some delay, you may not need to assume simultaneous flow from all BDV's.

 

- Ideally: yes, and this model should account for actuall piping arrangements (isometrics), piping dimensions, and all valves/fittings in the relief network. This is how a reliable and accurate model is built.

[flarenuf]

hi ayman

 

i agree with Dejan on the first point

 

Can we assume the blowdown load as the sum of peak flows?
 

historically this was the way it was done , add all three together and thats your peak load to size the flare system for.

In reality  this peak load i) only lasts a few seconds  and  ii) is not really seen at teh tip as no account of line packing is credited for
 

nowadays looking at these loads dynamically can usually show much reduced flows.

 

Shall we run simulation on ASPEN FLARE ANALYZER to confirm backpressure is not exceeding MABP during blowdown?

 

There is little chance of you exceeding the bP on the BD orifices themselves as you will initially have critical flow across them
however you may see a superimposed bP on some other non flowing valves that may cause you concern
So yes it is always good practice to run the dP case.

 

Flarenuf

 

 

 

-       Shall we run simulation on ASPEN FLARE ANALYZER to confirm backpressure is not exceeding MABP during blowdown?

[AymanAyadi]

Thanks All,