4 replies to this topic

[arianshahi]

Dear all,

 

I would appreciate, if you could help me out with my concern regarding the required relieving rate in case of control valve failure,

 

Is it against API 520/521 standards to take the influence of piping and fittings from the overpressure source (control valve failure in this case)  to the protected vessel to calculate the required relieving rate?

 

I have done a rigorous dynamic simulation (including hydraulic calculations through process piping) to study the impact of control valve failure at 100% open position on vessel overpressure. Control valve failure is definitely a credible scenario in this case. However, there is a T-fitting between the control valve and the pressure vessel, which will branch the flow to two different paths. Based on my dynamic study, in case of failure, 80% of the maximum flow will naturally go towards the safe path and only 20% of it will cause overpressure in the vessel. 

 

Piping pressure ratings are all safe and above MAWP and above upstream pressure of the control valve. I am trying to dig into the codes and standards to find if I am allowed or not to take this type of study into consideration as without considering the piping and T-fitting effect there is gonna be a massive difference in sizing.

 

Please note that valve is normally closed, so there is not such normal capacity credit for this case, and mostly I am focusing on the influence of piping after open failure.

 

Thanks a lot in advance.

[ankur2061]

Hi,

 

If you have a branched open path (the key word is "open") then obviously the flow will split at the "Tee" junction. However, when you are talking of overpressure of the vessel, it can still be over-pressurized but the time taken to reach overpressure would be longer since the flow rate to the vessel is reduced.

 

The relief rate in your case would be the maximum flow rate due to control valve failing open minus the flow rate through the open path at the "Tee" junction.

 

Open path means that the flow to the branch cannot be isolated by any means such as a manual block valve or a spectacle blind. If a manual block valve or spectacle blind provision is there, then it cannot be considered as an open path.

 

Regards,

Ankur.

[fallah]

Dear all,

 

I would appreciate, if you could help me out with my concern regarding the required relieving rate in case of control valve failure,

 

Is it against API 520/521 standards to take the influence of piping and fittings from the overpressure source (control valve failure in this case)  to the protected vessel to calculate the required relieving rate?

 

I have done a rigorous dynamic simulation (including hydraulic calculations through process piping) to study the impact of control valve failure at 100% open position on vessel overpressure. Control valve failure is definitely a credible scenario in this case. However, there is a T-fitting between the control valve and the pressure vessel, which will branch the flow to two different paths. Based on my dynamic study, in case of failure, 80% of the maximum flow will naturally go towards the safe path and only 20% of it will cause overpressure in the vessel. 

 

Piping pressure ratings are all safe and above MAWP and above upstream pressure of the control valve. I am trying to dig into the codes and standards to find if I am allowed or not to take this type of study into consideration as without considering the piping and T-fitting effect there is gonna be a massive difference in sizing.

 

Please note that valve is normally closed, so there is not such normal capacity credit for this case, and mostly I am focusing on the influence of piping after open failure.

 

Thanks a lot in advance.

 

arianshahi,

 

To get an accurate and proper response it's better to upload a simple sketch of the system you described...
 

[arianshahi]

 Screen Shot 2016-06-18 at 10.28.48 PM.png   63.97KB   2 downloads

 

 

Dear Fallah,

 

Here I attached the general sketch, focusing on the influence of t-fitting and open path after control valve failure.

Edited by arianshahi, 18 June 2016 - 10:26 PM.

[Pan Nata]

Dear Arianshahi,

 

As far as I know there is no industrial standard discuss about your case..

 

In my point of view, in the event of regulator failure, you also have to take credit for PSV (PSV2771) installed on the path (branch).

 

good luck