6 replies to this topic

[JamieG]

Hello - I am looking for a little guidance on calculating a relief load on a distillation tower during reboiler tube rupture. Ill preface by saying that I am by no means performing the "official" calculations. I just have some doubts about the existing calculations and want to check their validity. I also dont perform a lot of relief valve calculations so please excuse me if this has been covered ad nauseum.

 

The condensed question is: can I take credit for the overhead condenser during a reboiler tube rupture scenario?

 

In more detail:

The tower normally runs at 29 inHg vacuum. We are separating a mixutre of specialty amines that have a latent heat ~200 Btu/lb. We supply 225 psig steam to a control valve then to the reboiler.

 

Without going into great detail about determining the steam flow across the ends of the tube ruptures, can we use the condenser to reduce the required relief load? If so, do we also need to add the normal reboiler duty? Or do we simply use the volumetric displacement from the steam?

 

The existing calculations have taken credit for the condenser and there are some assumptions that bother me. The heat transfer coefficient during normal operating conditions is ~30 Btu/hrft2F. The heat transfer coefficient during relief conditions is closer to 90 Btu/hrft2F. The LMTD is also much larger at reflief conditions. The composition is also largely steam in these calculations. This has led to a condenser credit that is almost 10 times the normal operating duty of the condenser. It just doesnt seem right that the condenser can be given this much credit, completely negating the need for a PSV for this scenario.

 

Does anyone have an article they can reference or provide guidance on normal practice? API 521 doesnt provide guidance as far as I can tell.

 

Thanks in advance,

Jamie

 

 

[JamieG]

Anyone?

[breizh]

http://www.separatio...s/POT_Chp06.htm

 

Jamie G

Consider this document ! It may help you .

Calculation note of the column , trays  ( mechanical resistance) should be part of the analysis to give credit to the mitigation . To me PSV should be required to protect the equipment .

 

Good luck .

Breizh 

Edited by breizh, 26 January 2015 - 01:15 AM.

[gegio1960]

Jamie G,

 

My considerations:

1) a psv should be installed anyway;

2) the controlling scenario is probably the fire case;

3) the system pressure drop during the relieving could be quite high;

4) a dynamic simulation could be done to verify the various hypothesis.

 

Moreover,

5) The set pressure and design pressure of the steam should be not very different;

6) the tower should be designed for full vacuum;

7) vacuum conditions could be controlling on metal thicness;

8) systems of this kind are eligible to be designed as intrinsically safe.

 

Good luck

[gegio1960]

an article relevant to dynamic simulation of HE tube rupture has been published by Chemical Engineerig on Feb 2015.

[Zauberberg]

Also investigate if there are any actions from the SIS (system shutdown, unit/equipment shutdown) once when the tube rupture conditions are established, which would have a major effect on whether you should consider performance of any equipment in such a scenario.

[JamieG]

Thank you for your replies. I want to clarify that the PSV already exists and is sufficient for all scenarios per the existing calculations. I just question the validity of the tube rupture calculations because they show that the condenser negates the entire tube rupture scenario.

I believe that the dynamic simulation is the correct way to verify this scenario. IUnfortunately I have no experience in dynamic simulations and was hoping that there are some rules of thumb.

Zauberberg - you hit the nail on the head. The reason I am investigating this is because I am performing a LOPA on this scenario. I have decided to be conservative and assume that the PSV wont protect the vessel for a tube rupture. I will install a SIF to get the appropriate IPL.

 

Thank you again for your help.

Jamie