3 replies to this topic

[vvsastry66]

Dear Friends,

When we are using a continuous purging of flare header (with a automatic back-up source), do we need to install a water seal at stack bottom. Shell DEP says if water seal is not installed (for H2S falres), flare KOD has to be designed for 7 barg pressure. What H2S has got to do?

API 521 says only 3.5 barg. Does not talk of H2S.

Could any one pls through some light.

[ankur2061]

vvsastry,

Section 4.1.3.2- Design Pressure of DEP 80.45.10.10-Gen. states that a design pressure of 7 barg shall be used when no water seal vessel is used. The reasoning they have used is that in absence of a water seal vessel the KOD can safely withstand the overpressure due to an internal deflagration (flash back). A water seal vessel reduces the probability of a flash back and thus they say that if a water seal vessel is used the KOD may be designed with a lower design pressure of 3.5 barg.

Mind you, all of the above is the design philosophy adopted by Shell. If your client does not subscribe to Shell standards then you are under no obligation to go for a DP of 7 barg. You can simply state that a DP of 3.5 barg is recommended in API STD 521 is sufficient to take care of flash back issues.

The beauty of any engineering design is that in most cases there are multiple choices available to execute any design activity keeping in view the client's requirement, cost and the intended function of the designed equipment. Most of us tend to forget that and get stuck with what is written in any design manual or standard.

Hope this helps.

Regards,
Ankur.

[kkala]

Water seal drum offers protection (additional to purge gas flow) against atmospheric air ingression into the flare header through the tip. Mentioned air ingression could be created after a hot gas discharge, when header is getting cool, and could result in an explosion in the header containing flammable gases.
A labyrinth seal on flare stack also offers similar (additional) protection.
It is natural to call for higher header desigh pressure in case that no water seal drum is installed. My interpretation is that Shell considers internal explosion as a credible contingency in that case; and considers the theory that an overpressure due to explosion can be up to 8 times x operating pressure. The latter must be written in "Pocket guide to Chemical Engineering" by C. R. Branan (not at hand in my present location).
It is assumed that above is general for flare headers, not for the specific case of H2S. However if Shell requires it only for headers containing H2S, it may consider rupture of a header containing H2S (plus other flammable gases) as more dangerous.
It is noted that design pressure of 3.5 barg by API 521 should be considered as minimum requirement (actually being more or less minimum design pressure that a pressure vessel can have to support its weight, etc); there are flare systems designed for pressures higher than 7 barg, see http://www.cheresour...h__1#entry34880.

Edited by kkala, 12 August 2011 - 09:14 AM.

[Qalander (Chem)]

I endorse your analysis,

• 7barg design pressure for Flare KOD is better as it is wiser,as well as safer to ensure enhanced design integrated safety in-place.
• More over having said that I will still prefer to have water seal pot also provided in place downstream of KOD& upstream of Flare stack.
• Just a minor addition,Forum giants please correct me if wrong properly the KOD vessel should also be capable of withstanding certain external pressures(or slight internal vacuum;exact figures are not recalled right now)

Do Request all to correct my above gut feeling

Hope it prove helping.

Edited by Qalander (Chem), 08 September 2011 - 04:41 AM.