5 replies to this topic

[Nasiruddin]

Friends,

 

I am working on open drain system, brief description is as follows:

 

Continuously contaminated open (CCO) drain header collect the drains from sources such as flange leakages, pump seals, pump base plates, sample points through tundish collection points. Essentially it is open drain system and drain header ends up in CCO drain vessel (located in pit). Schematic is attached for better understanding. As per the operating philosophy Level indicator is provided with high alarm (in control room) as well as local display for the operator on the vessel, once vessel is filled up to high level, operator will arrange vacuum truck to empty out the vessel. This drain system is completely isolated from other drain systems on the facility (closed drain, AOC drains, etc.)

 

Vessel design pressure is 700 kPag and on vent line a flame arrestor is provided. During design review a question was raised on the requirement of Flame arrestor whether it is required or not as the vessel design pressure can withstand detonation pressure. Secondly the possibility of vapor cloud is also less, as vessel collecting drain from tundish which is open to atmosphere and first flash of hydrocarbon liquid will be at the tundish itself.

 

What is your opinion on the requirement of Flame arresotr on this vessel. Whether any code requirement for flame arrestor installation on this type of vessels.

 

Regards,

Nasiruddin

Attached Files

•  CC Drain Vessel.xlsx   97.39KB   56 downloads

[ankur2061]

Nasiruddin,

 

Your drain vessel is designed for a DP of 7 barg. Additionally it is an open drain system and any flash would first hapen at the open tundishes. Considering that you may not require a flame arrester.

 

There are other reasons which do not favor flame arresters to be installed. Flame Arresters require regular inspection and periodic cleaning to ensure that they do not get clogged with dirt. There is also a potential of insect nests on the narrow slots of the flame arrester which can clog the flame arrester and lead to a potentially dangerous back-pressure build-up.  

 

Only if regular monitoring and periodic inspection / maintenance can be ensured should a flame arrester be installed. This may not be possible on remote unmanned or partially manned stations.

 

There have been accidents related to failure of badly maintained flame arresters. Refer the section "Safety" in the link provided below:

 

http://en.wikipedia..../Flame_arrester

 

Hope this helps.

 

Regards,

Ankur

[Nasiruddin]

Ankur,

 

Thanks for quick reply, it certainly help me to resolve this action.

 

Regards,

Nasiruddin

[Art Montemayor]

 

Nasiruddin:

 

Also, please refer to the following authoritative API document as reference:

 

Flame Arresters for Vents of Tanks Storing Petroleum Products

Downstream Segment

API RECOMMENDED PRACTICE 2210

THIRD EDITION, MAY 2000

 

“Most companies have accepted the premise that a tight steel roof and a pressure-vacuum valve provide appropriate protection and that any potential additional protection afforded by flame arresters does not warrant their installation in addition to a pressure-vacuum valve.  API Standard 2000 (4.4.1.2) and NFPA 30 (paragraph 2 - 3.5.6) state that a flame arrester is not considered necessary for use in conjunction with a pressure-vacuum valve where the tank is normally closed except when venting.  This is consistent with OSHA requirements in 1910.106((2)(iv)(f).

 

The conclusion of the API is that a systematic evaluation based on engineering analysis and tests, supported by experience, show that there is no technical or experiential basis for requiring that an outdoor aboveground petroleum tank provided with a pressure-vacuum valve must also be equipped with a flame arrester.

 

For practical safety considerations the use of flame arresters for these vents is discouraged to avoid tank damage resulting from the introduction of a new failure mode, unless the user is able to institute the flame arrester maintenance necessary to ensure that the required venting capacity is maintained.”

 

[fallah]

 

API Standard 2000 (4.4.1.2) and NFPA 30 (paragraph 2 - 3.5.6) state that a flame arrester is not considered necessary for use in conjunction with a pressure-vacuum valve where the tank is normally closed except when venting.  This is consistent with OSHA requirements in 1910.106( (2)(iv)(f).

 

 

 

Just minor correction: API 2000 (Sixth Ed.) isn't included such statement in its 4.4.1.2 section and the point is in 4.5.2 where beside referring to good experience with tanks protected by PVRV without flame arresters it is also mentioned that based on recent testing, a flame can propagate through a PVRV into the vapor space of the tank mostly due to lightning strike...

[Art Montemayor]

 Fallah has brought out a good point regarding a recent edition of the API 2000 and I would follow it up with the following comment(s) that I have been reserving for such an occasion.  I have been following some recent threads regarding the changing and sometimes difficult interpretations of API documents (especially API 2000) and I am concerned by what is happening.  I believe there is a misconception of what the "new" interpretations of API 2000 (and similar issues) mean.

 

I continue to apply those API document EDITIONS that I feel apply to my projects and that I feel confortable with.  I don't feel confortable with the latest editions (or versions) of API 2000 because I identify them as ISO versions written and edited for the European engineering world and, consequently, written with European needs to fulfill their local needs and requirements.  When my projects are not in Europe and not subject to European jurisdictions, I see no need to increase the amount of calculated vapor generated in order to size a PVSV on a storage tank.  As engineers we are not bound nor obligated to follow the "latest" API version of a standard or RP.  Just because it is the latest doesn't make it the best for the application.  We are free to cite the specific edition of the API we intend to follow and with our client's acceptance or concurrance, we can design to that specific edition.  There is nothing wrong in that.  Unless it can be proven that prior editions pose dangerous or damaging results, we can do that.  However, if there were hazards involved in applying an old edition, that edition would have been banned - by the API and local rules and regulations.

 

I am not trying to be "old fashioned" here; I am trying to point out that there is nothing "safer" in the new API 2000.  The older versions worked very well for me (and most are still operating and without problems) and what works successfully for me is what I will continue to apply - until I find a new method that is more successful, faster, more economical, and safer.  When you don't have any failures, accidents, or hazards appear in your operations is what I consider safe.

 

The API has published API 2210 and affirmed their distrust about using flame arrestors.  Until I can prove otherwise, I would heed that warning because if it comes back to bite me, I will be guilty of not having heeded a warning.  The reason the API published their opinion is because incidents already occured prior to this document and they have had to take a stand and admit that indeed, there is a potential hazard involved in using flame arrestors - and they don't know how to resolve it except to admit it.