18 replies to this topic

[engg]

decarbonated gas washer outlet line (sweet gas line) has a vent valve.

It is proposed to connect this valve to the Flare header. Configuration involves a vent to atmosphere and a tee with blind and double block valve to Flare header. This is usefull during startup for purging and maybe during operations etc..

Question is that is a check valve needed in this arragement?  vent.png   160.76KB   80 downloads

[ogpprocessing]

Installation of check valve at flare lines is prohibited and you are not allowed to install it.

[ogpprocessing]

Installation of check valve at flare lines is prohibited and you are not allowed to install it.

[ogpprocessing]

Installation of check valve at flare lines is prohibited and you are not allowed to install it.

[ogpprocessing]

Installation of check valve at flare lines is prohibited and you are not allowed to install it.

[ogpprocessing]

Installation of check valve at flare lines is prohibited and you are not allowed to install it.

[engg]

Thanks for your reply and information, but what might be the reason?

[ankur2061]

Engg,

Failure of check valve can cause blockage of the relief path which is undesirable from the safety view point for a pressurized system.

Regards,
Ankur.

[ogpprocessing]

Engg,

Failure of check valve can cause blockage of the relief path which is undesirable from the safety view point for a pressurized system.

Regards,
Ankur.

Dear Ankur,

You mean check valve may be a cause of blocked outlet? Considering the components of check valves,I can not imagine this closure state of check valve as cause of blockage. Could you please clarify this point?

[engg]

is there any API or such standard guideline on flare line design?

[fallah]

Failure of check valve can cause blockage of the relief path which is undesirable from the safety view point for a pressurized system.

As far as i know,main concern about latent failure of the check valve (stuck open,flapper break,....) limited to possibility of reverse flow and i believe for this reason it has not been used in flare network not for possibility of forward flow blockage.
A little bit pressurizing with purge gas along with using liquid seal or velocity seal,... to be provided to prevent reverse flow in flare network.

Edited by fallah, 07 November 2010 - 02:20 AM.

[ankur2061]

ogpprocessing,

Common Industrial Check Valve types:

1. Swing Check or Tilting disc with disc that swings - Possibility of disc getting stuck in closed position causing blockage of relief path

2. Lifting plug or disc type Check Valve - Plug or disc getting stuck in closed position causing relief path blockage

3. Ball Check Valve - Ball getting stuck in closed position blocking relief path.

Company standards like Shell, BP, Mobil explicitly prohibit installation of check valves in flare lines

Regards,
Ankur.

[kkala]

Having followed the topic, opinion as below may be of interest.
1. API does not permit any restriction on the PRV discharge lines to eliminate risk of upstream overpressure. This is naturally extended to the flare header receiving branches of many PRV discharges. A check valve can get stuck closed (possibility considered in safety studies), so it is a potential restriction.
2. In the reference case a line vent is branched to flare header and the check valve will be installed on this branch. If stuck closed, downstream flare header will not be affected, only upstream gas washing tower will.
3. As indicated on the attached drawing, this vent is not related to pressure relief; its failure at close position will not cause overpressure in the gas washing tower, only operational problems (till the valve to atmosphere is opened). So installation of this vent can be acceptable. Situation in the rare case it gets stuck closed can be assessed, even though this contingency may be non credible (check valves are placed at pump discharge).
4. Other problems can be: (α) Corrosion due to CO2 + water vapor (β) Check valves are hard to detect when deteriorated or stuck opened.

Edited by kkala, 07 November 2010 - 10:50 AM.

[ogpprocessing]

Situation in the rare case it gets stuck closed can be assessed, even though this contingency may be non credible (check valves are placed at pump discharge).

Why this is non credible for check valve at pump discharge gets stuck closed?

[engg]

nice post kakla
Its a good point to use PRV instead to protect from overpressure. but in case of overpressure, we will still have some gases joining in from somewhere affecting the product gas composition....

ogp; i personally have not seen check valves at outlet of pumps getting stuck. if it happens then the pump will be damaged or even explode....I understand that if the valve is in gas service with H2S and CO2 then corrosive components of gas can affect the integrity of check valve.

[engg]

in any case, it is clear that Check valve should not be installed.

Thanks all for your posts.

[ogpprocessing]

ogp; i personally have not seen check valves at outlet of pumps getting stuck. if it happens then the pump will be damaged or even explode....I understand that if the valve is in gas service with H2S and CO2 then corrosive components of gas can affect the integrity of check valve.

What to do if the liquid to be pumped is corrosive and can affect the integrity of check valve? What to do with corrosive gas service? I am asking about corrective actions that is necessary to be considered for check valves.

[Poosticks]

Then tell your check valve supplier exactly what the service is and ensure the materials of construction are compatible.

[kkala]

Some additional info aiming at clarifying created queries:
1. Probability for a check valve to fail to open is 0.0009 /year, versus 0.0005 /year for a spring operated PSV (rough results from TNO CPR 12E, 1997). So check valve failure to open is probably a non credible contingency. Even so, corrossion may increase mentioned probability.
2. Once we assumed 304L for wet Natural Gas pipes containing ~20% CO2, a check valve may deserve 316L; but this is a duty of corrosion Engineer (Process does not have this duty here) and also depends on any other impurities (or solids) contained in the gas. Nevertheless corrossion risk will not be limited to check valve but will extend downstream as long as wet CO2 exists.
3. Will be a catastrofic failure, if the check valve sticks closed? This could be assessed by one knowing well the operation of upstream column removing CO2 (available data does not indicate that such failure would happen).
4. Even for an acceptable check valve on the branch to flare header, the original question is whether this is needed. Role of it would be to prevent (or at least limit) backflow from flare header to washing column (retaining CO2). Is there a case when pressure in flare header can be higher than pressure in washing column? If so, a check valve could be justified. No relevant data is given, probably there cannot be such a case. This would have to be assessed by the local team with your explanations.

Edited by kkala, 13 November 2010 - 06:10 AM.