21 replies to this topic

[Soothsayer]

Hello.

 

I am working for an engineering consulting company. One of our clients (offshore) wants to recover flare gas (for reducing CO2 emissions) without installing an FGRS (no compressor, ejector, blower, fan or any other equipment to compress the gas). The recovered gas stream would be re-injected into the main process stream (reception facilities) which is operating around 0,5 - 2 barg. For this, it is necessary to increase the normal backpressure of both HP/LP systems (headers, KOD). 

 

I understand that an extensive evaluation of all the sources is required as this could impact the mechanical integrity of the equipment on potential overpressure scenarios. In the same vein, it is necessary to perform an evaluation of every single relief device installed, the pressure of continuous sources, etc.

 

Such a strategy could affect the reliability of the recovering operation as there is not, from my point of view, operational flexibility (the frequency of flaring due to potential relief/blowdown scenarios could be high as there will be less volume available to accommodate more gas). 

 

My question is the following: Beyond the necessary evaluation I am briefly (poorly) describing above, I want to know in such a strategy targets process safety. 

 

Do you think that having this volume of gas (HC) always floating in the headers violates a fundamental principle of the flare system (ultimate protection layer)? 

[Bobby Strain]

This sounds like a bad idea. And dangerous.

 

Bobby

[Soothsayer]

Thank you, but why? I was told that a Process Safety evaluation is irrelevant during the concept phase.

Edited by Soothsayer, 30 November 2019 - 01:29 AM.

[PhilippM]

Thank you, but why? I was told that a Process Safety evaluation is irrelevant during the concept phase.

 

Regarding Process Safety evaluations as irrelevant (no matter what phase) sounds like another bad idea. Especially when dealing with a subject that is directly related to process safety (i.e. the flare gas system).

[Soothsayer]

Thank you, guys, for your response. My intention with this topic is to go to the root of this, back to basics and explain why you say is a Bad Idea...

Edited by Soothsayer, 30 November 2019 - 11:23 AM.

[Zauberberg]

Conceptually, you would have to confirm that ALL pressure safety valves are with adequate capacity to relieve into a higher pressure discharge (=flare) system.

 

Sonic flares are commonly used offshore but they are normally employed in high-pressure systems.

[breizh]

Hi,

Just curious what are the numbers ?

Breizh

[Soothsayer]

Both flare systems are operating at Atmospheric pressure. The intention is to keep a constant pressure of 1.0 barg to recover the flare gas. Could you be more specific? Thank you.

[Soothsayer]

Conceptually, you would have to confirm that ALL pressure safety valves are with adequate capacity to relieve into a higher pressure discharge (=flare) system.

Sonic flares are commonly used offshore but they are normally employed in high-pressure systems.

That's right. We will do that. It seems that’s the only major risk!

[breizh]

Hi,

My question is about the quantity you intend to recover .

Breizh

[Soothsayer]

Hi,

My question is about the quantity you intend to recover .

Breizh

 

It is just 500 - 700 Sm3/h of light MW HC gas.

[Zauberberg]

Sonic flare is used to enable lower size of the flare network piping, not to recover gas. I was mentioning sonic flare systems just as an example how flare systems can operate at higher pressures.

 

It's not enough just to pressure up the flare gas system and recycle gas back to process. You need to look at each production and relief scenario and see how recycling in the process will affect that scenario. There is probably several scenarios with multiple relief points (e.g. loss of power, loss of utility, etc.), including total site blowdown (emergency). I wouldn't advise to anyone to recycle the gas back to the process under those circumstances - you would actually be fueling the fire.

 

Open path to the flare must be maintained at ALL times - and there are no exceptions to this rule.

[Soothsayer]

Sonic flare is used to enable lower size of the flare network piping, not to recover gas. I was mentioning sonic flare systems just as an example how flare systems can operate at higher pressures.

 

It's not enough just to pressure up the flare gas system and recycle gas back to process. You need to look at each production and relief scenario and see how recycling in the process will affect that scenario. There is probably several scenarios with multiple relief points (e.g. loss of power, loss of utility, etc.), including total site blowdown (emergency). I wouldn't advise to anyone to recycle the gas back to the process under those circumstances - you would actually be fueling the fire.

 

Open path to the flare must be maintained at ALL times - and there are no exceptions to this rule.

 

Thank you, Zauberberg. 

 

I do understand, and I think it is crucial what you've said... ''To maintain an open path at all times''. In my opinion, this is the key for the evaluation, this is my safety concern, and a thorough evaluation is required.

 

However, I've seen facilities where recovery involves (treating)/reprocessing gas; this doesn't mean that you are fueling the fire. As in all flare gas recovery systems, there is a control logic. Simply, if the recovery is not possible (e.g., high pressure in the process line), valves will close, the flare header will pressurize (if the abnormal operation continues), the isolation valves (FOV) will open and flaring will occur. 

 

Regards.

[Bobby Strain]

You keep making a bad idea worse. You cannot have any valves or obstructions in the flare line. You should abandon the idea and proceed with other tasks.

 

Bobby

[Soothsayer]

You keep making a bad idea worse. You cannot have any valves or obstructions in the flare line. You should abandon the idea and proceed with other tasks.

Bobby

I understand that high pressure drops or pocket shall be avoided. But then how you explain Flare Gas Recovery Units? It is a package with a compressor or an ejector and a PCV downstream. FOV are installed for isolation of the flare header. I could show you systems with ON/OFF valves, check valves, flow meters in flare gas lines in facilities of major and worldwide recognize oil companies. Pressure drop does not necessarily mean obstruction. The flare system, and relief devices, can be design for that!

Once again, please explain why it is bad idea.

Regards.

Edited by Soothsayer, 02 December 2019 - 04:52 PM.

[PhilippM]

the isolation valves (FOV) will open and flaring will occur.

 

What isolation valves? Could you maybe provide a small (hand-drawn is fine) sketch of what you imagine that system would look like?

[PhilippM]

 

You keep making a bad idea worse. You cannot have any valves or obstructions in the flare line. You should abandon the idea and proceed with other tasks.

Bobby

I understand that high pressure drops or pocket shall be avoided. But then how you explain Flare Gas Recovery Units? It is a package with a compressor or an ejector and a PCV downstream. FOV are installed for isolation of the flare header. I could show you systems with ON/OFF valves, check valves, flow meters in flare gas lines in facilities of major and worldwide recognize oil companies. Pressure drop does not necessarily mean obstruction. The flare system, and relief devices, can be design for that!

Once again, please explain why it is bad idea.

Regards.

 

As you said yourself: An FGRS has a compressor. A compressor which you for some reason do not want to use.

Edited by PhilippM, 02 December 2019 - 04:18 PM.

[Zauberberg]

Soothsayer,

 

Flare gas recovery units are always designed in parallel to the continuously open flare header (i.e. the main header is not closed by any switching valves). The PCV feeding the flare gas recovery unit always maintains positive pressure inside the relief system, so as to prevent inward flow of Oxygen from atmosphere to flare header and/or into the Recovery Unit.

 

Imagine your FOV does not open in case of a major relief event, or in case of any significant relief event. That's a scenario with tragic consequences. No serious operator company in this world would approve such design. Even if you have a SIL 3 rated instrumented valve unit (or whatever the safety analysis calls for) for this interlocking system, it can hardly be justified because of unforeseen consequences in case of failure.

[Soothsayer]

Soothsayer,

 

Flare gas recovery units are always designed in parallel to the continuously open flare header (i.e. the main header is not closed by any switching valves). The PCV feeding the flare gas recovery unit always maintains positive pressure inside the relief system, so as to prevent inward flow of Oxygen from atmosphere to flare header and/or into the Recovery Unit.

 

Imagine your FOV does not open in case of a major relief event, or in case of any significant relief event. That's a scenario with tragic consequences. No serious operator company in this world would approve such design. Even if you have a SIL 3 rated instrumented valve unit (or whatever the safety analysis calls for) for this interlocking system, it can hardly be justified because of unforeseen consequences in case of failure.

 

The FGRU is aimed at avoiding continuous flaring to reduce CO2 (SOx/NOx) emissions.

 

Zero flaring implies that the main flare headers running to the flare tip shall be isolated! This is the best practice (and the only way to do it) when installing FGRU for assuring Zero Flaring. 

 

Zauberberg, the scenario you describe is not real because nobody would rely on a single layer of protection. What you usually do is to install a rupture disc in parallel to the FOV. If the FOV fails, the rupture disc will break, and flaring will occur. Liquid seal drums can be used as well for isolation of the flare header, but they have a larger footprint. Usually, a ballistic system is installed for ignition of the flare tip.

Not sure if you are familiar with this operation philosophy! 

 

I think we are missing the point of the topic. I will prepare an sketch ASAP 

 

Regards.

Edited by Soothsayer, 03 December 2019 - 06:39 AM.

[Zauberberg]

A sketch would definitely help, indeed.

 

Zero flaring normally refers to "zero operational flaring", which means no flaring under normal operating conditions and NOT no flaring under process or emergency relief condition. That's dangerous and I think everyone understood this is what you intend to do.

 

If you have a continuous flow of process gas (not relief gas) than you can definitely pressurize this stream and recycle it back to the process - but I fail to see why it has to be through the flare header.

[Soothsayer]

A sketch would definitely help, indeed.

 

Zero flaring normally refers to "zero operational flaring", which means no flaring under normal operating conditions and NOT no flaring under process or emergency relief condition. That's dangerous and I think everyone understood this is what you intend to do.

 

If you have a continuous flow of process gas (not relief gas) than you can definitely pressurize this stream and recycle it back to the process - but I fail to see why it has to be through the flare header.

 

Zauberberg,

 

You are right. I did not explain it well. The FGRU intends to assure zero flaring under NORMAL operating conditions. Of course, flaring will occur under emergency relief scenario. 

 

The reason for doing it through the flare header is because these continuous sources are connected to it. The design was conceived with this normal contribution of gas to the flare system. 

 

Regards.

Edited by Soothsayer, 03 December 2019 - 08:26 AM.

[Zauberberg]

You can modify the system, in a relatively simple way:

 

1) Connect the flare gas sources to a new process header, while maintaining their existing connections to the flare (sub)headers. Set a new pressure control valve in the header or, even better, in a new surge vessel downstream, and ensure the existing pressure control valves on flare gas sources will have sufficient delta pressure to discharge full flow while PCV's remain in the stable control regime (because they will discharge to a higher pressure).

2) Interlock the system in such way, that under normal circumstances the vent flow goes to FGRU, and in case of FGRU unavailability (e.g. valve to the process sub-header stuck closed, or FGRU unit shutdown) the flow is directed to flare (sub)headers.

3) Ensure that in case of interlocking system failure, the source vessels are protected from overpressure by other means, and there is unrestricted relief path (i.e. not valved) to the flare.

4) Run an independent safety analysis to confirm suitability of the design.

 

Note: edited a typo in point No 2.

Edited by Zauberberg, 03 December 2019 - 02:07 PM.