9 replies to this topic

[syn23]

Hi,

I am trying to size the nitrogen blanketing inbreathing and outbreathing valve (open to atmosphere) for Amine sump drum in LNG plant.
This is underground sump pressure vessel with a PSV installed.
It also has the submerged intermittent service pump.

The PSV on the vessel is designed for the blocked vapor outlet scenario. In the event the nitrogen outbreathing valve remains closed or fails closed during the fire event or does not have enough capacity during the event overpressure will occur.

I do not have a firm flow rate for the material coming in to the sump. The pump is sized for the intermittent service.

My concern is, should I size the outbreathing valve for the overpressure event (as the PSV sizing is in the event the control valve fails close) or assume the incoming flowrate under normal circumstances.

Please explain the best design for the case.

Thank you for all the help.

Attached Files

•  Sketch (1).pdf   185.49KB   204 downloads

[Art Montemayor]

Syn23:

In my opinion, you have not been clear in describing what you have as the existing relief scenarios (I assume this is an existing, operating installation). From your simplified sketch, I can quickly identify the immediate and important relief scenarios you should design for:

• The nitrogen make-up control valve (what you call “nitrogen inbreathing”) can fail open. This will cause a “blow-through” case, unless mitigated by properly designing the PSV on the tank;
• The nitrogen venting control valve (what you call the nitrogen “outbreathing”) can fail closed. This will cause a “blocked outlet” case and can easily be quantified by the amine pump-in rate (or any other fluid that is being introduced into the tank while it is inerted).

Of the above two scenarios, the first – the blow-through case – is always the worse case. The total flow rate of nitrogen going through the make-up control valve will that related to the C_V of the control valve. What is normally done in this type of application is that a Resistance Orifice (RO) is installed upstream of the make-up control valve to minimize the flow rate of the blow-through to a value a little above the maximum required for make-up nitrogen. Your diagram does not show this RO. That means if you don’t install an RO, then your PSV has to handle the C_V quantity of the valve.

The failure of both control valves shown at the same time is a double-jeopardy case, which is not a credible scenario. However, the PSV is there to handle an emergency situation while the venting control valve is a normal operations instrument – and not a safety device. I would not take credit for the capacity of the venting control valve during an emergency – such as a blow-through – simply because of the principle that it is not meant to function as a safety device.

You have not shown any vacuum breaker in your sketch and I would caution you on supplying one in the event the nitrogen make-up valve fails closed – or you run out of nitrogen.

[syn23]

Thank you Art for the valuable comments.

• The relief valve has been sized for the blocked vapor outlet. There is not much I can do to change that.
• There is no vacuum breaker install as the vessel is designed for vacuum conditions.
• I have to size the nitrogen blanketing valve. The first question I have is

• Should the blanketing valve be sized for any upset conditions or for normal operation. If for upset condition, should we want the outbreathing control valve to open before the PSV or we want it to close after a certain pressure in order to prevent HC and amine going to atmosphere.
• should the outbreathing valve be sized for the fully open inbreathing valve if there is no RO upstream of the inbreathing valve.

[Art Montemayor]

Syn23:

First, I will address your two questions:

• Should the blanketing valve be sized for any upset conditions or for normal operation. If for upset condition, should we want the outbreathing control valve to open before the PSV or we want it to close after a certain pressure in order to prevent HC and amine going to atmosphere.
I think I was very clear on this subject in my previous post. Perhaps you have not read it, so I will repeat myself. The PSV is there to handle an emergency situation while the venting control valve is a normal operations instrument – and not a safety device. Therefore, do not design the normal instrument to function as a safety device.

• Should the outbreathing valve be sized for the fully open inbreathing valve if there is no RO upstream of the inbreathing valve.
No. This is inherently wrong engineering logic. Cure the cause of the potential hazard – NOT THE EFFECT. Besides, it is safer and less expensive to install an RO than to oversize the venting nitrogen control valve.

Now, I will address the main topic of this thread.

I cannot allow myself, as a professional engineer, to participate in what I have clear and strong suspicions regarding an unsafe design. What you are stating is that although the blow-through case is a credible and possible scenario without an installed RO, you are going to go ahead and design for what may not be the worse scenario. That, in my opinion is unsafe engineering and I can’t support it or fail to call it what it is to my belief. If there is a possibility that the existing PSV has been sized wrongly, it is your duty to confirm that with calculations and research. If it turns out that I am right, you should immediately advise all concerned engineers and management what has happened and advise them to mitigate the hazard.

This is not a matter of who is right and who is wrong. This is a matter of doing the right thing – for the sake of personnel safety in the area - and that should take precedence. I sincerely want to help you out in this matter and that is the best that I can do for you: warn you of a possible hazard.

[DB Shah]

dear syn23
Some more data may be useful
What is the design pressure of the sump tank
What is the maximum possible pressure of the N2 (SV set point on N2 system)
Is there any pumped material input to the tank or every thing comes by gravity?
In case any pumped material entry, what is the maximum pressure possible of that fluid.

[fallah]

Of the above two scenarios, the first – the blow-through case – is always the worse case. The total flow rate of nitrogen going through the make-up control valve will that related to the C_V of the control valve. What is normally done in this type of application is that a Resistance Orifice (RO) is installed upstream of the venting control valve to maximize the flow rate of the blow-through to a value a little above the maximum required for make-up nitrogen. Your diagram does not show this RO. That means if you don’t install an RO, then your PSV has to handle the C_V quantity of the valve.

Dear Art,

Regarding above (red color) statement you certainly mean : "...installed upstream of the venting make-up control valve to maximize minimize the flow rate of the blow-through..." with brown color as correction. Am i right?

Fallah

Edited by fallah, 30 May 2012 - 11:42 PM.

[Art Montemayor]

Fallah:

Thank you for your valuable check and correction on a very important detail. Yes, that is exactly what I meant to write - and what I thought I had written on Word for Windows, before pasting it into this thread. Obviously, something went wrong in the process.
I will edit my post to reflect the correct recommendation.

[syn23]

dear syn23
Some more data may be useful
What is the design pressure of the sump tank MAWP 3.45 / -0.5 BARG
What is the maximum possible pressure of the N2 (SV set point on N2 system) 7.53 BARG
Is there any pumped material input to the tank or every thing comes by gravity? BY GRAVITY
In case any pumped material entry, what is the maximum pressure possible of that fluid.

[syn23]

Thank you Art.

I am more clear about the design of the blanketing system than I was before. I will suggest my lead to go for the "RO" approach. There are few layers who will want to approve this.

Since we do not want the out breathing control valve to function during the upset condition, is there a way we can set the logic of the pressure controller to shut off the valve once it exceeds certain pressure (close to set pressure of relief valve)

[fallah]

Since we do not want the out breathing control valve to function during the upset condition, is there a way we can set the logic of the pressure controller to shut off the valve once it exceeds certain pressure (close to set pressure of relief valve)

syn23,

I think it could be achieved by considering a PCV downstream of existed out breathing control valve (in series), set to be closed at a pressure between set pressures of out breathing control valve and PSV provided that there is adequate operating range and dead bands such that there would be no interaction in these devices due to adding this PCV.

Fallah