Is PSV required here? 2

[Orsiz]

Dear members,

I have a question regarding the need for a PSV. We're currently designing a nitrogen system for sweep through purging of some atmospheric equipment. The maximum nitrogen flow needed for purging is 200 kg/h at atmospheric pressure and 15°C.

The nitrogen supply pressure is 7 barg and a PCV is installed to reduce the pressure to 0,5 barg in order to deliver the required flow rate to the downstream equipment. The design pressure of the piping downstream of the PCV is 10 barg and the downstream equipment is atmospheric (sufficient large opening on equipment so that pressure build-up is not possible, blocking of this opening is also not possible). Shutoff valves between the PCV and equipment shall be foreseen. They will completely shut the nitrogen supply in case of failure of the PCV (a flow transmitter shall detect the higher flow rates and gives a signal to close the shutoff valve) or failure of the nitrogen extraction (interlock).

Do you think a PSV is recommended here? In my opinion, we don't need one but I could use some feedback here (I have read quite some topics on this forum and there are a lot of experienced members here). Please let me know if you need any more information.

FYI: We still have to do a HAZOP and LOPA study but I already want to encounter as much as possible to reduce the amount of actions coming out of the HAZOP.

BR

 

[pennpiper]

Have you ever heard of "Murphy's Law"?
That is "If anything can go wrong, it will"

- What would the cost of a small PSV (plus piping) be?
Vs
- What would the cost of a catastrophic failure (with injuries) be?

Your choice.


Sometimes its possible to do all the right things and still get bad results

 

[Latexman]

I think your "sufficient large opening on equipment so that pressure build-up is not possible, blocking of this opening is also not possible" is your relief device. Just document with calculations that this opening can handle a failed open PCV. Then, delete all those bells and whistles on shutting down the nitrogen. When the high flow alarm goes off, go investigate and close a manual valve, if needed.

Good luck,
Latexman

To a ChE, the glass is always full - 1/2 air and 1/2 water.

 

[MortenA]

i would agree with Latexman, the shut-of valve is not needed if the psv is not needed. The only real question you need to consider is: Could my N2 pressure go higher than 7 barg? If no then o think you are home free. A fire case is not needed for piping with small volume.

Best regards, morten

 

[georgeverghese]

Agreed, you have to run some calcs to show that, when the PCV goes wide open, the total backpressure built up at the downstream tank / vessel to be protected, does not exceed the MAWP of this component. A descriptor such as " sufficient large opening" wont be enough to get you past a process safety review. Add an RO downstream of the PCV to further restrict flow if required.

A FT / FAHH is more expensive than a PT / PAHH, so I'd go for a field installed PT at the downstream component to isolate the N2 PCV in case of pressure control failure.

 

[e43u8]

Let's focus on the OP query...

If all assumptions of OP with supportive documented calculation being provided; no PSV for over pressure protection is required...

 

[don1980]

As Latexman says, you already have a relief device - it's the open vent pipe. And that device is better, more reliable, and less expensive than a PSV. There's no logical reason to add a PSV to this vessel.

 

[Orsiz]

Great answers, thank you.

So if I understand it correctly, I should calculate the pressure drop across the equipment openings at max flow rate (PCV failure) to confirm my statement of 'a too large opening for pressure build-up'?

A detail I forgot to mention is that this equipment is located indoor. The main risk here is the escape of nitrogen in this area when the extraction system fails. In that case, the shutoff valve should close. A failure of the PCV can indeed be seen by a pressure transmitter with PAHH.

@MortenA: the maximum nitrogen pressure is specified and not higher than the piping design pressure.

 

[Latexman]

Correct, and I hope this "atmospheric equipment" actually has a positive gage pressure rating other than "atmospheric".

Good luck,
Latexman

To a ChE, the glass is always full - 1/2 air and 1/2 water.

 

[Orsiz]

@Latexman: Yes it has, I just use that term to indicate that it is open to atmosphere.

 

[georgeverghese]

As with all other built up backpressure calcs, one would start the calc from the most downstream component and work backwards. Assuming the PCV operates at critical flow (dp across PCV = 7barg to anything less than approx 3barg), you then will have the relief flowrate that should be used to begin this backpressure clc.

 

[choack]

I supposed that the vent of this "atmospheric equipment" be routed to a safe location. Normal purging of 200 kg/hr would be detrimental indoor.

As georgeverghese mentioned, please make sure to do a backpressure backwards calculation with CV failure flowrate. Do not account for the closing of the on-off valve. All the best!

 

[Orsiz]

@georgeverghese: why do you state 'dp across PCV = 7 barg to anything less than approx 3 barg'? Where is this value of 3 barg coming from?

 

[Latexman]

Link

Good luck,
Latexman

To a ChE, the glass is always full - 1/2 air and 1/2 water.

 

[Orsiz]

@Latexman: Thank you for the link. From what I read on that page, the 3 barg is mentioned here because for higher pressures, the Pu/Pd ratio is lower than 1,895 (so no choked flow). Can I then use this value in the choked flow equation for compressible gases?

 

[Latexman]

Orsiz,

I don't think you got what georgeverghese was saying. My guess is georgeverghese (georgeverghese correct me if I'm wrong please) assumed choked flow then did the rough math in his head and said, "approx 3barg". He did not say just "3 barg" or "exactly 3 barg". His intent was to convey the thought/fact that he believed flow to be choked and sonic. Therefore, it would be up to you to calculate accurately the downstream pressure at the PCV outlet.

Good luck,
Latexman

To a ChE, the glass is always full - 1/2 air and 1/2 water.

 

[georgeverghese]

@Latexman, thanks for filling in.
Yes, for a first iteration, you can assume choked flow and compute the flow based on Pd being 3barg or lower (based on the Wiki link, any Pd value less than 1.895 x 8bar abs will give the same flow when in choked flow). When you complete the built up backpressure profile calc, you will most likely get an actual backpressure at the PCV which is probably much lower than 3barg.

 

[TheChemengers]

Whats the design pressure of the nitrogen system? if it is higher than that of the downstream system connected for purging then there is a valid overpressure case and you need to determine if relief is necessary.

If you are going to rely on an open vent on the downstream system, then as a worst case you should calculate the max flow through the PCV - nitrogen system at design pressure and PCV fully open. consider also if there is a bypass on the PCV, this should be included.

Also you have to confident the vent will always be available - lock open any valves on the vent line, are there any filters or equipment that could get blocked in between the equipment being protected and the vent?

This requires detailed analysis and you cannot rely upon messages on a forum for this answer. You can't at present say in HAZOP that the system is adequately protected for overpressure unless you completed a suitable overpressure analysis.

 

[LittleInch]

I agree that you need to go back to the pressure source, i.e. the N2 supply. Stated as 7 barg, but make sure it has enough protection to ensure that is a good number to use.

Then your system - "atmospheric" is actually quite vague and needs to be defined. Any sort of flow in that system will introduce some sort of pressure in order to generate flow. Without knowing the precise definition of "atmospheric" it's rather hard to say. API 650 for "atmospheric" storage tanks actually defines this as a max of 18 kPa or 2.5 psi or weight of the roof, but often only inches of water guage.

From an engineering design view you really need to test the assertion that " blocking of this opening is also not possible". It is quite difficult sometimes to show that this is possible to check and confirm before you start the operation. Relying on procedures to implement safety critical activities is sometimes not allowed in a HAZOP scenario, but practicality can be introduced. Each situation is different.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[TheChemengers]

Just to add to this I had an issue recently on a plant where significant re-engineering was required right at the end of EPC. During HAZOP an overpressure scenario was not considered credible because there was a 11/2" vent on a compressor gearbox. This vent was not nearly large enough for the tube rupture scenario which was discussed. This type of thing can cause real problems in engineering design if you get complacent.