19 replies to this topic

[Unnatiyr]

In case of PSV, for fire case 21% overpressure is considered while for non fire cases 10% why so??
I read previous answers to this query and it states that we assume that the vessel is destroyed in fire but why so? Isn't the purpose of PSV is to protect the vessel?
(I don't want to know from where/how "21%" comes)

[breizh]

Hi,

You may find pointers here :https://www.eng-tips....cfm?qid=213081

also good to read :https://www.cheresou...me-due-to-fire/

Good luck

Breizh

[Bobby Strain]

These are decisions by a group of professionals who publish these rules. Just follow them. Focus on your job and career.

 

 

Bobby

[fallah]

Hi,

 

It's basically as per the code ASME Sec. VIII Div 1 requirement, but the concept behind is: PSV cannot so protect the pressure vessel against failure in fire case hence higher over pressure leading to smaller PSV and lower lost cost...

[latexman]

Unnatiyr:

 

Protect it how? Ultimately, it’s to prevent the vessel from exploding, and preventing loss of life.

[Art Montemayor]

Unnatiyr:

 

Latexman has hit the nail on the head.

 

You have the wrong idea or opinion about the reason for employing a safety relief device.  The first and basic reason for employing a PSV is NOT to protect a pressure vessel.  It is there to protect the human beings operating and maintaining your plant as well as the fire fighters who have to respond to your need should a fire occur in your facilities.  If any vessels are exposed to direct fire, depending on its intensity, it is doubtful whether the same vessels would survive the episode or ever be used again.  The name of the device in question says it all: Pressure SAFETY valve.   Bobby is absolutely correct.  Professionals involved in ensuring safety have made these decisions with the ultimate priorities in mind and these rules and standards should be heeded and followed.

[breizh]

Hi,

To add to this thread , consider to read this extract from Leser :

Edit : a link from Spirax Sarco :
https://www.spiraxsa...ty-valve-sizing

Good luck

Breizh 

[chocobang]

I would further assume that 21% is chosen by the codes committee with consideration that hydrotest pressure is at least 30%. They would not select anything higher than 30%. But this reasoning is not documented anywhere so it is just my assumption.

[Pilesar]

If the non-fire-case overpressure allowance is 110% and you allow an additional 10% overpressure for fire, the math works out to 110% times 110% which equals 121%.

[latexman]

What I heard was Fire is 10% more than Process Upsets, which are 10% over MAWP. So, 1.1 x 1.1 = 1.21.

[Bobby Strain]

Actually, the 10% additional allowance for fire was first additive to 110% for process upset. And later modified to reflect what we now have, 21% overpressure. This minor change from 20% was made by the committee. I always thought this was a dumb idea. Not many of you have been practicing long enough to know about this, with the exception of Art.

 

Bobby

Edited by Bobby Strain, 01 May 2022 - 10:35 PM.

[Unnatiyr]

Greetings Fallah sir,

Thanks for responding. Why is it that higher over pressure leads to smaller PSV?

(I am a fresher who has just started her career)

Thanks.
Unnati 

 

 

..

[Pilesar]

Vapor at higher pressure has higher density so lower volume flow can achieve the same mass flow.

[fallah]

 

Hi,

 

Same mass flow rate with higher pressure leading to lower volumetric flow rate, hence lower required orifice area and smaller PSV...

[Unnatiyr]

Greetings Art Montemayor sir,

Thanks for responding and clearing my doubt as well.

So we provide PSV not for protecting the vessel but to prevent it from exploding. So what if we provide just 10% overpressure instead of 21%? What will be the consequences?

.

[latexman]

Not many of you have been practicing long enough to know about this, with the exception of Art.

 

Clyde Neely was who I heard it from.  He served on ASME committee and sub-committees for over 40 years.  Did you ever meet/know Clyde?

[breizh]

Hi,

About safety valve , let you get a copy of : the Safety valve handbook by Marc Hellemans -IchemE

 

This will help you for today and tomorrow .

 

Good luck

Breizh 

[Bobby Strain]

Latexman,

 

     I didn't  know Clyde. I recall the change was in early/mid 1980s. We had to revise lots of stuff to accommodate this foolish revision.

 

Bobby

[latexman]

I started in 1979.  While I remember the change, it is a vague memory to me now.  Clyde was an alumni of my alma mater and I met him through local alumni gatherings/parties.  He was also high on the food chain (Director level), but when we ran into each other he morphed into a part-time mentor.  Good memories.

[astro]

Not having the benefit of the personal history associated with some of the contributors here, I get value out of reading the relevant historical narrative in API STD 521 (from the 7th Ed, refer Annex A - An Analytical Methodology for Fire Evaluations, A.2 Background of the Empirical Method in 4.4.13.2.4.2).

 

What I draw from that reading is an appreciation of the role of technical development to adjust, improve and refine design standards.

 

In that light, the paper by Andreasen et al., On the Adequacy of API 521 Relief-Valve Sizing Method for Gas-Filled Pressure Vessels Exposed to Fire, Safety 2018, 4, 11 presents an analysis with somewhat confronting conclusions. Note the abstract:

In this paper, the adequacy of the legacy API 521 guidance on pressure relief valve (PRV) sizing for gas-filled vessels subjected to external fire is investigated. Multiple studies show that in many cases, the installation of a PRV offers little or no protection - therefore provides an unfounded sense of security. Often the vessel wall will be weakened by high temperatures, before the PRV relieving pressure is reached.

From what I can gather, relatively thick wall vessels were tested "back in the day" and I'd bet that the push for economy and better alloys with material properties have seen thinner walled vessels become prevalent. So, it's not a surprise that the long standing practice, such as the 15 minute depressuring "rule" has gone by the wayside and forensic vessel survivability assessments drive depressuring system blowdown times.

 

API STD 521 appears to recognise the findings by Andreasen et al. coincidentally to an extent, refer 4.4.13.2.6 Protective Measures Excluding Insulation:

A PRD may not provide sufficient protection from vessel rupture due to open pool fire exposure for an unwetted-wall vessel or a vessel containing high boiling point liquid. Where a PRV alone is not adequate, additional protective measures should be considered ...

Where local jurisdiction permits, it can be appropriate to utilize these protective measures as an alternative to relief
devices sized for the fire case ...

(refer the standard's wording to fully appreciate the context).

 

I recognise that I'm not explaining the basis for 21% overpressure criteria. However, I do think that it's important to be aware of and recognise that there are cases where its application can be demonstrated to be irrelevant.

Edited by astro, 17 May 2022 - 10:30 AM.