20 replies to this topic

[J_Leo]

Dear Friends,

 

In the past, I seldom distinguish remote contingencies from design. I treated all of them as design contingencies. Our current client defines some remote contingencies in the guidelines, for scenarios such as tube rupture, both control valve and bypass wide open. For the remote contingencies, 130% overpressure is allowed based on hydrotest pressure.

 

For overpressure caused by NNF streams, such as startup streams and maintenance nitrogen streams, I plant to treat it as remote contingencies. Do you think it is a good practice to do so?

 

Thank you for any inputs.

 

-Leo

[Bobby Strain]

You should really analyze each case. Don't treat them all the same.

 

Bobby

[J_Leo]

Bobby,

 

Thank you. Any examples that you can think of which can or cannot be treated as remote scenarios? I think it is pretty arbitrary.

 

-Leo

[Pilesar]

  I would require the client document what they consider 'remote contingencies' and any other exception they take to API 520 recommended practice. If they do not define the scenario as an exception, then consider it no exception. I think the client could make a case for 'remote contingency' for scenarios of full tube rupture and both control valve and bypass wide open. I would not be comfortable putting continually connected NNF startup lines or maintenance nitrogen lines in the same category since these WILL be used according to plan.

  Some engineering companies would NOT allow their engineers to design to a lower standard no matter what the client wants. If you are part of a large, reputable engineering company, then your chief engineer and perhaps the legal team should be involved as it does not seem your authority to commit your company to possible legal hazards. If your company does not care about such things, then my suggestion is to document your calcs clearly showing the basis for allowable overpressure. Record calculations for both standard and exception to defend the later charge of ignorance and incompetence.

  It is of no consequence to anyone if the relaxed guidelines do not affect the pressure relief design. Where the relaxed guidelines result in 'less protection' then you need someone else (preferably the client) to accept the responsibility. Your company and the client should agree to the guidelines in writing with explicit acknowledgement that they violate recommended practice.

  Will you be able to defend your engineering actions before a jury if the occasion arises? Generally accepted engineering standards exist for a reason. Do not ever perform any engineering function that is reasonably likely to harm someone's health or safety. You have a responsibility to the public that overrides your responsibility to your employer or client.

[fallah]

 

In the past, I seldom distinguish remote contingencies from design. I treated all of them as design contingencies. Our current client defines some remote contingencies in the guidelines, for scenarios such as tube rupture, both control valve and bypass wide open. For the remote contingencies, 130% overpressure is allowed based on hydrotest pressure.

 

For overpressure caused by NNF streams, such as startup streams and maintenance nitrogen streams, I plant to treat it as remote contingencies. Do you think it is a good practice to do so?

 

Leo,

 

Allowable over pressure in the design code, say ASME Sec.VIII Div.1, of pressure vessels has been specified based on credible contingencies which could be either normal or remote one. Hence you cannot increase such allowable over pressure values just due to considering a contingency as a remote one.

[J_Leo]

Pilesar,

 

Thank you very much for your inputs. I do want to take them as the design scenarios. The client and the lead on my side tried to even not consider these NNF lines as credible overpressure sources. You cannot believe it.

 

I am trying to put a basis together for the client to approve.

 

-Leo

[J_Leo]

Fallah,

 

The design guidelines provided to us define 130% overpressure for the remote contingencies.

 

Thank you,

-Leo

[fallah]

 

The design guidelines provided to us define 130% overpressure for the remote contingencies.

 

 

Leo,

 

Then the design code isn't ASME Sec. VIII...right?

Edited by fallah, 30 November 2017 - 03:31 PM.

[J_Leo]

Fallah,

 

It is ASME VIII. 

So if it is a remote relief scenario, what is your practice if the overpressure is the same as design scenario? I don't remember I have done any sizing related to remote scenarios yet. I cannot think of anything else.

 

Thank you.

-Leo

[fallah]

Fallah,

 

It is ASME VIII. 

So if it is a remote relief scenario, what is your practice if the overpressure is the same as design scenario? I don't remember I have done any sizing related to remote scenarios yet. I cannot think of anything else.

 

Leo,

 

If a scenario is really remote it shouldn't be taken into account in sizing calculations of a PSV based on relevant credible scenarios hence specifying an over pressure value for such remote scenario(s), while isn't among credible scenarios, is meaningless.

[fallah]

Leo,

 

If for any reason you have to take a remote scenario into account; the relevant allowable over pressure cannot be more than what has been specified in its design code.

[J_Leo]

Fallah,

 

The client told us the design guideline overrides API. I guess I have to follow. I will read it more carefully in case there is any misunderstanding.

 

Regards,

Leo

[latexman]

By "remote contingencies", do you mean "far away from people" or "unlikely to occur" or both?

[fallah]

By "remote contingencies", do you mean "far away from people" or "unlikely to occur" or both?

 

Latexman,

 

It really means "Unlikely to occur"...at least i consider this meaning in my discussion...

[latexman]

fallah,

 

It can mean either, that's why I asked.  I think you are probably right, but . . . for those who put client's request above RAGAGEP/Code, I'm not sure.  Again, that's another reason I asked.  This situation doesn't smell right to me.

[J_Leo]

Hi Latexan,

 

My understanding is it means not very likely to occur or occur less frequently. API also mentioned remote contingencies.

 

Thanks,

-Leo

[Bobby Strain]

I don't believe the API Recommended Practices are a legal requirement anywhere in the US. ASME codes are a different matter. Code design in Texas is not mandatory, but only a fool would not comply. The catch with overpressure protection is that not all causes are enumerated in any code or recommended practice. And as far as I know, in Texas, nobody has to certify the causes are all inclusive. I don't ever recall having the need to certify any of my designs as a process engineer. I never bothered to obtain a license during 50 years of practice. A license seems to be required in Texas to be called an engineer. This law was passed at the insistence of lobbyists primarily for the training businesses, modeled after other professional practices. I never once heard of any chemical engineer being prosecuted for violation of the statute. Maybe your client can site a code case that allows the greater overpressure. It sound to me, however, that they have conjured a case using the misguided logic of the tube rupture where protection is not required if the low pressure side test pressure is at least equal to the high pressure side design pressure. They probably don't have a clue as to what the repercussion might be.

 

Good luck,

Bobby

[J_Leo]

Bobby,

Thank you very much for telling us your thoughts based on your experience.

- Leo

[latexman]

Leo,

 

Can you point us to where "remote contingencies" is referenced in API?  I assume you meant API 520 Part I and II or 521?  When I searched for that phrase in the .pdf's I have, I got nothing.

[J_Leo]

Hi Latexman,

 

Sorry I didn't visit the site the last couple of days. Here is what I copied from API 521 as an example of a remote contingency:

 

4.4.14.2 Shell-and-tube Heat Exchangers

4.4.14.2.1 Pressure Considerations

Complete tube rupture, in which a large quantity of high-pressure fluid flows to the lower-pressure exchanger side, is

a remote but possible contingency.

 

- Leo

[latexman]

Thanks, the phrase I searched with did not hit that.  Tube rupture is looked at similar to piping overpressure ( > 10% accumulation / < so many hours/year occurrence ).  That's because it's a scenario that is not typically due to process upsets, but due to corrosion.  I think it would be taking that language way out of context to apply piping-type overpressure rules to scenarios that have traditionally been considered process upsets (10% accumulation), like CV and by-pass wide open.