Pressure Relief Code Question 3

[PipingEquipment]

First post here, so hi everyone! :^D

When re-rating a vessel is it acceptable to set a new corrosion allowance based on available thickness above what is required? A senior member here at my company says he would not do this because we only take a limited number of UT readings and it could be significantly thinner than is implied by the lowest UT. But by that reasoning we shouldn't have any confidence in any thickness used in the re-rate calculations. I tend to think this could be done...Have any of you come across this scenario? Do you think this can be safely done?

 

[PipingEquipment]

My second question....

Is it the manufacturers responsibility to approve pressure relief devices which pertain to the vessel? It seems customary here at my workplace to simply put "safety relief devices provided by others" on the data reports with no method of checking the integrity of said devices. UG-125 says means of protection from over pressure need to incorporated into the vessels design, so should we the manufacturer be doing more to protect from over pressure? Or is our disclaimer on the data report sufficeint?

 

[zdas04]

On your second question. The key phrase is "provide the means". Building in a nozzle for a PSV is providing the means. As a vessel manufacturer you will often be putting vessels in places where the maximum credible pressure and the maximum credible flow is unknown at the time of order (this is true on nearly every wellsite for example). The operator of the vessel has the obligation to define the credible scenarios that could put a vessel in an overpressure situation and to specify the maximum flow under that scenario. From that point it is simple arithmetic to size a relief device.

You company's technique of marking the relief devices as "provided by others" is much better than some fab shops guessing.

David

 

[TD2K]

The manufacturer of the pressure vessel will typically have no idea of the possible overpressure scenarios. Those could be fire, blocked discharge, failure of an upstream pressure control valve from a high pressure source, etc as well as what credits the owner may be willing to allow or not willing to allow.

While the manufacturer 'could' be provided with the necessary information for them to identify and size a suitable pressure relieving device, this would be unusual in my experience. That's the responsibility of the process engineer designing or evaluating the overall system. I decide what size and inlet/outlet line size the relief device needs to be.

For your first question, if I'm understanding you correctly, yes, the owner can elect to reduce the corrosion allowance from (for example) 1/8" to 1/16" and use that extra 1/16th for pressure calculations if their corrosion results supports the reduction.

 

[PipingEquipment]

Thank you both for the fast responses.

TD2K, I believe you answered my first question, in that on a re-rate the corrosion can be adjusted. However, I was going the other way with it though instead of lowering the corrosion I was asking if the corrosion may be raised or kept the same?
In other words, they made the vessel and components extra robust in the beginning and according to my calculations the pressure can be raised by 20psi as well as keeping the original corrosion at 1/16", regardless of what corrosion has occurred. My QC manager said something to the effect of "you must adjust the corrosion to reflect what corrosion has taken place". So according to him the corrosion allowance should be lowered by 0.12" since this is how much the most corroded component has been reduced by, based on UT readings. So we would effectively be raising the corrosion above the original by 0.12" by saying there is still 1/16" available for corrosion.

Am I still correct in thinking we may adjust the corrosion to be greater than was originally allotted if the calculations support this?

 

[jte]

PipingEquipment-

Regarding CA and rerates: Yes, the CA can be modified up or down as required and supported by calc’s. For example, I may have a vessel with an MAWP of 100 psi and a CA of 0.125”. If my Process friends tell me that they need the MAWP increased to 110 psi, and my UT’s indicate that we’ve seen no more than 0.050” corrosion over the last 20 years of operation, I may consider dropping the future CA to a smaller value, perhaps 0.070” if that’s what it takes to “make it work.” On the flip side, I might be informed that Inspection is concerned with excessive corrosion, and Process tells me that they don’t have a problem resetting the relief valves to 70 psi. Now I can rerate the vessel to a lower MAWP to buy a bit more CA. I will say that I generally don’t like this approach as most rerate calc’s I see don’t consider structural loadings, and I’m always concerned that even if pressure governed the initial design, the engineer working the rerate may not recognize that perhaps something else (such as structural stability) begins to govern if the CA is increased.

I get the feeling that the discussion between your QC manager and yourself might be a result of confusion in when the start point is for determining the corrosion allowance. Say the vessel was fabricated from 1” plate. It has been in service for some years and has corroded by 0.125”. The original CA at the time of fabrication was 0.25”. Now we do a rerate (say for 516-70 material and a temperature increase from 200°F to 300°F). What is the CA? Well, I’d suggest you take a step back and recognize that it is not so much the CA which is important as the required thickness of the shell at retirement. Do you retire the vessel when it hits 0.75” wall thickness or do you retire the vessel when it has corroded by 0.25”? Do you care if the 0.25” corrosion occurred as 0.125” before the rerate and 0.125” after? No. Now, changing the CA to reflect the 0.125” corrosion which has taken place becomes confusing: Do you run your rerate calc’s with a “new nominal” thickness of 0.875” and a “new CA” of 0.125”? This can only lead to confusion unless all of the original documents and drawings are tossed out and new ones made to replace them. Of course you can’t do that: The weight of the vessel for wind and seismic calc’s must be determined on the conservative basis of new thickness, as the corrosion is not likely to be exactly uniform everywhere. Bottom line: Run the numbers as though the vessel was fresh from the fab shop and never seen service. Determine the max CA, verify that the existing wall thickness is above the minimum required, and let the user’s qualified inspection group determine remaining life.

As to the UG-125 question… Seems to me that UG-125(1) is clear that it is not the fabricator’s responsibility to install the PRD. UG-125(2) backs up TD2K’s comment about who has the “big picture” in order to set relief requirements. Finally, UG-125(3) seems to make it clear that the Manufacturer need not supply relief devices. Not quite sure where the confusion is coming from…

jt