Overpressure Condition of ASME U-Stamped Vessels

Without knowing any of the circumstances the only comment I can make is that if either the materials that you're working with, or the pressures can pose any type of Hazzard don't use it until its integrity has been verified.

I'm supplying a link to a a previous thread that discusses your problem, it should contain all the information you'll need, the code numbers here are from that thread.

Note 1. The firm authorized to carry out the repair must has a current copy of ASME code (ASME II, Parts A, B, C & D, ASME V, ASME VIII-1 and ASME IX) and ANSI/NBIC NB-23.

http://cr4.globalspec.com/thread/21581/U-Stamp-Vessel-Repair

Thank you for your reply.

We test our vessels at 35 psi. At what overpressure value is repair not allowed. What must done done to repair units that can be repaired. What does the AI have to look at? etc.

vipat

ASME is pretty adamant that you do not go above the prescribed test pressure. And you won't find wordage to allow it.

Having said that -

What was the pressure that was reached?

What is the MAWP of the vessel? At what temperature? What was your test temperature?

Pneumatic tests are prescribed as 1.1 times MAWP compensated for the stress ratio at temperatures.

The saving grace may be the point that hydro-tests are prescribed at 1.3 times MAWP compensated for the stress ratio at temperatures. If the pressure that you actually reached is below this value - you may argue with your AI/client that everything is fine.

However, if you went above the 1.3 X value - you are essentially out of luck, but, doubtfully, may be able to get by with an API 579 fitness-for-service examination with some additional NDT. I don't have time right now to go through the ASME interpretations, but I believe if you went above 1.3 X you are no longer allowed to specify the vessel was designed/built to ASME VIII-1 in any case. So, in short, depending on the size of the vessel - you would probably be better off scrapping the unit and building a new one.

Our MAWP is 9.6 psig.

Our ASME p(pneumatic) is calculated as 1.10 * (9.6 + 14.7) = 26.7 psig.

Our PED p(pneumatic) is calculated as 1.43 * (9.6 + 14.7) = 34.7 psig.

We do our test at room temperature at 34.7psig.

We actually overpressurized to 68 psig.

I wouldn't take the chance. The material was never expected to see that pressure and I believe you may have over stressed it. At any point, it's not worth the risk of a field failure.

Next, go to the test station and install a very inexpensive relief valve set to pop at 35 or 38 psig to keep this from ever happening again.

Why are you adding in 14.7 psi when calculating Test pressure ? Your MAWP and Test Pressures are both listed as Gage Pressure.

The test pressure is based on design and/or a stated MAWP value that is set by the designer. Most vessels are designed and a material thicknesses close to that calculated is selected. Due to availability of thickness, most often the thickness chosen is greater than necessary because they cannot be optimized because of availability, so it may be that your overpressure condition for the design does not exceed the true MAWP of the vessel.

I would suggest that you run the calculations solving for P (MAWP) based on the actual t (thickness) of the materials you used. This value is the true MAWP of the vessel and if your test pressure is below the test pressure for that MAWP, you're hoome free and the overpressure is of no consequence. If pneumatic test still exceeds the MAWP test pressure, I'd next look at what effect RT has on the MAWP and recalculate again. If you're still in trouble or if you have deformed the vessel in any way, scrap it.

Our ASME p(pneumatic) is calculated as 1.10 * (9.6 + 14.7) = 26.7 psig

ASME Pneumatic test Pressure = 1.1 MAWP = 1.1 * 9.6 psig = 10.56 psig (not 26.7 psig)

Also :Our PED p(pneumatic) is calculated as 1.43 * (9.6 + 14.7) = 34.7 psig,

Must be 1.43* 9.6 psig = 13.73 psig ( not 34.7 psig).

Only we add atmospheric pressure 14.7 psi to gauge pressure to find the absolute pressure. So we have to nominate the pressure as psig for gauge pressure, and psia for absolute pressure.

Note. There are a good information in the following CR4 Thread U-Stamp Vessel Repair.

The key phrase in UG-99 is "visible permanent distortion". Even then it is not automatic rejection - it is based on the judgement of the AI and your own.

In the design of a vessel, some components are usually overdesigned (shell / heads / nozzle necks / standard flanges). These are likely to have a MAWP that is higher than the stated vessel MAWP or even your (actual) test pressure are are of no issue.

On the other hand, some components limit the vessel MAWP (having a MAWP that is right at or a little above the vessel MAWP). These are potentially overstressed during your overpressure situation. These can be nozzle opening reinforcements, custom flanges, etc... These may be identified by calculation.

If you haven't previously calculated a MAWP and used the design pressure as MAWP, calculating a MAWP could help you. If the new calculated MAWP turned out to be higher than your actual test pressure divided by 1.1, you could set the stamped MAWP as your actual test pressure divided by 1.1 (with permission from your AI). This would demonstrate that an actual overstress condition did not occur.

You could also verify that the actual test pressure did not exceed 1.3 x the current MAWP. This would be the minimum pressure for hydrostatic test.

You can also PT examine these areas that are identified by calculation as MAWP limiting components. Absense of cracks in these areas would give a comfort level for using the vessel. If the areas were not PT examined before the test, this is probably not a good idea - there will be a high likelyhood of 'false negatives', of finding pre-existing cracks.