A hydrostatic test is non-destructive if the pipe passes the test and destructive if the pipe fails. The test pressure and hold time should be selected to achieve the desired confidence in pipe integrity without detrimental effects on pipe properties.
@SJones: I don't think so. Hydrostatic test is non-destructive (without any ifs).
A destructive test is not complete until the specimen undergoes permanent deformation, which is not always the case in hydrostatic testing.
Test parameters (pressure and duration) are not so selected as to protect the specimen, they are specified by design and it is up to the specimen to withstand or fail under the conditions.
As for physical changes in the metal, such as effect on molecular structure etc., it is true that the material suffers certain effects under the applied pressure. But the assumption is that within the limits of test pressure, all deformation is reversible (elastic only), at least theoretically. The material is expected to resile back to its original condition after the test pressure is removed.
In reality, some of the effects may remain and influence the physical properties of the material to a small extent for the rest of its service period. This may not be very significant.
It is qualified as non destructive by the Code!
anything else is just speculation. if you destroy the vessel is because you put too much pressure beyond the vessel test pressure.
it becomes destructive with proof tests.
genb
See Genb's reply above. Test parameters can also be selected by the user that go beyond the minima expressed in standards, in which case, they are NOT 'specified by design'. Think about what one is trying to achieve with the hydrotest and think about relationships between stresses and hold times.
SJones,
What Code or Standard might you be thinking of in this regard? The ones I'm familiar with specify a hydrotest with the intent that the vessel is not damaged.
I'm not thinking of any code or standard. I'm thinking of end user requirements. A specific example for you: when Shell buy line pipe, they require a hydrotest to a pressure that produces a stress of minimum 95% SMYS. This is a lot more than the basic API 5L hydrotest pressure and is designed to maximise the probability of failing a 'defective' pipe in the short hold time or increasing any imperfections to a size more readily detectable by the subsequent NDT.
SJones,
Thanks for that piece of info. That doesn't seem too much different that the ASME VIII hydrotest that would develop shell stressses as high as 90% of yield for some carbon steel materials. In a static load case such as a hydro test, how do you determine the time for short vs long term. It does not seem as though the test duration would have much influence on the results.
Hydro's can have an impact on material properties and gross structure.
I've always considered a hydro the first cycle in the shakedown cycle.
Hydro's can also provide a modicum of stress relief if high enough and if held for some period of time, can blunt defects if the defects don't result in failure.
I have often specified controlled cyclic hydrotests to "find" defectively made tube-to-tubesheet welds (not normally found during standard Code hydrotest)in fixed tube heat exchangers used in Sulfur Recovery Units.
It is supposed to be nondestructive if the components were designed and fabricated properly or have not been damaged over time to an unacceptable extent, and the test temp is above the min needed for required ductility.
It will affect material properties at local stress raisers- assuming the material has the minimum required ductility at the test temperature and the vessel was fabricated properly, there will be local yielding and some degree of hardening at the stress raisers. Many simplified stress concentration factors are on the order of 2.5-3 ( can be as high as 5 at the edge of a crack), but on a microscopic scale and considering yielding and fracture mechanics, the high stresses result in local yielding and not component failure.
Failure would occur if there was inadequate ductility due to too low a test temperature or incorrect heat treatment such that there was nill ductility, or if the component was not designed or fabricated correctly. The failure due to nill ductility is a rapid progression of the crack, and if it is massive component, can lead to serious injury to personnel who are directly under or alongside the vessel. It sounds bad, but could be much worse- if you were to pnuematically test the vessel and it failed with nill ductility the result can be an explosion and damage within a wide area of teh vessel.
I was at an IMechE conference recently where a paper was presented that described what happened when it was found that the PWHT had not been properly carried out. Instead of repeating the PWHT ahydrostatic pressure test was carried out that was said to relieve the stresses to the same extent as if the PWHT had been carried out.
There are guidelines buried in the various codes, and also in our company proprietory standards for a temperature safety margin between MDMT and hydrotest water temperature.
This is to gaurd against brittle failure during hydro.
Depending on wall thickness, the margin could be as high as 17degC.
For some reason, most fab shops in Australia don't record the test water and metal temperature, but we have it as a standard requirement in our job specifications and test forms.
