There is no an upper limit of hydrotest pressure in VIII-1 3

[mechengineer]

Hi All,
UG-99(d)mentioned: “…This Division does not specify an upper limit for hydrostatic test pressure. However, if the hydrostatic test pressure is allowed to exceed, … the Inspector shall reserve the right to reject the vessel….” It does not make sense to justify whether it is safety for using such pressure to perform a hydrotest. In fact, in case of the MAWP calculated equals or is closed to the design pressure, the stress calculated (a primary membrane stress) in the wall of cylinder may exceed the allowable stress, even beyond the 90% yield limit during hydrotest.
I think that the Code (VIII-1) should make a simple rule as a guidance to prevent the hydrotest from overstress. For example, the primary membrane stress in hydrotest condition shall be checked and the allowable hydrotest stress shall not exceed 0.9 yield stress. Does anyone know what is the reason that ASME VIII-1 do not give the upper limit of hydrotest pressure or the allowable test stress?

 

[BJI]

Why does it need to? It provides more flexible by allowing it to be left to the discretion of the AI, if it doesn't cause permanent distortion of the vessel then it is acceptable (and most likely beneficial).

You could pressurize to 90% of general membrane and get GPD around a nozzle opening (especially with stainless steels), so what stress limit would you propose to avoid this, where should it be measured, how should it be calculated and how do you assess GPD? For division 1 vessels it seems more pragmatic enforce a minimum limit and leave the upper limit to visible distortion/damage.

 

[TGS4]

I'm not sure that you fully appreciate the nature of the ASME BPV Code Committees. Asking a question such as "why doesn't the Code do this..." is a pointless exercise in futility. The answer for many of these question is that either no one has proposed such a clause/change and therefore the Code may be ripe for such a change, or that it has been proposed in the past (to a set of engineers different from the current membership) and it was rejected. The latter is not to say that it couldn't be re-proposed.

It is true that many owners/users already impose their own limits via their owners/users specifications. That is entirely their prerogative and nothing in the Code prohibits that. It is worth noting that VIII-2 DOES have a limit - that is discussed in 4.1.6.2 (which received a significant revision in the 2017 Edition, BTW). B31.3 also imposes a limit.

What also needs to be understood is that "yield" in the context of Code work is actually the engineering yield value, which is obtained using the 0.2% offset method. That means that there is already some permanent deformation going on at this "yield". If one wishes there to be no permanent deformation, one would need to use the proportional limit, but that stress value is typically much less than the allowable stress itself. In the end, these re all arbitrary points, anyway. The limit that is currently provided is a fairly robust criterion, if subject to significant interpretation.

 

[mechengineer]

TGS4, thanks for your valuable comments.