Allowable stresses and material test reports

[johcatrac]

If an analysis is done to the ASME I, III or VIII code for a loading that is extremely unlikely say with a probability of 1x10-6, and if the allowable stresses are exceeded, can we use the values specified in the actual material test reports as the true allowable stresses? If we are allowed to do so, what would be the Code basis?
As we know, the values in the material test reports are always higher than the code specified.

 

[metengr]

You use the code allowable stress values in design and for load extremes.

 

[BenStewart]

Unfortunately not.

What scenario are you trying to design for? You quoted a "low" probability, are you looking at blast over-pressure, earthquake? I seem to recall in pre '07 Div 2 that for linear-elastic analyses there was a k factor applied to the allowable stress and depending on the load combination being assessed that factor varied between 1 and 1.2. This no longer appeared in Div 2 post '07, instead it was acknowledged that time varying loads (say, maximum wind and earthquake) do not occur coincidentally and the load factors for these loads within the load combination were reduced to less than unity; see Table 5.3 in the current Div 2 for linear-elastic analyses.

 

[israr1]

For Sec I & III, i dont know.
But for Division VIII, you cannot go for allowable stress higher than that specified by code.

Better if you can share the loading for which you are apprehensive.

 

[johcatrac]

The loading is 'overpressure caused by heating of water trapped in piping between two valves'. The heating is by virtue of a very low probability steam line break in the room that this piping is in.

 

[TGS4]

Sounds like a double/triple jeopardy thing to me. Take care of that in your process safety management plan.