By CWI555 
Date 10-05-2007 05:05
Edited 10-05-2007 05:13
As a partial proof, I did find something of interest publicly documented. It's in API RP 571. 4.2.17 vibration induced fatigue.
"
4.2.17.3 Critical Factors
a) The amplitude and frequency of vibration as well as the fatigue resistance of the components are critical
factors.
b) There is a high likelihood of cracking when the input load is synchronous or nearly synchronizes with the
natural frequency of the component.
c) A lack of or excessive support or stiffening allows vibration and possible cracking problems that usually
initiate at stress raisers or notches."
Since API has listed this, It's reasonably safe to quote it. In one specific incident there was a failure of the socket much like drawing NDTIII provided that I was party to the failure analysis on.
It was determined that the excessive gap created an unstable fluid flow within the socket, setting up harmonic vibrations that eventually cracked the part out which subsequently dramatically failed injuring one worker with burns from the fluid. The power of harmonic vibrations cannot be underscored enough. It was found that there was sufficient support, the weld was made correctly except for the excessive gap, pressures and fluid were per normal and well within tolerance, but due to the slight change in the natural frequency created by the excessive gap and subsequent flow disturbance, it made the difference. Had that socket been made per normal industry standard, it was and still is believed this particular incident would not have happened.
As for this comment:
"And, if people are really dying, if people are really being seriously injured, how efficient could this secrecy be?
I do not believe that signing a secrecy agreement that would conceal information about practices that threaten life and limb would hold up in court."
"for a more robust and efficient form of secrecy"
My reference was to DOD and DOE work. "edited for clarity" simply put, if something is branded a national security matter, the same rules do not apply.
Back in the standard commercial world, the large multinationals do not enjoy the same protection, but they can make your ability to be employed rather difficult.
In the real world, chit happens every day that isn't right, that could be prevented, but for "cost" reasons, is over looked.
As for the inspectors duties, most often there are standard industry practice clauses in contracts, if I see that, I'll stick to it as it's standard to use 1/16". If there is nothing there, I'll sure elevate it to make damn sure the "engineers" who for whatever reason "missed" this, are aware of it, and get their answer in writing.
Early on in my career I've been victim to that "well you never said anything" crap from engineers when they screw up the specs, for which I follow, then somehow there lack of planning and foresight becomes my problem when I follow what they put in writing. I do not play that game with anyone anymore. Therein is the inspectors duty to raise the concern in writing and get an answer in the same, If they (engineers) insist, it's on their heads then, but they are not going to stick me with the silence is acceptance routine again.
I've heard this multiple times in various meetings and projects, and have had it happen to me. It happens with a higher frequency that some realize. For this reason, I keep detailed records of everything I do. 99 percent of which will never see the light of day unless something screws up and the engineers start looking to shift blame.
That is where I am coming from. Now if your talking about someone making up their own code and specs, thats another story. For some of the same reasons listed above, speaking up without a basis for either an engineer or an inspector can be a career ending decision. For me, my goal is to assure the items meet the minimum AQL as set forth in the codes and standards. If they do not, and the engineer wants it to go anyway, they are going to put that in writing or find themselves some other sucker to do their handy work.
