Further clarification of terminology

There has been additional correspondence relating to the use of the term ‘calibrating’ from Stewart McMillan, who has offered the following two cents’ worth.

Stuart had recently read an article on AWS D1.1 in an ASNT journal where the author quoted ASTM 1316, which apparently states that with regard to “the proper term for setting up a machine to perform a particular inspection technique, you are standardizing, not calibrating”.

There are obviously synonyms that can be looked at, for example setting or assimilating, but standardising is actually what one is doing when setting a distance amplitude curve (DAC) or time-corrected gain (TCG), etc. The Germans use ‘Justierung’, which in English would translate to ‘justification’.

Stuart’s personal experience when having multiple ultrasonic testing (UT) technicians on a job was to ensure that their sensitivity settings were identical, so as not to have a flaw rejected by one person that would have been accepted by another because of amplitude. Remember, 0.5 dB is the difference between ‘accept’ and ‘reject’.

On another theme not unrelated to the accept/reject issue linked to amplitude is the measurement of uncertainty (MOU) concept. This is not the same as probability of detection (POD) but relates to the measuring of a detected indication. For the MOU concept, a percentage of uncertainty is applied to a measurement and for the ease of mathematics I will use 10%. If the indication is measured at 20 mm and the 10% MOU is used as the range in which there is confidence, the value considered would be 18 mm to 22 mm (±10% either side). If the acceptance criteria states that indications of 20 mm or more are rejectable, what initially looked like a straightforward case of reject is no longer so as the uncertainty value could be under 20 mm at 18 mm, hence acceptable, or as high as 22 mm and rejectable. This means that the indication cannot be clearly sentenced when in the vague zone where the value from the MOU is either side of the accept/reject limit and the client is probably going to be most unhappy with the situation. How accurately can measurements be taken when hanging from a rope in poor weather conditions halfway around a 20 m-diameter oil rig leg? In laboratory conditions on smaller components the measuring will be more accurate but, again, when the method being used is penetrant testing, the indication size will change due to bleed out; what measurement would be the most accurate and when should it be carried out? I am interested to see if the MOU concept elicits any responses from readers.

Comments by members

This forum post has no comments, be the first to leave a comment.

Submit your comment

You need to log in to submit a Comment. Please click here to log in or register.

<< Back