Re: Is the GE specification good and still relevant? Rolf -
Near surface resolution remains a difficult area to
specify because the actual parameters you need depend on
what size defect you need to detect and on the signal-to-
noise you need to reliably detect it. Certainly the GE
specification will not be adequate in some cases, it
will be in others. So my answer is it depends.
One very good way to make a specification for a
particular test is to pin everything down. Specify the
near surface resolution in terms of a reference FBH a
certain distance from the surface and the dB of signal-
to-noise you need. The GE specification is of this type
but not completely rigorous because it doesn't specify
how you measure SN. If you want to detect 0.4 mm FBH 1.2
mm into steel. The GE spec. is good although putting a
more "scientific" definition on the two wave amplitudes
If you want to apply the GE spec. to another material
that is more attenuative for example, then the GE spec.
will be inappropriate. Specifying the near surface
independently from the application is as far as I can
see beyond our current knowledge. The "do everything in
steel and translate to other materials method" is of
limited use to most ultrasonic users because there are
no "translation" codes available to do that. Indeed
there is not much data on attenuation as a function of
frequency out "there."
I might also add that near surface resolution is more
complex than most ultrasonic measurements because it is
a function of almost every parameter you can think off.
Center frequency, band width (damping), material
attenuation, element diameter, focal length, instrument
setting, and water path all change near surface
resolution. So I think you are stuck with using the GE
Spec. and "translating" it experimentally to your needs
or of creating a spec. specific to your own needs.
: The GE spec describes the probe's and equipment's capability
:for its near to surface resolution.
: See also GE Spec in UT A-Z.
: For my opinion this specification let to much parameters open,
: for instant there is not any dB value mentioned about the
: signal noise ratio at the flaw's echo. Is this true?
: Are there better standards available?