Back to Basics

Limitations to the Use of Reference Blocks for Periodic and Preinspection Calibration of Ultrasonic Inspection Instruments and Systems

Introduction
Deficiencies are found in some specification requirements for the use of standard flat bottom hole reference blocks to perform periodic qualification calibrations of ultrasonic inspection systems. Such performance checks often provide little or no indication of the ability of a system to yield satisfactory results in actual material inspections. A legitimate reason for the performance of periodic checks is to discover signs of long term performance degradation that might lead to costly delays during future inspections. A stable test block assembly that has been used in such checking is described. Limitations to the use of standard flat bottom hole reference blocks to check instrument linearity or system sensitivity for material of other shapes, sizes and composition is noted. Appropriate use of such blocks is for the characterization of instrument performance without reference to any specific inspection task.

The most fundamental, and perhaps only, measure of the value of a nondestructive inspection system is the degree to which its indications are representative of discontinuities that affect the usefulness of the tested material for its intended purpose. The phrase "representative of discontinuities" is the key part of this sentence. It implies that the types, sizes, and orientations of potential discontinuities that will affect material usefulness (that is, quality) are known and that their seriousness may be estimated by observing some properties of the indications produced by the examining system. The properties most frequently measured for this purpose are the amplitude, timing, and, sometimes, frequency content of the indications as compared to the responses obtained from artificial targets in a reference standard. Such standards (and the targets within them) are valid only in so far as they represent the material to be inspected and actual discontinuities that may be encountered in them. Ideally, the standard is fabricated from an otherwise good sample of the material to be inspected.

---

Despite these limitations, there is merit to requiring a periodic instrumentation check.

---

Many ultrasonic inspection specifications dictate two types of calibration procedures to which an instrument or system must be subjected. These include the usual preinspection calibration, in which the responses of the system are measured to artificial targets in a reference standard.

The other type of calibration procedure often specified is for a periodic qualifying check of an instrument or system. This often involves measurement of the response to a set of reference block standards. Although such standards, and the responses to them, may have no correlation to the materials, number of discontinuities, or conditions of any actual material inspection to be performed, this type of checking procedure does have important uses as will be discussed below.

 

Preinspection Calibration and Checks During Inspection
The relationship of system indications to seriousness of the discontinuity (size, location, etc.), that is, the sensitivity of the system, must, of course, be estimated before inspection is begun. This is the purpose of the preinspection calibration process.

Theoretical and experimental studies have been made of the possibility of achieving accurate sensitivity threshold selection for ultrasonic examinations without the need for conventional reference standards (Gilmore and Czerw, 1977; Rogovsky and Rose, 1979 and 1980). These methods use calculations based on the relationships among the characteristics of transducers, instruments, typical discontinuities, and materials. Although providing good insight into the interrelations of these properties, this work has not led to widespread adoption of the techniques described for actual inspections.

One way to calibrate the sensitivity of a system for a given inspection might be to observe system responses while scanning a large group of material samples that are somehow known to contain discontinuities representative of various degrees of seriousness. Since this is an obviously impractical, if not impossible, task to perform prior to every inspection, an alternate procedure must be employed.

The most appropriate preinspection calibration consists of observing system response to one or more artificially created anomalies, or targets, of prescribed dimensions, locations, and orientations, in an otherwise good sample of material of the same size, shape, surface finish, and composition as the material to be inspected. Standard ASTM type flat bottom hole reference blocks are sometimes specified to be used as the standard. They are seldom appropriate for this function since their shape, dimensions, and material properties are unlikely to represent the actual material to be inspected and the types of discontinuities to be detected.

This calibration procedure presupposes that the artificial target characteristics have been selected from a database somewhere that has established the validity of the rejection level(s) so defined. The validation might have been established by correlating (empirically or analytically) typical system responses to various artificial discontinuities to those obtained from a group of known discontinuities. The actual seriousness of the known discontinuities would preferably have been itself established by subsequent destructive testing of the samples observed.

In most cases, the characteristics of artificial targets to serve as representative standards for system calibration are selected on the basis of knowledge gained through many years of successful screening of a given type of material for a given purpose using systems calibrated by the response to particular types and sizes of artificial targets.

Of course of equal importance to the selection of artificial targets to represent minimum rejectable discontinuities is having identical equipment and conditions during preinspection calibration as are to be used for the actual testing. This obvious requirement is a part of most test specifications.

 

Periodic System Qualification
The rationale for requiring a periodic instrument or system qualification check is to ensure that the equipment has not suffered performance degradation since its last check that might render it unable to detect the types of discontinuities of interest.

System characteristics required to assure a desired inspection sensitivity are verified automatically each time a preinspection calibration, or calibration check during inspection, is made. This is a much more realistic way of ensuring adequate performance than by depending upon the results of periodic qualification checks using arbitrary reference blocks.

An example of the type of qualification requirement that does not relate significantly to system requirements for real discontinuity detection is, for instance, the measurement of instrument linearity by observing the amplitudes of signals from flat bottom hole (FBH) targets of increasing area in a set of flat ASTM type reference blocks, even though the system configuration and values that are necessary conditions for many required inspections ignore well established criteria for the legitimacy of such measurements (ASTM E 127, 1997; Beck, 1992):

• The entry surface of the material to be inspected is curved while that of the blocks is flat.
• The size, frequency, and damping of the transducer (as well as the bandwidth of the instrument) that are required to produce an approximately linear area amplitude relationship, are not appropriate to the type of inspections to be made.
• The actual accept/reject decisions to be made do not require a linear amplitude relationship but involve only the comparison of the amplitude of the signal from an indication to a fixed threshold level determined by a reference target in the calibration standard.

Other examples of typical system qualification requirements that bear little or no relationship to the performance required for discontinuity detection in actual test situations to be encountered include:

• Calibration of gain control accuracy over a greater range of output levels or instrument display indications than will be used in any actual inspections to be performed.
• Checking instrument resolution under conditions that do not duplicate inspection situations to be encountered.
• Measuring instrument output noise level under conditions that have little connection with actual inspection conditions.

These potential shortcomings notwithstanding, it might be argued that it is useful to employ some type of arbitrary reference block to furnish standard targets for evaluating, on a periodic basis, instrument sensitivity, resolution, and noise performance under fixed conditions. Data so obtained would apply to a specific instrument only, with a specific transducer continually coupled to the reference block. Performance figures obtained for one instrument would not apply to another instrument of different pulse shape, damping, and spectral response, or to the results obtained from the same instrument used with another transducer of different size, frequency response, damping, piezoelectric coupling factor, orientation, and coupling to the block.

Because of the possibilities for changes in transducer characteristics with time, and the variations that are possible in transducer orientation and coupling efficiency (and even the possibility of changes in block surface condition due to handling) it is doubtful that a given block and transducer (contact or immersion) can furnish a very accurate series of readings to be used for evaluating the stability of a given instrument's characteristics over an extended time period, much less its suitability for performing inspection of a particular material for a particular type and size of discontinuity.

Despite these limitations, there is merit to requiring a periodic instrumentation check. One benefit of periodic checking of instrument performance is that it may reveal a gradual degradation in certain areas that, while not serious enough to affect the ability to detect discontinuities, as verified by calibrations taken before and during inspection, does indicate the possible need for maintenance procedures to be carried out. This could prevent future breakdowns or failures to stay in calibration between checks made during inspection, either of which could result in costly delays.

One method that has been used to overcome the difficulties outlined above in obtaining a suitable means for long term evaluation of changes in instrument performance consists of the construction of a test block, as shown in Figure 1. This block has a permanently bonded quartz transducer on one end and several side drilled holes at different distances from the transducer end. The block is housed in a case to protect it from external influences. It provides both an amplitude and timing reference. This type of block has long been used to evaluate several features of instrument performance for various types of instruments. Although there is no way in which such a block can be traced to any known NIST standard, it has proved to be a remarkably consistent device for evaluating the stability of a given instrument's performance from one calibration period to the next. No relationship is implied between instrument performance so measured and its suitability to perform any given inspection.

Figure 1 - Test block for periodic instrument checks.

It may be desirable to perform periodic checks on the amplitude linearity of an instrument even though accept or reject decisions are made without reference to multiple amplitudes. Changes in this characteristic may often reveal deterioration in electronic components in the instrument that might ultimately lead to failures. The accuracy specified for this characteristic should not exceed what is needed for test validity or what is theoretically possible with the checking method specified. It is impractical to do this reliably within close tolerances by using the area amplitude characteristic of signals obtained from flat bottom hole targets. This is made very clear by observing that, even with the very specific transducer and instruments specified for this purpose by ASTM E 127, that practice, based on careful NIST studies, only specifies linearity as measured by standard ASTM area amplitude blocks to be within +2, -3 dB accuracy limits (ASTM E 127, 1997).

A superior method for measuring instrument linearity is a calibrated attenuator (Bailey, 1977). ASTM Practice E 317 also provides for this method,which measures true instrument linearity to a high degree of accuracy, is traceable to NIST standards through the attenuator calibration, and removes the uncertainties and inaccuracies due to variations in transducer coupling, orientation, wavelength, and beam shape.

 

Appropriate Use of Flat Bottom Hole Reference Blocks
There is no absolute correlation between the amplitude of the signal response obtained by a given ultrasonic discontinuity detection instrument from a flat bottom hole (FBH) target in a standard ASTM type reference block and the signal that might be produced by the same instrument from an actual discontinuity in material of another size, shape, and composition. There appears, therefore, to be no justification for using such blocks to set rejection criteria for the inspection of material of shapes, sizes, and compositions other than that represented by the blocks themselves. This point is explicitly made in ASTM Practice E 428 for steel reference blocks in the section titled "Material Selection," and throughout the entire text of ASTM Guide E 1158 for material selection for reference blocks.

Exceptions to the forgoing statement may be made in cases where extensive successful use has generated enough data to validate sensitivity calibration with a standard block to set rejection criteria for certain types of discontinuities in certain specific materials of other shape and composition when procedures are adhered to that limit the characteristics of the transducers and instruments employed.
The most important use of these blocks lies in their application to the measurement of the characteristics of one instrument or transducer as compared to another under controlled, standardized (within limits), and traceable target conditions. General sensitivity, resolution, and noise performance can thus be assessed and used as a common basis for instrument or transducer specification. These data must, however, be carefully analyzed to determine their significance with respect to a judgement of suitability for a given inspection application.

 

Conclusions
Requirements for periodic instrument or system qualification may often be overstated in inspection specifications. As sometimes specified, they can be impractical to perform, or worse, may be based on mistaken assumptions about relationships among instrument, transducer, and material properties. These requirements may also be misinterpreted as having a direct bearing on the ability of a system to detect discontinuity of a specified type in a specified material. The real use of qualification checking should be to reveal long term changes in system performance that might lead to future breakdowns or delays during calibration or testing, or which might require frequent recalibrations to maintain sensitivity during testing. For this purpose, care must be exercised to specify checking by means that have long term stability and accuracy.

It must again be stressed that, whenever possible, reference standards used to set sensitivity and rejection limits for inspection of a given material should be fabricated from samples of the same size, shape, and composition as the material to be examined. There is also ample evidence to indicate that targets in reference blocks should not be used to measure instrument linearity.

 

References
ASTM E 1158, "Standard Guide for Material Selection and Fabrication of Reference Blocks for the Pulsed Longitudinal Wave Ultrasonic Examination of Metal and Metal Alloy Production Material," Annual Book of ASTM Standards, Vol. 03.03, 1997.

ASTM E 127, "Standard Practice for Fabricating and Checking Aluminum Alloy Ultrasonic Standard Reference Blocks," Annual Book of ASTM Standards, Vol. 03.03, Section 11.4. 1997.

ASTM E 317, "Standard Practice for Evaluating Performance Characteristics of Ultrasonic Pulse-Echo Testing Systems Without the Use of Electronic Measuring Instruments," Annual Book of ASTM Standards, Vol. 03.03, Section 5.3.3. 1997.

ASTM E 428, "Standard Practice for Fabrication of Steel Reference Blocks Used in Ultrasonic Inspection," Annual Book of ASTM Standards, Vol. 03.03, Section 4, 1997.

Bailey, J. A., "Ultrasonic Pulse-Echo Instrument Linearity Measurement," Materials Evaluation, Vol. 35, No. 5, May 1977, pp. 64-68.

Beck, K. H., "Ultrasonic Area-Amplitude Limitations," Materials Evaluation, Vol. 50, No. 8, August 1992, pp. 978-985.

Gilmore, R.S., and G. J. Czerw, "The Use of Radiation Field Theory to Determine the Size and Shape of Unknown Reflectors by Ultrasonic Spectroscopy," Materials Evaluation, Vol. 35, No. 1, January 1977, pp. 37-45, 56.

Rogovsky, A. J., and J. L. Rose, "Additional Aspects of the Standardless Sensitivity Selection Technique," Materials Evaluation, Vol. 38, No. 9, September 1980, pp. 47-52.

Rogovsky, A. J., and J. L. Rose, "Ultrasonic Sensitivity Selection Based on Probabilistic Techniques," Materials Evaluation, Vol. 37, No. 4, March 1979, pp. 47-55.

* Chair, Technical Instrument Corp., 152 Mercer County Airport, Trenton, NJ 08628-1392; (609) 882-2894; fax (609) 882-3147; e-mail tactices@aol.com.

 

Copyright © 1999 by the American Society for Nondestructive Testing, Inc. All rights reserved.

IRRSP, NDT Handbook, The NDT Technician and www.asnt.org are trademarks of the American Society for Nondestructive Testing, Inc. ACCP, ASNT, Level III Study Guide, Materials Evaluation, Nondestructive Testing Handbook, Research in Nondestructive Evaluation and RNDE are registered trademarks of the American Society for Nondestructive Testing, Inc. ASNT exists to create a safer world by promoting the profession and technologies of nondestructive testing.