Nondestructive testing techniques are increasingly applied to equipment, systems, and components to insure the continued safety and performance reliability of operating chemical plants and refineries. Most pressure vessels, piping and equipment components operating at temperatures ranging from sub-zero temperatures, where metals can become brittle, to very high temperatures, where they can fail by creep, have been subject to extensive nondestructive testing during the original construction of a plant. As these plants are operated, however, the materials utilized are subject to metallurgical changes as a result of the service conditions.

Obvious conditions of deterioration involve corrosion, erosion, fatigue cracking, and creep or combinations of these which essentially involve a reduction in the cross sectional thickness of the material. When severe reductions in the thickness of the material have occurred, the strength of the vessel, pipe, or component may become inadequate to withstand the applicable static or dynamic loading conditions. More difficult to detect by nondestructive testing techniques are conditions involving changes in the properties of a material involving embrittlement as a result of hydrogen embrittlement, graphitization, carburization, strain aging, etc., which can also affect significantly the properties of a pressure vessel, a heater, fan, compressor or other equipment.

If these changes are not detected, they may result in catastrophic failures of the component in question. However, even leak-type failures in some of the systems may not only cause costly interruptions of production, but may also result in undesirable environmental consequences or, on occasion, in fires or explosions.

Meaningful nondestructive testing and inspection techniques must be carefully selected and utilized during the operations of these plants not only during periodic outages, but also between outages, to detect potential deterioration in the materials utilized. Early meaningful detection of deterioration by nondestructive examination techniques may result in recommendations for changes in the operating conditions, which will either entirely stop or at least reduce the rate of the deterioration.

The various nondestructive examinations utilized include not only the conventional methods such as ultrasonic, radiographic, magnetic particle and liquid penetrant examinations, but also techniques such as eddy current testing, pipe crawlers for internal visual pipe inspections with video cameras, infrared techniques, replication, capacitive strain gage testing, etc. Special procedures may be required for the early detection of leaks applicable to vessels and equipment provided with weep holes or other leak detection systems. Examples of the applications of these various techniques and of the results detected will be presented in the paper.

In all these examinations, it is important to separate those indications that represent deterioration which potentially can lead to failures from other inconsequential indications involving minor corrosion, superficial surface fissuring, or original manufacturing or welding defects, etc.

Nondestructive testing principally involves surface examinations and volumetric examinations. The surface examination techniques primarily involve liquid penetrant examination generally utilizing red dyes, and magnetic particle examination. Magnetic particle examination may involve wet fluorescent magnetic particle liquids, or dry magnetic powder. Although magnetic particle examination may detect subsurface flaws, these techniques are primarily used for the detection of defects or discontinuities extending from the surface into the wall thickness of the component being inspected.

Volumetric examinations involve radiographic examinations (isotopes or x-rays), ultrasonic examinations and eddy current testing.

Significant improvements have been made both in nondestructive examination equipment and in specific techniques utilized. Thus, defects that may not have been revealed by nondestructive testing performed five or ten or forty years ago, or even currently, may then be detected by more highly sophisticated nondestructive testing equipment or techniques available currently. One of the problems associated with nondestructive examinations then becomes the evaluation of a defect condition that was not apparent when the original inspection was performed, but is discovered with more sensitive equipment or a refined procedure.

An operator may thus believe that the absence of recorded defects during prior inspections and the detection of a defect or discontinuity during a current inspection represents a condition such as progressive cracking or progressive corrosion deterioration. An erroneous interpretation of this may then result in the shut down and repair or replacement of the vessel, pipe or equipment, when, in fact, the defect represents an original condition that would never have resulted in a leak or a rupture.

There also are many types of vessels and equipment that cannot be inspected in a meaningful manner while the equipment is in operation. In other words, a proper inspection would then require the equipment to be shut down and opened or disassembled so that access could be gained for the visual and nondestructive examination of critical areas or parts in the specific vessel or component being questioned.

For example, recently a urea reactor vessel exploded catastrophically in an ammonia plant, resulting in the death of a number of plant employees. Even though the welds in this equipment were reported to have been inspected by ultrasonic examination techniques, the techniques actually utilized did not detect major cracks that existed at weld locations of dissimilar compositions at the time the vessel was fabricated. Had these cracks been detected by more refined nondestructive examination techniques performed by more highly experienced NDE personnel, proper repairs could have been made.

Similar conditions may exist in equipment that is currently in operation as service related deterioration involving cracking, corrosion or erosion may or may not be detected by the inspection techniques or equipment utilized, and, in particular, by the inspection procedures utilized. Not enough, can therefore be stated about the importance of the experience and qualification of the engineering-inspection personnel utilized in the performance of the inspections necessary and in the interpretation of the inspection results to confirm the continued integrity and safety of the equipment involved.

For example, at another plant a urea reactor vessel recently failed catastrophically as a result of cracking through the thickness of the stainless steel liner alongside a girth weld. The stainless steel liner represented a separate shell that had been press fit to the inside of the carbon steel pressure vessel shell. Leakage through the crack of the liquid maintained at a high pressure on the inside of the stainless steel liner resulted in corrosion of the carbon steel shell. Although the pressure vessel which consisted of many layers of steel was provided with weep holes, the weep holes did not reveal that leakage had occurred through the liner. The corrosion that occurred was not detected by the inspections of the vessel shell, which due to the multilayer construction were performed externally. The loss in shell strength caused by the corrosion subsequently resulted in a catastrophic failure of the vessel. In this respect it must be recognized that layered pressure vessels, when they are accessible only from the outside surface of the vessel, do not lend themselves to the type of volumetric nondestructive testing that can be performed by ultrasonic examination of the shell thickness of an entire vessel cross section.

To demonstrate the importance of meaningful nondestructive examinations, various examples will now be presented illustrating typical defect conditions that may be encountered with interpretations provided as to the effects of these conditions on the safety and operability of the equipment involved.

A visual examination of a multi-layered urea reactor vessel, indicated that an overflow condition had occurred. Removal of the insulation from the vessel, which had been constructed of 21 layers of carbon steel, followed by liquid penetrant examination, revealed a type of cracking. This cracking was interpreted as caustic stress corrosion cracking. The early detection by the visual and liquid penetrant examinations performed confirmed that only the outermost two shell layers had cracked. This permitted replacement of the cracked shell section at the plant site, instead of dismantling the vessel and shipping it to a pressure vessel fabrication shop for repairs.

A liquid penetrant inspection was performed of a 2-1/4 Cr 1 Mo low-alloy steel valve body to confirm its suitability for service. A crack was revealed by this inspection. This crack was interpreted as a casting defect and thus was not repaired by welding.

Two reinspections were performed after operating periods of approximately 30,000 hours each. For reference purposes, the operating temperature was 9500F, and the pressure was 2000 psi. The crack was inspected again after an operating period of 60,000 hours. The results of the liquid penetrant inspection, when compared to the results of the two previous inspections, confirmed that crack progression had not occurred. Thus repair welding or replacement of the valve at the crack location was not necessary.

Deaerators, as extensively installed in large chemical plants generating their own power, are subject to cracking both alongside welds and transverse to welds. This cracking is generally associated with water hammer, which can introduce stresses as high as 30,000 psi. To ascertain the continued integrity of deaerators and deaerator storage tanks, insurance companies and interested trade associations, such as NACE, recommend inspections at 1-, 2-, or 3-year intervals. A proper inspection of deaerators to detect this type of cracking would normally utilize the wet fluorescent magnetic particle examination technique. The cracking tends to occur in the heat-affected zone either along the edge of the weld or in the heat-affected zone perpendicular to the weld. When cracking occurs, it normally is noted near every weld in the deaerator. To determine the significance of this cracking, exploratory grinding is performed. Typically, this would involve grinding a "slot" across the indication to determine its depth Where the depth of indications is relatively minor and does not extend below the minimum design wall thickness, repairs by welding may not be necessary. When the conditions are more severe, repairs by welding may be necessary.

Similar conditions can occur in flash tanks and other equipment containing water subject to water hammer and significant pressure cycling.

A routine inspection of a cast iron flywheel on a large compressor in a chemical plant revealed extensive linear indications. The compressor had been in service for a period of several years. The owner was advised by the manufacturer to purchase a new flywheel which would have required the compressor to be shut down for a period of several months. Subsequent magnetic particle examination revealed indications in an area that had been previously ground. The indications in the ground areas were interpreted initially by plant personnel as representing new cracks that had occurred where cracks had previously been removed by the foundry.

Careful engineering evaluations of the indications resulted in the interpretation that these represented casting defects which had not been subject to progression. On the basis of an engineering report subsequently issued, the continued utilization of the flywheel was recommended. The plant operator thus did not replace the flywheel.

The compressor with the original flywheel has now been in service for a period of 12 years without any evidence of crack progression or new cracking.

Radiographic examinations are frequently required on welds in pressure vessels, piping, and other components constructed to the ASME Boiler and Pressure Vessel Code, the ASME B31.3 Code for Pressure Piping on "Chemical Plant and Petroleum Refinery Piping", and by other piping Codes that may apply to specific sections of a chemical plant. Similarly, radiographic examinations are required to tanks constructed to the applicable API Standards.

During the past forty years, radiographic examination techniques have improved significantly. Thus, welds in pressure vessels, tanks, and piping that had been radiographed in the 1950s or even 1960s, may contain defects that at that time were not detectable or, when detected, not considered significantly severe to be rejectable under the applicable Standards. Although the acceptance/rejection standards of the various ASME Pressure Vessel and Piping Codes and API Standards generally have not changed during the past forty years, the delectability of defects has improved significantly. Thus, where radiographic examinations are performed with current techniques, defects may be found that are considered rejectable to the Standards that apply currently or that applied when the original radiographic examinations were performed some twenty, thirty, or forty years earlier. Questions then arise if the defect is in fact related to a condition of deterioration involving progressive cracking, corrosion, or does the defect represent an original manufacturing, casting, or welding condition that has not changed during the entire operating period of the equipment.

Two pipe welds with lack of penetration and incomplete fusion defects were not apparent on the radiographic films taken about 40 years ago with radium radiation sources. Reradiography performed during the past 5 years with Iridium 192 sources and high contrast sensitive films then revealed these defects. However, crack progression or corrosion had not occurred. Thus, in the absence of actual deterioration of the welds, the weld joints were acceptable.

Various types of ultrasonic examination equipment and techniques are utilized to evaluate the condition of equipment, pressure vessels and piping in the chemical industry.

One of the most widely utilized techniques involves ultrasonic wall thickness determinations. The purpose is to detect any loss in wall thickness in pressure vessels, piping, tanks or other equipment. The equipment most widely used involves digital readouts that are then recorded manually or by computer techniques; i.e. data logging.

Since in many chemical plants corrosion can occur in piping either during normal operations or when the plant is shut down, inspection of piping thus is particularly important where unusual corrosive conditions can occur. A pipe rupture occurred as a result of general corrosion that developed along the 6:00 o'clock position in a piping system, where, during shutdown periods, water collected along the bottom of the pipe. Because of relatively high sulfur levels contained within the petroleum product flowing through the pipe, sulfuric acid formed. In time, the acid corrosion reduced the wall thickness of the pipe sufficiently that it was unable to withstand the operating pressure and ruptured.

Chlorination tanks generally are fabricated of carbon steel. Since the chlorine solution is highly corrosive, the inside of these tanks generally are lined with ceramic. In some instances, overflow has occurred, allowing the chlorine solution to run down on the outside of the carbon steel tank. Severe rains can also leach through the insulation and activate the chlorine deposits on the outside of the carbon steel shells from prior overflow incidents. In time, over a period of years, this can result in a significant loss in wall thickness accompanied by heavy scale formation. If the reduction in wall thickness is not detected, tank ruptures can and have occurred. Corrosion had occurred along the outside surface of a chlorination tank which extended through the roof of a building. At specific locations representing a 611 grid, the heavy scale layers which had built up on the vessel were removed by flaking off scale segments until all loose scale had been removed from the shell at the grid locations. The surface was then carefully scraped with a hardened steel edge to expose shiny metal surface areas on the tank shell. Ultrasonic wall thickness determinations were then made. Although the original tank shell at this location involved wall thickness values of 0.250" and heavier, the remaining shell thickness showed values ranging from approximately 0.120" to 0.176". Although these represented values below the minimum design wall thickness, the tank was considered sufficiently safe for continued operation until a replacement tank could be erected within a period of two months. In the meantime, further corrosion was stopped by maintaining the outside surface of the chlorination tank free of moisture.

Shear wave techniques are also utilized extensively to detect flaws and defects, as well as progressive conditions of corrosion and cracking. Although these techniques are well established, interpretation is still subject to considerable experience by the ultrasonic examiner. An oscilloscope image was interpreted to represent a 3-1/2" deep crack in a 5-1/4" thick vessel shell. Subsequent removal of a 4" diameter plug from the vessel to evaluate the reported crack, showed that the indication was due to intermittent stringer-type laminations which had been misinterpreted as cracking.

Improved techniques and interpretations are made with the utilization of digital ultrasonic inspection techniques in conjunction with a PC-based signal analysis system to evaluate critical components.

The system uses a standard ultrasonic pulser receiver with a high-speed analog to digital converter. This allows the normal analog UT signal to be digitized along with transducer position data and recorded on a computer disc. Once recorded, several analytical tools can be used to evaluate and characterize the signal. These include 3-D imaging, filtering, frequency analysis, power spectrum and many others. This system also provides a permanent record of the inspection results and calibration data which is valuable for comparison with future inspection results.

There are three primary factors that affect the accuracy and repeatability of ultrasonic inspections. They are the calibration procedure, the inspection report and operator effectiveness.

The calibration procedure involves both the instrument and the operator. The calibration of the instrument should be performed using a standard that is as similar as possible to the material and geometry of the part to be inspected. Whenever possible, a section of a similar part that has been removed from service should be used. It is important for the plant to try to develop and maintain an inventory of these standards such that they can be utilized for subsequent inspections. When possible, the calibration information should be digitally recorded and stored on computer discs for retrieval during future inspections. The "calibration" of the operator involves informing him of the type of defects that may be encountered. Whenever possible, samples from failed parts should be provided. This will allow the operator to develop signal recognition skills.

The accuracy of the inspection report is extremely critical for comparison with future results. These reports must be very accurate and detailed. Whenever possible, all reportable reflectors should be digitally recorded and archived on computer disc for comparison with future inspection results.

Finally, the effectiveness of the equipment operator must never be overlooked, even with the most sophisticated, state-of-the-art equipment. The inspection is only as good as the operator. The operator must be qualified and should have experience performing the same or very similar inspections. In addition to qualified operators, there should always be experienced engineering personnel available to provide evaluations of all indications that are identified.

In recent years, increasing utilization is also being made of videoborescopes for the internal inspections of vessels and particularly pipes. More advanced techniques utilize pipe crawlers that can travel up to 600 ft. through the inside of pipes. This allows the detection of cracks and corrosion deposits.

Considerable experience is necessary to the interpretation of cracking, corrosion deposits, and other defect conditions on the inside of pipes. Videoborescopes tend to magnify significantly any defect. Thus a relatively insignificant defect may appear far more severe when detected with a videoborescope.

For best interpretation, it is generally advisable to compare the results of videoborescopic inspections with photographs of similar defect conditions or samples containing actual defect conditions. The centerline shrinkage crack in a weld detected with a videoborescope represented an original crack that occurred in the root pass of a weld of a 1-3/4" thick cross section. Subsequent ultrasonic examination confirmed that the weld, as expected from other engineering evaluations, had not been subject to crack progression. Had a repair of this weld been required, the cost would have exceeded about $200,000.

Pressure vessels that involve multi-layer shells with liners, or a single shell with a liner, and that are designed to Section VIII of the ASME Boiler and Pressure Vessel Code, generally are provided with weep holes.

The purpose of the weep holes is to allow detection as early as possible of any leak which develops in the inner liner of the vessel. Since inner liners may involve stainless steel, nickel alloys, titanium, or other materials, any leak that develops through these liner materials, either as a result of a crack, or a defective weld, can be followed by corrosion of the carbon steel or low-alloy steel shell or multiple shell layers. If this corrosion, which sometimes may be relatively rapid, is not detected, and continues to occur, a major vessel rupture may result.

A stainless steel liner on the inside of a pressure vessel developed a crack through the wall thickness of the liner. This crack was detected by visual examination and confirmed by liquid penetrant examination of the inside of the liner during a shut-down inspection of the vessel. As the crack had progressed through the wall thickness of the liner, gradual corrosion of the carbon steel shell took place. Fortunately, since the cracking through the wall thickness of the liner was noted in the early stages, significant corrosion of the shell had not occurred. After the corrosion of the shell had been confirmed to be minimal, the vessel was readily repaired by welding the liner.

If the liner cracking and leakage and the corrosion of the shell or shell layers is detected in the early stages of deterioration, repairs by welding generally can be readily accomplished.

In a 10,000 psi acetic acid pressure vessel, cracking occurred through a 1/2" thick stainless steel liner after about two years of operation. This resulted in corrosion extending deeply into a 2" thick shell plate which was surrounded by an additional 8" of thick layered steel bands intended to place the shell under compression. The corrosion produced grooves and penetrations approximately 1-1/2" deep into the 2" thick shell. The leakage was detected by an inspection of the weep holes.

The subsequent evaluation confirmed that repairs by welding were entirely feasible. The repairs took approximately one month to accomplish. The replacement time of the vessel, if repairs had not been possible, would have involved the period of approximately two years.

If the leakage had not been detected by the very thorough examination of the weep holes, and the corrosion had continued, a major catastrophic failure could have occurred.

Another urea plant involved a pressure vessel of multi-layer construction with cracks at a weep hole location. This cracking was the result of stress corrosion cracking of the carbon steel shell layer by the process liquid on the inside of the vessel.

Thus, during any vessel inspection that is being performed, including vessels of layered construction, or with internal liners, an inspection of all of the weep holes is considered very important. Inspections should be very carefully performed by experienced engineering inspection personnel and include the following:

1. Visual examination
2. Checking for moisture
3. Soap bubble tests
4. Detailed weep hole examinations with lighting sources
5. Borescope examinations (if diameter of weep holes is sufficiently large so that borescopes can be inserted).
6. Utilization of special tubular bore scrapers to collect deposits for determining if the deposits are the result of external atmospheres or represent corrosion products resulting from leakage from the inside of the vessel. Examinations to determine locations of deposits from OD to ID of vessel.
7. Analysis of deposits collected from OD to ID.
8. Evaluation of deposits or contaminants for determination of source of deposits (external or internal leakage).

If weepage is noted the equipment should be shut down and opened for an internal inspection. It should be recognized in multi-layered vessels that weep holes may involve continuous holes or may represent staggered holes with staggering varying from one layered vessel plate to the next layered vessel plate. In some instances it may be necessary to drill a special hole into the shell to verify leak conditions at suspected areas. This drilling must be performed extremely carefully and must not extend into the inner shell.

In any case, internal vessel inspections are extremely important to confirm the location of leaks. Such inspections require highly qualified and experienced engineering/inspection personnel as there have been a number of instances where inexperienced inspectors have missed minute cracks or weld defects that, when the liner was pressurized, allowed leakage through the liner. However, when the pressure was released, as applies during internal inspections, the cracks became very tight and were not apparent by routine liquid penetrant examinations.

When a component is subject to localized overheating, either as a result of a fire or an upset operating condition, severe damage can result. This damage may result in physical distortion of the component in question. It may also involve metallurgical changes to the component which, in the worst case, will result in failure. Although fires or upset operating conditions can result in significant damage, the damage is generally less severe than appearances would indicate. Thus, it is essential to perform a comprehensive inspection of a component subject to such an incident to ensure that those repairs which are necessary are performed, but that unnecessary repairs are avoided.

On a general note, it should be realized that components such as pressure vessels, piping, heat exchangers, etc., may be heated to temperatures of 1600F to 2000F during original manufacture as part of normal shop operations including rolling, forming, extrusion, etc. Welding will heat the base metal in the localized heat-affected zone to temperatures of over 20000F. Some postweld heat treatments may involve normalizing at 16500F. Stress relief heat treatments, depending upon the composition of the steel, may be performed at temperatures of 1150F to 1350F. It is thus recognized that steel generally will not be damaged when heated to these elevated temperatures during manufacture, forming and shaping, fabrication and/or erection. On the other hand, it should be realized that when components are heated during normal shop operations, the heating is a closely controlled process. Fires or other upset conditions which result in localized overheating, by their very nature, are uncontrolled processes. Thus, extreme care must be taken to evaluate the effects of overheating which occurs as a result of such upset conditions.

Generally, the most effective methods of evaluating the effects of these conditions are in-situ metallurgical examination in conjunction with hardness testing.

As a case in point, a refinery in Pennsylvania had a fire in the Fluid Catalytic Cracking Unit (FCCU). The structural steel was severely damaged. In addition, the insulation had been burned off a number of the vessels. The integrity of the vessels in this unit was in serious doubt.

To confirm their integrity, they were subjected to replication and hardness testing. In the replication process, a small area on the surface of the vessel in question is ground successively smoother with a flexible shaft grinder in conjunction with a progression of abrasive floppy wheels ranging from 40 grit to 320 grit.

Subsequent to the grinding, the area is then polished with a series of felt wheels impregnated with diamond spray ranging in size from 9 microns to 3 microns. This is followed by final polishing with chromium oxide jeweller's rouge.

The surface area is then etched and covered with a piece of emulsified acetate tape. The tape, when pressed firmly into place, will harden to form an exact image of the microstructure.

Subsequent to hardening, the tape is removed and shadowed or sputtered with gold to provide sufficient contrast for viewing under a high power microscope. It is then examined comprehensively at magnifications ranging from 50 diameters (50X) to 1000 diameters (1000X) to detect any evidence of metallurgical deterioration.

The two replicas were removed from the stripper vessel. One of the replicas, which had been removed from an apparently unaffected area, exhibited a microstructure considered typical of carbon steel plate material. The microstructure consisted of equiaxed grains of ferrite and pearlite. The other replica, removed from an area that was in close proximity to the fire, exhibited changes associated with overheating. Despite the fact that the microstructure had changed, it was still ferritic/pearlitic and entirely suitable for the intended service conditions.

The microstructures observed can also be compared to reference standards to determine the temperature to which steel was exposed and the time it remained at that temperature.

Some conditions of overheating may result in severe conditions of embrittlement. In carbon steels and carbon 1/2 molybdenum steels this is associated with graphitization. In carbon steels, low alloy steels and even stainless steels, depending upon the temperatures and exposure times, conditions of hydrogen embrittlement and hydrogen damage can occur. If undetected catastrophic vessel, pipe and equipment failures have occurred. Replication techniques generally are effective in detecting the extent of these conditions. In some instances, and if detected early, the vessel, pipe or component can be restored to satisfactory levels of integrity by heat treatments.

To provide additional confirmation that the vessel or component in question is suitable for the intended service conditions, hardness testing should also be performed. The mechanical properties of the material in question can be extrapolated for the results of the hardness testing. Any deterioration of the mechanical properties arising from the high temperature excursion will be readily apparent.

In this manner, i.e., using replication and portable hardness testing, the integrity of the vessels subject to high temperature excursions can be determined with a much greater degree of certainty than previously possible. Moreover, these types of tests do not require any destructive testing.

Considerable equipment in chemical plants is subject to operation at temperatures of 800F and higher. Equipment that operates at these temperatures for prolonged periods of time is subject to creep. Creep represents the slow flow of metal subjected to high temperatures, or great pressures. Metallurgically, creep is defined as the time-dependent strain occurring under stress. The creep strain occurring at a diminishing rate is called primary creep; that occurring at a minimum and almost constant rate, secondary creep; that occurring at an accelerating rate, tertiary creep.

At some stage during secondary creep, voids will develop in the microstructure of the metal followed by void linkage and fissuring. If not detected, the fissures will grow to cracks and, in time, to rupture. Conditions of creep occur in many pressure vessels' furnaces, boilers, reformers and other high temperature equipment. Creep also occurs in turbines, compressors, and other equipment where the metal may reach temperatures above 800F.

Because of a number of catastrophic failures, detection of creep by nondestructive examination techniques has become increasingly important. In the past, such detection involved the removal of metal samples. In more recent years, creep has been measured by metallurgical techniques referenced as replication. These methods, generally, are costly in that they require removal of insulation and cooling of the equipment to room temperature at the location where creep is suspected. A more effective nondestructive examination technique that recently has been developed involves direct creep measurements with capacitive strain gages attached to the surface of the pressure vessel or pipe. A thermocouple is also attached adjacent to each capacitive strain gage.

These gages provide an accurate predictive method for estimating the residual life of high temperature pressure vessels, nozzles at high stress locations, piping, headers, and other components. Creep strain rates and temperatures were measured on a 2-1/4 Cr - 1 Mo alloy steel pipe operating at 950F. These creep measurements are particularly advantageous for components that have been subjected to 50,000 or more hours of operation, where creep damage involving void formation is slight. Even where initial void formation is discovered, the vessel or component, if properly monitored, may provide another 150,000 to 250,000 hours of safe and satisfactory service.

If the direct measurement shows an excessively high creep strain rate which may lead to cracking and premature failure of equipment, the creep rate can be reduced significantly by lowering the operating temperature at the location of creep. Similarly, the creep rate can be reduced by reinforcing the pipe, pressure vessel or nozzle section, by means of additional weld buildups or reinforcements.

Actual measurements, in conjunction with the evaluation techniques referenced, provide a significant improvement over previous methods for evaluating creep damage resulting in void formation, fissuring, and cracking, and for monitoring the remaining safe life of high temperature components.

There are occasions when defects can occur during manufacturing that are not detectable by routine nondestructive testing methods.

A reactor vessel of layered construction ruptured catastrophically. The rupture originated in a forged nozzle. The initiation of the cracking preceding the rupture occurred at a location of a weld overlay repair applied by the manufacturer of the forging. The forging initially had a rounded corner along the inside of the nozzle. Since a square cross section was desired, the manufacturer applied a weld overlay over the rounded surface and then ground the weld overlay to provide a square corner. As the forging was not preheated during the application of the weld overlay, small subsurface cracks developed. These subsurface cracks were not detectable by normal inspection techniques such as liquid penetrant or magnetic particle examination. They also would not have been readily detectable by ultrasonic examination. Because of the high residual stresses associated with the repair weld without preheat, after an operating period of several years, one of the cracks with the high residual stresses suddenly propagated extremely rapidly to result in the catastrophic vessel rupture.

Even with the nondestructive testing techniques currently employed, this type of high stress condition generally would even now not be detectable.

Quite obviously, in this type of short paper only a few examples can be presented to illustrate the types of conditions and problems that apply to nondestructive examinations of equipment in chemical plants, refineries and natural gas plants.

Even the nondestructive examination methods described are incomplete, as time did not permit detailed discussions of eddy current testing, acoustic emission testing, hydrostatic and pneumatic pressure testing, infrared testing, as well as other testing techniques.

The primary purpose here is to illustrate the importance of recognizing the limitations of repeatability that apply to each type of nondestructive testing equipment utilized and to the variations in techniques performed. It must also be recognized that defects indicated by ultrasonic examination may not be apparent when the same casting, weld, or component is inspected by radiographic examination. Thus, frequently a number of nondestructive examination techniques must be utilized in combination to minimize the possibility of erroneous conclusions involving either the acceptance of a vessel, component or equipment that may subsequently rupture, or the rejection of the same type of equipment because of an inherent defect that may never result in a failure.

Nondestructive testing, when performed in a meaningful and effective manner is essential to the safe and cost effective operation of chemical plants, refineries and gas processing plants.

Improper inspections, and incorrect or inadequate utilization of nondestructive testing equipment can prevent detection of conditions of deterioration that may reach levels where repairs no longer are feasible. Where deterioration then continues, and is not detected, serious and even catastrophic failures may result.

Equally as important in nondestructive examinations are the engineering analyses and interpretations of the indications, signals or conditions identified, denoted or suggested by the nondestructive examinations performed.

Improper interpretations may then lead to rejection or scrapping of a pressure vessel, weld, casting or component that is entirely suitable for service or, if minor defects or minor progressive conditions of corrosion or cracking are apparent, could be readily repaired during a regularly scheduled or budgeted outage period.

Many billions of dollars are wasted or lost by industry because of the lack of understanding and/or the lack of experience with the performance of engineering materials with respect to the existence or observation of defects discontinuities, and indications under various operating conditions. These may or may not involve corrosion, erosion, mechanical fatigue or shock, creep, embrittlement, design notches or weaknesses, dissimilar metal combinations, hard spots and other potential metallurgical or mechanical conditions that may or may not be detectable by any specific nondestructive testing technique and any one of a multitude of types of environment.