This section provides an overview of the role of specifications as a means of communication between all parties concerned in the purchase of steel castings.
One needs only a superficial knowledge of a few of the existing steel casting specifications, and of metallurgy in general, to understand that stating what one needs is not a simple matter. All requirements must be clearly and accurately stated with nothing taken for granted. This is best accomplished by the use of standards and specifications.
As you read this section, you will notice that the use of nationally recognized standards and specifications is recommended, while the use of proprietary specifications is strongly discouraged.
In the specification process for steel castings, there are three key words which should be understood. These are specifications, standards, and codes.
A specification is a form of standard, which precisely states a set of requirements to be satisfied by the casting. Some of these requirements might be chemical composition, mechanical properties, repair procedures, or any other requirement that is necessary to develop the quality of the casting needed for its end use. Specifications for steel castings are sometimes expanded or limited by standards and codes.
A standard can be defined as a specification, test method, definition, or recommended practice that has been approved by a nationally recognized specification-writing body such as ASTM (American Society for Testing and Materials), ISO (international Organization of Standards), or SAE (Society of Automotive Engineers). A standard can be further be defined as a document which details properties, processes, dimensions, materials composition, relationships, or concepts. This connotation follows Webster's definition of "something set up and established by authority as a rule for the measure of quantity, weight, extent, value, or quality." It can be seen then that there is some overlapping between specifications and standards, and for that reason the two terms are often used interchangeably where steel castings are concerned.
The word " code " is a term of much broader meaning than either specification or standard and can best be described as a set of rules established by a recognized authority such as the American Society of Mechanical Engineers' (ASME), Boiler and Pressure Vessel Code or the United States of America Standards Institute's (USASI) code for pressure Piping. In adopting the rules that make up the various codes, consideration is generally based on health, safety, and environmental protection. The code-formulating bodies, in addition to writing their rules, usually adopt ASTM material specifications either in whole or in part to be come a part of the code.
A specification, as previously defined, is a precise statement of requirements. Therefore, any requirement can be specified or incorporated into the specification. The most common are listed and discussed in the paragraphs that follow.
It should be emphasized that ASTM specifications take into consideration all of these requirements and more, so when using an ASTM specifications, there seldom is a problem with omissions.
Most steel casting specifications take into consideration the chemical analysis of the casting either directly by specifying the analysis, or indirectly by specifying properties that are related to the analysis, such as hardness or tensile strength, and leave the choice of composition to the foundry. But no matter how considered, the composition is important. For example, the chromium in the various stainless grades of ASTM A351 must be within prescribed limits for predictable corrosion resistance. Other elements in those same grades such as carbon, nickel, and molybdenum, must also be within limits in order to maintain a balanced microstructure necessary for the mechanical strength of the alloy and to insure proper corrosion resistance and performance in different environments.
There are also carbon and low alloy grades with specified chemical ranges found in some ASTM specifications such as A216, A217, A487, and several others. Most of the grades found in those specifications have been evaluated as to weldability and mechanical properties at various elevated temperatures, and approved for ASME code use by the ASME Boiler and Pressure Vessel Committee. Since any change in composition may have some effect on weldability, and on the high temperature performance of the casting, other grades whose chemistry may deviate only slightly from the approved grades are not acceptable for ASME code use.
Structural and engineering grades of high strength cast steel are covered by A148 (High Strength Steel Castings). The only chemical requirement in that specification is for sulfur and phosphorus. Other chemical requirements have been avoided because no foundry can cast all steels in the many modifications. The strength levels in A148 run from 80 ksi to 175 ksi (552-1207 MPa) tensile strength. The chemistry chosen for each grade should take into account the strength level, section sizes to be cast, the complexity of design, heat treating methods, and the end use of the casting.
Tolerances for chemical analysis are relatively new to steel casting specifications, although they have been in use for other steel products for some time. The product analysis tolerance, if one is given, merely specifies the amount by which an analysis of a sample, taken from a casting, may deviate from the specified composition range.
Variations occur in dimensions and weights of parts made by any metal-shaping process. Tolerances are the expression of the expected or acceptable variation. Dimensional tolerances should be included on any casting drawing. Quotations and acknowledgments from the foundry will often refer to a variation in weight or weight tolerance.
Steel castings, whenever possible, should be purchased to property requirements rather than to chemical analysis specifications. most of the national specification are written in terms of tensile properties and in some cases hardness values, impact values, and hardenability ranges. This permits the foundry engineer to select the alloy compositions which best satisfy mechanical property selection.
Mechanical properties of steel castings can be categorized as follows:
For many years, most casting specifications, including those issued by ASTM, contained very ambiguous wording in regard to surface inspection and integrity. For instance, castings were to be clean and "free from injurious defects." There was no definition for defect: and no basis for judgment as to what was "injurious." If the purchaser's inspector said a discontinuity was injurious, no matter how small, it had to be removed and repaired. Furthermore, the "injurious defect" had to be "completely removed to sound metal." Again, there was no definition for "sound metal" and no basis for judgment for "completely." Requirements of this type can easily be misunderstood, and misapplied; they can cause no end of grief for both the foundry and the purchaser. This problem has been rectified in the ASTM specifications.
Unfortunately, many other specifications today still contain the same or similar ambiguous wording. Whenever requirements such as these are discovered, every attempt should be made to rewrite them in a manner similar to those found in the latest ASTM specifications. The writers of the ASTM specifications, which include both producers and users of steel castings, have replaced such ambiguous wording with requirement such as " The surface of the casting shall be examined visually and shall be free of adhering sand, scale, cracks, and hot tears. " This preferred wording goes on to say that " unacceptable visual surface discontinuities shall be removed and their removal verified by visual examination of the resultant removal verified by visual examination of the resultant cavities. " If more stringent examination of the cavity is required, the ASTM specifications allow for that option, and the purchaser may so specify in his purchase order, and then both parties will have a clear understanding of what is expected.
To help define "unacceptable visual surface discontinuities," the ASTM specifications state that "Visual Methods MSS-SP-55", which is issued by the Manufacturers Standards Society of the Valve and Fittings Industry, may be used. This standard contains photographs of various casting surfaces and defines them as acceptable or unacceptable.
ASTM A802 has a 31 piece set of surface comparators which has many more categories of surface finish from which the purchaser will specify the level of acceptance he requires.
When higher levels of surface integrity are needed, other inspection techniques such as magnetic particle or dye penetrant examination may be specified. With these methods, cracks can be revealed that would go undetected with the unaided eye.
The soundness of a casting is most often determined by radiographic methods, although ultrasonic inspection is also used, especially on heavy sections. Pilot castings of small size or those preceding large production runs, in addition to being radiographed, are often destructively examined by sawing into slices and examining the pieces. Once the casting procedure and foundry technique have been established, the internal integrity will remain relatively unchanged.
However, when a specific internal quality level is required, it should be stated in the order. Even though a pilot casting might meet all radiographic requirements, there is no guarantee that all others will meet the same level. Some of the controlling factors, no matter how closely monitored, may change even slightly and affect the end result.
There are three principal means of inspection in common use to detect internal and surface discontinuities in steel castings. These are:
Standard methods or recommended techniques for carrying out the inspection have been developed and are published in ASTM documents listed in the Testing and Inspection Specification Table. In all cases, the pertinent ASTM document should prevail over any individual company specification unless it is proven to be inadequate for the specific application.
Test results, frequency of testing, and the sampling procedures for obtaining specimens to be tested are usually specified in material specifications. The actual methods and procedure for performing the tests, such as tension, bend, hardness, and impact, are generally specified in a testing specification such as ASTM A370.
Complying with the requirements of a testing specification assures that all testing is conducted in a standard and reproducible manner.
The use of " how to do it " or process specifications in the manufacturing of steel castings is to be discouraged for several reasons. There are wide variations between methods used by various foundries, yet each is capable of achieving the same end results. For a customer to decide which method should be used would seriously hamper the development of new manufacturing techniques. Such specifications might be justified in certain well-established areas, that by their very nature are not subject to further development or change, but in the field of foundry science new developments are constantly being made. As a result, there are very few outside the field who are qualified to write such a document, even if it were desirable to do so. One inherent problem with process specifications is that they often contain requirements which cannot be checked by the buyer. This weakens any specification.
Welding methods, likewise, should be left to the foundry and not dictated by a process specification. Upgrade welding is just as much an operation in the manufacture of steel castings as is the molding operation or any other operation involved in casting manufacturing, and all freedom possible should be granted the foundry. However, since the weld will become a part of the casting and go into service with the casting, it is perfectly justifiable, for metallurgical reasons, to specify that the welding procedure be qualified. In fact, almost all of the ASTM casting specifications require that procedures and welders be qualified in accordance with the recommended practice described in ASTM A488. Procedures and welders qualified to ASME, Section IX, are automatically qualified to A488.
All orders should reference a nationally recognized specification, preferably an ASTM specification. The requirements of ASTM cover many widely diversified applications of steel castings, including carbon, low alloy and high alloy (corrosion- and heat-resistant steels. They are prepared by knowledgeable representatives of both purchasers and producers working together to develop specifications of proven usefulness and compatibility.
These specifications are subject to review and discussion twice a year by the subcommittee on steel castings. Anyone, whether an ASTM member or not, who has a problem with an ASTM specification can write to ASTM and describe the difficulty, or meet with the subcommittee. All problems and solutions are openly discussed. Overlapping, redundant, and contradictory requirements are eliminated whenever they are discovered.
Many non-standard specifications, such as those prepared by individual company organizations, may have conflicting requirements. A common example is a Brinell hardness requirement which is not always compatible with the specified tensile strength. Since there is no absolute conversion from hardness to tensile strength and a maximum hardness. Even then, care should be taken to be sure there is a workable range between the two. Another example is specification of chemical analysis when mechanical properties might be the only requirement really needed.
Unfortunately in some cases, changes are made to existing ASTM requirements, and then incorporated into customer specifications even though they may not apply to the customer's casting needs. These changes frequently contain provisions that have been previously rejected by ASTM as impractical or unnecessary. The net result is partial duplication of specifications and some unnecessary restrictions. Thus, castings for similar end use may have requirements for two or three different quality levels.
Such multiplicity of specifications results in confusion and misunderstanding, and unnecessarily increases the cost of a casting without affecting its serviceability.
There are numerous organizations, public and private, that promulgate specifications and hence have full jurisdiction over them. By far the largest specification writing body in the United States is the American Society for Testing and Materials (ASTM) whose standards are used worldwide. Other specifications in common use for special products and under the jurisdiction of their respective trade associations are those issued by the Association of American Railroads (AAR) and the Society of Automotive Engineers (SAE). In addition, there are a number of military and other agency specifications. However, for complete details of any of the specifications, it will be necessary to refer to the complete and latest document.
One other major specification writing body whose specifications are beginning to be used in the US is the International Organization for Standards (ISO). Because of the worldwide use of the ASTM and ISO specifications and standards, an understanding of each is helpful.
Specifications and standards predominately used by the steel casting industry are those issued by ASTM and can be grouped into three general groups:
To obtain complete details, the original specification should be consulted. The additional number following the ASTM designation, e.g., A216-80, indicates the year of adoption or latest version. The most recent revision of any specification should always be used.
Material specifications for structural and engineering grades of steel castings are covered by A27 and A148. Carbon and low alloy steel valves and fittings for elevated temperature service are covered by A216 and A217, while castings for low temperature service are covered by A352. Other material specifications cover castings for specific applications, such as steam turbine castings (A356) and bridge castings (A486).
Each specification has been written for a particular type of service or environment which should be described in the title and in the scope of the specification. In all ASTM specifications, the scope is stated in the first paragraph. The requirements of any specification should be compatible with the intended use of the casting. Castings should not be produced under a particular specification if the intended use of the casting is outside the scope of that specification. For instance, carbon steel valve castings intended for high temperature service should be ordered to A216 and not A27. Also, high strength structural castings should be ordered to A148 and not to A87. Ordering castings for use outside the scope of the specification may result in additional and unnecessary requirements or in the omission of requirements that are necessary for the particular application.
Welding specifications are listed in the Testing and Inspection Specification Table.
These two specifications, ASTM A488 and Section IX of the ASME Boiler and Pressure Vessel Code, are the documents most often used for the qualification of procedures and welders.
Not all ASTM grades of steel have been adopted for pressure service under the ASME Boiler Code and referenced in Section IX. To try to qualify those grades under Section IX may result in confusion and possibly misunderstanding with the customer as to interpretation of the qualification rules. For that reason it is best to qualify to Section IX only those grades of steel actually approved for Code use and referenced in Section IX. All other grades should be qualified under the rules of ASTM A488.
There need be no duplication in qualification because A488 states that welders and procedures qualified to Section IX are automatically qualified to A488.
Testing specifications for steel castings are included in the Testing and Inspection Specification Table.
Standard test methods for the purpose of obtaining mechanical property data, such as tension, bend, hardness, and impact, of a cast steel are specified in ASTM A370. Test coupons, specimen dimensions, and exact testing procedures are detailed for each type of mechanical test. Whenever mechanical testing is required and is not covered by the material specification, the exact type of cast coupon and type of specimen should be spelled out in the purchase order or contract to be in accordance with the provisions of ASTM A370.
If the determination of the nil-ductility transition (NDT) temperature is required, the drop-weight test method described in ASTM E208 should be used.
The steel casting industry has numerous specifications and standards that are concerned with nondestructive testing of its products. A list is included in the Testing and Inspection Specification Table.
The chief criteria of casting quality are surface appearance and integrity, soundness of sections, and accuracy of dimensions. For this reason, most inspection standards are concerned with these properties.
Often, the designer, if he is unfamiliar with the foundry process, may specify a quality level higher than the design really requires, which serves no purpose except to increase the cost. A more favorable price and delivery can be obtained by first selecting the material specification (preferably ASTM) which meets the mechanical test requirement, and whose scope encompasses the service for which the part is intended. The necessary quality level can then be established by specifying special inspection procedures such as visual, magnetic particle, liquid penetrant, radiography, ultrasonic, and dimensional tolerance.
Surface discontinuities are the irregularities, imperfections, or cracks that are found on the surface of the casting. Although some are of such size that they can be seen visually, others are either not visible or go unnoticed without special inspection methods such as magnetic particle or liquid penetrant examination. For the examinations to be meaningful as a basis for purchase, all parties concerned must use inspection methods that are standard.
Most ASTM specifications contain a requirement stating that the surface of the casting will be examined visually and free of adhering sand, scale, cracks, and hot tears. Visual Method MSS-SP-55, available from the Manufacturers Standardization Society of the Valve and Fittings Industry may be used to define acceptable surface discontinuities. This standard consists of a series of photographs which are defined as acceptable and unacceptable. Any other visual standard may also be used as long as both parties agree to it.
ASTM A802 is a 31 plate set of comparators depicting various degrees of surface discontinuities in several categories such as wrinkles, porosity, veining, etc. Under this new standard the purchaser can specify surface requirements by quoting category numbers and levels of appearance. The standard specifies nothing as being acceptable or unacceptable. The comparators are merely points of reference used in communicating a requirement.
Magnetic particle inspection is used to detect surface discontinuities. Under ideal conditions certain discontinuities lying just below the surface can also be detected. However, this is primarily a surface inspection method and caution should prevail in attempting to ascribe other capabilities to it. Also, any conclusion with regard to depth or extent of the interior nature of the discontinuity must be based on exploration by other test methods. Magnetic particle techniques methods for dry powder and wet inspection are set forth in ASTM E109 and E138, respectively.
A set of reference photographs has been assembled by ASTM as document E125 depicting the appearance of different types of casting surface discontinuities as revealed by the dry power magnetic particle technique. Each type of discontinuity is classified in five degrees of severity, except porosity, where two examples are shown. To avoid any misunderstanding, it should be pointed out there is no correlation between degrees of the various type of discontinuities. For instance, degree 3 of type I is not equivalent to degree 3 of type II.
By prior agreement between the purchaser and the producer, these photographs may be used as standards to accept or reject castings. The acceptable degree of severity for each type of discontinuity must be spelled out in the purchase order or contract. Different types of discontinuities do not have equal effects on the serviceability of the casting and an effort should be made to assign realistic acceptance levels to each area of the casting, based upon the type and magnitude of stresses to which each area is subjected in service.
Admittedly, it is difficult to rigidly interpret magnetic particle indications on castings against a set of photographic references; consequently there is a need for close cooperation between the manufacturer and the purchaser.
One example is in the linear discontinuity of degrees 1 and 2. The degree 1 indications are approximately 1/2 in. (13mm) long, while those of degree 2 are approximately 5/8 in. (16 mm) in length. Also, the degree 2 causes the powder to cling in a wider pattern. In the interpretation of indications, however, their width is seldom considered; only the length of the indication is compared to the photographic references.
Although the separation between degree 1 and degree 2 is completely arbitrary and in no way related to service performance, there is often great concern as to whether the indication is greater than 1/2 in. (13 mm) or less than 5/8 in. (16 mm).
To overcome some of the interpretation problems, some purchasers specify acceptance standards to various degrees in E125 and then add that cracks and not tears shall not exceed 1/2, 1/8, or even 1/16 in. (13, 3, or 1.5 mm) in length. Although these dimensions are somewhat arbitrary, their being specified does eliminate much misunderstanding.
Misunderstanding can be minimized if the inspectors for both parties are least ASNT (American Society for Nondestructive Testing) Level 1 inspectors. This is a rating of the level of competence to which the individual is certified by training and completion of a prescribed number of classroom hours in inspection techniques and interpretation. The ASME Boiler and Pressure Vessel Code requires that inspectors making interpretations be certified to Level 2, which is a higher qualification than Level 1.
Additionally, orders should not state both visual and magnetic particle standards because of the possibility of overlapping or contradictory requirements. If magnetic particle examination is needed, then visual methods of inspection need not and should not be specified.
Magnetic particle inspection has probably led to more misunderstanding than any other inspection tool. It has made possible the selection of castings for critical applications by greatly assisting the upgrading effort with its outstanding ability to detect surface discontinuities. It can also, when improperly applied, increase the cost of a casting without improving its performance. Therefore, it is essential that standards of acceptance be applied with discretion.
Liquid penetrant inspection is another surface discontinuity detection method. It is not generally used on the " as cast " or shot blasted surfaces because of the likelihood of obtaining false indications. The penetrant method is best suited for use on machined, ground, or very smooth " as cast " surfaces. Liquid penetrant inspection is of particular importance for austenitic alloys because they are non-magnetic and therefore their surfaces cannot be examined by magnetic particle inspection. ASTM E 165 describes the standard method for conducting this test.
A set of reference photographs for acceptance or rejection is contained in ASTM E433. It should be pointed out that there are no degrees of severity, as in E125 for the dry powder magnetic particle technique. Each of the documents must specify actual dimensions including maximum length of indications and number of indications per unit area. Also, no attempt has been made to establish the metallurgical cause of the discontinuity.
When E433 is specified, there should be a prior agreement of interpretation and acceptance to prevent subsequent misunderstanding.
Dimensional tolerances are permissible deviations from the nominal dimension. Deviations from the nominal, or aimed for dimension, may occur for several reasons. The primary source is the contraction of the liquid metal as it solidifies and cools in the mold.
An experienced foundryman can estimate the metal contraction that will occur on any dimension, but only trial by actual production will show precisely how the metal will behave. Tolerances for the production of a single casting, therefore, tend to be liberal. On the other hand, with castings produced in large numbers, the opportunity exists to make changes in pattern equipment and manufacturing processes to compensate for abnormal casting contraction behavior. In this situation variation will be minimized but a slight variation will still exist.
Upgrading, by grinding and gagging, coining, straightening, and other measures are available to achieve any desired tolerance level that cannot be achieved by the casing process alone. Upgrading of this type adds to the price and should be specified only where required to minimize the cost of the component.
A series of tolerance classes is a practical means of communicating the needed tolerance to the foundry and of explaining to the purchaser the ability of the foundries' process, or processes, without subsequent dimensional upgrading. The new system of tolerance classes is therefore desirable, and five tolerance classes have been suggested.
The tightest grade, or class, represents the best tolerance encountered in a recent SFSA study. It is named T3 to provide room for additional tolerance classes in the future, and for specialized processes such as investment and ceramic molding techniques. Tolerance grades T5 and T7 represent the average, and the widest tolerance encountered, respectively. The intermediate classes T4 and T6 were selected to provide for a geometric progression from T3 to T7.
There are three basic groups of reference radiographs issued by ASTM for evaluation of steel castings as seen in the Testing and Inspection Specification Table.
E446 applies to castings up to 2 in. in thickness (51 mm), E186 to 2 to 4-1/2 in. (51-114 mm) thick sections, and E280 to wall thickness of 4-1/2 in. to 12 in. (114-305 mm). Each group is available in a choice of sets based upon the source of radiation employed, such as low-voltage X-rays, iridium -192, cobalt -60, 1 to 2 MeV X-rays or 10 to 24 MeV X-rays.
A special set of reference radiographs for investment castings is available as ASTM E192.
Reference radiographs of discontinuities common to steel welding are categorized in ASTM E390. Repair welds should be inspected to the same standards employed for the original casting, i.e., E446, E186, or E280. E390 is applicable to inspection of welds used for cast-weld inspection.
Reference radiographs become standards for acceptance and rejection only after the purchaser and the producer have agreed, in the purchase order or contract, to the acceptable severity level for each individual type of discontinuity. The choice of discontinuity severity level should ideally be based upon realistic evaluation of design and stress analysis criteria under anticipated service conditions. Generally, low severity levels are specified for pressure-containing castings with high pressure rating and wall sections of 1 in. (25 mm) or less. Likewise, low severity levels are specified for machinery or dynamically loaded casting subject to high fatigue and impact stresses, and with wall sections of less than 1/2 in. (13 mm). As wall sections increase and as the fatigue and impact stresses are reduced, severity levels become somewhat relaxed. For structural castings which are not dynamically loaded, moderate severity levels are usually specified, and again, for heavier sections about 3 in. (76 mm) higher severity levels are usually called for.
To require quality levels in excess of those justified by actual service conditions adds needlessly to the cost of the casting. Also, requiring a single across-the-board severity level for all types of discontinuities should be avoided. Some types are more detrimental than others, depending upon he nature of the stresses to which the casting is subjected in service. For instance, severity level 2 might be specified for shrinkage, and severity level 3 for gas porosity, since the latter is generally much less deleterious to tensile properties. It should also be kept in mind that the entire casting need not necessarily require radiographic inspection and that the same severity levels need not apply to all areas of the casting. This again is governed by the type of stress and the stress levels in the given casting section. Careful analysis or, at least, good judgment can affect sizable cost savings. In any case, the areas to be radiographed with the required severity level should be indicated on the casting drawing.
It should be borne in mind at all times that the severity rating is strictly arbitrary and based on little more than opinion. None of the reference radiographs are based on any kind of test data, and the severity levels are not graded to any basis of acceptability as to service performance. They only serve as a reference point in communicating the purchasers' requirements.
Consistent quality of the radiograph itself can be readily achieved if the recommendations and methods outlined in these two ASTM documents are followed: ASTM E94 is a guide for radiographic testing and E142 is a guide for controlling the reliability or quality of the radiographic images. Both are completely adequate in that internal discontinuities of any significance can thereby be detected. Except for a very few isolated cases, no deviation need be made. Reference radiographs in E242 show how such factors as radiation energy, specimen thickness, and film properties affect the radiographic images.
There is a tendency on the part of some individual company standards to specify films, unsharpness ratios, densities, and other details aimed at producing perfect films with cost being no object. It should not be forgotten that the radiographic film is a means to an end and not the end in itself. It is simply not logical to specify a technique capable of sensitivity which will show discontinuities smaller than the minimum size for rejection.
Although the ultrasonic method of inspection has not been in common use for as long as radiographic methods, it nevertheless is a valuable tool for examining heavy wall castings for internal discontinuities. The first ASTM specification for ultrasonic inspection of steel castings was issued in 1970 and is for longitudinal-beam ultrasonic inspection of heat treated carbon and low alloy steel castings. This inspection method is in general not useful for austenitic steel castings due to large grain size of these castings.
It is well recognized that ultrasonic inspection and radiography are not directly comparable. However, the technique is invaluable in detecting discontinuities in heavy sections, where radiographic methods would be considerably slower. Since no picture, in the usual sense, of the discontinuity is obtained, considerable judgment must be exercised in interpretation of results.
One approach in the examination of large heavy wall castings when ultrasonic evaluation may not be acceptable to the purchaser is to first inspect by ultrasonic, then to radiograph only those areas where a suspicious ultrasonic indication is found. Another possibility, since radiography does not reveal the depth of a discontinuity, is to follow radiography with ultrasonic in order to determine and evaluate the depth of the discontinuity.
The ASTM steel casting specifications contain the requirement that the surface of the casting shall be free of visual cracks and hot tears. The meaning of this statement is quite clear and there is seldom any disagreement concerning this requirement when cracks and hot tears can be seen visually. However, when they are detected by other methods, the severity and relevance become a matter of judgment. In fact, even defining acceptance levels was once something of a problem.
To help overcome this deficiency, ASTM issued specification E125, which has become the standard for surface quality for the industry. The document consists of 37 reference photographs of surface discontinuities divided into five classes of graded severity. However, neither castings nor sections of castings were tested to determine the relationship of various degrees of discontinuity observed to the service requirements of the casting. The different degrees of severity are based on nothing more than opinion. Although the photographs show magnetic indications on steel castings to various levels of severity, the castings were never available for study.
Steel casting buyers often specify a severity level across the board and in many cases, severity level 1 is arbitrarily selected. Some buyers request wet magnetic particle inspection rather than dry powder inspection, and others specify liquid penetrant inspection. The dry power reference photographs of E125 are often employed for all three types of inspection because ASTM has not supplied reference photographs for wet magnetic particle; and the reference photographs in E433 for liquid penetrant inspection are presented in a manner different from and less accepted than those in E125. In fact, E433 show only examples of discontinuities and makes no attempt to classify them as to severity.
For a clear understanding of current radiographic standards, it is necessary to go back to the beginning of such standards. The first, " Gamma Ray Radiographic Standards for Steam Pressure Service, " was issued by the Navy's Bureau of Engineering in 1938. Personal opinion was the criterion for determining that a certain radiographic quality level was considered acceptable and another rejected. Primarily, government personnel decided, from viewing the radiographs, whether they were acceptable or rejected on the basis of the way the defect appeared to them. Some experience was available regarding valve leakage related to varying degrees of shrinkage in steam pressure service casting.
The present reference radiographs are available for the casting buyer to set his own severity levels. In other words, a buyer could select severity level 3 for gas porosity, severity level 4 for sand inclusions, severity level 2 for shrinkage, severity level 1 for linear discontinuities, and so. However, this has not been employed by most casting buyers. Most have been specifying a single severity level across the board. There is no basis provided by ASTM for concluding that severity level 2 for porosity and severity level 2 for shrinkage are related in any manner as to the ability or inability of the casting to perform the service for which it has been designed. In fact, information made available by SFSA would indicate that severity level 2 shrinkage and severity level 5 porosity is a much more comparable relationship.
To add to the confusion, it is pointed out, for example, that severity level 2 shrinkage in E446, E186 and E280 are not the same severity. This seems to cause some concern even though the committee that prepared the standards was of the opinion that severity levels could be relaxed somewhat as the section size increased.
A summation of the past and present on reference radiographs would indicate that their use has grown extensively. However, they are not based on factual test data, and the severity levels are not graded to any basis of acceptability of the casting as to its load carrying ability. They are based on the opinion that anything that is less than perfect is questionable.
Steel casting are specially designed and manufactured parts, and therefore, the cost of castings will depend upon the complexity of the design of the part and upon the purchasers' requirements. The cost of one casting cannot necessarily be compared to the cost of another casting similar in weight, shape, and design, because differences in quality requirements may exist. Two castings which may look alike may have different costs because the service requirements of the two are entirely different; dictating that the quality and tolerance requirements of one be of a different order than those of the other.
Steel casting costs reflect variations in material specification, tolerance limits, inspection requirements, acceptance standards, affidavits, and certification requirements. The purchasers should always rely on value analysis in the specification and buying of steel castings.
A wide range in estimated casting costs from several foundry bidders often reflects that the purchaser was not specific as to the properties and requirements desired. Specifying minimum quality requirements is necessary if castings of minimum cost are desired.
A cost area often overlooked is one of maintaining the most current editions of specification and reference standards. To produce valves, fittings, or other pressure castings, a foundry would have to have as a bare minimum, ASTM Reference Radiographs E186, E280, E446, and E99, in addition to the ASME Boiler and Pressure Vessel Code and the USASI B-31 Standards. The foundry would also have to have documents such as the ASTM standards, the American Petroleum Institute Standards, the American Welding Society Standards and the Standards published by the Manufacturers Standardization Society for the Valve and Fitting Industry.
All of these documents, costing several hundred dollars each, are considered necessary to properly process an order for parts for the construction of steam power plants, refineries, and chemical plants. There is also the necessity of maintaining files of specifications for the military and for countless other customers, some having 30 to 40 separate specifications. Some power plant contractors have specifications several hundred pages long. All of these documents must be kept up to date, since revisions are constantly being made. This is no small task and even a specification specialist cannot remember the details of every document with all of the variations in formats and requirements. As a result, a foundry doing extensive work of this nature must have a large quality control department.
The principal costs in this area are incurred when the castings are processed through the shop. Testing and inspection to ensure exceptional quality levels requires the careful efforts of a relatively large quality control staff along with reams of paper work. A separate book of procedures must be followed for each specification in addition to numerous special handing procedures to cover various customers' requirements.
Coincident with any extra testing and inspection is the cost of upgrading by skilled workmen, followed by further inspection and additional production time. Narrow ranges of acceptability are usually congruent with high quality levels, and a higher percentage of rejected and reworked casting is probable. Naturally, these costs must be reflected in the price of the casting.
Levels of quality which are higher than demanded by the end use are excessively costly, and add nothing to the serviceability of the casting. If the requirements of the casting are overstated, the cost of the casting will be higher than it should be. Necessary quality requirements should not be compromised in order to obtain a lower price, but it must recognized that the more requirements specified to attain higher levels of quality, the more costly the product will be.
To produce castings to rigid specifications requires skilled and qualified personnel working in an adequate facility. The qualification of both facility and personnel is nothing more than being assured that the producer has the capability of supplying castings to the specified requirements.
Inspection personnel are often required to be certified as a Level 1, 2, or 3 inspector in accordance with the American Society for Nondestructive Testing Recommended Practice No. SNT-TC-1A. Radiographic facilities must be certified and licensed by city and state agencies. Welders and the procedures they use must be qualified to ASTM 488 or to Section IX of the ASME Boiler code. Although castings for ASME Boiler and Pressure Vessel Code use and castings for nuclear use may be produced without the foundry having to obtain the ASME " U " stamp or " N " stamp, the surveillance costs are extremely high when those stamps are required.
Additional costs are incurred for the approval and certification of equipment such as tensile testing machines, impact testing machines, magnetic particle inspection equipment, heat treating furnaces and temperature controllers, calibration standards and numerous other items that are used to prove conformance to the specifications.
The determination of an economical specification range, whether it be for chemical analysis, mechanical properties, hardness, dimensional tolerances, or any other range, requires careful study, much statistical information, and common sense. The preparation of a specification is an exacting undertaking in which buyers and producers should collaborate. Specification control can be obtained only when the normal expected value and the standard deviation are known.
The proper creation of a specification is much more time consuming than is often supposed. Averaging the results of a few tensile tests, or thumbing through the pages of handbooks and selecting average values and adopting them as specification limits is never satisfactory. When the distribution curve is normal, half the results are higher than the average and half the results are lower. If the average value is taken as the specification limit, half the results will be immediately rejected. Specification limits are never based on averages.
A specification range should be as narrow as necessary and practicable. However, if it is too small, rejections become excessive. Therefore, a balance must be maintained between the value of establishing a narrow specification range and the cost increase resulting from the more exacting quality control required in holding the process to narrow limits. On the other hand, if the range is too wide, additional processing costs in other areas may be incurred. For instance, if the chemical ranges for a low alloy heat treatable steel are too wide, the hardenability of the castings from different heats of that grade of steel might have a wide variation which will result in excessive heat treating costs when heat treating to a narrow range.
It is probably much more economical and advantageous in the long run for purchasers, producers, and engineering groups to discard private specifications and replace them with specifications created by nationally known specification writing bodies.
In the ASTM specifications, there will probably be found a closer balance between process capabilities and purchasing requirements than in any other group of specifications.