Building Research Advisory Board and the subcommittees formed within it met with members of the staff of the National Bureau of Standards at selected intervals to review the progress of the program and to advise on its broad phases. By mutual agreement reached during the progress of the work, the actual testing was limited to flat specimens and bathtubs. 1.3. Authorization and Funding The investigation undertaken by the National Bureau of Standards was authorized by the ad hoc committee in June 1964, in accordance with a proposal prepared by the Bureau. The work was financed jointly by multiple sponsors under the auspices of the Building Research Advisory Board and by the National Bureau of Standards. 1.4. Work to be Performed By mutual, written agreement, the work to be performed by the National Bureau of Standards was to include the following: 1. Inspect manufacturing facilities, testing laboratories, and typical installations of fixtures to familiarize project personnel with current operations, use conditions, and test procedures. 2. Review existing test proceduress of ASTM, ASA, and other organizations, as well as commercial standards requirements, Federal specification requirements, and selected regulatory requirements for suitability, reproducibility of test results, and applicability to a wide range of construction materials.2 3. Identify, based on this review, existing test procedures that should be tried out on various materials; and develop new test procedures where needed. 4. Carry out trials of existing and new test procedures on flat specimens and bathtubs. 5. Prepare a final report to the ad hoc committee and include in this report (a) a description of the laboratory work that was performed on each property of interest, (b) a description of each recommended test procedure and apparatus, (c) the rationale for the recommended test procedure, and (d) a discussion of performance requirements that might be pertinent to sanitary fixtures. Each of the properties of interest is discussed in the succeeding sections, based on the above-mentioned report. However, because of both time and budget limitations, it was not possible to recommend test procedures for all 28 properties that were judged important by the ad hoc committee. Those tests requiring additional work are discussed in the appropriate sections. It should be pointed out in this connection that the original 1 This language was intended to indicate plumbing code provisions. Several codes were reviewed. 2 This refers to the various materials that might be used in the manufacture of plumbing fixtures. Actual test development was carried out only on bathtubs and flat specimens. project agreement did not require that test procedures be recommended for all of the 28 properties. 1.5. Test Specimens Two types of test specimens were utilized: flat squares, either 4-in × 4-in or 414-in × 44-in in size, and whole bathtubs. The specimens were furnished by various manufacturers in accordance with arrangements made by the Building Research Advisory Board and the ad hoc committee. The plan adopted for specimen procurement called for 100 squares from each of three or more manufacturers of vitreous-china, enameled-steel, enameled cast-iron, fiberglass-reinforced polyester (FRPE), and stainless-steel fixtures, respectively. Four whole bathtubs were to be furnished by each of three or more different manufacturers of enameled-steel and enameled cast-iron fixtures, respectively, and four whole bathtubs from each of four or more manufacturers of fiberglass-reinforced polyester fixtures. This plan for specimen procurement was largely adhered to, except that (1) no stainless steel specimens were obtained, (2) three brands of polyester (FRPE) bathtubs rather than four were obtained, (3) only two manufacturers provided flat FRPE specimens, and (4) two rather than three makes of enameled steel flat specimens were provided. It is not believed that these changes in the original plan for specimen procurement had any substantial effect on the outcome of the investigation. All specimens were furnished with a white finish except for one FRPE bathtub, which was green, and one lot of flat cast-iron specimens which was furnished in five different colors. However, because these cast-iron specimens were received late in the investigation, it was not possible to include them in the testing work. 1.6. Specimen Identification Each lot of bathtubs and each lot of flat specimens were assigned identification numbers on arrival in accordance with table 1.6-1. The reason for including this table is to provide a more complete identification of specimens referred to in the tables and figures in later sections of the report. 1.7. Field Surveys Appendices A and B summarize the results of two limited surveys of bathtubs that had been in service for varying periods of time. Appendix A relates to fiberglass-reinforced polyester bathtubs and Appendix B relates to porcelain-enameledsteel bathtubs. These surveys were made to acquaint the project staff with typical manufacturing, installation, and use conditions, and to provide some guidance in arriving at realistic test conditions. Reference to Appendices A and B at this point will be help ful in understanding various sections of this report. Originally it had been expected that an extensive field survey would be made to obtain statistically significant results relating to various use factors, physical properties, length of service, etc. However, this survey was not undertaken, principally because competent statisticians advised that a meaningful survey would be prohibitive in cost. 1.8. Discussion of Performance Tests and Performance Requirements For a piece of equipment as intimately involved in family living as a bathtub, the desirable functional characteristics include not only requirements such as water-tightness, drainage characteristics, strength, and safety, but also the more subjective concepts such as ease of cleaning, appearance, durability, resistance to staining, burn resistance, and resistance to chemical attack. These latter subjective requirements are likely to be expressed in qualitative terms involving such phrases as "acceptable appearance," "normal service life," and "typical use conditions," that have no precise definition in physical terms. Developing test methods that are meaningful requires conception of a test procedure that simulates to an acceptable degree the kinds of physical, chemical, and mechanical exposure received by the fixtures in actual use. While simulative service testing is not new, simulation of the interaction. between human beings and household fixtures in a way that will provide a basis for fair competition among materials of unlimited variety presents additional complexities. It involves careful analysis of the important processes of use, wear, and deterioration that are brought to bear on the fixtures in service, and translation of these processes into a piece of laboratory equipment that can be described and reproduced and that can measure the effects of these processes in quantitative terms. Once an acceptable test procedure and a test apparatus have been developed, the performance of typical fixtures can then be measured under the selected test conditions. Although fixtures or specimens of materials prepared especially for test purposes can sometimes be obtained on which to collect quantitative data, more frequently only commercial items are available for testing. In this latter case, test results on the commercial items must be used for comparison with user requirements. The selection of acceptable performance levels involves evaluation of human aesthetic reaction as well as the frequency and duration of exposure to a variety of human activities. In the case of properties that may be expected to deteriorate gradually with use over a period of years, valid performance levels cannot be set without extensive statistical data on use conditions, and on users' concepts of long-term acceptability. Such statistical data are usually not available because of the prohibitive cost of obtaining them. Where the measured values of a given property vary over a wide range for various materials using a selected test method, it may not always be logical to set a single performance level for all materials. The factors of use conditions and tolerance level of users are intimately related to standard of living, psychology, and the effect of long experience with conventional materials, Thus, a given performance level might be acceptable in the view of some users, but entirely unacceptable to others. Similarly, after a period of experience with a new material, users might come to accept or require a different level of performance than formerly. For example, it is not difficult to imagine the attitude of users toward the introduction of glass for windows if the material previously used had been a transparent resilient material such as plexiglas. It might take many years to obtain consumer acceptance of the low impact resistance of glass under these conditions. In the absence of adequate statistical data on use conditions and tolerance of users, one practical approach to performance levels is to select levels that either upgrade, downgrade, or maintain the existing quality of a class of products in current use, based on the measured performance of a suitable comprehensive sampling of contemporary products. Decisions with respect to upgrading, downgrading, or maintaining present quality require the studied judgment of experienced persons acting together. Performance levels arrived at in this way would, of course, be subject to later adjustment as more extensive service data or user reaction became available. Recommended performances levels are not included herein. However, comments on performance requirements have been added at the end of each section in the hope that these comments may be helpful to code and specifications groups who may wish to set performance levels that are based on the recommended test procedures. 1.9. Use of Trade Names for Identifying Instruments, Materials, and Equipment Certain commercial instruments, materials, and equipment are indentified in order to specify the experimental procedure adequately. In no case does such identification imply recommendation or endorsement by the National Bureau of Standards, nor does it imply that the instruments, materials, or equipment identified are necessarily the best available for the purpose, nor that items not identified but having the necessary characteristics cannot be used. 1.10. Test Titles The test procedures to be discussed in what follows are titled in a fashion consistent with the identification system used by the ad hoc committee. The word "bathtub" in parentheses indicates that the test is limited to bathtubs in its present state of development. If the word "bathtub" does not appear in parentheses in the test title, this signifies the test is applicable to fixtures in addition to bathtubs. The number in parentheses signifies the test classification adopted by the ad hoc committee. For example, S101 means test No. 101 in the Structural Series, M203 means test No. 203 in the Mechanical Series, T303 means test No. 303 in the Thermal Series, etc. 1.11. Applicability of Findings to Various Types of Fixtures Since no test development was conducted in this investigation on actual fixtures other than bathtubs, some of the recommended tests must be limited to bathtubs. In these instances, further work will be necessary to adapt the tests to other fixtures. However, a number of the tests are applicable to more than one fixture type, because certain per formance characteristics are primarily related to the material and are affected little or none by the type of fixture the material is used in. 1.12. Statistical Limitations of Findings Many of the test procedures discussed herein should be looked upon as in need of further refinement. This is particularly pertinent from statistical considerations. For example, some tests are obviously affected by several sources of variation, but limitations on the amount of work that could be performed in the present investigation precluded the collection of adequate statistical data for the computation of the variances corresponding to these various sources. Among the general problems that need further attention in applications of the work reported are (1) required number of test specimens and replicate tests, (2) sampling procedures, and (3) effects of variation among observers, equipment, and laboratories. In summary, the problem is: How much testing is required as a minimum to assure a just acceptance or rejection of a single fixture or a lot of fixtures? This is particularly important where the measured performance by one or a few measurements is close to specified minimum performance levels. The selection of fixtures and flat specimens for use in test development was handled by the manufacturers through the auspices of the Building Research Advisory Board. The National Bureau of Standards made no study regarding the extent to which these fixtures and specimens represented current production, and therefore makes no claim in this respect. Thus, in any application of the findings of the present investigation, it is important to recognize the statistical limitations described above, and to recognize that a very considerable amount of additional work will be required to obtain an altogether correct solution to the statistical problems. 1.13. Units of Measurement Results of this investigation are reported in terms of conventional units, for the most part. This was done for two reasons: First, conventional units are ordinarily used by those groups most concerned with the results. Second, most of the existing standards and test methods extensively referenced and evaluated herein utilize conventional units. However, because of the increasing importance of international standards in foreign commerce, it is strongly recommended that groups which utilize the results of this investigation in the promulgation of standards and specifications assume the responsibility for appropriate conversion to International Standard (SI) units. For example, 0.500 in equals 1.27 cm, 2.1 lb equals 0.95 kg, 120 °F equals 48.9 °C, etc. 2. Test Procedures 2.1. Uniform Loading (Bathtub) (S101) No test method exists for the uniform loading of bathtubs and no test method is recommended. This type of loading would be the result of the weight of the bath water. The amount of water necessary to fill each of eight different bathtubs to the overflow outlet hole was measured. The average weight of water was found to be 339 lbs. A 300-lb concentrated load at the center of the sump 3 would usually be more severe than a uniform load of 339 lbs. Therefore, if the recommended concentrated static-load test (Sec. 2.2) is used a uniform-load test would not be needed. Although a uniform-load test was not considered essential, some data were obtained on sump deflections from a type of load similar to that imposed by a person sitting or standing in a bathtub partially filled with water. Load-deflection tests were made on three FRPE tubs using 150 lbs of water and three 50-lb cast-iron weights centrally placed in the sump. Deflections with time were measured by using the procedure recommended for the concentrated static-load test (sec. 2.2). These data, for hot and cold water, were obtained for general information only. Figure 2.1-1 presents the results of two tests on one of the FRPE tubs. The three rectangular weights covered a combined area of approximately 132 in' in both tests, but were placed slightly differently in the two tests. In Test 1, the weights were placed end-to-end along the longitudinal centerline of the tub, but were placed side-by-side with their lengths transverse to the longitudinal centerline of the tub in Test 2. Room temperature (72 °F) water was added for Test 1 and hot (104 °F) water for Test 2. For Test 1 the three 50-lb weights produced a center deflection of 0.039 in and the addition of the 150 lb of cool water produced an immediate additional deflection of 0.007 in. For Test 2 the three 50-lb weights produced a center deflection of 0.032 in, but the addition of the 150 lb of hot water caused an immediate upward movement of 0.026 in at the center of the sump (indicated by a downward movement on graph). It should be noted that the immediate upward movement of the sump from the hot-water load was about four times the downward movement from the cool water in Test 1. However, with time the center deflections for both tests approached the same value. Immediately upon removal of the load in Test 2 the expected upward movement occurred. However, as the tub cooled the thermal gradients caused a downward movement. It is unfortunate that measurements were not continued for a longer period, but from other test results there is not much doubt that the sump would gradually return to a position close to the original. 2.2. Concentrated Load, Static (Bathtub) (S102) a. Purpose and Scope 2. A loading device for applying a 300-lb test load without shock to the center of the tub bottom. NOTE: Either calibrated weights or a suitable mechanical or hydraulic load applicator may be used. 3. A distribution pad to distribute the load over a 5- × 10-in area. The pad shall consist of a sheet of 34-in-thick sponge rubber (Shore Durometer Hardness of 8 to 14) topped with a plate of plywood or stiffer material having a minimum thickness of 3/4 in. This plate shall be of sufficient thickness so that there is no more than 0.01 in deflection of the ends when the 300-lb load is applied during the test. 4 The "center deflection" is defined, as the deflection at the mid-point of the longitudinal centerline of the sump. 5 (2) Support of Specimen During Test The bathtub shall be mounted for test in a wood frame simulating normal installation. Design of the supporting frame shall be such as to allow for clearance under the tub for deflection-measuring dial gages. The manufacturer's installation instructions when obtainable shall be used in spacing the 2-X4-in stud of the frame and in fastening the tub to the frame. In lieu of explicit manufacturer's instructions the tub shall be fastened along the water bead to each stud of the test frame by 12-in long No. 6 steel wood screws using 3%-in steel washers. The spacing of the studs in the supporting frame shall be no greater than 16 in on centers. The top of the supporting frame shall be at least 12 in higher than the top of the installed tub. If legs or other component parts of the tub are within in of the floor line, after installation, rigid vertical support shall be provided for these components. The front apron shall be supported in a continuous bed of plaster of paris. For tubs without integral wall surrounds the back ledge of the tub shall be supported by a continuous horizontal 2-X4-in wood framing member that is fastened securely to the studs. (3) Test Procedure Maintain the temperature of the testing laboratory and bathtub at 75 ± 5 °F. Before starting the test, inspect the finished surface of the bathtub in accordance with section 2.8 of this report. Note all defects detected. Center the distribution pad over the horizontal center lines of the sump with the 10-in dimension along the length of the sump. Determine the sump center lines by using the average length and width of the sump. Place the three micrometer dial gages on a rigid base beneath the longitudinal center line of the sump, with one directly beneath the center of the distribution pad and the other two within 2 in of each end of the bottom of the sump. NOTE: The areas where the tips of the dial gage bear on the tub must be flat and smooth so that a small lateral movement of the tub will not change the gage reading by more than 0.001 in. These bearing areas can be prepared either by grinding the under surface or by rigidly fastening a small flat, level plate to the tub. In case the drain hole of the bathtub should interfere with the placement of a gage, a standard drain-fitting spud can be inserted in the drain-outlet hole and a flat area on the spud used for the dial-gage bearing. Preload the specimen by applying the test load of 300 lb at the center of the distribution pad and leaving the load in place for 5 min. Then remove the load and make the initial dial-gage readings. Reload the tub and make the gage readings immediately following load application and also 5 min later. Finally, remove the load and make the dial readings immediately after removal and again 10 min later. Determine the average settlement of the tub and supports by averaging the deflections measured by the two end gages. Determine the center deflection of the sump by subtracting this average settlement from the deflection measured by the center gage. After load removal, inspect the surfaces of the tub for cracks and spalls in accordance with section 2.8. Note any other damage that develops from the testing. (4) Information To Be Reported Include the following in the test report: 1. Specimen identification; 2. Description of specimen including overall height, width, and length; 3. Description of test frame; 4. Method of suporting and fastening bathtub in test frame; 5. Defects, if any, prior to test; 7. Deflection at the center of sump; a. Immedicately after application of load; c. Test Results and Discussion (1) Discussion of Existing Methods A test method for evaluating the structural integrity of FRPE bathtubs is described in the August 1964 proposed revision of Commercial Standard CS221-59, hereafter referred to as the "proposed revision of CS 221-59" [1]. This test method requires that a full-size unit be installed according to manufacturer's directions and an applied load of 300 lb be distributed over a 3-indiam area near the center of the bottom of the bathtub. The applied load is allowed to remain on the tub for not less than one nor more than two minutes. The maximum allowable deflection under the load point is 0.150 in. The residual deflection which is determined 10 min after removal of the applied load is not to exceed 0.008 in. (2) Test Data The relationship between the applied loads and defiections are shown in figure 2.2-1. Each pair of curves represents the maximum and minimum deflections observed during tests of bathtubs manufactured from the materials designated. The observed deflections for individual units tested appear in table 2.2-1. Table 2.2-1 also includes the results of tests performed on units using a 3-indiam disk as the applied load area. As can be expected, the measured deflections are larger for these tests. Table 2.2-2 is a summary of static load tests to destruction. The table includes results of tests on Figures in brackets indicate the literature references at the end of this publication. |