The United States Federal Railroad Administration (FRA) allows the use of six (6) NDT methods (AE, MT, PT, RT, UT, or VT/RVT) for the inspection of in-service railroad tank cars. New regulations require contractors to use one or any combination of these methods, which replaces hydrostatic testing for checking the integrity of tank cars. The inspections help evaluate high stress areas including welds, on the bottom of the tank. The results are used to qualify tank cars for a ten-year inspection cycle. This paper will address the advantages and limitations of each method as it relates to railroad tank car inspection and the problems created for each individual method by the inspection surface and environmental conditions of in-service tank cars.
The structural integrity test for in-service tank cars had been a hydrostatic test every ten (10) years. This involved filling the tank up water and pressurizing to a predetermined pressure (e.g. 690 x 10³ Pa (100 lb./in.²) for general service cars) and performing a leak test. Recent structural failures in tank cars have caused regulators to consider alternatives to hydrostatic testing. In one recent accident, in Dragon Mississippi, a tank fractured at the weld seam between two tank shells causing a release of commodity. An accident investigation revealed the crack slowly propagated until catastrophic failure. This particular tank had just received a hydrostatic test, which did not identify the crack. Federal regulators quickly saw that more stringent inspections were needed to evaluate the integrity of in-service tank cars.
The FRA implemented, through the Code of Federal Regulations 49CFR 179.7 and 180.501-517, an entirely new and major change in qualifying inspection of in-service tank cars. The use of hydrostatic testing is being eliminated and new inspection methods were mandated. These new criteria are summarized below.
- Any tank car facility doing repairs on tank cars has to have a quality assurance program (QAP). Depending on their QA certification, shops are either annually audited for QA compliance or audit for just cause.
- All nondestructive testing personnel have to be qualified and certified in accordance with the Association of American Railroads (AAR) Manual of Standards and Recommended Practices (MSRP) M-1002 Appendix 'T'. This standard establishes personnel training hours, NDT procedure requirements and inspection reporting requirements. Companies are required to have a written practice of qualifying and certifying their NDT personnel and has to be in accordance with Appendix 'T'. The written practice must be modeled on the American Society for Nondestructive Testing (ASNT) SNT-TC-1A.
- The implementation of new inspection areas with designated NDT methods for qualifying tank cars as follows:
- A visual inspection of the interior and exterior of the tank. These inspections include inspection of tank welds, shells and heads for weld defects, corrosion, distortions or any other condition that would make the tank unsafe for transportation.
Figure 1 shows typical areas to inspect on a tank car using a combination of NDT methods.
- A structural integrity inspection using one or a combination of six (6) methods Magnetic Particle (MT) Liquid Penetrant (PT), Radiographic Testing (RT), Ultrasonic Testing (UT), Visual/Remote Visual Testing (VT/RVT) and Acoustic Emission (AE) by exemption. The inspection involves a structural integrity inspection of all transverse fillet welds greater than 0.64 cm (0.25 inch) and termination of longitudinal fillet welds within 121.92 cm (4.0 feet) of the bottom longitudinal centerline of the tank car. The structural integrity inspection also includes an inspection of butt welds within 60.96 cm (2 feet) of the bottom centerline. A typical structural integrity inspection is shown in Figure 2.
- A thickness check of the tank shell, heads and nozzles. The minimum allowable thickness is mandated and will vary with car design. If the tank has a lining or coating the thickness check is not due till the lining or coating is removed.
- A safety system inspection that assures all safety equipment (e.g. thermal protection, safety relief devices and nozzle protection systems) are in place and functional.
- A lining and coating inspection. A general inspection of interior coatings and linings, if applicable, are in proper conditions.
- A leak test of service equipment such as manway covers, valves, pipes and fittings is performed to ensure they are leak free.
Fig 1: Typical areas to inspect on tank car (side view)
Fig 2:Typical structural integrity inspection (bottom view)
Tank cars vary by design. Two major classifications of exterior designs are jacketed and non-jacketed tank cars. Jacketed cars receive an application of insulation or thermal protection over the tank exterior followed by an outer metal jacket. The non-jacketed cars receive no additional application of material and the tank is exposed to the elements. One difficulty with a qualification inspection of non-jacketed tank cars is the exterior coating that is applied to the tank car for corrosion protection. A complete inspection of the tank exterior has to be through the coating. The jacketed tank cars proposed an additional problem for the structural integrity inspection. Metal jacket, insulation and in some case primer, rust, and exterior corrosion create a very poor inspection surface. Jacket and insulation have to be completely removed from the inspection areas or inspection ports have to be cut in the jacket and remote visual methods have to be used.
Previously in the tank car industry NDT was used to evaluate original welds or new repair welds. The new requirement creates a new problem. Car owners are now required to inspect in-service welds (some 30 or more years old) to some criteria. The acceptance criteria stated in CFR180.509 is any condition that would make the tank unsafe for transportation until the next inspection. Acceptance criteria for new welds on tanks and tank attachments is located in the AAR MSRP M-1003 Appendix W and the American Welding Society (AWS) D15.1-93. Tank owners have to establish in their inspection programs what acceptance criteria is proper for their cars designs.
Probability of detection (POD) studies have been funded by the FRA and conducted by the Association of American Railroads (AAR) at the Transportation Technology Center, Inc. (TTCI) using the six (6) methods allowed. Preliminary results have showed a variety of minimum flaw sizes found (see Table1). Visual was the least sensitive showing a 50% probability with 90% confidence level of showing a 7.62 cm (3 inch) crack. Although the results are preliminary, the POD for visual compared to the other methods was expected.
|Crack length cent.(inches)
|Table 1: POD percentages for four NDT methods. (preliminary data)|
The new code allows equal inspection intervals regardless of the NDT method use to qualify the tank. A low sensitivity method such as visual is weighted the same as higher sensitivity methods such as ultrasonics and radiography. Because the methods are equally weighted inspection cost becomes a factor. Remote visual inspection is usually the method of choice of car owners because with the use of inspection ports removal of jackets is not required. Inspection ports are cut into the jacket eliminating costly jacket removal. More sensitive but more costly NDT methods will not be used. POD's and minimal flaw size should dictate inspection cycles. At this point the industry has not established critical flaw size in critical areas of inspection.
Regulators have put responsibility on the car owner to set inspection cycles. The industry mandated inspection cycles of ten (10) years is the maximum interval set by the FRA. Car owners have to demonstrate the effectiveness of their method(s) and accompanying inspection procedures. The inspection intervals may change as critical flaw size and POD studies are still being conducted by the AAR. The final results of the POD studies may require regulators to vary the inspection interval allowed depending on the sensitivity of the NDT method used.
The industry is now at a crossroad as to where the structural integrity is headed. When considering factors such as cost, sensitivity, surface conditions and environmental issues, the six (6) NDT methods display a wide range of success when inspecting railroad tank cars. The industry may have to revisit the regulations that allow equal inspection cycles when qualifying tank cars with NDT methods of unequal sensitivity. Car owners original equipment manufactures (OEM's) shippers and regulators will have to work together to establish critical flaw size and effective maintenance programs.
- The Association of American Railroads, Manual of Standards and Recommended Practices Section C, Part III Specification for Tank Cars M-1002
- Code of Federal Regulations, 49 CFR 179.7 and 180.501-517.