By Randall Cobb, P.E. Inside Sales Manager, NTS Detroit.
Vibration testing is a necessary part of many programs, and there are a wide variety of vibration types, but the one thing every test has in common is that the energy must be transmitted into the test sample with some kind of fixturing. Fixturing can be as simple as mounting bolts into a flat plate, or it can be an elaborate weldment with responses tuned to match the final assembly. The best fixturing is rigid, lightweight, and simple. The following is a list of general guidelines to aid in choosing the best fixturing for your test sample.
Rigidity: A fixture should be very stiff in order to transmit the vibration without adding extra noise. A flexible fixture can resonate at test frequencies, affecting the amount of energy transmitted to the test sample, and thus test quality. A very flexible fixture may be difficult to control and will require a more powerful vibration system, affecting both quality and price. At 2000 Hz, a 1 g sinusoidal input only requires 0.000005” of shaker displacement. It doesn’t take much fixture bending to affect that!
Weight: A fixture with extra mass will require more force to vibrate at a given amplitude (g-level). Low mass fixtures will allow multiple samples to be tested simultaneously, improving costs per part and test throughput. Fixtures should be low profile and keep the sample as close to the system input as possible. Eliminate unnecessary features, but don’t skimp on material if it will affect stiffness.
Material Selection: Unfortunately, vibration fixtures are not always made of the most inexpensive materials available. Steel and Aluminum have roughly the same strength to weight properties, so it might make sense to pick steel as the less expensive option. However, aluminum has a density much lower than that of steel, and features can be made much larger and stiffer with no weight penalty. This makes it the superior material for high-frequency vibration.
For very high-performance fixtures, magnesium will sometimes be used as its performance is even better than aluminum. The tradeoff is difficulty in machining and much higher fixture costs.
Complexity: The best vibration fixtures are simple, with the minimum number of features. Vibration testing is not only stressful to the sample, but also to the fixturing. Extra features and thin sections can add more potential for unwanted resonances, and more components that can fatigue and break. Simple fixtures are usually less costly up front, and cost less over the long run.
It is advisable to use through-bolts for sample mounting instead of tapped holes whenever possible. Threads fatigue after heavy use and a nut and bolt can be cut away and replaced at the test site, but extracting a damaged threaded insert may require a trip to the machinist’s shop. If you must have a tapped hole, use a steel threaded insert (E-Z Lok or Keensert type), rather than a helical insert (Heli-Coil or STI type), and never thread a fastener directly into aluminum!
Other Considerations: Most vibration testing is performed in a single axis, and repeated in each orthogonal axis (X, Y, & Z). Vibration systems can either be vertical or horizontal, and fixtures are often rotated 90° to perform two of the three axes on a horizontal surface. If you only have one vibration system available, it may make sense to design a fixture with mounting holes on more than one surface so it can be rotated for each axis of testing.
Since few test parts are so simple they can mount directly to the shaker, most fixturing is a compromise of the qualities above. Samples may need to be mounted in a specific orientation, or have features that must be supported above the vibration table. Welded fixtures are often necessary since it is not always practical to carve a fixture out of a single block of aluminum. Be sure weld joints are continuous and smooth, with two or even three passes if possible.
Validating a Fixture: When a fixture is used for the first time, it is good practice to perform a resonance scan in order to check for any unwanted responses. This is done by instrumenting the fixture with multiple accelerometers, and sending a low-level random signal that covers the entire frequency range intended for the test. If you imagine a tuning fork that rings at a certain frequency (pitch), and that frequency was in the range of the test, then when the test hits that frequency the fixture will start ringing and add more energy into the sample than it should receive. A resonance scan will identify any problem frequencies, and can be performed with or without the test sample in order to observe any effects.
Repeatability/Liability/Fixture Ownership: The standard offering for many test labs is to build a custom fixture for your program, and they will then maintain responsibility for fixture responses. You are paying for the fixture design and build, as well as a guarantee from the test lab this fixture will perform reliably from one test to the next. The major caveat to this service is that the test lab retains ownership of the fixture, because they cannot be responsible for its use and measurements made at other facilities. If you choose to make your own fixture, be sure to review it with the lab during the design phase in order to confirm it will match the test system.
For more information about vibration testing, visit our webpage. You can also contact the author of this article at our Detroit location: 313.835.0044. For more details including specifications on NTS Detroit vibration testing, see our brochure.