If you would like to watch other IEST presentations by Wayne via Webex, please click on the links below:

October 15, 2002 - What is Resonance all about?
December 3, 2002 - Measurement and Analysis
January 21, 2003 - Vibration Aspects of Reliability Enhancement via HALT, ESS and HASS
May 6, 2003 - Vibration Testing and Screening of PCBs
December 9, 2003 - Resonance can damage your hardware

Fixtures for Vibration and Shock Testing
by Wayne Tustin

(Nov 18, 2003)

Many people involved in environmental test belong to IEST – the Institute of Environmental Sciences and Technology. Members and non-members are invited to participate in writing RPs or Recommended Practices such as this one dealing with vibration and shock test fixtures. This RP is available from IEST headquarters in Rolling Meadows, IL.

What is a fixture? It’s an intermediate structure, bolted to and driven by a shaker or shock test machine and some device under test (D.U.T.).

Let’s examine the moving part of Figure 1, the shaker armature, the fixture and the D.U.T. (which here we are pretending is just one mass). At higher test frequencies the shaker armature behaves like two masses joined by a spring. An important question: where, on the shaker table at A, on the fixture at B, or on the DUT at perhaps C or D, shall we locate our control accelerometer?

That’s an important decision. See Figure 3. The control accelerometer informs your computer that controls your test what’s happening out at your shaker. Where shall we attach the sensor (control accelerometer)? We must decide at what “input” location(s) we want to control our test. Another necessary decision: at what location(s) on the D.U.T. do we want to record responses? If you are responsible for testing, be sure that these locations are specified.

At left, Figure 4, we have a family of transmissibility (mechanical amplification) curves for an idealized, very simple, spring-mass system. On the magnified graph at right, a 3000 Hz resonance is shown at frequency ratio 1. Perhaps we are testing up to 2,000 Hz, which falls at frequency ratio 0.67. Notice that our fixture has boosted shaker motion by half.

With our control accelerometer at “B” (back to Figure 2), that resonant buildup will have prompted our computer to reduce electrical drive to the shaker to about 0.67. That’s one reason “B” is so popular. With control at “B”, our computer electrically corrects for our fixture’s mechanical resonance.

Now let’s talk about our fixture, Figure 5. Probably the most popular fixture shape is the cube. DUTs can be attached to five sides and variously oriented with respect to shaker motion.

Another popular shape is the “ell” or “bookend” fixture. The fixture of Figure 6 suggests fixturing a car radio. “Input” is one of Figure 2’s many “B” locations. “Response” is one of Figure 2’s many “C” or “D” locations. Test results can change dramatically – depending upon which “input” and which “response” location you choose. These (as well as intensity and frequency) should be specified.

Do you attempt to hold your DUT in the same manner it will be held in service? The fixture in Figure 7 supports a torpedo in somewhat the same way an aircraft would support it. Figures 8 and 9 suggest alternatives.

How shall we fabricate our fixtures? Beginners try to bolt plates together. This might work below 50 Hz, but not to 500 or 2,000 Hz, where vibratory displacements get down into the microinches and parts can move about, can rub and rattle. Casting is best, because then we can use highly damped K1A alloy, the same alloy as used in many shaker armatures. Casting requires time for design, for building a model, for casting in sand and for machining. Welding is the most-used compromise.

If we are welding, need we chamfer plates, in order to fully penetrate with our weld material? Generally no. A fixture is not a pressure vessel. Have your weld shop clamp the parts. Tack weld, then lay down a heavy bead. Width “A” should be comparable to plate thickness. When the bead cools, it will strongly and rigidly secure the fixture parts.

Place inserts such as shown in Figure 12 into drilled holes in your aluminum or magnesium fixtures (soft material) for strongly bolting your DUTs.

How might a test change if you relocate your control accelerometer? This is a fairly lengthy story, and this slide lacks sufficient space.

When an adapter plate or a fixture has served its purpose and is full of holes, recycle it. Don’t compromise your testing by continuing to use junk.

Do you need to shake a number of circuit cards that are destined to serve in a “card cage”? You might want to strengthen an actual “card cage”, as in Figure 15.

Another way to hold circuit cards so that all receive the same vibration is shown in Figure 17.

I’d like to make sure that everyone out there knows that I teach about vibration and shock testing, as well as measurement, analysis and calibration. Here is a listing of “open” courses presently scheduled. Just one more this year, at Chatsworth, near Los Angeles, commencing tomorrow morning. Where else would you like me to teach?

Some of you know that for the last couple of years I’ve been writing a new book, shown at the right. A few copies of my 1984 book are offered at Amazon.com. Chapter 26 deals with fixtures.

Thanks again to B&K and to IEST!