Random Vibration Tests

What is Random Vibration

The first specifications for Random type vibrations were introduced following the analyzes carried out after missile launches carried out in the 1950s.

These analyzes revealed that the vibrations produced by the missiles being launched contained a large band of randomly varying frequencies.

Some examples of Random type vibration can be associated with the behavior of a vehicle during a journey on the road or with the vibrations generated on the structure of an aircraft during the flight phase.

A fairly explanatory example of the difference between a Random and a Sine type signal is the following:

Imagine exciting a structure composed of beams of different lengths with a sine-type signal by sweeping in frequency or with a broadband random-type signal.

In the first case the beams will vibrate clearly only when the frequency scan will be of a value equal to the resonant frequencies.

In the second case, by exciting the structure with a Random type signal, the beams will all vibrate simultaneously at their own resonance frequency.

The Power Spectral Density Function (PSD)

To run random tests, you need to define a random test spectrum. Real-time data acquisition uses spectrum averaging to produce a statistical approximation of the vibration spectrum. Typically, the random vibration spectrum profile is displayed as a power spectral density

This graph shows the mean squared acceleration per unit bandwidth (acceleration squared per Hz versus frequency Fig. 1). The shape of a PSD graph defines the average acceleration of the random signal at any frequency. The area under this curve is called the mean square of the signal (g²), and its root square is equal to the overall value of the root mean square (RMS).

Demand: 4.063 G RMS Fig.1 Acceleration Spectral Density

This type of graph allows you to view the distribution of vibration power, but does not allow you to highlight the instantaneous and maximum acceleration values, which can be calculated using probabilistic methods. Instead, you can calculate the total RMS value by calculating the square root of the area bounded by the PSD curve.

Consequently the RMS displacement will be calculated directly from the acceleration value.

While for the maximum peaks a conventional value of 3 times the RMS value will be used.

A signal of type Random is of a random type with a Gaussian distribution.

If a Sine signal is analyzed in the time domain, it is noted that the signal stays around zero for less time than the time it stays around the peak. (Fig. 2)

Fig. 2

In a Random type signal analyzed in the time domain, most of the time is spent at the amplitude value around zero. (Fig.3)

The fact that the Random vibration spends a lot of time at low amplitude values leads to the manifestation of little damage to the test piece compared to a SINE type profile.

Therefore, in cases where the excitation signal of the component is not of the periodic type, the Random type test allows for a reproduction of the stresses more responsive to reality.

For more information visit our site on laboratory tests where you will find the vibration tests you are looking for!

A special thank you goes to Noris Vivarelli, a great expert in the vibration and fluid-dynamic sector, as well as Plastlab Laboratory Manager, who raises our awareness on this issue so as to provide us with new content for these #laboratory pills!

You can also find us on social media with our #labpills – Facebook, Instagram and LinkedIn