Types of Vibration Testing

On a fundamental level, vibration testing can be categorized into three broad categories, sine vibration, random vibration, and shock. There are more complex forms of vibration testing, however, that would be a topic for another day.

Sinusoidal Vibration Testing
Sinusoidal vibration is typically encountered when dealing with rotating machinery. This could include an engine, electric motor, lathe, or anything that has a repetitive waveform.

The testing might be at a single frequency, or maybe over a range of frequencies called a bandwidth (frequencies between which the testing is to be performed.) i.e 10Hz (cycles per second) to 1000 Hz.
If it is a single frequency, the testing lab will ask for the displacement. Displacement is normally stated as peak‐to‐peak displacement or peak displacement and can be given in inches or in mm. Displacement is needed because some older vibration tables are restricted to 1‐inch peak‐to‐peak displacement. Modern ones are at 2‐inch peak-to-peak, and the recent builds are at 3‐inch peak‐to‐peak. If they have a 1‐inch peak‐to‐peak machine, they might not be able to run a task that requires 2 inches peak‐to‐peak. 

If the testing is performed over a range of frequencies, the testing lab will ask for a sweep rate. The sweep rate refers to how quick the frequencies are ramped from lowest to highest. The sweep rate may be linear or logarithmic. If it is linear, it will likely be stated as an increase in Hz per minute. i.e. 10Hz to 55Hz to 10Hz in 2 minutes. If it is logarithmic, it is going to be given as octaves per minute. As in music, an octave is a doubling of frequency. Therefore, a logarithmic sweep of 1 octave per minute over the frequency range 5-1000Hz could take 5‐10 (1 minute) 10‐20 (1 minute) 20‐40 (1 minute) 40‐80 (1 minute) ... (640‐1280) last minute. It would take almost 8 minutes to move from 5 Hz to 1000 Hz at one octave per minute.

Some tests may have one or multiple breaks at some point of the test where displacement is increased or reduced. Consequently, the specification could read 5Hz to 1000Hz, a logarithmic sweep of 1 octave per minute, displacement 0.5mm from 5Hz to 500 Hz and .05mm from 500 Hz to 1000 Hz. In summary, the testing lab will specifically request for the following information for a sine vibration test:

• Range over which testing will occur (bandwidth) in Hz. (5‐500 in the example above)
• Sweep type octaves per minute (logarithmic) or time per sweep cycle (linear)
• Displacement (peak to peak) and /or acceleration in g's. this could be split over the bandwidth, i.e., acceleration for half and displacement for half.
• Specific displacement in bandwidth and acceleration in two.
• Number of axes to test.
• Number of hours to test in each axis.

Random Vibration Testing
Random vibration is the type of vibration that happens typically in the real world. An example of random vibration is an automobile driven down a washboard gravel road. It is just what it sounds like, there is no rhyme or reason to the resultant wave shape. It is random in nature. One way to generate a random vibration profile on a shaker is to position accelerometers on the testing object, place it in the testing environment, and then play back the recorded waveform. That is called field‐data replication. In this case, the x‐axis is time, and the y‐axis is the acceleration amplitude in g's. This technique is not the most practical, nor the most portable. Therefore, engineers have invented a power spectral density (PSD) curve. It is considered that the waveform from the field data has frequency content and that it may be transformed using mathematical techniques into the PSD.
The PSD is a graph that allows the vibration table controller to reproduce a simulation of the field data which was initially collected, although it is not precisely that data.
It has a good probability of developing the same frequencies and amplitudes as the original data, but the sequence may not necessarily be identical. Compared to the sine vibration where only a single frequency is evident at any point in time during random vibration all frequencies in the test range are exceeded. The PSD describes the amplitude of energy which is to be controlled at each frequency.

Shock Testing
shock testing is less complicated to specify when compared with both sine or random vibration testing. A shock test is closely similar to hitting a deep pothole with your car or slamming a door shut. For shock testing, the following will be required by the testing facility:

• The pulse shape

• The pulse width in milliseconds

• The amplitude in g's (number of times the force of gravity).

• How many shock pulses are required and in what direction? This is commonly 3 in a forward and aft direction for a total of 6 pulses in each axis but may in some tests extend to thousands of pulses.

• How many axes are to be tested? generally, this is the three orthogonal axes and usually in both the positive and negative directions.

In need of dynamic testing services, vibration testing services or design & fabrication of vibration testing fixtures?
Contact Paragon Systems Testing.