What is Random Vibration Testing?
Real world vibration is inherently random. A random vibration test, then, is a realistic method of bringing a product to failure. A random test excites all frequencies in a defined spectrum at any given time. By exciting all the resonances of the device under test, engineers can determine the interaction between multiple resonances.
Read the full paper here.
Over 9999 separate frequency/amplitude breakpoints
Up to 26,000 Lines of Control
VR controllers come standard with 50 to 26,000 lines of control to provide the frequency resolution required for your test. With the VR9203 High Frequency option, analyze up to 50,000Hz VR9203 High Frequency option, without sacrificing resolution.
Input Configuration
Use a single input channel for control or configure 2 to 256 input channels with multichannel averaging or multichannel notching. The standard frequency range is DC4,900Hz which can be extended up to 50,000Hz with the VR9203 HighFrequency option.
One to 4 control loops can be run simultaneously to independently control up to four shakers with four individually configurable and statistically independent waveforms.
Independent Channel Notching Profiles
Assign maximum limiting breakpoint profiles to individual channels. The drive output will be notched, if necessary, to keep the input spectrum for the channel below the defined profile. Up to 9,999 breakpoints can be entered for each channel, or the limit can be entered as a dB level relative to the demand profile.
In addition, minimum limits can be defined to boost the drive output if a channel is below a defined profile. The notching profiles may also be used as spectrum abort limits so the abort limit breakpoint profiles can be assigned to each individual channel. If the channel reaches the defined limit, the test will safely shut down.
Test and Level Scheduling
Tests can be scheduled to run a userdefined length of time, and the spectrum level can be scaled by a specified dBlevel, percentage, or for a specified RMS acceleration. Level schedules can be entered to run for various durations at different acceleration levels. Levels can be changed while the test is running.
Easy test entry
Enter frequency/amplitude breakpoints in an easytoread, tabular format. The operator can select to control constant or ramped acceleration, velocity, or displacement. Automatically calculate and enter the frequency of intersection between any combination of constant acceleration, velocity, or displacement lines. Over 9,999 separate frequency/amplitude breakpoints can be entered, allowing entry of virtually any test specification.
Should I use a sine or random vibration test?
A sine test outputs a single frequency sine tone at a defined amplitude and time. A random vibration test outputs random vibration that, throughout the test, include the entire frequency range of the test. Random testing is more realistic than sine testing because realworld vibration is inherently random. However, sine is a more suitable test method for:

 Identifying resonances in a test item.
 Identifying issues with the shaker system (cracked armature, loose bolts, etc.). Variations in the response, especially if there are resonances that shift significantly, indicate there is an issue in the shaker system.
 Validating Fixtures. If the fixture has resonances in the test frequency range, the results of the test will be questionable.
 Qualifying a product if product failure arises from the excitation of a known resonance.
Random Graph Display Options
Available graph display options include power/acceleration spectral density (PSD), output voltage spectral density, and channeltochannel transmissibility. Graphs can be easily auto scaled, and cursor display can be adjusted. Data and text annotations can be easily placed on the graphs, and data values update live with changes.
When Should I Select a Random Test?
Realworld vibration displays random characteristics and is not repetitive or predictable like a sine wave. A random vibration test can be generated to represent one or multiple environments where the device under test will be used. A random test can also be accelerated to represent a longer field life than can be run in a lab.
A random vibration test excites all the frequencies of the device under test and is a more realistic representation of the enduse environment than a sine test. Additionally, a device under test may have multiple resonances. A sine test only excites one resonance at a time. By exciting all potential resonances at the same time, a random vibration test reveals the interaction between multiple resonances. The random test often reflects a greater damage potential than a sine test is capable of.
Vibration test engineers should use a random vibration test to:

 Test a product against vibrations that will occur in the field environment
 Determine how a product will respond to the excitement of multiple resonances at the same time
 Test a product to failure
What is the Power Spectral Density (PSD)?
Random vibration is often analyzed with the power spectral density (PSD). The PSD represents the distribution of a signal over a defined frequency spectrum. It reveals resonances and harmonics that may not be visible in a timehistory graph.
Normally distributed timedomain data is transformed into frequencydomain data (i.e., a PSD) using the fast Fourier transform. An understanding of the PSD will help engineers to determine the correct parameters to use when generating a PSD from a waveform. To learn more about generating a PSD, please enroll in our Random Testing course on VRU.
What is the Probability Density Function (PDF)?
The probability density function (PDF) is another method of quantifying random vibration from a digitized signal. It is a histogram that is computed by counting the number of samples within a set of frequency bins, which cover the range of the signal’s amplitude.
The PDF has a Gaussian, or normal, distribution. A normal distribution has a skewness of zero and kurtosis of 3. The area under the PDF curve is equal to one.
What is Kurtosis?
Kurtosis is a dimensionless measurement pertaining to the average deviation of a signal from its mean value. Presentday random testing assumes Gaussian distribution of random data and, therefore, a kurtosis value of 3. However, there are data that do not fit Gaussian distribution. In such instances, a kurtosis value of 3 may omit higher peak values and result in undertesting.
Vibration Research recommends using kurtosis control in some testing situations to produce a more realistic random test. Learn more about our kurtosis control innovation Kurtosion.
Read also: Kurtosis – the Missing Dashboard Knob
Advanced Testing Options
Random Import (Optional) VR9204
Import analog or digital time data to automatically generate a random profile. Import PSD data to analyze statistics from a timehistory recording.
Fatigue Damage Spectrum (Optional) VR9209
Measure a product’s fatigue and calculate its lifespan. For years, engineers have used methods to calculate the lifespan of a product based on the material s/n curve. With FDS, rain flow analysis techniques are applied to measured vibration, equivalent fatigue is quantified, and time to failure is calculated. Commonly used life acceleration techniques are applied to shorten test time.
Dual Phased Output (Optional) VR92RDP
Perform dualaxis testing with phase control. The drive and the aux (COLA) output are used to drive different shakers with the same profile. Matched phase between the two outputs is auto controlled.
Random Fundamentals Webinar