Shaker Selection: What You Need to Know

Selecting a shaker can be both challenging and perplexing: challenging because it is neither a straightforward nor mundane task, and perplexing because of the nagging thought “What if I make the wrong selection?”

Some say a shaker cannot be too powerful. True, as long as you have an inexhaustible supply of money. It is not uncommon, however, for a company to farm out test work because the technical people involved in defining requirements either over-specified a system (exceeded the available budget and couldn’t buy anything) or bought an undersized one.

Good decisions begin with the basics. Three types of shaker systems—electrodynamic (E-D), servo-hydraulic (hydraulic) and mechanical—are typically used for random vibration, sine vibration and many types of shock tests. The major specifications to consider are frequency range and waveform, displacement and velocity, the number of axes to be tested, payloads and vibration/shock levels, and force rating.

Frequency Range and Waveforms

Evaluate both current and future test needs in terms of frequency range and waveforms. Various technologies are used as energy sources for shakers and each has a different frequency range with considerable overlap among them.

The list of available technologies may be reduced considerably for certain applications because they cannot handle the frequency range or waveforms. Table 1 defines frequency ranges and waveforms.

Displacement and Velocity

Shakers have definite limits on the vibration levels they can produce. These limits are quantified in terms of acceleration, velocity and displacement (related by frequency-based conversion factors). With sinusoidal motion, simple formulas define the relationship between D (displacement in inches pk-pk), V (velocity in inches/s [ips]), g (acceleration), and frequency, f, in Hertz.

When a sine-test specification clearly defines displacement and velocity limits, it is simple to compare the requirements to the shaker system’s capabilities. It can be disastrous, however, when large displacement or large velocity requirements are “hidden” by the way the test specification is presented.

For example, this requirement appears rather benign: “Perform a sine sweep at 1″ (pk-pk) from 10 Hz to 17 Hz, then hold 20 g(pk) from 17 Hz to 1,000 Hz.” What is not stated can be just as important as what is: at 17 Hz, the 1 inch (pk-pk) of displacement equates to a velocity of nearly 54 ips (pk). If you select a shaker with a 40-ips velocity limit, it will trip off-line at 14 Hz regardless of how big its force rating is.

Historically, “canned” specifications were used extensively. These specifications ran on a particular type of shaker and therefore you seldom encountered specifications with nasty surprises. Today, more tests are based on actual field or end-use measurements, so you no longer can rely on others having already run a particular test. Operating environments can easily exceed the limits of an existing shaker or a low-priced shaker system.

Displacement and velocity limits apply to random vibration and shock tests as well as sine tests. A wise user will check all known tests for hidden requirements. Table 2 shows typical velocity and stroke limits for each of the major types of shakers.

Number of Axes To Be Tested

The decision to test in one, two or three axes is application specific. Many tests are performed in only one axis: sometimes vertical (V), sometimes horizontal axis #1 (H1), sometimes horizontal axis #2 (H2). But others require two or even three axes of vibration/shock. Multi-axis testing can also be done sequentially or simultaneously, which can significantly affect the selection of the shaker system. It is critical that these factors be defined before the shaker system is selected.

Payloads and Vibration/Shock Levels

A certain amount of force is required to cause a particular payload and armature to vibrate at a given vibration level. Newton’s Law states that F = ma (force = mass ´ acceleration).

The form in which this is almost always used is F = wg. F is the force (in pounds) required to vibrate the weight w (in pounds) at a certain vibration level g (which is dimensionless). To eliminate confusion, since both force and weight are measured in pounds, pounds-force (lbf) is often used to refer to force and pounds-mass to refer to weight.

Everything that vibrates is represented by w. While visualization seems to help most people, high-frequency vibration is not visible, so imagine a low-frequency (5 Hz), large-amplitude (1 inch) sine test. Even if this is nothing like the test you intend to run, you will still find the exercise useful.

Now that you are visualizing the UUT vibrating at 5 Hz and 1 inch, notice that all attachment fixturing is vibrating along with the head expander, the slip plate and the attachment bolts. The armature of the E-D shaker or the piston of the hydraulic shaker is also vibrating.

Since all these items vibrate, you must include the weight of all of them in the w of the equation F = wg. If you overlook any of the items that move during this visualization, you will underestimate the amount of force required to create a certain g level. That could cause some very nasty surprises when you start to run the test.

Force Rating

Even though sine tests are decreasing in popularity, shaker sine force ratings are still very important. There are considerably fewer assumptions necessary when assigning a sine force rating as compared to ratings for random vibration or shock. So when you compare shakers, it is best to use sine force ratings.

Sine force ratings for shakers are almost always defined in pounds (force) pk for sine tests. However, the label “pounds (force) pk” is often implied, so it can be quite confusing if you are not accustomed to industry jargon. For example, if a certain shaker is called “our 10,000-pound shaker,” the implication is that the shaker will produce a sinusoidal force of 10,000 lb pk for a sine test.

Random vibration ratings have quite a few built-in assumptions, and different suppliers use different sets of assumptions. The major factors affecting random vibration ratings are payload size, bandwidth, power spectral density (PSD) shape, sigma value (crest factor) and dynamics of the payload. The details of these factors are beyond the scope of this article, but we recommend you contact the supplier of the shaker system to fully understand the assumptions made for a particular system.

Shaker force ratings are derived using lumped masses as payloads. When a payload (fixture or UUT) has resonances in the test bandwidth, it can reduce the shaker’s capability. This is undoubtedly the most subjective area of shaker sizing, but there are some rules of thumb using a concept known as the mass multiplier. It is based on the fact that a resonating payload appears at certain frequencies to be much heavier than its deadweight.

Defining Target Tests

The best method to address the issue of payload and vibration levels is a concept known as target tests. ISO5344 is being promoted as a good way for you to compare one shaker with another. Unfortunately, the test parameters in ISO5344 are not real-world tests—a flat PSD spectrum from 100 Hz to 2 kHz, for example, is one of the rarest tests imaginable. As a result, you may think you are getting a good bargain on a shaker.

We conclude that ISO5344, while helpful to shaker manufacturers (by relieving them of the responsibility that their equipment will perform as needed), offers you little assurance that your target tests will be run.

The target test can help you select a shaker system. There is, however, a tendency to resist this concept, to ask for a really big shaker in anticipation of future testing needs.

As previously defined, this bigger-is-better approach can be totally wrong. For this reason, it is critical to go through the target-test approach even though the variables of UUT weight and velocity levels must be estimated. It is much better to estimate the target tests than to estimate a force rating.

It is usually sufficient to use extreme cases for the target test. This reduces the number of target tests. For example, define the complete set of target-test information for tests having:

The highest vibration level.

The highest level shock test.

The heaviest UUT.

The heaviest fixturing.

The UUT with the largest footprint.

The worksheet that accompanies this article can be useful.

What Not To Do

Don’t pick a shaker system out of a catalog.

Don’t buy a shaker without a controller and fixturing unless you are highly experienced.

Don’t specify a force rating for a shaker—specify target tests.

Don’t wing it when it comes to fixture design and specification. Get advice from an applications expert.

Acknowledgments

The Institute of Environmental Sciences originally identified the need for guidelines for selecting a shaker system, and several of its members contributed many good ideas. We thank them.

About the Author

Edward L. Peterson is part owner of and Director of Applications Engineering for MB Dynamics. From 1971 to 1985, Mr. Peterson held numerous technical and test management positions with Structural Dynamics Research, both in the United States and the United Kingdom. He has B.S.M.E. and M.S.M.E. degrees from the University of Cincinnati, and is a Professional Engineer. MB Dynamics, 25865 Richmond Rd., Cleveland, OH 44146, (216) 292-5850.

Table 1

Type of Energy Source	Typical Low- Frequency Limit	Extreme-Case Lowest Frequency	Typical High- Frequency Limit	Extreme-Case Highest Frequency	Waveform
Electrodynamic	5 Hz to 20 Hz	2 Hz⁽¹⁾	2 kHz to 3 kHz⁽²⁾	5 kHz to 10 kHz⁽²⁾	Any
Servo-Hydraulic	DC	DC	100 Hz to 500 Hz	500 Hz to 1 kHz⁽³⁾	Any ⁽⁴⁾
Mechanical	15 Hz	5 Hz	60 Hz	80 Hz to 120 Hz	Sine only, usually fixed frequency ⁽⁵⁾

Table 2

	Displacement				Velocity
Type of Energy Source		Typical Maximum	Extreme Case	Typical Maximum		Extreme Case
Electrodynamic		0.5″ to 2″ pk-pk	24″ pk-pk	40 to 70 ips (pk)		330 ips (pk)
Hydraulic		2″ to 12″ pk-pk	24″ pk-pk	20 to 30 ips (pk)		50 ips (pk)
Mechanical		0.1″ pk-pk	0.3″ pk-pk	30 to 40 ips (pk)		50 ips (pk)

Suggested Worksheet for Determining Target Tests

(Use one copy for each test.)

1. Description of UUT (without fixturing)

a. Deadweight:

b. Footprint Size (at attachment points):

c. Height:

d. Number of UUT per Test Run (may want to define both minimum and desired number):

2. Test Description

a. Purpose (circle all that apply): design qualification, environmental simulation, transportation simulation, ESS, other (specify)

b. Type of Waveform (circle all that apply): fixed frequency sine, sine sweep, random (smooth PSD shape), random (real-world PSD shape), shock (classical), shock (real-world shape), other (specify)

c. Test Level(s) and Frequency:

1. Random—minimum and maximum frequency, grms (overall), and sigma value

2. Sine—minimum and maximum frequency, maximum g, maximum velocity and maximum displacement

3. Shock—g level, time duration of biggest peak and maximum frequency

3. Number of Axes

a. Single-Axis Vertical:

b. Multi-Axis Sequential (define which axes):

c. Multi-Axis Simultaneous (define which axes):

4. Description of Fixturing (always depends on UUT and test description)

a. Fixturing (including general-purpose fixturing such as slip plate, head expander, cube and T-plate, and special-purpose fixturing such as holddown/attachment fixtures):

b. Deadweight of General-Purpose Fixturing and Special-Purpose Fixturing (for each axis of the test): vertical horizontal #1 horizontal #2

5. Mass Multipliers for UUT/Fixturing

a. Random Vibration and Shock Tests—add at least 10% to the deadweight of the UUT/fixturing if it has any significant resonances in the test bandwidth; add at least 20% if the UUT/fixturing has especially deep anti-resonances:

b. Sine Vibration Tests—add at least 25% to the dead weight of the UUT/fixturing if it has any significant resonances in the test bandwidth; add at least 50% if the UUT/fixturing has especially deep anti-resonances:

May 1996