Comparing different vendor products for your custom vibration test system is a tricky and often confusing task. You have to wade through technical specifications that frequently do not correlate from vendor to vendor because different conditions are used to determine the ratings. As a result, you need a method to ensure you have appropriate comparison data for all the vibration products.
To begin, you must know what test environment–sine, random or shock–is required because each is rated differently, said John Raymond, Applications Engineer for Unholtz-Dickie. The dynamic behavior of the test specimen and fixturing play major roles in determining the requirements for a given system. The specific limitations imposed by these variables are significant and cannot be estimated by the shaker manufacturer.
To eliminate some of the system variables, shaker manufacturers have adopted a set of accepted conditions for determining a shaker’s performance, conditions that make comparison among competing products possible, said Mr. Raymond. These standard conditions are defined either by the manufacturer of the equipment or by industry standards such as ISO 5344 Electrodynamic Test Equipment for Generating Vibration – Method of Describing Equipment Characteristics. In general, these conditions include testing with a nonresonant (stiff) payload and testing with a heavy payload, usually twice the armature weight.
For sine testing, the conditions include testing with a flat frequency spectrum limited by the maximum peak acceleration, peak velocity and peak-to-peak displacement. Ratings for random testing per ISO 5344 use a flat spectral density from 100 Hz to 2,000 Hz; however, some companies offer a rating as low as 20 Hz for light payloads because the lower range contains frequency components common to most test applications.
The ISO 5344 standard specifies a sharp roll-off of -20 dB/decade below 100 Hz , which allows manufacturers to cite a higher overall rating for light loads or bare-table operation. This rating may not apply for applications that require vibration excitation below 100 Hz, said Mr. Raymond.
At the heart of every vibration system is the armature: the test table on which the UUT is clamped and the coil assembly in which an electromagnetic force is produced. The diameter of the armature dictates the size or the volume of the UUT that can be vibrated at one time. The lighter the armature, the heavier the UUT can be for a given acceleration or force. When calculating the maximum load capacity, account for the total moving mass, which is the UUT, the vibration fixture, the test table, the coil assembly and all the bolts.
The greater the armature speed needed, the greater the voltage swing required. The greater the force or acceleration needed, the more current required. Additionally, electrodynamic (ED) shakers driven with sine and random signals at full power for long periods make severe demands on their power amplifiers. An amplifier system should be rated to allow for continual operation.
Choosing the best amplifier for your vibration system should be done with special attention to the harmonic distortion of the instrument, said Christopher Williams, Joint Managing Director at Ling Dynamic Systems. If harmonic distortion is high, there is a risk of creating unwanted resonances during the test process. To get optimum efficiency, reduced power consumption and reliability, look for current peaks at least three times the rms value, said Mr. Williams.
For ED shakers, force ratings are usually defined in pounds force peak for sine tests, said Ed Peterson, Director of Applications Engineering at MB Dynamics. The vibration for a sine test is generally defined in gravitational units (or m/s2) peak, so it is convenient to express the force level in pounds-peak, eliminating conversion from peak to rms or even peak-to-peak.
Look more closely when comparing random-vibration ratings from different vendors, said Mr. Peterson. You have to make assumptions when comparing random-vibration ratings because there are so few industry standards that many suppliers develop their own rating formulas.
When comparing ED shaker random-vibration ratings, consider the payload size, bandwidth, power spectral density, the crest factor and dynamics of the payload. The demands on ED shaker armatures usually increases as the payload increases, due to potential resonant frequencies. Most suppliers require the payload to be at least three times heavier than the armature to achieve the random rating.
Most random-vibration ratings for ED shakers assume a crest factor of three, said Mr. Peterson. But it is not a fixed limit because tests are performed with data from the operating environment with a crest factor of greater or less than three.
Another factor affecting random-vibration ratings is the dynamic quality of the payload, continued Mr. Peterson. Force ratings assume a nonresonant payload and do not account for the dynamic characteristics that make it easier or harder for the shaker to produce a given level of vibration.
A standard force rating is then determined from the equation:
F = (Warm + Wtw)A
where: Warm = weight of the shaker armature
Wtw = test weight
A = maximum test level in gravitational units (g)
This equation makes no provision for the dynamic behavior of the fixturing or test specimen encountered in the test, said Mr. Raymond. Nor does it reflect a specific method of control or placement of accelerometers. These factors significantly affect the force requirements and should be discussed with the vibration test-system manufacturer.
The impact that the test specimen, total payload and control strategy have on the final requirements for a vibration test system is significant, said Mr. Raymond. For example, if a particular set of requirements falls outside of the standard rating envelope, the manufacturer should be advised so an expanded system rating can be provided. Estimating the performance by applying a generalized derating function is not always recommended and may be misleading, he said.
Some manufacturers have developed target tests to help customers select a shaker system. The number of different vibration tests will determine how many target tests to run. For example, a test lab that serves as a resource for many customers will perform several target tests to ascertain the best shaker system.
Many test engineers, however, lament that they have no way of knowing what they will be testing or at what levels the system will have to operate, said Mr. Peterson. Often, they get a big shaker to cover all possibilities.
This bigger-is-better approach can lead to problems because there are many target tests that a 5,000-lbf shaker can perform and a 10,000-lbf cannot. For this reason, it is important to perform a target test even if the variables of the UUT weight and velocity must be estimated. It is better to make estimates related to target tests than it is to make estimates of force ratings, said Mr. Peterson.
To keep the number of target tests at a reasonable number, it is sufficient to use extreme cases, continued Mr. Peterson. For example, you could define the complete set of target tests covering the highest vibration level, the highest level shock, the heaviest UUT, the heaviest fixturing and the largest UUT footprint. It is prudent, however, to define at least a few intermediate target tests that have a medium-size payload and level of vibration, because they occasionally require a more capable shaker than is used for the extreme cases, he said.
In the case of the quasirandom pneumatic shaker, the force is produced by reciprocating pneumatic hammers. As a result, the force rating is not calculated in the same manner as the ED system, said Gilber Bastien, New Products Development Engineer for Screening Systems. However, an approximate method may be used to obtain an equivalent force for the pneumatic quasirandom shaker and the ED shaker.
The force rating of the ED shaker is multiplied by the maximum payload weight to obtain the payload force pound rating. The equivalent of the ED rating to the quasirandom force-pound rating is calculated by multiplying the maximum vertical grms by the maximum table payload weight. A modified version of the quasirandom force-pound rating is performed by integrating the flat portion of the input spectrum for the grms value between 20 Hz and 2,500 Hz and multiplying this value by the maximum payload weight, said Mr. Bastien.
Multiaxis Vibration
Conventional vibration generators, including hydraulic, ED and mechanical, are typically single-axis devices. They can be transformed into a multi-axis system with the addition of plates and bearings.
The most common form of multi-axis conversion uses a slip plate combined with a vibration generator. It utilizes a vertically mounted ED generator which is rotated 90° and connected to a magnesium slip plate. The slip plate is restrained from cross-axial movement by bearings, which are often load-carrying.
The dynamic behavior of the plate is improved by using a granite block. The slip plate slides on the granite block, lubricated by a pressurized hydraulic film between the adjacent surfaces.
Although this is the most common approach to multi-axis testing, it is not without its weaknesses. For instance, the UUT must be moved three times from the vertical to the horizontal and then rotated on the horizontal. Also, each axis is tested sequentially, which is unlike actual conditions.
For automotive applications, you can mount triaxial accelerometers on the UUT as it would be installed in the field, and then record the acceleration responses of typical operating loads, said Mr. Peterson. Through sensible data reduction and analysis, these acceleration-time histories in the three orthogonal directions can be used on a selective basis either as direct time-domain input to control the shaker in each direction or converted to acceleration spectral densities for creating or using random-vibration profiles for excitation in each of the three directions, said Mr. Peterson.
Mechanical multi-axis testing systems suffer from a huge disparity between the desired and the actual spectrum, said Gerry Priebe of Data Physics. An FFT analysis system is required to check them thoroughly. A digital controller/FFT analyzer combination can also be used to measure the actual vibration expected and then to control the shaker to reproduce the desired spectrum.
Vibration Test Products
Accelerometer Offers
Sensitivity to 500 mV
The EGE-73 Series Accelerometer uses silicon-chip technology. The sensing element allows voltage sensitivity to 500 mV full scale. It is offered with half-bridge or full-bridge wiring and standard frequency response envelopes, and is compatible with uniaxial and triaxial systems. The unit is made for vibration systems, and automotive, aerospace and aircraft testing. It is available in force ranges from 20 g to 5,000 g. Entran Devices, (800) 635-0650.
Input Subsystem Features
92-dB Dynamic Range
The High-Performance Input Subsystem features a 92-dB dynamic range and an autoranging function. The 16-bit A/D converter on each channel enables users of the GenRad 2550 Vibration Control System to increase measurement accuracy. The subsystem combines four parallel measurement channels with digital signal processors on one board computing a 1,024-point FFT waveform in less than 4 ms. The sampling rate is 51.2 kHz. GenRad Structural Test Products, Division of Spectral Dynamics, (408) 970-1600.
Air-Cooled System Performs
Automotive, Electronic Testing
The combination of the Series 800 Shaker, the SPA-K amplifier and either the DVC 4000 or DSC4 Controllers provides the capability to perform vibration testing for automotive components, electronic assemblies and avionics equipment. The shaker series supports force ratings from 2,200 lbf to 8,000 lbf. The amplifier series has a power range from 5 kVA to 50 kVA and has a signal-to-noise ratio of >68 dB. The DVC 4000 Controller includes sine, random and transient control. The DSC4 Controller provides 16 user-defined sine test programs with up to eight vibration levels. Ling Dynamic Systems, (800) GO TO LDS.
Software Helps Tailor
Single-Axis Test Specs
The LMS Mission Synthesis is a software module that helps tailor single-axis vibration testing to account for different environments and loads. It incorporates random, sine and transient inputs to customize testing. The software helps characterize mechanical environments for fatigue or limit load damage potential, and synthesize an environment based on the fatigue-damage equivalent or maximum-load equivalent. The module runs within the company’s CADA-X Time Data Processing Monitor package. LMS North America, (810) 952-5664.
System Runs Vibration
Programs in Parallel
The VCP9000 Vibration Control System runs random programs, performs setups and plots, and synthesizes a response spectrum in parallel. The multitasking X-Windows-based system enables closed-loop random control as well as closed-loop swept and fixed sine control. Sine dwell, classical shock, shock response synthesis control, sine-on-random, random-on-random, and combined vibration, temperature and functionality tests also are offered. The control system provides notching in the random and the sine modes. Channel mapping is also supplied in all modes of operation. M + P International, (201) 239-3005.
Vibration, Shock Controller
Runs Under Windows
The Win2001 Vibration and Shock Controller features a Microsoft® Windows-based interface. It allows simultaneous data inputs in two, four and eight channels and controls one or two shakers separately. The controller software includes random, sine, classical shock with time, frequency and SRS analysis, sine-on-random, random-on-random, FFT analysis, second-shaker control, and random excitation and analysis with 1,600 and 3,200 lines of resolution. MB Dynamics, (216) 292-5850.
System Incorporates
48″ x 48″ Vibration Table
The OVS-4 Temperature and Vibration Screening System incorporates a 48″ x 48″ vibration table and offers simultaneous excitation and exposure to temperature extremes. The vibration frequency range extends from 2 Hz to 10,000 Hz. The company’s OmniAxial™ Vibration System excites three linear axes and three rotations with broadband random vibration. The UltraRate™ Thermal System allows temperature ramping to 60° C/min over a range from -100° C to +200° C. QualMark, (303) 254-8800.
Chamber Provides
16 Pneumatic Actuators
The QRS-600V Power Screen System has a triaxial quasirandom vibration system that uses a damped 44″ x 44″ shaker table with 16 precision pneumatic actuators suspended on four air bags to isolate vibration from the system enclosure. Six accelerometers provide feedback to the controller, which updates the valves to maintain programmed vibration characteristics. Vibration levels of 30 grms are attainable. Loads up to 600 lb can be placed on the shaker table. The temperature chamber provides transition rates to 60° C/min. Screening Systems, (714) 855-1751.
Vibration Controller Offers
Hardware/Software Modules
The DP550Win Vibration Controller is a series of PC-based cards that consist of a digital signal processor, AD and DA hardware, and software for processing all modes of vibration and shock control. Each board contains 16 channels of simultaneous input configured for control, response or limit. The 16-bit DAC has a reconstruction filter and provides an 80-dB dynamic range. Unlimited test sequencing via switch closure or remote access is provided. Real-time display of 16 graphic or text windows is supported. Data Physics, (408) 371-7100.
Digital Control System
Supports Parallel Inputs
The UD-VWIN Digital Vibration Control System operates within a Windows™ environment running on a Pentium®-based workstation. The system uses digital signal processing and powerful shaker control algorithms for real-time test control. It supports random, swept/discrete sine, transient, random-on-random and multitone sine-on-random vibration environments. Multichannel parallel inputs, each with 16-bit resolution, are provided. The graphical interface supports eight independently configurable displays. Unholtz-Dickie, (203) 265-3929.
Copyright 1996 Nelson Publishing Inc.
January 1996