There are many challenges in running a successful vibration control test. In this article, we hope to provide a guide for troubleshooting vibration control tests including common problems, mistakes, and how to fix them.
A basic understanding of vibration control is assumed. For more background on what is vibration control testing, how it is applied, and the benefits of testing, see the knowledge base article: What is Vibration Control Testing?
Vibration Test System
Before we discuss the vibe controller or vibration control system (VCS) let’s define the Vibration Test System (VTS). As shown in Figure 1, the VTS actually consists of a few systems: control system, shaker system, fixtures, signal conditioning, and of course the test item. Each of these “subsystems” have their own dynamic characteristics which contributes to the VTS control loop. A problem with any of them can affect the startup, control, and/or test shutdown.
This article leaves the vibration control system “controller” for a discussion later and focuses on some common problems associated with each of the other “subsystems” and how to avoid them:
FAQ 1: Why can I not pass self-check?
The first place to start when you have an early challenge is the output Drive signal from the DAC. You want to drive the shaker/fixture/test item combination with sufficient energy to excite nonlinearities but not too much to create unnecessary fatigue. It’s very common to be over cautious with the gain setting. Having operated numerous shakers in numerous labs around the world experience dictates using an initial amplifier gain of 30 to 50%. This will assure the output DAC voltage is not too low.
Another historical guideline is trying to achieve a 1 Vrms drive for random control and a 1Vpk drive for sine. This will help maximize the dynamic range of the measurement hardware an allow headroom to increase or notch the drive signal as required to stay in control.
Learn more about self-check here Understanding Self-Check
FAQ 2: What does a DAC overload mean?
Most digital to analog (DAC) outputs, the drive signal to the amplifier, have a 10Vpk maximum output capability. When exceed or estimated to be exceeded a DAC Overload Error will occur. The required test level, with the given masses and amplifier gain settings, cannot be achieved.
If the specifications for the shaker system have been defined properly along with all masses (test item, fixture and armature ….) and no warning has been displayed, increase the amplifier gain if possible and rerun the test. If a warning is displayed you have to consider reducing the moving masses or test on a higher force rated shaker system.
FAQ 3: I hear a hum and feel a vibration when I turn on my amplifier and shaker.
The VTS consists of many electronic pieces. In a system of electronics (i.e. the subsystems of the VTS), like to have a common ground. When these systems have different grounds the differential can produce 60Hz (50Hz in Europe) noise in the control loop. The amplitude of this signal is often proportional to the gain setting on the amplifier.
To resolve and reduce this unwanted noise first check that your transducers are isolated from the fixture and test item as discussed in FAQ 14. If the noise still exists change the power source for the controller. In some labs, there may be a clean power source (usually with orange receptacle covering as shown in Figure 2). You can also try floating any signal conditioning equipment if present.
FAQ 4: I pass self-check but abort on start up.
Review the drive signal spectrum. Good chance it is too low. Increase the self-check level as shown in Figure 3. If you do not hear some rattling during this pretest check good chance your amplifier gain is too high. Decrease the gain setting, check your transducers and rerun the test.
FAQ 5: What is the difference between the amplifier gain and the gain measured after a self-check?
The simple definition of control loop gain is the relationship of the magnitude of the input to the magnitude of the output. Each piece or ‘subsystem” of the VTS has its own signature or transfer function. Each of these pieces has its own inherent gain, some controllable, some fixed. The gain measured by the control system is the complete system gain, all the pieces together. This gain is defined as the number of g’s per volt. How many g’s will one can expect per 1V. (Figure 4)
Independently, the amplifier has its own controllable gain which does have a direct effect on the overall system gain. This should explain the sensitivity of the total system gain on the amplifier gain.
FAQ 6: I can hear the self-check but I get an open channel afterwards.
Too often the focus of the test is on the control system and not the parts that make up the VTS. Transducers (accelerometers) plus their cabling and how they are powered can be considered the most critical part of the control loop.
Besides noting the calibration date and sensitivity one has to pay very close attention to how the transducer is mounted to the fixture and test item. Whether it is mounted with a screw, glue, or wax it must be electrically isolated from the mounting location. Double check how the transducers are mounted (Figure 5).
FAQ 7: What does an input channel overload error mean?
The measurement hardware part of a typical control system has a 10Vpk maximum input range. Any signal larger than 10Vpk will trip an overload to protect the internal electronics. It should be noted that an AC signal riding on top of DC signal can cause an overload if the 10Vpk threshold is exceeded. If an overload is possible during the test, it is often flagged in the selfcheck as shown in Figure 6.
An overload can also occurred if the wrong transducer is selected or the sensitivity is entered incorrectly. If the predicted response at full level with the defined transducer sensitivity will produce a signal response greater than 10Vpk an overload warning will occur.
FAQ 8: Too much background noise detected.
As discussed above in FAQ 3 and FAQ 6 above there is a number of possible sources of unwanted background noise. Review these Frequently Asked Questions. When too much background noise is detected, a message similar to Figure 7 may be shown.
FAQ 10: What sensitivity accelerometer should I use for my test?
This is a “Catch 22”. You want a reasonable voltage response from the transducers but too high a voltage can cause an overload.
The larger the sensitivity value, more mv per engineering unit (mv/EU), the greater the possibility you have to pass self-check and run your test but pay attention to the maximum target levels and expected responses.
Unless you are going to experience very high responses 100mv/EU is always to good initial choice.
Avoid running low level tests such as a 0.5g sine test with a 10mv/g accelerometer. Do the math, it can drive you crazy trying to complete a test.
FAQ 11: What is TEDS and do I need to use it?
Transducer Electronic Data Sheet (TEDS) is an IEEEE standard that defines how to store data on an analog measurement sensor (Figure 8). Once programmed the data can be downloaded to automate the test set-up process. Data includes calibration value, model number, serial number and other specifications.
The answer to this question is "No", you do not need it, but it can make test set up easier if you have a lot of measurement channels and it can help reduce the possibility of a data entry error. The vast majority of transducers worldwide today do not have TEDS capabilities but the market for TEDS is growing as older transducers are replaced and new ones purchased.
FAQ 12: Where do I mount the transducers?
How was the data you are trying to replicate measured? If measured in the field and analyzed in the lab there is a very good chance you can mirror the measurement points for control and response channels (Figure 9). Unfortunately this is rarely the case.
Most tests reference some specification, see the Standards for Vibration Control section of What Is Vibration Control Testing? , as a guideline for performing the test(s). In this case the item under test must be subjected to the required vibration or shock levels. You should control at or near the input interface of the test item. It is also very common to measure on or inside the test item to assure predicted design levels are not exceeded.
FAQ 13: How long can I run my transducer cabling?
The shorter the better is always the best answer. Cables from the transducer to the signal conditioning equipment should be kept to a minimum especially if running low level tests i.e. 1Grms with 10mv/g sensitivity. The typical microdot cables used to connect the transducers can work like an antenna especially near the electromagnetic field around the shaker. Utilizing coax cable for long runs to the control room / area is recommended whenever possible.
FAQ 14: What is the proper method of attaching my transducer and cabling?
As outlined above in FAQ 6 no matter how the transducer is mounted it must be electrically isolated from the structure it is mounted too. Each transducer cable should allow for the displacement – movement of the shaker during vibration and provide strain relief for the transducer – cable connection point.
Figure 10 shows a common recommended method for routing cables from the shaker.
FAQ 15: Do I need external signal conditioning?
This all depends on the type of transducers you are using.
Most systems today provide the choice of ICP or Voltage inputs. In the ICP mode transducers with integrated electronics are powered and supply a voltage proportional to the defined sensitivity back to the measurement hardware. In the voltage mode it is assumed an external power supply and/or charge amplifier is in the feedback loop.
FAQ 16: What’s the difference between a Charge Amplifier and a Power Supply?
Accelerometers that measure vibration can be either charge based or voltage based:
Voltage based ICP accelerometers and charge based accelerometers are very similar in performance. Charge based transducers have an advantage for high temperatures.
Charge based accelerometers require extremely good electrical contact connectors, so connectors are typically threaded (Figure 11 - Microdot cable), while voltage based can use simpler co-axial connections (Figure 11 - BNC cable).
Be careful if mixing charge based transducers and voltage based ICP transducers in the same test. While the amplitude measured is the same, the phase is inverted between charge and ICP transducers.
FAQ 17: Do I have to calibrate my signal conditioning hardware and my transducers?
Unfortunately Yes. This is why many labs have move to ICP transducers whenever possible and eliminate the need to track and calibrate another piece of hardware.
ICP transducers with integrated electronics have made significant advancement in recent years especially in the applications of combined environments and shock. They have also decreased in size reducing mass loading effects.
Figure 12 depicts the broad range of ICP transducers available today.
Shakers - Exciters
FAQ 18: My customer wants to know if I can run their high level sine test. How do I determine if it is possible with my vibration test system?
Most shaker systems (amplifier and shakers combination) are delivered with manuals containing detailed specifications for their random, sine and shock capabilities. For a quick estimate you can use the famous Force = Mass x Acceleration, equation PROVIDED you use the correct sine force rating for your shaker system.
Today it is more common to enter all the shaker specifications along with all the masses into the control system software along with the target profile, see Figures 13 and 14.
Once entered and target profile defined you will receive feedback on a number parameters including the percentage of force, velocity and displacement that is required.
FAQ 19: Why do I have to be concerned with the “CG”?
Center of Gravity (CG) for the test object on a shaker test is a single point that gravity would not cause the object to fall. This would be if the object was resting on this single location. Ideally, the center of shaker would be aligned with the center of gravity of the test object as shown in Figure 15.
Most shakers are designed for singular axis movement. Manufactured to provide energy in one direction and minimize vibration in other directions, what is called off axis vibration. If the combined fixture and test item alignment (CG) is off centered you can induce harmonic distortion and cross axis vibrations.
FAQ 20: I have a small shaker we use for modal testing, can I use it for vibration control?
Yes. Look up the shaker system specifications and enter them into your control system. Define the target spectrum and you will have a comparison between what is needed and what is possible. See FAQ 18 for more details.
FAQ 21: Do I need to have a slip table and/or head expander?
Every shaker manufacturer believes so but …… No. It depends on what you’re testing today and in the future. If you’re designing, manufacturing and testing small components and do not anticipate this changing in the future, you can test all three axis on something like a cube fixture as shown in Figure 16.
A head expander is shown in Figure 17. It can be very useful since it increases your testing surface area.
A head expander needs to be very stiff to transfer the shaker energy so the total system mass increases which reduces the force capabilities of the shaker system.
FAQ 22: How thick should I make my shaker mounting plate?
The common engineering practice is 1 inch of thickness for every inch the mounting plate exceeds the diameter of the shaker armature. Most common is a 2 inch thick plate which would allow the mounting plate diameter to be 4 inches greater than that of the armature. Remember every inch means more mass!
FAQ 23: Should I use magnesium or aluminum for my new fixture?
In the equation F = ma, you purchase the F, you are dictated a, and in most cases you can only control the m. Magnesium is the material of choice but it is expensive and challenging to work with in fabrication.
Moore’s Law applies to shaker system too. Force ratings have increased, the market is competitive, and pricing has gone down. Aluminum is a good alternative. But, again you must consider your needs today and in the future, and select a shaker system that is rated for your needs. In most cases, more force is good.
FAQ 24: Is a fixture survey important?
Only if you want to run a good test and replicate the true environment.
As discussed above, the VTS consists of different parts or subsystems, each with their own transfer function. This includes the test fixture. Other than the transducers, the fixture is probably the most important in the system control loop. You need to know the dynamic characteristics of the fixture to ensure you are not inducing unwanted structural vibration into the unit under test. For example, the first modes of the head expander fixture (Figure 19) should be higher than the maximum control frequency.
Ideally the fixtures transfer function would be flat in the testing region, the energy in, equals the energy out. Since fixtures are used repeatedly, it is good practice to perform fixture surveys on a regular basis.
Related Siemens Testing Community Articles: