Simcenter Testing Solutions Vibration Control: Understanding SelfCheck

Direct YouTube link: https://youtu.be/V-J8rwo3UNQ

*** Note: Simcenter Testlab 2206 and higher, the SelfCheck was replaced by System Identification and Verification.  ***


What is a SelfCheck? And why is it required? The SelfCheck is a pre-test check, like having an extra pair of eyes to verify the setup of the vibration control test before it is run. SelfCheck is a feature in Simcenter Testlab Vibration Control software.

There are many components and connections in a closed loop vibration test (Figure 1) that need to be checked for potential issues.
 

Figure 1: Multiple connections and components to troubleshoot in a vibration control test setup

While the SelfCheck may not find every possible problem, this “Pre Check” can uncover many common issues with transducers, cables, amplifiers, shakers, and drive profiles. This saves time and effort by ensuring tests run smoothly, and prevents unintended damage to test articles due to improper setup.

This article has the following contents:
1. How Does SelfCheck Work?
     1.1 Background Noise and DC Offsets
     1.2 Broadband Noise Buildup
     1.3 Prediction of Full Levels
2. Selfcheck Report
3. What are Common Errors/Concerns Found?
4. Channel Overloads
4.1 Possible Causes of Overloads
4.2 Possible Solutions for Overloads
5. Open Channels
     5.1 Possible Causes of Open Channels
     5.2 Possible Solutions for Open Channels
6. DAC Issues
     6.1 Possible Causes of DAC Issues
     6.2 Possible Solutions for DAC Issues
7. Conclusion
 

1. How Does SelfCheck Work?

How does SelfCheck determine if there are any issues in the test setup? It tests the complete shaker system with a low level, broadband input looking for overloads, open channels, DC offsets and other possible sources of operational issues.

There are multiple steps in the SelfCheck process:

1.1 Background Noise and DC Offsets

With no signal from the Simcenter SCADAS DAC driving the shaker system, the background noise level on all channels is measured. The background levels are used as a reference when broadband excitation signals are sent to the shaker system during the SelfCheck. The measurement channels should have significantly higher levels when the broadband signal is present than when no excitation is applied. DC offsets are also checked on all channels which could indicate ground loop issues.

1.2 Broadband Noise Buildup

The SelfCheck starts by sending out a low level, broadband vibration from the Scadas DAC output. This starts at a very low voltage level, as not to damage the shaker, and reduce the amount of fatigue the test item experiences.

The output level is then built up in stages (Figure 2) through multiple voltage levels until the signal to noise ratio and coherence criteria is achieved. These output levels are below the voltage required to reach the full test level. By default, these voltages range from 20 mV to 100 mV over five steps. The user can change the minimum and maximum voltages, along with the number of steps.

Figure 2: Right: Broadband SelfCheck spectrum, Left: SelfCheck buildup schedule.

In addition to calculating levels on all channels, transfer functions are also computed between the following:

• Control Channels and Measurement Channels
• SCADAS DAC Output

These transfer functions are checked for consistency via a coherence function, and used in the prediction of levels for the full scale test.

1.3 Prediction of Full Levels

The full scale test levels are predicted. Even if issues like overloads did not occur during the low level SelfCheck, they can be identified by the prediction of the full level test, before it is actually started. The SelfCheck uses the transfer functions and the target profile to project the voltage levels required to run at full level test. Using the Signal to Noise (S/N) ratio defined under the Advanced Parameters menu, the predicted levels are compared to the measured background noise. The predicted levels should be well above the background noise for a successful test execution, otherwise a open loop flag is shown.

2. Selfcheck Report

After the SelfCheck is finished, a report is generated as shown in Figure 3.

Figure 3: SelfCheck Report 

It contains several items for each channel:

1. Channel Name
2. Background Noise – Level measured on the response channels with no drive signal output to shaker system.
3. SelfCheck Level (milliVolts) – The response signal level measured during the maximum SelfCheck drive level
4. S/N Ratio – The Selfcheck Level divided by the Background Noise must be greater than the Signal to Noise (S/N) ratio set under the “Advanced…” button in SelfCheck.
5. Expected Level (Volts) – Level expected during the full scale test based on the maximum value in the reference profile.
6. Expected Level (EU) – Same as 5, but expressed in Engineering Units (EU).
7. Status – SelfCheck result, possible values include “OK”, “Overload”, and “Open Channel”.

3. What are Common Errors/Concerns Found?

The errors that are commonly reported during SelfCheck fall into three broad categories:

• Channel Overloads
• Open Channels
• DAC Issues

Possible causes and solutions for each category follows:

4. Channel Overloads

During the SelfCheck, all channels are checked for actual (during the buildup) and potential overloads (at the full level test) as shown in Figure 4.

Figure 4: Overload error during SelfCheck buildup

Overload checks are done both on control and measurement channels.

4.1 Possible Causes of Overloads

Actual or predicted overloads can be caused by a few different issues:

• DC or Ground Loop issues. Accelerometer could be grounded to the test structure.
• Predicted vibration level during the full test is expected to create overload.
• The channel range is too restrictive for the expected levels during the test.

4.2 Possible Solutions for Overloads

When overload errors occur, some possible solutions include:

• Check the lights on the SCADAS frontend. When the shaker system is not being driven, no red overload lights should be visible. With no excitation signal, a red overload light on ICP accelerometer channels indicate a problem with the transducer or the transducer cable. Tighten the accelerometer cable. If the red light persists, replace the cable. If still not resolved, replace the accelerometer.
• Eliminate any ground loops in the test setup. For example, isolate the transducer from structure via isolation mount rather than screwing directly to test structure.
• For a predicted overload at full level, use an accelerometer with less sensitivity. For example, replace 100 mV/g accelerometers with 10 mV/g accelerometers.
• Increase the input range of accelerometer channel. This is done in the Channel Setup Locate a column called Range. Normally the range cannot be increased beyond 10V, which is the maximum voltage level signal a SCADAS can accept.
• Sometimes it might be decided that overloads cannot be avoided, for example on some measurement channels. In this case, under "Advanced Control Setup", the "Measurement Channel overload action" can be set to "Ignore" as shown in Figure 5.

Figure 5: Overload ignore setting in "Advanced Control Setup" menu

5. Open Channels

Open channel messages, as shown in Figure 6, indicate that there is not a stable relationship between the control accelerometer and DAC output. One reason for this would be that the accelerometer channels are not producing significant output above the background noise level.

Figure 6: Open channel error during SelfCheck buildup

A channel is considered “Open” if the measured level during the final stage of SelfCheck is less than the background noise multiplied by the user defined signal to noise ratio. I.e., if the broadband white noise signal is not higher than the background noise.

In addition to an Open Channel error, a “Low Coherence Warning” could also be given.

5.1 Possible Causes of Open Channels

An open channel error means that the signal from an accelerometer is very low, or of poor quality. It is not significantly above the noise floor. Some example causes are shown in Figure 7.

Figure 7: Possible causes of open channels during vibration control test

Open channel errors could be caused by issues with control and measurement accelerometers:

• Accelerometers are not well attached, may be only partially attached to test structure.
• Accelerometer is dead or malfunctioning, but gives no ICP overload error.
• Wrong conditioning mode is selected in Channel setup, for example, Voltage AC mode set for ICP
• Accelerometer fell off the test structure and is not attached. In a test setup with many channels, this may not be obvious.
• Amplifier gain may be too low.
• SelfCheck maximum voltage level is too low. If the levels are too low, the SelfCheck does not exceed the background levels measured on the channels.

5.2 Possible Solutions for Open Channels

• Make sure all accelerometers are attached securely to structure.
• Check channel setup and make sure ICP input mode is selected for ICP accelerometers.
• Increase the amplifier gain.
• Use an accelerometer with more sensitivity. For example, replace a 10 mV/g accelerometer with a 100 mV/g accelerometer.
• Decrease the input range of the channel, for example from 10 Volts to 1 Volt.
• Under the “Advanced…” button in SelfCheck, increase the “Maximum maximum drive level” for the SelfCheck as shown in Figure 6.
• Under the “Advanced…” button in SelfCheck, lower the value “Selfcheck S/N ratio to drive update” parameter as shown in Figure 8. This should be thought of as a last resort, if the previous suggestions did not work. By lowering the S/N ratio, the measurements will definitely be of poorer quality and contain more noise.

Figure 8: Parameters from the “Advanced Button…” in SelfCheck worksheet
 

6. DAC Issues

Based on the SelfCheck’s low level white noise, the full test levels are predicted. If the DAC output would exceed 10 Volts (minus a safety factor) to drive the shaker, the self-check fails as shown in Figure 9. The global status will read DAC Overload.

Figure 9: DAC Overload warning message from SelfCheck
 

6.1 Possible Causes of DAC Issues

There is a high likelihood of an issue with the shaker amplifier. It may not be capable of outputting enough to run the full scale test. The capabilities of the shaker system may be exceeded.

6.2 Possible Solutions for DAC Issues

• Check if shaker amplifier is ON.
• Increase the shaker amplifier gain. For example, if the gain is half way up, turn it to three quarters.
• Check the test profile and make sure it is correct. If any values were accidentally entered at higher levels than they should have been, this would be a potential cause.
• If the amplifier is turned all the way up, a higher performance amplifier or shaker capable of larger displacements/velocity/accelerations may be needed.

7. Conclusion

If all overloads, open channels, and DAC requirements are met, the SelfCheck will pass as shown in Figure 10. In the upper right corner of the SelfCheck screen the Global Status indicator will be set to “OK”. Each channel will have an individual status that should also be “OK”.

Figure 10: Successfully passed selfcheck result

If utilized properly, a SelfCheck is another pair of eyes checking and quickly identifying possible problems and issues in a vibration control test campaign. Some of the benefits of running SelfCheck include:

• Saving Time - The SelfCheck tries to make sure that the test will run without a problem from startup to the full level, and complete the required test time duration. This often prevents tests from having to be rerun.
• Preventing Damage – The SelfCheck protects the test article and shaker system from unwanted damage due to unexpectedly high levels from an improper test setup. This can be especially important when testing a one-of-kind prototype.

In the Simcenter Testlab Documentation, there is extensive documentation on the inner workings of the SelfCheck in the “Testlab Random Control” manual. The documentation is located in “Start->Programs-> LMS Test.Lab {rev}->Documentation” or in “Start->Programs-> Simcenter Testlab {rev}->Documentation”

Questions? Email william.flynn@siemens.com or contact Siemens Support Center.

Related Vibration Control links

• Index of Testing Knowledge Articles
• What is Vibration Control Testing?
• Vibration Control FAQs
• Power Spectral Density
• Fatigue Damage Spectrum
• Breakpoint Table Generator
• Getting Started with Random Control
• Random Control: Weighting, Frequency Resolution, and DOFs
• Combined Modes: Sine on Random (SoR), Random on Random (RoR), Sine on Random on Random (SoRoR)
• Kurtosis
• Random Control: Measurement Channels
• Sine Control: Estimation Methods
• Sine Control: Closed Loop Control Parameters
• Sine Data Reduction
• Sine Control: Notching
• Simcenter Testlab: Tracked Sine Dwell
• Vibration Control: Multi-Point Control Averaging
• Shock Response Spectrum
• Classical Shock Pulses
• Shock Response Analysis and Synthesis
• Single Axis Waveform Replication (SAWR)
• Vibration Control: Test Sequencing
• Direct Field Acoustic Noise Testing
• "Excessive DC Level" message
• Inverse Transfer Function (ITF) in Random Vibration Control
• Vibration Control Seminar
• Vibration Control YouTube Playlist

Related Durability links

• History of Fatigue
• Stress and Strain
• Calculating Damage with Miner's Rule
• Rainflow Counting
• Mean Stress Corrections and Stress Ratios
• Difference between 'Range-Mean' and 'From-To' Counting
• Some Thoughts on Accelerated Durability Testing
• Goodman-Haigh Diagram for Infinite Life
• Measuring strain gauges in Simcenter Testlab
• Rosette Strain Gauges
• Calculating Damage in Simcenter Tecware Process Builder
• Setting up a SN-Curve in Simcenter Tecware
• Strain Gauges: Selecting an Excitation Voltage