Simcenter Testing Solutions Getting Started with Random Control

2023-11-02T00:29:57.000-0400

Simcenter SCADAS Simcenter Testlab

Summary

Details

Direct YouTube link: https://youtu.be/8tVmqPmVT6U

This article explains how to perform a complete test with Simcenter Testlab Random Control. It spans controller configuration, run-time, to post processing.

Contents include:
1. Shaker, Amplifier, and Controller Setup
2. Starting Simcenter Testlab Random Control
3. Test Setup
3.1 Channel Setup
3.2 Setting Test Profile
3.3 DOFs: Averages per Loop and Weighting
3.4 Sigma Limiting
3.5 Schedule
3.6 Kurtosis
4. System Identification and System Verification
5. Random Control Worksheet
5.1 Random Control Pane Header Options
5.2 Random Display Area
5.3 Random Control Panel
5.3.1 Run Name Panel
5.3.2 Status Panel
5.3.3 The Alarm/Abort Indicator Panel
5.3.4 Recording Panel
5.3.5 Action Buttons Panel
6. Runtime Procedure: Random Control
7. Post Test: Batch Printing

1. Shaker, Amplifier, and Controller Setup

A closed loop random vibration control test consists of several components as shown in Figure 1:

Figure 1: Diagram of closed loop shaker control items.

Components include:

Controller Computer: PC with Simcenter Testlab installed. Vibration signals are shown on the screen in real-time during the test.
Simcenter SCADAS Hardware: Hardware that digitizes incoming analog accelerometer signals and outputs analog signal to amplifier/shaker.
Amplifier/Shaker: The SCADAS analog output is input into the amplifier and shaker to vibrate the test object.
Test Object: Product or device under test (also called DUT).
Control accelerometers: The control accelerometer is actively monitored during the test to maintain a specified vibration profile.
Measurement accelerometers: Accelerometers on test object that monitor vibration at other locations aside from the control.

The Simcenter SCADAS must be equipped with a control card ending with “-V”. The “-V” option (short for vibration control) indicates that the sources are equipped with an extra safety that will always shutdown gradually as not to damage the shaker or test object.

There is a “STOP” connector that must be in place for the Simcenter SCADAS to be able to output an analog signal from the source. The stop is shown in Figure 2 below:

Figure 2: STOP must be in place for SCADAS to function.

The stop can be hooked up to a DAC shutdown as shown in Figure 3.

Figure 3: DAC Shutdown.

The DAC shutdown has a large yellow button that can be used to quickly stop a test in progress.

2. Starting Simcenter Testlab Random Control

To start the Simcenter Testlab Random Control software, look in the “Testlab Environmental” folder. With the Simcenter SCADAS connected and turned on, double click on the “Random Control” icon as shown in Figure 4:

Figure 4: To start Simcenter Testlab Vibration Control, double click on Random Control in the Testlab Environmental folder.

After the control software is started, choose “Tools -> Options -> Shaker” to make sure the correct shaker is selected for the test as shown in Figure 5.

Figure 5: The shaker system being used in the vibration control test is selected under “Tools -> Options -> Frontend” in Simcenter Testlab. Limits should be entered based on shaker table specifications.

The shaker configuration menu lets important physical limits of the shaker be entered (maximum displacement, maximum acceleration, frequency range, etc). The random vibration control software will check any vibration profiles to ensure that no shaker limits will be exceeded during a vibration test.

If the appropriate shaker is not listed, it will need to be setup. Press the “Configure” button and enter the appropriate information for the shaker being utilized.

3. Test Setup

After starting the software, choose “File -> Save As” to create a project file as shown in Figure 6:

Figure 6: From the main menu, choose “File -> Save As” to store the project. After saving, the worksheets along the bottom are used to setup and run a test by moving from left to right.

The project (*.lms) can be stored in any directory.

While using the Simcenter Testlab software, the test setup parameters (frequency range, settings, vibration profile, etc) and any data collected will be stored in the project file.

The bottom ribbon of the software interface has several worksheet tabs. The worksheets can be utilized from left to right to setup, perform, and then document a test.

3.1 Channel Setup

In the “Channel Setup” worksheet, all accelerometer sensor information can be entered as shown in Figure 7.

Figure 7: The “Channel Setup” worksheet.

Each row in the table corresponds to one sensor input:

Control/Measure: At least one channel must be set as the control channel. The control channel is used as the reference measurement location to recreate the target vibration profile. Multiple channels can also be selected as control, in which case control is based on the average, minimum, or maximum of the selected channels.
Measurement Information: The location of the channel and engineering unit.
Transducer Information: Transducer model, serial number, sensitivity.

Some additional channel setup resources:

3.2 Setting Test Profile

The right side of the Random Setup worksheet is where all the test settings are entered.

Starting from the top right, in the "Control" parameter area, enter the desired frequency range and frequency resolution for the test as shown in Figure 8.

Figure 8: Frequency information for random test is entered in upper right of Random Control screen.

A typical frequency resolution used in a random control test is between two to four hertz. This is to keep the total control loop time at a reasonable interval. While both higher and lower frequency resolutions are possible, the minimum loop time must be considered when programming a random test. Simcenter Testlab calculates the minimum loop time and displays this in the Advanced Control Parameters pop-up shown in Figure 9.

Figure 9: Minimum Loop Time.

The desired amplitude levels across the frequency range are entered in “Edit Reference Profile…” button shown in Figure 10:

Figure 10: Press the button “Edit Reference Profile” to enter the target vibration levels.

The “Edit Reference Profile” button opens a the Profile Editor dialog box where the target vibration profile is entered. Frequency and corresponding amplitude breakpoints are entered. The frequencies entered must cover the minimum and maximum frequencies specified in the upper right of the “Random Setup” worksheet.

The engineering units for the target are typically in g^2/Hz (although Simcenter Testlab can use any type of measurement unit for control). The g^2/Hz is the target vibration level in RMS amplitude, squared, divided by the frequency resolution. This is a Power Spectral Density format for specifying vibration.

Target profiles are sometimes provided via standards for certain industries. Some examples:

MIL-STD-810: United States Department of Defense environmental standards
ASTM D4728: Random Vibration testing of shipping container
ISO 16750-3: Automotive electronics
EIA-RS-186: Passive electronic component parts- Method 8- Vibration, High Frequency
And more…

Alternatively, for some products a customized profile based on specific customer usage may be desired. In this case, a test tailoring or Mission Synthesis approach would be helpful.

More information in the following knowledge articles:

3.3 DOFs: Averages per Loop and Weighting

In addition to the Minimum and Maximum frequency, the upper right area of the “Random Setup” worksheet contains an important setting called “Degrees of Freedom”. The higher the Degrees of Freedom (DOF) setting, the narrower the amplitude variation of control Power Spectral Density as shown in Figure 11.

Figure 11: Top – Control Power Spectral Density (PSD) with low number of DOF (blue), medium number of DOF (magenta), and high number of DOF (black).

While there is less average amplitude variation with a higher DOF number, the trade-off is that the control is less responsive to changes that happen during the test. Some specifications, like MIL-STD 810, recommend a minimum of 120 DOF.

The “Degrees of Freedom (DOF)” is based on “Averages per Loop” and “Exponential Weighting Factor”. “Averages per Loop” and “Exponential Weighting Factor” are set as shown in Figure 12:

Figure 12: The “Degrees of Freedom” setting is calculated from “Averages per Loop” and “Exponential Weighting Factor” found under the “Advanced…” button.

The equation that relates the Degree of Freedom (DOF) parameter to the “Averages per Loop” (M) and “Exponential Weighting Factor” (W) is given in Equation 1.

Equation 1: Degrees of Freedom (DOF) as a function of averages per loop (M) and the weighting factor (W).

It is also possible to enter the desired DOF. The weighting and averages per loop will selected by Simcenter Testlab accordingly to achieve the desired DOF.

For more information, see the knowledge article: Random Control: Averages per Loop, Frequency Resolution, Weighting, and Degrees of Freedom

3.4 Sigma Limiting

The vibration signal generated by the Simcenter SCADAS hardware is true random. This means that in theory, a data sample at some point in time could approach infinity. A sudden large amplitude could be a problem – it might exceed the capability of the shaker or amplifier for example, a vacuum tube based amplifier in old days were very sensitive to this). More practically, it might cause the test to abort as it exceeds a safety limit.

The “sigma limiting” and “sigma iteration” settings prevent these large amplitude samples from occurring. They are set using the “Advanced…” button in the upper right of the Random Setup worksheet as shown in Figure 13 below:

Figure 13: Sigma limiting and sigma iteration settings.

The settings do the following:

Sigma Limiting: The signal peak limit in standard deviations (according to equation below) from the signal RMS.
Sigma Iteration: The amount of smoothing applied (higher means more smoothing) applied to a sample that exceed the sigma limit. Avoids clipping a signal.

The sigma limiting is set by the Equation 2 below:

Equation 2: Sigma Limiting

If sigma limiting was set to 5 and the random vibration signal was 5 g rms, this would limit the highest peak in the control signal to 25 g (5 sigma times 5 g rms). Note that these peaks are only evident in the time domain signal. Because sigma clipping only affects the peaks in the time domain while maintaining the overall desired g RMS value of the target profile, it will not have an effect on the shape or amplitude of the PSDs in the frequency domain. Examples of the effect of sigma clipping on the time domain control signal are shown in Figure 14 below.

Figure 14: Time domain control signal with sigma clipping of 4 (blue) versus sigma clipping of 2 (magenta).

Sigma limiting will clip peaks in the time signal that exceed the max peak value. This results in those peaks having flat spots in the time domain, much like a square wave. This can affect the frequency content of the calculated drive signal. To compensate for this, sigma iterations are performed to smooth the peaks that are clipped in the time domain while maintaining the overall g RMS and frequency content of the shaker drive signal. For example, if a very sensitive transducer is used, then to reduce the effect of the maximum output amplitude being exceeded the sigma iteration specifies that the process be performed a number of times in an iterative loop. To have the best results, it is recommended to define a maximum of 5 sigma iterations. More than 5 iterations are usually not necessary, since it will only marginally improve the results, at the cost of more calculations.

3.5 Schedule

On the middle of the right-hand side of the Random Setup worksheet, the test schedule can be entered (Figure 15).

Figure 15: Test Schedule.

The schedule area defines:

Startup Level: Typically in the range of -12 to -9 dB below full-level.
Number of Steps: The number of steps before full level. In Figure 15, with 3 steps, the test will ramp up from -9 dB to -6 dB to -3 dB before ramping to 0 dB full-level.
Equalization Time: This is the amount of time Testlab will spend at each lower-level step.
Test Duration: The time spent at 0 dB full level. In Figure 15, this is 1 minute 30 seconds. The total test time will be 2 minutes, 10 seconds at each of the 3 lower-level plus 1 minute 30 seconds at 0 dB.

3.6 Kurtosis

Kurtosis control is an option that is sometimes used to increase the severity of the test. This is done by increasing the number of peaks in the time history while keeping the overall g rms of the random vibration signals the same.

When using Simcenter Testlab Random Control, the generated vibration signal has a gaussian random amplitude distribution (red curve) by default as shown in Figure 16 below:

Figure 16: Gaussian random vibration signal (red) has no spike events (kurtosis of 3) compared to vibration signal with kurtosis value of 6 (green).

The vibration experienced by an object in the real world does not always have a gaussian random amplitude distribution. Often the real-life vibration might have some “spikey” behavior (Figure 16 above, green curve).

A signal that has a gaussian random distribution of amplitude has a kurtosis value equal to 3 in the Simcenter Testlab Random control application. The higher the kurtosis value, the more “spikey” the signal will become, while still maintaining the target PSD RMS level and frequency content.

If desired, the random signal normally generated by Simcenter Testlab Random Control can be augmented with spikey behavior using the “Kurtosis Control” add-in located under “Tools -> Add-ins” as shown in Figure 17:

Figure 17: Tools -> Add-ins -> Kurtosis control

After turning on the Kurtosis control add-in, a new tab called “Kurtosis” is available under the advanced settings as shown in Figure 18.

Figure 18: After turning on the “Kurtosis Control” add-in, a new tab appears under the “Advanced...” button. Turning on “Use Kurtosis control” adds a new field called “Desired Kurtosis value” to the main menu of Random Setup.

Turning on the checkbox on “Use Kurtosis control” in the Kurtosis tab adds a “Desired Kurtosis value” field in the Control Pane in the Random Setup worksheet. The available values vary between 3 and 12. Default value is 3 corresponding to a Gaussian distribution signal generation.

By enabling the Kurtosis control, the sigma limiting and the sigma iteration parameters are no longer used to calculate the drives. Instead, the sigma clipping value can be edited in the Kurtosis tab. By default, it is set to 20 (peaks higher than 20 times the RMS value are clipped). When the Kurtosis Control method is used, limiting (both RMS and spectral), defined in the Channel Parameters pane of the setup sheet, is not available either

Note: When performing analysis on throughput data with throughput processing in Simcenter Testlab, the values for kurtosis are excess kurtosis values, which is the kurtosis value minus 3. This leads to a Gaussian Random signal having an excess kurtosis value of zero. More background on Kurtosis and Kurtosis Excess in the knowledge article: Kurtosis.

4. System Identification and System Verification

In a typical random control test, the complete vibration control system is usually “checked over” by using the System Identification and System Verification (previously called SelfCheck prior to Testlab version 2206). Once the system check is performed, then the test can be run.

Go to the System Identification worksheet. The System Identification worksheet is used to make sure the test setup is controllable and that the shaker system can perform the test.

Press the START button (middle, right side of screen) as shown in Figure 19:

Figure 19: System Identification.

During the system identification:

A low-level broadband random signal is sent to the shaker system. This signal will be increased in stages until the programmed signal-to-noise ratio (S/N ratio) is met or exceeded.
Once the S/N ratio condition is satisfied, the programmed number of acquisition blocks are measured, and results such as PSDs, coherence, and system FRFs are calculated and averaged together (a default of 5 blocks are averaged together, but this is adjustable).
Once averaging is completed, the excitation signal is ramped down and Testlab calculates linear, forward predictions of each sensor location and the drive signal at 0 dB.

The Simcenter Testlab Vibration Control software will then use this information to determine if the random control test can be run successfully given the parameters specified in the System Identification/Verification settings. If all checks are passed, the software will allow the full level test to be run.

Some possible error or concerns that might be flagged:

Poor Coherence between Drive Output and Control accelerometer: The accelerometer may not be securely fastened, in a poor location for control, the test article may not be sufficiently secured, or other issue. Coherence varies between 0 and 1. A value close to 1 at all frequencies indicates that drive and control accelerometer have a fixed, linear relationship. This means that the test can be well controlled. Dips in coherence (<1) are indications of non-linearities and possible indications of difficult-to-control frequency bands. While this is sometimes unavoidable due to anti-resonances, poor coherence must be scrutinized before proceeding with the test.
Channel Overloads: At the full level of the test, a measurement or control accelerometer may overload. A less sensitive accelerometer may be needed (ie, swap a 100 mV/g accelerometer for a 10 mV/g accelerometer), or channel ADC ranging may need to be increased to a higher voltage level.
Open Channel: Accelerometer is not responding sufficiently above the noise floor of the sensor. This may be caused by a broken or loose cable, bad or loose sensor, or other issue.

If desired, the functions calculated during the system identification can be reviewed further in the System Verification workbook (Figure 20):

Figure 20: System Verification workbook.

Functions that can be displayed include coherence, transfer functions, responses, etc. This can be helpful especially when trying multiple reference random control.

For a complete description of the system check or self check:

Vibration Control: Understanding SelfCheck (Version 2021 and lower)
Vibration Control: System Identification and Verification (Version 2206 and higher)

If all checks are passed, go to the Random Control worksheet to start the test.

5. Random Control Worksheet

The Random Control Worksheet is used to initiate a test run, monitor it and take action outside of the programmed test schedule, such as taking manual measurements, increasing or decreasing the test level, or holding the test at the current level for an additional period of time.

The Random Control worksheet is shown in Figure 21 below.

Figure 21: Random Control Worksheet with main areas labeled.

As noted above, before entering this worksheet a selfcheck procedure should have been satisfactorily completed.

Each section of the Random Control worksheet is explained in further details below:

5.1 Random Control Pane Header Options

In the header at the top right of the screen (Figure 22) are control pane options that can be turned on and off:

Figure 22: Random Control Pane Header Options.

Available options are:

Show/Hide Control Displays: This button will show or the hide the control display panel.
Show/Hide user defined layout: This button will show or hide the user defined layout panel. In the user defined layout panel functions just like in the Navigator worksheet.
Show/Hide responses viewer: This button will show or hide the response viewer.

5.2 Random Display Area

Displays take up the majority of the screen (starting on left side):

Control Signal Display: This window shows the reference profile and the observed average control signal. The “Show first drive” checkbox allows for the visualization the spectrum of the drive which will be sent out when starting the test. Note: when Limiting is selected in the Random Setup worksheet, a display window showing the defined limit profile and the response signals will appear in the Control Signals display.
User Defined Layout: This window functions identically to the Navigator worksheet.

Both the Control Signals Display and the User Defined Layout allow for additional signals to be displayed. While the test is running, other live or previously recorded measurement channels can be displayed besides the control channel as shown in Figure 23.

Figure 23: Use the “Data Explorer” and online data folder to see other measurement channels while test is in progress.

Click on the Data Explorer icon in the top bar (or choose Data -> Data Explorer from main menu). Go to the Online data folder into the “Current Run/Measured PSD” folder. Drag the Power Spectral Density (PSD) measurements into the display to view them while the test is in progress.

The rest of the display area includes:

Responses Display Panel: This panel offers a means to scroll through the response data. Much less flexible than the user defined layout, it has the advantage that it is preconfigured, offers Channel overview information and automatically adapts some display properties to the displayed signal.
Channel Overview: Checking this "on" brings up an overview display (below the responses display) of the levels on all channels. What is shown in these bar displays will depend on the current function on display. A tool tip display with the channel number, point id, direction and function will appear by hovering with the mouse over a particular bar display. Unselecting the "Channel overview" option can slightly increase the screen refresh rate for systems with a very large number of channels.
Bar Displays: The area below the Responses display contains a set of bar displays that show the input level of each channel in terms of the full scale value of the currently selected function.
Function Selection: This dropdown allows the selection of which function to display in the Response display windows. The offered selection is dependent on the workbook and the activated Online Data functions.
Display Windows: The display windows show the selected function on the selected range of the active channels. Some display properties are set based on the selected function and possible user interactions will be limited.

5.3 Random Control Panel

The Random Control Panel (right side of Random Control screen) contains information and action buttons important to the execution of the programmed test. The Control Panel may have more or less of the following panels depending on what features are enabled.

The default Control Panel versus the Control Panel with full features enabled is shown in Figure 24.

Figure 24: Default Control Panel (left) vs. Control Panel view with Limiting and Throughput activated (right).

The panel displays key information about the test in progress (duration, level, etc). The measurement run name can also be set in addition to starting or ending the test.

5.3.1 Run Name Panel

The run name is shown in the upper right (Figure 25). This name is given to the measurement to be performed:

Figure 25: Run name in upper right.

Run name: When the Random Control worksheet is entered, a new run name (e.g. Random_X) is automatically propagated, and the system needs to be armed to enable test start functionality. To change the name of the run, delete the current name and enter a new one before arming the system.
Arm button: This button initializes the system using the current settings. It is possible to change some parameter settings after “Arming” and before starting the next run, but these changes will not be used as settings for this run. They will only be valid for the run that follows the next “Arming” of the system. This button has a green background when it is sensitive and active. During arming and test run, the background turns grey and becomes insensitive and inactive.

Recommended best practice to avoid confusion is to avoid changing settings after the system is armed and before the test is stopped.

5.3.2 Status Panel

The status panel provides information about the progress of the current run (Figure 26).

Figure 26: Random Control Status Panel.

The status panel contain the following information:

Status: This field displays the current state of the run. When the frontend is armed it will indicate that the system is Ready. Once the run is underway, it will display a series of messages indicating the current process or which action button has been hit.
Drive RMS: This indicates the RMS level of the SCADAS output in Volts.
OpenLoop: This field is grey when the test is in closed loop mode and highlighted yellow when the controller is set manually to open loop mode.
Level: This shows the current level in dB as defined in the schedule or activated via the action buttons.
Hold Level: This field is grey when the test is leveling automatically and highlighted in yellow when the manual HoldLevel is activated in the test schedule or via the action buttons.
RMS: The first field shows the RMS level of the measured control PSD (averaged depending on the control mode defined) on the left-hand side and the reference RMS level for that level on the right-hand side.
Control: This field displays the Control method. It shows "Average Control" when no Limiting is currently active on any channel. It shows "Spec. limit: {pointname}" when an algorithm is limiting the spectral response of any channel, as defined in the Limit profile editor. If the software is limiting the response of several channels, the channel creating the most restrictive (i.e. deepest) notch will be displayed.
RMS lim: The RMS level can be limited on any channel. This is done by a global scaling down of the drive. In the RMS lim field, this scaling factor is displayed in dB. If RMS limiting is not active, 0.0 dB will be displayed (scale factor of 1). If RMS limiting is active, the scaling factor and the {pointname} will be displayed in red.
Time: The time that has elapsed at the current level in hh:mm:ss format is displayed in the field. The progress indicator below shows the elapsed time proportionally to the total level time as defined in the schedule.

5.3.3 The Alarm/Abort Indicator Panel

Alarms and aborts are shown in the lower right of the Random Control worksheet (Figure 27):

Figure 27: Alarm and Abort Panel.

The number of points/lines that exceed their limits in proportion to the maximum number allowed are shown in several bar indicators in this area.

Alarm lines: Shows the number of frequency lines that exceed the alarm levels.
Repeats: Shows the number of frequency lines that are repetitively in abort every time the control PSD is updated.
Abort lines: Shows the number of frequency lines that exceed the abort levels.
Averages: Shows the number of averages taken as defined in the 'automatic measurements' parameter of the Random setup.

5.3.4 Recording Panel

This Recording panel appears when the “Time Data Recording During Random Control” add-in is loaded (Figure 28).

Figure 28: Recording Panel

This panel displays the actual state of the throughput recording during the test.

Status: This field displays the current status of the throughput recording. Possible Status states are:
Throughput not available: The “Time Recording During Random Control” add-in must be activated using the Tools Add-ins menu.
Throughput deactivated: The corresponding setting is switched off in the setup worksheet.
Waiting for start: Time recording is not yet running; it will start together with the start of the test.
Active: Time recording is running.
Stopped: Time recording is stopped.
Time: The time field shows the elapsed time of throughput recording measurement for the current test. An estimate of how much recording time is available based on the current amount of free disk space is also provided.

6. Runtime Procedure: Random Control

To run the test, go to the “Random Control” worksheet. Press "Arm" in the upper right corner, and then press "Start" in the middle right (Figure 29).

Figure 29: Random control worksheet.

After starting the test, a control PSD will be displayed against abort and alarm limits.

In the lower right, the test will start at a lower level (usually -9 dB from full test level) and will step up to full level (0 dB). The test will run the specified test duration once full level is reached.

7. Post Test: Batch Printing

After the test is finished, a report can be made quickly using the Batch Reporting worksheet (Figure 30):

Figure 30: Batch Reporting worksheet.

After highlighting the test, click on the Print button to make the report. Use “File -> Print Options” to select the desired printer or a Powerpoint/Word file for the report.

There are two ways to change the Siemens logo to a different logo in Batch Printing:

Edit the files directly in C:\Program Files (x86)\Simcenter\Testlab {revision}\central\Application Resources. This will change the logo for anyone using the computer.
Copy them to your local Application Resources directory (C:\Simcenter \UserConfiguration\{user}\Testlab {revision}\Application Resources) and edit the logo file. This will change the logo for the user login only.

Logo.bmp is the Siemens logo in the top right corner. LmsHeading1.bmp is the Header you see in the top left corner in some versions of Simcenter Testlab.

Questions? Email chris.sensor@siemens.com.

Related Links:

KB Article ID# KB000123093_EN_US

Summary Details

Associated Components

SCADAS Durability SCADAS III SCADAS Lab SCADAS Mobile SCADAS PBN SCADAS RS Configuration App SCADAS RS Hardware SCADAS RS Recorder App SCADAS Recorder SCADAS XS