Simcenter Testing Solutions Random Control: Measurement Channels
2023-02-28T02:47:38.000-0500
Simcenter SCADAS
Simcenter Testlab
Summary
Details
When discussing random shaker vibration, most of the focus is typically given to control channels, control strategies, notching, and other settings and features related to how Simcenter Testlab can make the shaker to perform the vibration test according to specifications. However, measurement channels, while often taken for granted, are no less important to the customer than the control channels.
Measurement channels (Figure 1) are used to acquire data on all other locations of a test object that are not being controlled. While the levels of the control channel are closely monitored and targeted, the measurement channel levels are not. It is useful to measure these locations to be able to compare them to field data.
Figure 1: Measurement accelerometers (blue) and control accelerometer (magenta) in a vibration control test.
This article will explain how measurement channels are implemented in Simcenter Testlab, where to find various settings, and other considerations.
Article Index
1. What is a Measurement Channel?
2. Channel Setup
3. Random Setup
4. Automatic Measurements
5. Automatic Measurements Advanced Settings
6. Scheduled Measurements
7. Manual Measurement
8. Measurement Channel Safety Settings
9. Other Notes on Measurement Settings
10. Throughput Recording
1. What is a Measurement Channel?
A measurement channel is a Simcenter SCADAS input channel that records data but is not required to reach the programmed target reference Power Spectral Density (PSD). These channels do not typically affect how the test runs, unless programmed to do so (See Measurement Channel Safety Settings below). Measurement channels are often used to capture the vibration levels on key components of the Device Under Test (DUT) to ensure that vibration levels during a shaker test adequately simulate field conditions.
Measurement channels can also be used during development testing to determine what vibration levels subcomponents might experience, or for curiosity’s sake because a bunch of accelerometers were available anyway. Whatever the reason, the important thing to remember is that measurement channels do not affect the control loop, with a few exceptions.
2. Channel setup
The first step to taking a measurement with a SCADAS input channel is to navigate to the Channel Setup Workbook in Testlab, as noted in Figure 2 below.
After all the other Random Setup Parameters are set, the settings that affect measurements can be manipulated. All settings that affect measurement channels are found in the Schedule, Automatic Measurements, and Throughput Recording sections, as indicated in Figure 3 above.
There are a few options for measurement channels:
4. Automatic Measurements
The “Automatic Measurements” section is used to set up the measurement schedule when the advanced schedule table is not being used (See Scheduled Measurements section below) and will save data only at 0 dB test levels.
Automatic Measurements is not enabled by default. To enable Automatic Measurements, click on the box next to “Measurements:” as indicated in Figure 4.
When enabled, the box will be checked and the “Initial Offset:,” “Period:,” and “Averages:” input boxes will become active. Testlab will wait for the “Initial Offset:” time to elapse before starting the first set of averages for the measurement. While Simcenter Testlab is averaging the number of frames programmed in the “Averages:” input box, the “Period:” countdown begins. Once the “Period:” time has elapsed, Simcenter Testlab will begin the next measurement.
This process will repeat until the test ends. A graphic of an example measurement schedule is in Figure 5.
Note that if the test ends before the indicated number of frames can be measured, Simcenter Testlab will store a partial measurement. All measurements taken at a particular time are placed in a folder that is labelled with the time at which the measurement is made.
5. Automatic Measurements Advanced Settings
The advanced settings in Automatic Measurements are accessed by clicking on the “Advanced” button as seen in Figure 6.
Figure 6: Advanced Automatic Measurements settings.
Advanced Automatic Measurements settings include three options for averaging as shown in Figure 7:
The three averaging methods:
Additional settings include:
Figure 8: Measure FRF options.
In the Measure FRF dropdown, the reference for the FRF calculation can be selected. FRFs can be calculated versus the Drive, versus the Average Control channel, or versus any SCADAS input channel that has been activated in the Channel Setup Worksheet. Only one reference can be selected.
6. Scheduled Measurements
The next option to cover is using the advanced time level table to schedule measurements. This allows the user to schedule measurements per level. This is different than using Automatic measurements, which applies the measurement schedule uniformly across all equalization levels and full 0 dB level. To enable scheduled measurements, click the box for “Use advanced time level table:” and the click on the “Advanced” button to access the advanced time level table (Figure 9).
In the Time Level Editor (Figure 10), each level can have measurements enabled independently, and each enabled measurement can have an Offset, Period, and Averages set up respective to that level’s measurement.
These options have the same definitions as explained above in Automatic Measurements. All measurements defined here use the same averaging type that is set in the Advanced Automatic Measurements settings.
7. Manual Measurement
During the test run, a manual measurement can always be started at any level. The manual measurement uses the settings in the Advanced Automatic Measurements dialogue as discussed above in the Automatic Measurements section. Remember that the number of averages for the manual measurement is set separately from other measurements as seen in Figure 10 above.
8. Measurement Channel Safety Settings
If desired, Simcenter Testlab can be set to either ignore or abort the test should a measurement channel overload. By default, Testlab is set to “Abort” should a measurement channel overload. This behavior can be changed in the Advanced settings box in the Safety tab (Figure 11). This is the only setting where a measurement channel influences the control loop, and the only effect is to trigger a test abort.
Figure 11: Measurement Channel Safety settings.
More information on channel overloads can be found here: Overloads.
9. Other Notes on Measurement Settings
All measurements are taken using a Hanning window, 0% overlap, and uses the same resolution, spectral lines, bandwidth etc. as the control settings. If any of these options need to be different than the control settings for a particular measurement channel, Throughput Recording can be used to capture the raw digitized time histories which can be processed using the desired parameters.
It is prudent to note that, in general, an ideal frequency resolution for most random tests is between 2 to 4 Hz resolution. This range tends to be the most ideal compromise between PSD resolution and control inner loop speed. While it may be tempting to set a finer frequency resolution to meet a measurement requirement in the test specification, setting the test frequency resolution too fine may result in overly long control loop times and unsafe testing conditions.
For more information on frequency resolution and the Random Test control loop see here: Random Control: Averages per Loop, Frequency Resolution, Weighting, and Degrees of Freedom
10. Throughput Recording
A final option to mention is using Throughput recording to record the raw digitized time signal. This can then be post processed using Siemens Testlab Signature Throughput Processing, for example. To enable Throughput recording, the “Time Recording during Random Control” Add-in must be enabled as seen in Figure 12. Then the “Activate recording:” must be checked.
Figure 12: Check the box next to “Time Recording during Random Control” to activate the Throughput Recording Add-in.
Advanced Throughput settings allow for an oversampling rate to be set above the sampling rate required for an alias-free test bandwidth (Figure 13).
Figure 13: The oversampling factor can be set in the Advanced Throughput Recording setup.
Additional information about the test can then be determined that would otherwise be unavailable, such as enabling the user to more accurately analyze any transients that occur or measure out-of-band energy.
Re-processing the time data can be done with Simcenter Testlab Signature Throughput Processing. More on Throughput Post-Processing can be found here: Simcenter Testlab Throughput Processing Tips
Questions? Email chris.sensor@siemens.com or contact Siemens Support Center.
Measurement channels (Figure 1) are used to acquire data on all other locations of a test object that are not being controlled. While the levels of the control channel are closely monitored and targeted, the measurement channel levels are not. It is useful to measure these locations to be able to compare them to field data.
Figure 1: Measurement accelerometers (blue) and control accelerometer (magenta) in a vibration control test.
This article will explain how measurement channels are implemented in Simcenter Testlab, where to find various settings, and other considerations.
Article Index
1. What is a Measurement Channel?
2. Channel Setup
3. Random Setup
4. Automatic Measurements
5. Automatic Measurements Advanced Settings
6. Scheduled Measurements
7. Manual Measurement
8. Measurement Channel Safety Settings
9. Other Notes on Measurement Settings
10. Throughput Recording
1. What is a Measurement Channel?
A measurement channel is a Simcenter SCADAS input channel that records data but is not required to reach the programmed target reference Power Spectral Density (PSD). These channels do not typically affect how the test runs, unless programmed to do so (See Measurement Channel Safety Settings below). Measurement channels are often used to capture the vibration levels on key components of the Device Under Test (DUT) to ensure that vibration levels during a shaker test adequately simulate field conditions.
Measurement channels can also be used during development testing to determine what vibration levels subcomponents might experience, or for curiosity’s sake because a bunch of accelerometers were available anyway. Whatever the reason, the important thing to remember is that measurement channels do not affect the control loop, with a few exceptions.
2. Channel setup
The first step to taking a measurement with a SCADAS input channel is to navigate to the Channel Setup Workbook in Testlab, as noted in Figure 2 below.
Figure 2: Channel Setup Workbook in Simcenter Testlab Random Control.
In the Channel Setup Workbook, select the entry in the “ChannelGroupId” column for the channel being set to a measurement channel. Then select “Measure” in the dropdown. Note that by default, all channels are initially set to “measure” type channels.
3. Random Setup
The next thing to do is set up the Random Vibration Test in the “Random Setup” Workbook as shown in Figure 3..
3. Random Setup
The next thing to do is set up the Random Vibration Test in the “Random Setup” Workbook as shown in Figure 3..
Figure 3: The Random Setup Workbook. All settings that affect Measurement type channels are found here.
After all the other Random Setup Parameters are set, the settings that affect measurements can be manipulated. All settings that affect measurement channels are found in the Schedule, Automatic Measurements, and Throughput Recording sections, as indicated in Figure 3 above.
There are a few options for measurement channels:
- Automatic Measurements.
- Scheduled measurements in the Advanced Time Level Table.
- Throughput Recording and post process the measurement channels.
- The manual “Measure” button during the test run.
More information on setting up a random vibration test can be found here:
- Getting Started with Random Control
- Random Control: Averages per Loop, Frequency Resolution, Weighting, and Degrees of Freedom.
4. Automatic Measurements
The “Automatic Measurements” section is used to set up the measurement schedule when the advanced schedule table is not being used (See Scheduled Measurements section below) and will save data only at 0 dB test levels.
Automatic Measurements is not enabled by default. To enable Automatic Measurements, click on the box next to “Measurements:” as indicated in Figure 4.
Figure 4: Automatic Measurements settings.
When enabled, the box will be checked and the “Initial Offset:,” “Period:,” and “Averages:” input boxes will become active. Testlab will wait for the “Initial Offset:” time to elapse before starting the first set of averages for the measurement. While Simcenter Testlab is averaging the number of frames programmed in the “Averages:” input box, the “Period:” countdown begins. Once the “Period:” time has elapsed, Simcenter Testlab will begin the next measurement.
This process will repeat until the test ends. A graphic of an example measurement schedule is in Figure 5.
Figure 5: Example timeline of a possible Automatic Measurements schedule.
Note that if the test ends before the indicated number of frames can be measured, Simcenter Testlab will store a partial measurement. All measurements taken at a particular time are placed in a folder that is labelled with the time at which the measurement is made.
5. Automatic Measurements Advanced Settings
The advanced settings in Automatic Measurements are accessed by clicking on the “Advanced” button as seen in Figure 6.
Figure 6: Advanced Automatic Measurements settings.
Advanced Automatic Measurements settings include three options for averaging as shown in Figure 7:
Figure 7: Three types of averaging are available: Stable, Exponential, and PeakHold.
The three averaging methods:
- Stable: This is a linear averaging method which averages the number of frames input into the “Averages:” box in the main Automatic Measurements input GUI.
- Exponential: This averaging type weights each frame based on the “Weighting:” factor in the “Weighting:” box. This averaging type is the most similar to the weighting used for the control channels. More information on exponential weighting can be found here: Random Control: Averages per Loop, Frequency Resolution, Weighting, and Degrees of Freedom
- Peak Hold: This averaging type retains only the peak values from each frame measured according to the “Averages:” box in the main Automatic Measurements input GUI. Each successive frame will overwrite any line values in the PSD that are higher than PSD of the previous frame.
More about the methods in the knowledge article: Averaging types: What’s the difference?.
Additional settings include:
- Weighting: This is the weighting factor used when the averaging type is set to exponential.
- Man. Meas. Nb. of Avgs.: This is an abbreviation for “Manual Measurement Number of Averages.” This is the number of averages that will be used when a manual measurement is taken. This number of averages overrides the value set in the Automatic Measurements panel.
The last option to point out is the ability to calculate and store Frequency Response Functions (FRFs). To calculate FRFs click the “Measure FRF:” box in the Automatic Measurements section as seen in Figure 8.
Figure 8: Measure FRF options.
In the Measure FRF dropdown, the reference for the FRF calculation can be selected. FRFs can be calculated versus the Drive, versus the Average Control channel, or versus any SCADAS input channel that has been activated in the Channel Setup Worksheet. Only one reference can be selected.
6. Scheduled Measurements
The next option to cover is using the advanced time level table to schedule measurements. This allows the user to schedule measurements per level. This is different than using Automatic measurements, which applies the measurement schedule uniformly across all equalization levels and full 0 dB level. To enable scheduled measurements, click the box for “Use advanced time level table:” and the click on the “Advanced” button to access the advanced time level table (Figure 9).
Figure 9: How to access scheduled measurements.
In the Time Level Editor (Figure 10), each level can have measurements enabled independently, and each enabled measurement can have an Offset, Period, and Averages set up respective to that level’s measurement.
Figure 10: Options for scheduled measurements.
These options have the same definitions as explained above in Automatic Measurements. All measurements defined here use the same averaging type that is set in the Advanced Automatic Measurements settings.
7. Manual Measurement
During the test run, a manual measurement can always be started at any level. The manual measurement uses the settings in the Advanced Automatic Measurements dialogue as discussed above in the Automatic Measurements section. Remember that the number of averages for the manual measurement is set separately from other measurements as seen in Figure 10 above.
8. Measurement Channel Safety Settings
If desired, Simcenter Testlab can be set to either ignore or abort the test should a measurement channel overload. By default, Testlab is set to “Abort” should a measurement channel overload. This behavior can be changed in the Advanced settings box in the Safety tab (Figure 11). This is the only setting where a measurement channel influences the control loop, and the only effect is to trigger a test abort.
Figure 11: Measurement Channel Safety settings.
More information on channel overloads can be found here: Overloads.
9. Other Notes on Measurement Settings
All measurements are taken using a Hanning window, 0% overlap, and uses the same resolution, spectral lines, bandwidth etc. as the control settings. If any of these options need to be different than the control settings for a particular measurement channel, Throughput Recording can be used to capture the raw digitized time histories which can be processed using the desired parameters.
It is prudent to note that, in general, an ideal frequency resolution for most random tests is between 2 to 4 Hz resolution. This range tends to be the most ideal compromise between PSD resolution and control inner loop speed. While it may be tempting to set a finer frequency resolution to meet a measurement requirement in the test specification, setting the test frequency resolution too fine may result in overly long control loop times and unsafe testing conditions.
For more information on frequency resolution and the Random Test control loop see here: Random Control: Averages per Loop, Frequency Resolution, Weighting, and Degrees of Freedom
10. Throughput Recording
A final option to mention is using Throughput recording to record the raw digitized time signal. This can then be post processed using Siemens Testlab Signature Throughput Processing, for example. To enable Throughput recording, the “Time Recording during Random Control” Add-in must be enabled as seen in Figure 12. Then the “Activate recording:” must be checked.
Figure 12: Check the box next to “Time Recording during Random Control” to activate the Throughput Recording Add-in.
Advanced Throughput settings allow for an oversampling rate to be set above the sampling rate required for an alias-free test bandwidth (Figure 13).
Figure 13: The oversampling factor can be set in the Advanced Throughput Recording setup.
Additional information about the test can then be determined that would otherwise be unavailable, such as enabling the user to more accurately analyze any transients that occur or measure out-of-band energy.
Re-processing the time data can be done with Simcenter Testlab Signature Throughput Processing. More on Throughput Post-Processing can be found here: Simcenter Testlab Throughput Processing Tips
Questions? Email chris.sensor@siemens.com or contact Siemens Support Center.
Vibration Control Related Links
- Index of Testing Knowledge Articles
- What is Vibration Control Testing?
- Vibration Control FAQs
- Vibration Control: Understanding Selfcheck
- Vibration Control: System Identification and Verification
- Getting Started with Random Control
- Random Control: Averages per Loop, Frequency Resolution, Weighting, and Degrees of Freedom
- Combined Modes: Sine on Random (SoR), Random on Random (RoR), Sine on Random on Random (SoRoR)
- Kurtosis
- Power Spectral Density
- Breakpoint Table Generator
- Fatigue Damage Spectrum
- 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
- What is Total Harmonic Distortion (THD)?
- 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
- Stress and Strain
- Calculating Damage with Miner's Rule
- What is a SN-Curve?
- Rainflow Counting
- Some Thoughts on Accelerated Durability Testing
- Mission Synthesis Seminar
- Vibration Control YouTube Playlist