Simcenter Testing Solutions Simcenter Testlab Neo: Calculating Signal Rise Rate
2021-11-26T19:00:15.000-0500
Simcenter Testlab
Summary
Details
Direct Youtube link: https://youtu.be/pLbuYC4cI3c
A measured signal can start in a very abrupt manner. For example, a vehicle chime sound might activate so fast that it startles a listener. Or the smooth combustion of an engine might have a metallic clanging that is less than desirable.
In this article, an analysis method will be outlined for determining the initial “rise rate” of a sound. The methods and steps to perform this analysis in Simcenter Testlab Neo: Process Designer are also described.
Index:
1. Overall Process
2. Envelope
3. Differentiate
4. Statistics
5. Example 1 – Rise Rate
6. Example 2 - Clicks
1. Overall Process
The process involves creating an envelope of the signal. The envelope is then differentiated, and the peak value of the differentiated data is extracted.
A process from Simcenter Testlab Neo Process Designer is shown in Figure 1.
Figure 1: Process for determining the starting abruptness, or rise rate, of a signal.
Filtering is an optional step that can be used to smooth the results.
2. Envelope
The envelope is a smooth curve outlining the extremes of an oscillating signal as shown in Figure 2.
Figure 2: The envelope (red) of an oscillating signal (green).
Even a signal that “starts up” still oscillates. The envelope can be useful to understand the overall start up as shown in Figure 3.
Figure 3: The envelope (red, top) of a signal that is starting up from zero (green, bottom).
The envelope has a clear upward slope as it goes from zero to the steady state level.
The envelope method is made available in Simcenter Testlab Neo by turning on the “Interactive Analysis” add-in (File -> Add-ins). If using token licensing, 16 tokens will be occupied.
In the method, a low-pass filter can be specified as shown in Figure 4.
Figure 4: Parameters that can be set for the envelope method, including a low-pass filter.
The low pass filter ensures a nice smooth curve that is less affected by the oscillating components of the signal as shown in Figure 5.
Figure 5: Envelope of same signal with a 2 Hertz low pass filter (black) and no filter (gray).
Without the filter, the maximum rise rate would be difficult to determine. The unfiltered envelope (gray) has much variation while the filtered envelope (black) follows the overall trend of the signal.
For sounds, using Time Varying Loudness in place of the envelope is also useful. More about loudness in the knowledge article: Sound Quality Metrics: Loudness and Sones.
3. Differentiate
By differentiating the envelope versus time, the maximum rate of change can be determined, as shown in Figure 6.
For sounds, using Time Varying Loudness in place of the envelope is also useful. More about loudness in the knowledge article: Sound Quality Metrics: Loudness and Sones.
3. Differentiate
By differentiating the envelope versus time, the maximum rate of change can be determined, as shown in Figure 6.
Figure 6: The differentiated signal (blue, top) of the envelope (red, bottom) determines the maximum rate of change.
Because the envelope is the overall level of the signal (without the influence of individual oscillations) the result is the overall rise rate or ramp rate of the signal.
The differentiate method is also part of the Interactive Analysis add-in. There are no parameters available to set in this method.
4. Statistics
The Statistics method is used to isolate a single number from the time data traces.
To get the maximum rise rate of the signal, “Maximum” should be checked on in the method (Figure 7).
The differentiate method is also part of the Interactive Analysis add-in. There are no parameters available to set in this method.
4. Statistics
The Statistics method is used to isolate a single number from the time data traces.
To get the maximum rise rate of the signal, “Maximum” should be checked on in the method (Figure 7).
Figure 7: Maximum select in Statistics method.
The resulting output is displayed as single numbers as shown in Figure 8:
Figure 8: Output of Statistics method is single values.
These numbers can be copied into Excel by right clicking and selecting “Copy Values” under the Table section as shown in Figure 9.
Figure 9: Copy data to Excel with right click and select “Copy Values” under Table.
After selecting "Copy Values", open Excel and choose "Paste".
5. Example 1 - Rise Rate
The signals shown in Figure 10 have three distinct rise rates.
Figure 10: Signals with three different rise rates – Slow (red, top), Medium (green, middle), and Fast (blue, bottom).
The respective envelopes of the three signals are shown in Figure 11. They were calculated with a 2 Hz low pass filter.
Figure 11: Envelope of the three signals with three different rise rates – Slow (red, top), Medium (green, middle), and Fast (blue, bottom).
Next the envelope functions are differentiated (Figure 12):
Figure 12: Differentiated envelope of the three signals with three different rise rates – Slow (red, top), Medium (green, middle), and Fast (blue, bottom).
Then the maximum values can be extracted from the differentiated envelopes. Results are shown in Figure 13.
Figure 13: Differentiated envelope of the three signals with three different rise rates – Slow (red, top), Medium (green, middle), and Fast (blue, bottom).
The maximum value is higher for the fast rise rate signal, and lowest for the slow rise rate signal.
5. Example 2 - Clicks
Two clicking sounds are shown in Figure 14.
5. Example 2 - Clicks
Two clicking sounds are shown in Figure 14.
Figure 14: A longer, subtle click sound (top, red) versus a short, distinct click sound (bottom, green). The bottom trace has a higher rise rate than the top trace.
One of the clicks is more subtle (longer duration, less distinct high frequency) than the other.
Because these signals have no oscillating component, the signals can be differentiated directly. The envelope calculation is by-passed as shown in Figure 15.
Because these signals have no oscillating component, the signals can be differentiated directly. The envelope calculation is by-passed as shown in Figure 15.
Figure 15: Click on the left side of the method to bypass it.
The differentiated signal results are shown in Figure 16.
Figure 16: Differentiated results of the two traces show that the rise rate of the short click is higher than the long click.
One of the clicks is more subtle (longer duration, less distinct high frequency) than the other. This is confirmed by looking at the rise rate of the signal. The long click has a lower rise rate while the short click has a higher rise rate.
Hope this is helpful! Questions? Email peter.schaldenbrand@siemens.com.
Related Links:
- Index of Simcenter Testing Knowledge Articles
- Simcenter Testlab Neo Introduction
- Simcenter Testlab Neo: Process Designer
- Simcenter Testlab Neo: Modulation Metrics
- Simcenter Testlab Neo: Audio Replay
- Kurtosis
- Sound Modulation Metrics: Fluctuation Strength and Roughness
- Sound Quality Metrics: Loudness and Sones
- Sharpness in Simcenter Testlab
- Tonality
- Tone-to-Noise Ratio and Prominence Ratio
- Articulation Index
- Sound Metrics: Speech Interference Level
- Describing Sounds with Words
- Sound Metrics Unleashed Seminar
Associated Components
Simcenter Testlab Digital Image Correlation
Testlab Environmental
Testlab Acoustics
Testlab Data Management
Testlab Desktop
Testlab Durability
Testlab General Acquisition
Testlab General Processing & Reporting
Testlab Rotating Machinery & Engine
Testlab Sound Designer
Testlab Structural Dynamics
Testlab Turbine