9.1 Plot types

Data plotting is one of the core functions of SOURCE, present in the Operational Data, FRF Data, and Analyze modules. The plotting functionality is set up to allow you to review large datasets very efficiently.

A typical plot from the Analyze module is shown. The specific plot settings vary slightly between the different modules but are generally very similar. In this section, we present some of the typical settings and commands that are useful for working with plots.

SOURCE currently offers the following plot types, which can be created by right-clicking on the dataset and choose any of the following options (if available). Alternatively, you can drag a dataset onto the graph canvas in order to create the default plot, which is either a time or spectrum plot depending on the dataset type.

Time plot

This is only available for time data. A context menu is available on right-click, allowing to style the curve or change the curve to become an area type.

Spectrum plot

This type provides all standard spectral plots, including also sum level and n^th-octave plots. Options regarding amplitude scaling, spectral density, acoustic weighting etc. can be controlled through the axes options control on the top-left.

In case you have selected a time series or time blocks dataset, the Fourier transform is performed automatically. The settings that determine how the spectra is computed, i.e. the block length, overlap and window type, can be set in the Preferences in the home screen.

Octave, sum and other spectral plot types

You can make multiple spectral plot types, such as n^th-octave, sum, real, imaginary, etc.

To make such plots, complete the next steps:

1. Make a spectrum plot.
2. Right-click on the curve to visualize the curve pop-up.
3. Click to expand the pop-up.
4. Select the spectral plot type from the dropdown.
5. If necessary, type the minimum and maximum frequency values.

Order plot

Order plots are a specific type of spectrum plots, in which the complex-valued spectra is expressed against an order axis rather than a frequency axis. The order axis, together with the RPM channel, gives access to the instantaneous frequencies for a given speed. This type of data visualization is useful when working with components that have rotating parts, such as e-motors. In addition to the plotting, order spectra can also be used to calculate order-based blocked forces and TPA syntheses.

Order plots and order analyses can only be done if you import order spectral data in the ATFX format. When such operational data is used in an analysis with an FRF matrix, SOURCE interpolates the FRF matrices to match the order frequencies of the operational dataset. This way, you can easily create order blocked forces as well as order TPA syntheses, which can be plotted in Analyze or exported to ATFX for further analysis.

Order cut plot

SOURCE accepts order cuts in ATFX formats from Müller-BBM PAK or Siemens TestLab. Order cuts can be used in analyses to calculate order-based Blocked Forces and TPA syntheses.

Order cuts can be exported to ATFX only.

1. Order cut magnitudes can be plotted over RPM, time, or derived frequency. You can also create strip plots in which the channels are shown in the vertical axis.
2. The derived frequency is computed as $$ f(t) = \mathrm{ord} * \mathrm{rpm} (t) / 60 $$.

Waterfall plot

Waterfall plots are functionally similar to the spectrum plot. They are by default displayed with the columns shown on the vertical axis. By right-clicking, you can change the orientation between horizontal and vertical.

In the right-click menu, you also find an option to change the Z-axis. This determines the value that gets shown on this vertical axis, and indeed allows to select something other than time or index. This then becomes the tracking value and the data will be displayed and scaled according to the values on this axis. The resulting diagram is often referred to as a Campbell diagram.

Whether such tracking values are available depends on the origin of the input dataset. If the spectra are computed by SOURCE and you select the dataset in Analyze, you will find every channel that was marked as tracking channel in the channels card.

Strip plot

To show data in strips, choose the strip plot option. You can show multiple channels for one dataset, a channel for multiple datasets or even a combination of them. New datasets can be added by dropping them on an existing strip plot graph.

Sum level plot

The sum level plot shows sum levels over time. This operates on both time and spectral data. The sum levels are calculated according to the spectral settings in the top left of the axes.

Operational consistency plot

You can use operational consistency to do quick quality checks on your operational data. With the operational consistency, you can detect overloads, non-linear phenomena, moved or loose sensors and inconsistent sensor movements. This is done by evaluating the consistency of the response data through the Virtual Point Transformation.

To plot the operational consistency, complete the following steps:

1. Map the channels of the operational data to the master channels
2. Switch to Edit in the master channels
3. Check the Virtual Point groups of the master channels
4. Plot the operational consistency

FRF plot

In the FRF Data module, the Graphs card will display the FRF (including magnitude, phase and coherence) for whichever entry is selected in the Matrix Viewer, as shown in the figure. For VP-transformed data ($$ \mathbf{Y_{um}} $$, $$ \mathbf{Y_{qm}} $$), the coherence plot will be empty, as coherence is only valid for raw FRF measurements.

To plot an FRF, complete the following steps:

1. Go to the FRF data module.
2. Select a dataset to plot.
3. Right-click and select Plot.
4. In the Matrix viewer, select one entry.

CMIF plot

In SOURCE, it is also possible to show the singular value breakdown, often referred to as Complex Mode Indicator Function (CMIF) of any subset of the FRF matrix. Any time you try to plot more than one row along with more than one column of the FRF matrix, the CMIF will automatically be displayed, rather than the individual FRFs. In this mode, the CMIF is displayed in the upper subplot, and the condition number is displayed in the lower subplot, as shown in the figure.

The number of curves shown in the CMIF will be the minimum amount of the number of rows or columns. Reviewing the CMIF is useful for setting the matrix inverse settings in the Analyze module and can be plotted in two ways:

1. Select the desired entries for the CMIF directly in the Matrix Viewer. Create a box around the FRF data of interest by clicking on one entry, holding SHIFT, and selecting another entry. The two entries should be selected to form the opposite corners of the subset of interest.
2. In the plot in the Graphs card, select the drop-down menus in the title and select multiple channels and reference channels by holding SHIFT or CTRL while clicking.