AmesimKnowledge

Order tracking

Creating an order tracking curve

The order tracking menu item is available only on a spectral map, by a mouse right click on the spectral map area.

Figure 32: Order tracking menu item

The starting point of this example is the spectral map generated in Creating a spectral map.

Click on the spectral map with the mouse right button and select the Order tracking item in the menu. The following dialog box appears:

Figure 33: Set up order tracking parameters

Order(s): order that must be extracted from the spectral map. Note that the order can be a floating number.

Bandwidth(s) Bandwidth : specifies the width of the interval used to compute the frequency RMS of each FFT (see Bandwidth)

Display in a new plot: by default, the order curve appears in a graph below the spectral map. Selecting this box will create the order curve in a new plot.

Force the reference variable on X axis Force the reference variable on X axis : in case of non-monotonic RPM, the spectral map can use time to scale the Y axis (see Example with non-monotonic reference velocity). In such cases, the order curve can recover the reference variable RPM and use it as the X axis variable. If the check box is not selected, time will be used instead.

We enter orders 1 to 4 in the Order(s) field with the bandwidth set to 10 Hz:

Note

You can enter multiple orders or bandwidths in the Order Analysis dialog box.

You can separate multiple orders or bandwidths with spaces, commas or semicolons.

The result is shown in the following figure:

Figure 34: Orders 1 to 4 of the analyzed signal

The curve clearly shows the resonance peaks, appearing at 3000 RPM for order 1, 1500 RPM for order 2, 1000 RPM for order 3 and 750 RPM for order 4.

The amplitude of the resonance peaks corresponds to the settings of the model: orders 1 and 3 had a gain k=1, whereas order 2 had a gain of 0.5 and order 4 a gain of 1.5.

The comparison of the order curves with the original signal is performed using a post processed variable. As shown by the figure below, the output from second order lag is dragged and dropped to the Post processing tab, as A1. A second item A2 is created, and set with the expression A1-100, to get rid of the constant term in the analyzed signal.

A2 is plotted on a graph also containing the 4 order curves.

Figure 35: Comparison of order curves and the original signal

Note that the order 1 (red curve) perfectly matches the amplitude of the original signal, whereas other order curves seem be too low. This comes from the fact that the signal is a bit more complex than a simple sine wave below 2000 RPM. The zoom at 1000 RPM shown below illustrates this:

Figure 36: Zoom of figure Figure 4.7

Source: https://docs.sw.siemens.com/en-US/doc/254352342/PL20250521841123434.amesim_collection.Spectral_Analysis/Creating_an_order_tracking_curve · retrieved 2026-07-17