Order Cut

An order cut extracts the amplitude of a single order (or a narrow band around an order) as a function of rotational speed. This produces a 2D curve showing how the vibration or noise level at a specific rotational harmonic evolves during a run-up or coast-down, making it the primary tool for identifying critical speeds and evaluating the severity of individual excitation mechanisms.

Concept

While the Campbell Diagram provides a complete overview of all orders simultaneously, the order cut focuses on a single order, extracting its level from the waterfall data. Mathematically, the order cut tracks the frequency:

f(RPM) = Order × RPM / 60

At each RPM step, the algorithm reads the amplitude of the spectral component at this speed-dependent frequency. The result is a curve of amplitude (dB or linear) versus RPM.

Bandwidth Consideration

Because spectral bins have a finite width (Δf), the order cut can either take the value of the single bin closest to the theoretical order frequency, or integrate over a narrow band centered on the order. The band integration approach is more robust when the order is not perfectly aligned with a spectral bin:

A_order(RPM) = √( Σ |X(f_k)|² ), for f_k ∈ [f_order − ΔB/2, f_order + ΔB/2]

where ΔB is the integration bandwidth, typically specified as a fraction of the order spacing.

Order Tracking Methods

Computed Order Tracking (COT)

COT is the standard method implemented in V-Listen. The time-domain signal is resampled to the angular domain using tachometer information, and the order spectrum is computed at each RPM step. The order cut is then extracted from these spectra. This method provides the cleanest separation between adjacent orders.

Vold-Kalman Filter

The Vold-Kalman (VK) order tracking filter is a time-domain filter that extracts individual orders directly from the time signal. It provides both amplitude and phase as functions of time (or RPM) and can resolve closely spaced orders that overlap in the frequency domain. The VK filter is particularly useful for operational deflection shape (ODS) analysis at specific orders.

The Vold-Kalman filter requires more computation time than COT but provides superior order separation, especially during rapid speed changes. Use it when standard COT shows order smearing or when phase information is needed.

Parameters

Parameter	Options	Description
Order(s)	Any positive number	Order number(s) to extract. Multiple orders can be tracked simultaneously.
Integration Bandwidth	0.1 – 2.0 orders	Width of the band centered on the order for amplitude integration.
RPM Range	Min – Max RPM	Speed range for the order cut.
RPM Resolution	1 – 100 RPM	RPM step between successive data points on the curve.
Tracking Method	COT / Vold-Kalman	Order tracking algorithm.
Amplitude Scale	dB / Linear	Y-axis scaling for the output curve.
Tacho Channel	Any pulse channel	Reference tachometer signal.

Interpreting Order Cut Results

Resonance Identification

When an order cut passes through a structural resonance, the amplitude shows a clear peak at the corresponding critical speed. The shape of this peak provides information about the damping of the resonance:

Sharp, narrow peak: Lightly damped resonance. Even a small order excitation can produce large response amplitudes.
Broad, rounded peak: Heavily damped resonance. The response amplitude is moderate but spread over a wider RPM range.
Asymmetric peak: May indicate nonlinear behavior or interaction between closely spaced modes.

Level vs RPM Trends

Beyond resonance peaks, the general trend of the order cut provides diagnostic information:

Rising with RPM: Expected for aerodynamic or flow-induced excitation, which typically increases with speed.
Approximately constant: Common for mechanical unbalance after balancing, where the centrifugal force effect is compensated by the frequency response roll-off.
Decreasing with RPM: May indicate that the excitation mechanism weakens at higher speeds or that the measurement bandwidth is limiting.

Multi-Order Comparison

V-Listen allows multiple order cuts to be displayed on the same graph for direct comparison. This is useful for:

Comparing harmonics (1×, 2×, 3×) to distinguish unbalance from misalignment.
Overlaying gear mesh orders with shaft orders to understand modulation effects.
Tracking bearing defect orders alongside dominant structural orders.

Practical Tips

Start with the 1st order cut to verify the tacho signal and RPM tracking are correct. The 1st order amplitude should be smooth and free of artifacts.
Use a narrow integration bandwidth (0.2–0.5 orders) for well-separated orders. Widen the bandwidth only if the order is not well defined or shows frequency jitter.
Compare run-up and coast-down order cuts for the same order. Differences may indicate thermal effects or speed-dependent behavior.
Export order cut data to a table or CSV for post-processing, limit evaluation, or comparison with simulation results.
When the order cut shows unexpected peaks, cross-check with the Campbell Diagram to verify that the peak corresponds to a genuine resonance crossing and not a measurement artifact.

If the integration bandwidth is set too wide, adjacent orders may leak into the order cut, producing artificially elevated levels. If set too narrow, the order peak may be partially missed when the actual order frequency does not exactly align with the bin center.