Understanding Cascade Plots
A cascade plot — also called a waterfall plot, 3D spectrum, or spectral map — is a three-dimensional display showing how vibration frequency spectra change over time, speed, or another variable. Frequency runs along the X-axis, the changing variable (time or speed) along the Y-axis, and vibration amplitude along the Z-axis, rendered as height, colour intensity, or both. Successive spectra are stacked one behind another like a series of cascading waterfalls, building a picture that exposes patterns no single 2D spectrum can reveal.
This added dimension makes the cascade plot indispensable for two jobs in particular: rotor dynamics analysis, where it pinpoints critical speeds during a startup or coastdown, and long-term fault tracking, where it lets an engineer watch a bearing defect frequency first emerge and then grow. The terms cascade plot and waterfall plot are used interchangeably throughout the field.
1. How a Cascade Plot Is Built
Axes and Dimensions
- X-axis (horizontal): frequency, in Hz, CPM, or orders.
- Y-axis (depth): the variable being swept — time, speed, or load.
- Z-axis (vertical or colour): vibration amplitude.
- Perspective: typically viewed from a front-top angle so that near traces do not entirely hide those behind them.
Types Based on the Y-Axis Variable
What the Y-axis represents defines the plot’s purpose:
- Speed-based cascade (startup/coastdown): the Y-axis is rotational speed, generated during a run-up அல்லது coastdown, with speed usually increasing from front to back. This is the most common form for critical-speed identification.
- Time-based cascade: the Y-axis is calendar time, showing fault development over days, weeks, or months — recent records at the back, older ones at the front — which makes it ideal for monitoring progressive failures.
- Load-based cascade: the Y-axis is load or power, revealing how vibration responds to loading and exposing load-dependent phenomena on variable-duty equipment.
2. Reading and Interpreting Cascade Plots
The whole technique hinges on one visual rule: components that track shaft speed slope diagonally, while components fixed in frequency stand vertically. Learn to read that geometry and the plot interprets itself.
Speed-Tracking Components
These appear as diagonal lines, because their frequency rises and falls with speed:
- 1× line: a straight diagonal running from the origin — the signature of unbalance.
- 2× line: a steeper diagonal, commonly misalignment அல்லது looseness.
- Higher orders: steeper diagonals still, the harmonics of running speed.
Fixed-Frequency Components
These appear as vertical lines, holding constant regardless of speed:
- Natural frequencies: vertical features marking structural resonances.
- Electrical frequencies: twice line frequency (120 Hz on a 60 Hz supply, 100 Hz on a 50 Hz supply) stands perfectly vertical.
- External vibration: constant frequencies bleeding in from nearby equipment.
Identifying a Critical Speed
The payoff comes where a diagonal 1× line crosses a vertical natural-frequency feature. At that intersection the amplitude peaks — rising into a “mountain” on the plot — because the rotor is being driven through a resonance, and the sharpness of that peak gives a direct, visual read on damping.
3. Applications
Critical Speed Analysis
This is the classic use, central to commissioning and troubleshooting. A speed-based cascade lets an engineer locate every critical speed in the operating range, verify the separation margins from running speed, judge damping from peak sharpness, and compare the measured critical speeds against those predicted by a Campbell diagram or rotor model.
Bearing Defect Monitoring
A time-based cascade is the natural way to follow a deteriorating bearing: watch the BPFO, BPFI, and BSF peaks emerge and climb, note the harmonic development that signals advancing damage, and estimate a failure timeline from the growth rate — a foundation for predicting remaining useful life.
Order Analysis
Plotting the frequency axis in orders rather than Hz flips the geometry usefully: speed-synchronous components line up vertically while non-synchronous ones (such as bearing tones or oil whirl) slope away diagonally. This is especially powerful on variable-speed machines, where a conventional Hz axis would smear every order into a band.
Fault Development Visualisation
More broadly, the cascade plot is the format of choice for watching a fault evolve — new peaks appearing, existing peaks growing, harmonics multiplying, and sidebands emerging — all laid out on a single picture.
4. Creating Effective Cascade Plots
Data Collection
- Sufficient slices: a minimum of 10–20 spectra is needed for a clear, readable surface.
- Consistent increment: even spacing in the Y-axis variable keeps the geometry undistorted.
- Adequate resolution: enough frequency resolution to separate the peaks of interest — a choice the FFT Resolution Calculator can help make.
- Full range: cover the complete operating speed range or the entire trending period so nothing important falls outside the plot.
Display Settings
- Amplitude scale: linear or logarithmic, chosen to suit the data’s dynamic range.
- Colour map: selected to make the features of interest stand out.
- Perspective angle: usually 20–30° of elevation for clarity.
- Peak retention: some software draws a peak envelope across the slices to sharpen the picture.
5. Where Field Instruments Fit
Capturing a usable cascade demands an instrument that can record a series of spectra synchronised to shaft speed throughout a run-up or coastdown. A portable two-channel analyser such as the Balanset-1A measures vibration together with a shaft tachometer reference, so a field engineer can collect the speed-tagged spectra needed to spot a critical speed on a machine in its own bearings — then, if the diagonal 1× line proves dominant, move straight on to field balancing without ever leaving the site.
6. Advantages and Limitations
Like any visualisation, the cascade plot is a tool with a defined sweet spot rather than a universal answer.
Advantages
- Presents multidimensional data in an intuitive, single view.
- Reveals patterns that are simply invisible in isolated 2D spectra.
- Cleanly separates speed-dependent from speed-independent components.
- Gives a comprehensive picture of dynamic behaviour — and reads well in reports and presentations.
Limitations
- Can become cluttered when too many components are present.
- Requires experience to interpret correctly.
- Fine detail can be hidden behind nearer peaks in the 3D view.
- Makes it hard to read off precise numerical values, so it complements rather than replaces conventional 2D spectral analysis.
Cascade plots are powerful visualisation tools that add the dimension of time or speed to frequency analysis, exposing the dynamic patterns and progressions that static spectra miss. Mastering their interpretation — distinguishing diagonal from vertical features, spotting critical-speed intersections, and tracking fault progression — is a core skill for advanced vibration analysis and rotor-dynamics assessment.
