Industrial Mixer

Figure 1. Left –Solid CAD model of one of two peripheral mixer assemblies support by a solid steel mixer deck. Right – Cross-sectional view of model details. 

Key terms such as natural frequency, resonance, amplification factor, and separation margin are explained in this one-page document. 

The key deliverable from this analysis is determining whether the machinery will operate free from resonant vibration problems, as documented in a report that summarizes the methods and results. If not, possible modifications are discussed with the mixer manufacturer and re-analyzed until a specific, workable solution is designed and agreed upon before manufacturing and installation proceed. 

What the Structural/Lateral Rotordynamic Analysis Report Should Include:

1. A table listing natural frequencies and their separation margins relative to running speed and other potential excitation sources (up to 2x VPF) is provided. Any natural frequencies falling within the specified separation margin are highlighted and evaluated for possible design modifications. For example, the draft tube inlet baffle thickness may need to be adjusted to improve separation margins
2. A Campbell Diagram providing a clear summary of the results. In the example shown in Figure 2, acceptable separation margins exist between the structural natural frequencies and the excitation sources (motor and mixer running speeds, and VPF). Although four mixer-related natural frequency lines in the 30-35 Hz range fall within the motor speed's ±20% margin, experience shows that these modes, operating at 285 RPM, are sufficiently separated from the motor running speed (1750 RPM) to be non-problematic.
3. Mode Shape Animations of natural frequencies, both perpendicular and parallel to the draft tube inlet, are typically provided. Figure 3 shows an FEA animation of a mixer bending mode. Care must be taken when evaluating these modes, as the vast majority of natural frequencies are benign. To be resonant, a mode must be both (1) lightly damped and (2) close to, or at, an excitation source (such as 1x running speed). Natural frequencies may shift as a plant ages or after maintenance/repair, so maintaining awareness of separation margins is critical throughout the life of the equipment.

Figure 2. Structural/lateral rotordynamic analysis Campbell diagram showing acceptable separation margins between the structural natural frequencies and the motor and mixer running speed excitation sources, as well as the mixer vane pass frequency. The horizontal lines represent the natural frequencies predicted by FEA. The two vertical dashed lines indicate the motor rotational speed (purple) and the mixer running speed (red). Specified separation margin requirements are overlaid. 

Figure 3: FEA animations of the 1st mixer bending modes, perpendicular (left) and parallel (right) to the draft tube inlet, respectively. Be cautious when reviewing mode shape information, as the animations exaggerate motion and are typically for informational purposes only (i.e., no problem is currently predicted). In this example, there are no excitation sources near these frequencies, so these natural frequencies should not become excited (i.e., resonant) and cause high vibration.

Torsional rotordynamic analysis

In addition to structural and lateral rotordynamic vibration concerns, torsional rotordynamics must also be considered when evaluating a rotating train . The goal of a torsional rotordynamic analysis is to determine whether any torsional natural frequencies (critical speeds) are too close to known excitation sources. If the separation margin is insufficient, the system may experience torsional resonance, which can lead to accelerated fatigue or, in severe cases, shaft failure.

 What the Analysis Involves

The analysis calculates the torsional rotordynamic natural frequencies of the complete drive train and compares them against potential excitation sources, including:

• 1× running speed (over the entire speed range for variable speed equipment)
• Vane Pass Frequency (VPF)
• 1× and 2× line frequency
• Variable Frequency Drive (VFD) control frequencies (if applicable)

Industry standards such as ANSI/Hydraulic Institute Guideline 9.6.8 recommend maintaining at least a ±10% separation margin between natural frequencies and excitation sources. Figures 4 and 5 show a representative torsional model and the corresponding Campbell Diagram..

 Figure 4. An example torsional analysis model of a mixer drive train. 

Figure 5: A key deliverable is the torsional rotordynamic Campbell diagram (see Figure 2 for the structural/lateral rotordynamic Campbell diagram).

Example Findings – Torsional Rotordynamics Analysis

The Campbell diagram (Figure 5) shows the first torsional mode (solid green line) falling close to the 1× line frequency excitation (dashed green line). At first glance, this suggested a potential issue. However, when forced response calculations were plotted on a Goodman diagram, the results indicated that the rotor system would still maintain a sufficient operating life. In this case, design modifications were not necessary.

Practical considerations

At MSI, one of our greatest strengths is that all of our analysts also have significant field experience. This combination is critical: hands-on experience enables analysts to make better assumptions, validate models against real-world conditions, and perform "bracketing evaluations" that account for uncertainty in natural frequency predictions.

While advanced forced response analysis can estimate vibration levels and evaluate complex scenarios, the foundation of reliable operation still rests on good installation and maintenance practices. Even the best system design and the most thorough risk reduction analysis can be undermined by basic mechanical issues, such as imbalance, misalignment, soft foot, loose bolts, poor fits, or inadequate foundation stiffness. Another challenge is that analysis is based on the original design intent, but what is designed is not always what gets built. For example, changing a machine support from a solid 1-inch steel plate (as specified in the design) to grating in the actual installation can significantly shift natural frequencies, undermining the original analysis.

Part 1of this blog discussed the overall vibration risk reduction process and the importance of proper specification wording. Part 2 focused on the analysis itself. In Part 3, we outline best engineering practices for performing a Vibration Acceptance Test (VAT), including the role of Experimental Modal Analysis (impact) test in confirming natural frequencies and separation margins after installation.

If you would like more information or assistance in reducing the risk of post-installation rotating machinery vibration issues, please  contact MSI. Our combination of analytical expertise and field-proven experience enables us to help clients avoid costly vibration problems before they occur.