This paper describes the general practices recommended when measuring piping vibration. Differences are drawn between vibration measurements on rotating equipment versus those on piping. The need to measure piping vibration in units of displacement is emphasized due to its close correlation with stress – high piping stress being the primary cause of piping failure. Due to the transient nature of flow-induced vibration inherent in piping, the need to measure vibration amplitudes from time waveform data ≥ 10 seconds long is mentioned.
This paper serves as an introduction to finite element analysis as a tool for the vibration analyst. A general description of the technique is given along with factors that affect its accuracy such as mesh size and boundary conditions. The FEA technique is then compared with experimental modal analysis (EMA) to show differences and to illustrate its strengths. FEA strengths over EMA such as determining natural frequencies and mode shape prior to equipment fabrication and installation, estimating the effects on natural frequencies & mode shapes following structural modifications, and modal studies on very large structures are given.
Orbital analysis of machine vibration data is the topic of this paper. It begins by defining what an orbit plot is and how it is properly constructed. The keyphasor, blank-bright spot on the plot and how to identify forward versus reverse precession are discussed. Both 1x rpm compensation and unfiltered compensation of orbits are described. The “loop rules” are defined with examples to determine the frequency content of an orbital plot. How loading or a restriction to shaft motion effects the orbit shape is explained. Within the paper, orbital plots of both shaft vibration data (proximity probes) and bearing pedestal vibration data (accelerometers) are presented. The usefulness of orbit plots when transient events occur in machinery is shown. For someone wanting an introduction to orbital analysis, this well organized, brief paper is the ticket.
This paper as the title suggests addresses the topic of shaft vibration (not bearing or casing vibration) standards and specifications. It begins with a brief history of how machine shaft vibration has been measured in the past and proceeds to describe current sensors and measurement conventions. The difference between relative and absolute shaft vibration is defined and ways to measure both are described. The remainder of the paper presents some of the different standards available and how each is applied; these include ISO 7919, API, Erskin/Sohre (R/C), Dresser-Clark & the McHugh Method. The paper concludes with seven case histories where these standards are applied to vibration data from real machines of various types.
This paper gives a concise overview of proximity probes and how they are used to monitor machine condition. Included is a short history of the development of these probes. In addition, the development and description of the API 670 specification that applies to the use of these probes is discussed. Emphasis is made on the fact that these sensors measure both vibration and position. Common problems encountered in the use of these probes are discussed such as improper grounding, mechanical run-out, electrical run-out or glitch, and shaft scratches. Common solutions to these problems are given. Typical procedures and practices employed when using these probes for machinery protection are described. This paper is of great value for someone using proximity probes for the first time, but could also serve as a refresher for those who use them infrequently.
There are a number of phenomena that affect the frequency content of observed spectra. Most are some type of non-linearity like misalignment, looseness, and rubs. Nyquist and Bode’ plots can present valuable clues about the type of non-linearity. Also, system and local resonances can amplify or attenuate parts of the spectrum. In most cases, their amplification, damping, and phase change can be determined from the spectra alone without formal response testing. Therefore, their effect can be backed out. A third class is path dependence. That is, the vibration signal can take multiple paths from the source to the transducer and arrive at different times. Furthermore, signals can interact as seen in the several types of modulation where one signal becomes part of the system response as seen by other signals. All these things can operate both independently and in concert to produce a very complex spectrum. Although the mathematics is simple (nothing more than high school trig) and gives a deeper and broader understanding, the field analyst usually does not have the luxury of dwelling in that area. Therefore, this paper will attempt to shed some light on these subjects in a qualitative way so that the reader can begin to develop an intuitive ability to recognize and account for them. There is no such thing as noise… only data that we don’t know how to handle yet.
This paper presents the topic of variable frequency drives from the viewpoint of electrical design with the goal of minimizing damaging harmonics. The concept of impedance as the combination of resistance, capacitance & inductance is described. Different circuit designs are discussed along with how they affect the development of harmonics. The three different types of harmonics are described and the negative effects of each. Tolerances for harmonic amplitudes in current spectra are given. Factors that affect harmonic amplitudes are presented such as circuit design, impedance, machine speed, temperature & cable length. The concept of standing waves or voltage reflection within transmission lines is discussed and how matching motor impedance with transmission line impedance minimizes them. The switching frequency or pulse width modulation rate (PWM) is defined and how a change to this parameter affects both heat buildup & standing wave formation. The paper concludes with general recommendations for VFD system design that should minimize the formation of harmonics.
This brief case history discusses the uncommon topic of gear assembly phase (GAPF) and its appearance in a gearbox following rebuild. Case was not taken to match mark the meshing gears prior to repairs and the predictable result was the appearance of gear assembly phase frequency (GAPF) vibration & harmonics thereof following repairs. Calculations for this uncommon fault frequency are provided.
This brief paper describes the application of a dynamic absorber to a pump with chronic vibration problems. Piping and base problems conspired to place this pump in a state of resonance at its operating speed. The resonant condition predictably resulted in mechanical seal failures, coupling failures and worn pump bearings. Installing the dynamic absorber reduced pump vibration by more than ten times!
This paper covers the topic of vibration absorbers: what they are, what they do, when they should be considered, and how to design them properly. Formulas and plots are provided to aid in the readers understanding. Even stress calculations are provided to aid the designer in prevention of an absorber fatigue failure. For someone wanting to learn more about dynamic vibration absorbers, this brief three page paper is the ticket.
This is an excellent paper for describing both mechanical instabilities and their similarities and differences to resonance issues. This is a very good paper for beginner and intermediate analysts and engineers (or advanced personnel that have not had very much experience with this subject) to learn about mechanical instabilities and corrective action options. There is a table showing the similarities and differences between resonances and instabilities. Illustrations are used throughout the paper to assist in descriptions and discussion of the subjects. The paper is written for practical use and includes the basic math on the subject for fundamental understanding. Two case studies are presented at the end of the paper for examples of identifying and correcting mechanical instabilities. The descriptions of the corrective action options/categories are well written and can be used for guidance on all variations of instability.
This excellent paper lays down the foundation for performing impact testing properly. A large number of illustrations and graphs are used to ensure that the subject material is clearly described. This is a very practical paper that walks the analyst through the steps and information needed to be successful in performing impact testing. While explaining the underlying theory, this remains a practical paper and does not explore the depth of the mathematics needed to develop the tooling. Two case studies are presented at the end of the paper that illustrate impact testing and this is where the paper focuses on machine tool spindles. Even though machine tool spindle testing is the focus of the case studies, the material in this paper would be applicable for all impact testing applications.
This paper examines the effect that the Hanning window and other windows have on signal resolution. The WNF (Window Noise Factor) equation is shown to be inadequate when attempting to resolve two closely spaced signals. The paper demonstrates this through a series of tests with in-close signals using three different windows. Further testing is done with the in-close signals set at different areas of the sample window. The relative signal amplitudes were also shown to influence required signal spacing. All test results are clearly shown in multiple graphs. The paper concludes with the comment never try to skimp on resolution.
This paper describes the many types of signal clipping and the resultant spectrum displays. The introduction prefaces one possibility of signal clipping that can occur is when the operator disables the analyzer’s auto-ranging feature. Both symmetric & asymmetric clipping causing A/D convertor overflow are described with both time and spectrum graphs showing resultant displays. Accelerometer overload, clipping prior to the anti-aliasing filters, transient clipping and other forms of clipping are also covered in this paper. Multiple graphs are included displaying the results of various forms of clipping.
This paper compares motor current signature analysis & vibration signature analysis techniques on the most common induction motor defects. It would be a good reference for beginning and intermediate analysts and engineers that are looking for additional information comparing different PdM technologies for monitoring induction motors. There are several spectral charts and a table showing the comparisons of the differing technologies. This paper is more practical in showing the results of different monitoring technologies. This paper thoroughly describes the technologies, their applications, and the results. The paper is not intended as a case history but uses real data for the development of the comparisons.
The paper presents the results of an experimental program established to conduct accurate measurements of energy lost due to misalignment and unbalance of rotating machinery. Test item was a 30 HP, 3-Phase motor driving a 20 kW, 400 cycle AC generator mounted on a commercial foundation. Testing included multiple levels of unbalance, offset & angular misalignment using various couplings. Testing results displayed very small levels of energy savings due to misalignment or unbalance.
Analysis and Solution of Reliability Problems with a Steam Turbine-Feed Water Pump Train in a Nuclear Power Plant
This is an excellent case study paper for obtaining a basic understanding of rotor dynamics and bearing analysis. The paper is intended for an intermediate level analyst or engineer that has a thorough understanding of the basics of vibration analysis and has an interest in a more in depth understanding of the dynamics of rotors and bearings in operating machinery. There are illustrations of the equipment and plots of vibration data and modeling analysis results for the discussion. This is a more practical paper, but does provide the basic formula for the calculations discussed. This is a descriptive paper in that there is good discussion of what varying the stiffness and damping of the bearing will do for the rotor dynamics of the system.
This paper discusses four prominent roll dynamics interaction problems. The first discussed is Multiples inherit to the machine design. Undesirable multiples can develop sized imperfections on mating rolls. The second is Barring characterized by alternating magnitude differences in a roll’s circumferential surface parameters. Barring can exist due to imperfections in the roll or resonance. The third is Roundness an ideal roll condition that can change due to operating forces. The fourth is Wandering Alignment caused by outside forces making the roll centerline wander. Examples of these four problems are presented with graphs and corrective measures. The paper concludes with how roll dynamics influences vibration and recommends avoidance of components with simple multiples and repetitive imperfections.
A full description of the Spike Energy signal processing technique is the subject of this paper. Differences between this high frequency processing technique and others such as acceleration enveloping are described. The need for proper sensor mounting to obtain consistent measurements is stressed. Three case histories are presented at the end to illustrate Spike Energy’s effectiveness in detecting problems in machines.
This second in a two part series on pump vibration deals with the analysis of common pump problems. A suggested list of common pump fault frequencies is given. As in the first paper on this topic, an emphasis is again made on the need to monitor not only pump vibration, but also its pressure & flow. For pumps operating on fluid film bearings like boiler feed pumps, the use of thrust probes to monitor the position of the rotor is emphasized.
This is the first of two papers by the author on the topic of pump vibration. The first paper discusses the most common causes of pump vibration: flow-related issues. The need to operate a pump at or near its best efficiency point (BEP) is emphasized and why it is important. Descriptions of the opposite flow problems of low-flow and cavitation are given. The need to gather periodic data on pressure and flow along with vibration is stressed. Suggested machine locations for pump vibration monitoring are given.
The Use Of Non-Intrusive Phase & Amplitude Demodulation Techniques To Identify Torsional Modes On Machinery Under Constant Speed Conditions, Part 2 Applications
This paper is the second in a two part series on torsional vibration where a unique measuring and processing technique is used involving the Hilbert Transform and both frequency and amplitude demodulation. The first paper dealt with explaining the theory and mathematics of the signal processing technique. This second paper presents two applications of the technique to real machines with real torsional problems. The first application is on a turbine-generator where a gear toothed wheel is used in conjunction with a proximity probe, accelerometer and tach to gather torsional data. The second application is on a paper machine roll drive where two accelerometers measuring the existing gear mesh frequencies generated by the gearbox are used instead of a gear toothed wheel or encoder to gather torsional data. In both cases, the signal processing technique described in the first paper is used to pull the desired information out of the signals. The torsional vibration frequencies, torsional natural frequencies, and estimates of angular displacements and mode shapes can be found using this approach.
This last of three papers on the topic of modal analysis presents two case histories where both experimental modal analysis (EMA) and operating deflection shape analysis (ODS) were performed in an effort to solve machinery vibration problems. The first case history involved high vibration on skid mounted cooling water pumps at a nuclear plant. The second case history involved high vibration on a vertical pump at a nuclear plant. Both problems and their solutions are described in detail.
This second of three papers on the topic of modal analysis describes experimental modal analysis, what it is, why you would do it and the basics on how to do it. The differences between experimental modal analysis (EMA) and finite element analysis (FEA) are described. How to validate the results of a modal test using the modal assurance criterion (MAC) is discussed.
This first of three papers on the topic of modal analysis presents the theory and basic equations involved. Impact testing of equipment is described in detail and recommendations are given on how to perform it properly. Variables involved with impact testing such as hammer tip type, response sensor selection, frequency range, frequency resolution, force window type & transfer function selection are discussed in detail.