Blog

June 2011 • RELIABLE PLANT

A VIBXPERT® vibration data collector was used to collect data on a DDM 750 AC 1M drilling machine at the workshop to check for factory acceptance. The VIBXPERT allows for high-resolution time signal and spectrum measurements to be collected. The VIBXPERT can collect more than 100, 000 lines of resolution, which allows for the VIBXPERT to capture any transient events that might occur. The vibration data was collected while there was no load on the machine. The main shaft speed was 216 rpm, and the motor itself was turning at 1762 rpm.
After the vibration data was collected, the data was transferred back into the OMNITREND® software for analysis. Vibration analysis on the machine during the Factory Acceptance Test (FAT) detected impacts from the cage (train) in the pinion upper bearing.

See plots and entire case study Condition Monitoring on Drilling Platforms: A Case Study

June/July 2011 • UPTIME

Analyzing only vibration response spectra is difficult since they often don’t clearly match wall chart and textbook examples.

As anyone who has practiced vibration analysis knows, vibration signatures obtained on routes are often far from the wall chart examples. The reason for this is that the vibration signatures collected and analyzed represent the response of a system due to a variety of different forces that act simultaneously to produce one signature.

Unfortunately, vibration analysts are actually interested in determining the individual forces that cause the response. Once the forces are accurately identified, only then can they be reduced or eliminated.
Take for example the force of unbalance. Wall charts and texts on vibration analysis represent mass unbalance as a running speed peak in the spectrum that dominates all other content. Also, these theoretical, or textbook, examples indicate the vibration amplitudes will be equal in the horizontal and vertical planes. However, experienced vibration analysts know this is often not the signature we see. This is due to the fact there are multiple forces acting on the system, and it may have asymmetric stiffness resulting in highly directional vibration. In these situations, following the wall chart examples without additional phase analysis may send an analyst down the wrong path. In order to be effective in vibration analysis, it is necessary to first resolve the most dominant problem and then reanalyze the machine to determine if there are any further forces that need to be minimized. Properly identifying the most dominant problem can be difficult, so make sure to use all tools available. This case history illustrates a situation in which the vibration signature was far from being textbook due to multiple sources simultaneously acting on the system to produce one on-textbook signature. Getting to the root causes of the problem took multiple iterations.

Read the entire article Balancing Out the Root Cause by Chad Wilcox • http://www.pioneer-engineering.com/

June 2011 • PUMPS & SYSTEMS

Ensure proper bearing assessment and maintenance with this proven method
Of the methods used to assess the operating condition of rolling element bearings, one of the most successful and popular techniques is that of Shock Pulse evaluation. Shock Pulses are a special type of vibration that can be clearly distinguished from ordinary machine vibrations:
• The actual Shock Pulse is the pressure wave generated at the moment when one metallic object strikes another.
• The bulk of the impact momentum, however, acts to deform the target object, which then oscillates at its natural frequency. This vibration ultimately dissipates primarily as heat due to internal friction material damping.

Shock Pulses in Bearings
Shock Pulses occur during bearing operation when a rolling element passes over an irregularity in the surface of the bearing race. Of course, there is no such thing as a perfectly smooth surface in real life. Therefore, even new bearings emit a signal of weak Shock Pulses in rapid succession. This Carpet Value rises when the lubrication film between rolling elements and their races becomes depleted.
A defect on the surface of a rolling element or bearing race produces a strong Shock Pulse reaction with up to 1,000 times the intensity of the Carpet Value. These clusters of high amplitude peaks or Maximum Value stand out clearly from the background noise and are ideal indicators of bearing damage.

Measurement
Shock Pulses propagate within a much higher frequency range than that of ordinary machine vibration, and their energy content is relatively low.
Therefore, the accelerometer used for Shock Pulse measurement is tuned with a 36 kHz resonance frequency that lies precisely within the Shock Pulse frequency range. In addition, a 36 kHz bandpass filter is applied to the accelerometer signal to help filter out lower frequency mechanical vibration. When Shock Pulse is present the tuned accelerometer resonance is excited and amplifies the Shock Pulse signal resulting in an excellent indication of bearing lubrication and damage.
Shock Pulse is responsive even when far more energetic machine vibration is present. Therefore, lower frequency mechanical conditions such as unbalance, shaft misalignment, or vibration from adjacent machines have little effect on Shock Pulse. In addition, high-frequency signals tend to dissipate rapidly so very little interference is encountered from adjacent bearings.

Read the article Reliable Shock Pulse Evaluation of Anti-Friction Bearing Condition

Learn about our VIBXPERT II Portable Vibration Analyzer —with Full Color Display, Fast Data Acquisition and Powerful Vibration Diagnostic Tools. VIBXPERT uses the Shock Pulse method to detect lubrication condition and bearing damage.

Some vibration software and vibration data collectors allow the use of frequency bands to help the analyst measure and identify specific equipment faults.  These frequency bands are actually measured in the vibration data collector.  The results can be used for trending and alarm purposes.  You can create specific bands around specific faults that may occur in the equipment being monitored.  For example, bands for imbalance, misalignment, specific bearing defects, electrical defects, and much more can easily be measured.  Alarm thresholds can be created to alert you when attention is required.  These bands can help you identify equipment issues in your vibration data before you look at the FFT or time waveform data.  A review of the FFT and time waveform data should always be completed.  However, band alarms can keep you from overlooking an issue and reduce the amount of analysis that may be required.

June 2011 • TPO Magazine

ACING IT IN OKLAHOMA

Strong skills, preventive maintenance and good planning lead to success at the Coffee Creek Treatment Plant in Edmond.

The Coffee Creek Wastewater Treatment plant in Edmond, Okla., has had near-perfect compliance for 38 years and has won several awards, most recently 2010 Large Wastewater Plant of the Year from the Oklahoma Water and Pollution Control Association (OWPCA).

It all has happened with a staff of five, despite rapid population growth, several upgrades, and various episodes with collection system inflow and infiltration. “When you have only five staff, you have to focus on working smarter, not harder,” says Fred Rice, water resources superintendent for the city. This means preventive maintenance, SCADA monitoring of critical alarms, and ongoing equipment and safety training.

It’s also a matter of teamwork. Kris Neifing, chief plant operator, hired in 2004, supervises two operators, a maintenance specialist, and a lab technician, and is also responsible for one lift station at the plant and nine lift stations located throughout the collection system.

Rice credits Neifing and his staff for the plant’s track record. “Kris has really pulled everyone together as a team,” he says. “All the credit for what we’ve achieved in the last six years is due to Kris and his staff. My role is like coaching a sports team. You can coach them, but the team executes the plays.”
Says Neifing, “What makes us successful is that everybody has different skills that collectively make us the best we can be. Some are better at maintenance while others prefer operations. We believe that no one knows how to do their jobs better than the ones who do it every day.”

Continuous improvement

The plant’s compliance and safety record do not mean the staff is complacent. “We strive to continuously improve,” says Rice. “There is no process out there that can’t be improved, especially on the maintenance side.”
Rice and Neifing frequently attend the WEFTEC conference and other trade shows to check out the latest equipment. This has led to several innovations, such as vibration analysis and laser alignment equipment to help ensure that pumps and motors operate normally with the lowest possible maintenance. Rice and Neifing also read trade journals and network with others in the wastewater treatment business to glean ideas.

“The staff comes to us with ideas, like getting air compressors for maintenance, and suggesting better equipment or ways of doing things,” Neifing says. “We empower them to make suggestions, and we listen.” Adds Rice, “The city started a program based on the general concepts in the Good to Great book because we believe that organizations that excel are successful from the ground up. We give our employees responsibility and then hold them accountable.”

Read the entire article Acing it in Oklahoma

March 2011 • IMPO MAGAZINE, iPurchase Supplement

As the American economy recovers, how aware, or active, are manufacturers concerning predictive maintenance?

Most manufacturers never lost the desire to increase their overall reliability and predictive maintenance efforts during the recent economic slowdown.  Some companies did postpone purchases of predictive maintenance-related products.  However, a lot of companies realize that an investment in predictive maintenance technologies is a viable means to decrease overall maintenance expenses, so they do it the right way.  A lot of manufacturers, as a result of this understanding, continued to invest in predictive maintenance-related technologies during the recent economic slowdown.  This allowed them to reduce overall maintenance costs and place their company in a more competitive position once the economy recovers. Interest in these products is higher this year as companies continue to invest in vibration- and alignment-related products to reduce their costs. increase competitive advantages, and manufacture equipment reliability.

Read the entire interview Q&A Roundtable from iPurchase, a supplement from IMPO Magazine including:
Are there any interesting trends occurring in the maintenance market?
Why do you think laser alignment is important to a manufacturer’s maintenance strategy?
How would you recommend a manufacturer approach the creation of a more robust maintenance plan?

MYTH:  A well-trained technician can predict within a window of a few hours when a machine will fail.

TRUTH: This myth is common throughout industry and poses a danger to machinery.  The myth often leads to operating machinery with known defects for periods longer than is safe for the equipment.  Let’s explore the elements that brought about this myth and why it remains so prevalent.
One reason for the birth of this myth is the curve fitting plots found in many PDM software programs. Figure 1 shows such a plot. Those not trained in the use of the software may think that the time projected for the condition to reach the alarm level indicates “time to failure”. In reality, the projected time is “time to alarm”. The machine may run quite sometime after exceeding an alarm. These plots are very valuable if used as intended.

Another reason for this pervasive myth is that published PF curves are almost always shown as downward exponential curves. Figure 2 shows a typical curve.  If PF curves really are this shape, finding the “time-to-failure” would simply be a mathematical process.  However, PF curves are seldom this shape because many variables may influence the shape of the curve.  Figure 3 shows how a real curve may look and also presents some reasons why the shape is erratic. PF curves may have hundreds or thousands of shapes.

Even the best-trained technicians, using the best tools can’t possibly know all variables that may lead to component or machine failure.  Load, speed, temperature, and environment are only a few of the many variables that may affect “time-to-failure” for a given defect.  A well-trained technician can detect defects and may even be able to state with a fair amount of confidence that a machine will operate or will not operate until the very first opportunity to take it out of service for repairs. Doing otherwise and operating with known defects is tantamount to rolling the dice and gambling with your machinery.

February 2011 • WORLD CEMENT

Robert Schmaus, Prüftechnik Condition Monitoring GmbH, Germany, outlines the benefits of applying online condition monitoring systems.

What is condition monitoring?

Condition monitoring is the process of monitoring machinery health or machinery condition, such that a significant change is indicative of a developing failure. It allows maintenance to be scheduled, or other actions to be taken to avoid the consequences of failure before the failure occurs. The main goal of condition monitoring, therefore, is to improve plant production capacity and make the process more profitable. Different maintenance strategies have been adopted by maintenance personnel – starting from pure ‘breakdown maintenance’ (run the machine until breakdown without intervention), preventive or time-based maintenance(scheduled maintenance stops), to predictive maintenance. Having adopted a predictive maintenance strategy- in a plant, maintenance stops for interventions required are scheduled according to machine conditions and not by mere periodical interventions, as is the case in time-based maintenance. Predictive maintenance enables operators to take advantage of the lifetime of machine components and minimizes required spare parts stock, as repairs are effectuated according to the condition of machines and not on a periodical basis. As developing failures can be detected in time, consequential damages are avoided. Improved machine reliability minimizes interruptions to the production process.

Download the entire article Online CMS – Value for Money?

Oil data reports made available by laboratories to their customers are usually delivered in MICROSOFT EXCEL® files or text files.  The data is generally made available via an online download.  These reports contain information on many different substances found in the oil such as iron, aluminum, copper, and many more.  Other factors such as viscosity are included as well.

Oil analysis data provides a great insight into the health of your equipment and should be a routine part of any good reliability program.  It can be overwhelming to process all this data in a meaningful way.  You may not wish to trend and alarm on every single parameter contained in the oil report.  So what do you do?  The answer may be your vibration analysis software!

A good vibration analysis program like OMNITREND will allow you to create customizable imports for your oil analysis data.  A routine and automated process can be set up to import only the data of interest for specific machines.  Once the data is imported, then automated alarms can be created to alert you of any oil parameters that are of concern for your equipment.

Oil data can be visually trended over time and you will automatically be alerted when an oil parameter reaches a value of concern for a specific machine. This can allow you automatically manage selected oil parameters of concern and greatly reduce the amount of time required to manually review each parameter measured by your oil analysis lab.

Furthermore, it will allow you to use the routine vibration data collected and oil analysis data provided by your laboratory to detect and confirm equipment problems.  You should not underestimate the power of having your vibration and oil analysis data integrated.

March 2011 • WindPower Engineering

Condition-monitoring systems have made it clear that wind turbines are complex machines in which overall vibration values must be systematically determined and evaluation references made available. Points to consider when writing evaluation guidelines for wind-power plants include:

• Function and structural design of wind turbines and their components
• Interaction between the individual drivetrain components (modules) being tested
• Information and experience regarding the possible faults and damages occurring in the individual testing modules during operation and their economic impact
• Knowledge of operation-related and machine-related vibration influences, the diagnostic procedures that must be adhered to, and their respective limits

The recent VDI 3834, established and released in 2009, takes into consideration the special requirements for evaluating wind turbine components. The guideline is set for turbines ranging from 100 kW to 3 MW.
Read our article Finally, ISO Guidelines for Condition Monitoring

Water/Waste Processing Magazine • February 2011

A southeastern US water company has been developing a condition monitoring program with LUDECA as their partner, to monitor their on-surface equipment with VIBSCANNER^® and VIBCODE^® systems. The current program has provided significant benefits and cost savings by preventing expensive pump and motor failures, as well as optimizing their current maintenance planning and procedures. However, sometimes not all equipment can be monitored using a portable device, either for safety reasons or location.  A recent problem arose at the wastewater treatment plant.  The facility experiences high flows and because of the small relative space available, they use chemical and aeration treatment processes to meet code and final water approval levels.  The flow works by deep tunnels that run beneath the city terminating at a deep pump well 121 feet in depth (37 meters).  It is pumped by four submersible pumps to the plant’s inlet. These pumps have caused extreme revenue and capital cost in both repairs and operations.

Download the entire article “Online Condition Monitoring – Monitoring Waste the Easy Way”

I recently used an audible recording of a time domain waveform acquired with the VibXpert to help demonstrate the value of the actual “audible” sound of a bearing fault during a recent condition monitoring survey (baseline measurements) of a multistage circulation pump. As we know, the time domain waveform signal is truly the “raw” data of vibration measurements. In this particular case, the frequency domain (FFT plot) measurements showed very low vibration amplitude levels and the determining factor for the early detection of the bearing fault was primarily the use of time domain waveform signals. Frequency domain (spectra plots) along with time domain (waveform plots) was presented to the customer for their review.  Also, an “audible” recording of the time domain waveform signal was presented to the customer for review with very favorable results. The customer conveyed that the “actual” bearing impacting sound represented by the time waveform recording was easier to comprehend then reviewing the time signal plot. The customer stated that he felt like he was getting a combination of “ultra sound” technology along with vibration analysis and was very satisfied with the results.The VibXpert® data collectors allow you to listen to the raw data collected on equipment.  The OMNITREND® software allows time domain waveform data to be played back in the comfort of your office.  In certain situations these capabilities can be a great analysis tool. I have successfully used these capabilities in the past to capture “random” impacting events of ultra-slow operating equipment such as gear reducers on conveyor drives and building structure movement in connection with sensitive types of hospital equipment. Bottom line, I am planning on utilizing these capabilities much more in the future for illustrating time domain waveform signals to my customers. —Ray Wonderly with ADVANCED MAINTENANCE TECHNOLOGIES – http://www.amt-vibration.com/

1. Functionality: Do the Predictive Maintenance (PdM) tools you are considering have the ability to make all the measurements required by your physical asset management strategy? Are displays easy to see and interpret? Are the tools easy to learn and easy to use? Learn about our PdM tools. For Software, can it interface with your CMMS system? Can you import data from other systems such as oil data? Learn about our OMNITREND software.
2. Durability: Will the tools hold up to your plant’s environment? Are they rugged enough for multiple users? IP ratings such as water–, dust– and shockproof are very important when dealing with industrial tools.
3. Service: Will your vendor be available to answer questions or address problems should they arise? What is the vendor’s reputation for customer service? If you have a problem with a tool how soon can you expect a “loaner” until yours is repaired? Are the tools repaired and/or calibrated locally? Learn about LUDECA Repair and Calibration.
4. Training: What are the training costs associated with learning how to use the tools? Is training included with the purchase? What training resources are available? Learn about LUDECA Training.
5. Support: What level of support do you need? Does the vendor have a call-in tech support center, is it free or paid? Will the yearly costs of maintenance agreements make the tools considerably more expensive than competitors’ tools having similar capabilities? Do they offer free updates? Learn about LUDECA Technical Support.

The LUDECA Condition Monitoring team just completed a seminar on Modal and ODS techniques with Tony DeMatteo of 4X Diagnostics LLC.  The seminar was highly successful and participants came from as far as England to attend.   Everyone involved in vibration analysis can benefit from phase analysis. Operational Deflection Shape (ODS) is a software-based enhancement of phase analysis and can be used to analyze the vibratory motion of rotating equipment and structures. ODS is a non-intrusive test that can be completed during the normal operation of a machine.  A computer-generated model of a machine, structure, etc is animated with the collected data.  This process can be used to prove or disprove theories about failure modes in equipment.

The VibXpert® data collector can be easily used to collect this type of data.  The data is collected with the VibXpert and imported into the computer-generated model of the machine or structure for animation and analysis.  The new VibXpert II data collector was used in the class to demonstrate the process and power of ODS measurements.  Everyone was amazed at how fast this data can be collected with the VibXpert II versus other instruments currently available.

Problem: In a number of real-world applications the speed of a machine cannot be held to a fixed RPM while vibration data is collected.

Solution: The order tracking capability of the VIBXPERT® allows accurate data collection on machines that experience constant speed changes during data acquisition.Applications such as a winder that spools paper coming off the end of a paper machine, by its design continuously decrease in RPM as the diameter of the paper accumulated on the winder increases.  Other applications such as a pump that feeds a production process may be required to continuously change speed as the process demand dynamically changes.  A spectrum collected during an RPM change will result in smeared and skewed data.  During the spectrum measurement process a digitized time waveform of the vibration is collected.  The time waveform is collected for enough time to gather repeated cycles of vibration.  The FFT process converts the waveform into a spectrum that displays vibration amplitude versus frequency.  If the speed of the machine changes during the waveform collection process the peaks generated in the waveform will not be evenly spaced.  The resulting spectrum will have vibration peaks widened or smeared and frequency information from low frequency to high frequency that is frequency shifted or skewed. The resulting spectrum data taken from a machine whose speed is changing is most often unusable.  In this changing speed scenario if frequency peaks are displayed in the spectrum both the frequency and amplitude will be inaccurate at best. Order Tracking Spectrums allows one to successfully collect spectrum data from a machine that is continuously changing speed. The resulting spectrum amplitudes will be accurate and the frequencies will be displayed in orders of running speed. The collection process requires a tachometer to track the machine’s speed as the time waveform data is acquired. As the equipment speed changes during the measurement process, the start of the actual measurement has to be synchronized from average to average.  The number of averages is user selectable. Otherwise, the 1× on one average will not be the 1× on the next average, etc., for the measurement.  This results in the same peak showing up in multiple places or being merged with other peaks. This can create analysis havoc.  Order tracking results in an accurate spectrum with no data skewing or smearing.

Setting Up Order Tracking Spectrums
There are 5 preset Order Tracking Spectrum setups in OMNITREND®.  To view these setups in OMNITREND simply select the menu item <Tools> then <SetupManager> from the drop-down menu.  Measurement setups 260 through 264 are examples of Order Tracking Spectrums. The preset Order Tracking Spectrums are available in 5 different units of measure.  Like all preset measurements in OMNITREND, these are fixed and cannot be changed.  If the preset measurements are not suitable for your application, they may be modified per the following instructions:1) From the <Setup Manager> window, highlight and right mouse click on the Order Tracking Spectrum you are interested in modifying.

2) Click on <Duplicate Setup>

3) An identical measurement will be created and show up at the bottom of the setup manager list.

4) The spectrum parameters of this newly created spectrum may now be customized to better fit your needs. Once you have finalized your Order Tracking Spectrum setups we simply add them to a Measurement Location on a Machine in the equipment database or in the Template Editor. The Machine can then be downloaded to your VIBXPERT data collector as part of a route or as a machine template. At the present time, Order Tracking Spectrums are not available via the Multi-Mode area of the VIBXPERT.

Collecting Order Tracking Spectrums
When collecting Order Tracking Spectrums the appropriate accelerometer and tachometer are required per the measurement setup in OMNITREND.  The tachometer must be aimed at a speed reference such as reflective tape mounted on the turning shaft.  When the machine is running, the RPM may change while you are taking the measurement.

To the left is a spectrum taken as the speed constantly changed between 1775 RPM and 910 RPM.  Notice the waveform time units are recorded as the number of rotations.  The spectrum frequency units are orders of running speed. The lower half of the display can be changed to list the maximum 10 amplitudes in the spectrum. Toggle to the lower screen by pressing the F key and then select the peak you would like to see.  In the spectrum, the cursor will jump to the corresponding frequency. In OMNITREND the data is equally simple to view and analyze.

www.facilitiesnet.com • November 2010

Many manufacturing facilities use vibration analysis to detect early signs of machine failure, allowing technicians to repair or replace machinery before a catastrophic failure occurs. All rotating equipment vibrates, but as components begin to fail or reach the end of their serviceable life, they begin to vibrate more and in unique ways. Ongoing monitoring of equipment allows technicians to identify these indicators of wear and future damage well before the damage becomes a total failure. When technicians use condition monitoring correctly, it can result in significant cost savings compared to traditional maintenance approaches. In traditional maintenance approaches using preventive maintenance, technicians would replace these faulty components on a fixed schedule. In a reactive-maintenance scenario, technicians would repair or replace these components only after they have reached total failure.

Read the entire article Vibration Analysis Avoids Equipment Failure by Dave Bertolini, People and Processes, Inc.

Defects occur at specific frequencies in relation to the running speed of the equipment. Most vibration analysis software will allow these specific frequencies (bands) to be measured and trended over time. Trending this information will help identify problems as they occur in your equipment. This results in more accurate analysis of equipment problems that will help determine the severity and repair urgency of the problems identified.

For example, if the vibration trend is increasing slowly, then the failure may not be progressing rapidly. However, a sharp increase in a specific vibration trend over time indicates that a defect may have developed and failure is more imminent.­

We often hear about the need to acquire vibration measurements at precisely the same location each time we measure a point. Why is this important? In order for vibration measurements to be trendable they must be closely repeatable, and we need to eliminate measurement error. With high-frequency measurements, the vibration attenuates rapidly as it travels away from its source. The author has seen readings vary by as much as 50% when the collection transducer was moved by as little as ¼ inch.  If the transducer is not placed in the same location, the trended data will show an error that may be mistaken for a change in machine condition. When the collection point is different, the transmission path is either longer or shorter. This affects the amount of energy perceived by the transducer. Standing waves also exist in vibrating machinery. The transducer may sometimes be located at a nodal point of one of these waves; and if care isn’t taken in transducer placement, the next measurement may be at an anti-node. This is more apparent in larger machines because of the amount of surface area available for standing wave formation. There are several ways to precisely mark data points for measurement with a magnet-mounted transducer. Paint, glue-on-pads, stud-mounted pads, machined surfaces, and dimples made by a small drill bit are all used with success. Best of all are permanently mounted coded attachment studs (such as VIBCODE®) that guarantee precise re-placement of the transducer every time, at the same location, angle, and pressure. Regardless of the method employed, it is important to always precisely identify data collection points.