Archive | July, 2003


3:16 am
July 2, 2003
Print Friendly

Maintenance Information Systems

Directory of EAM/CMMS software form maintenance and reliability organizations.

Enterprise asset management (EAM) and computerized maintenance management systems (CMMS) are essential to most maintenance and reliability strategies irrespective of plant size. Maintenance information systems run on multi-platforms using mainframe, client/server, thin client, or browser-based applications. Smaller, stand-alone systems run on PCs or local area networks. Because some powerful packages can run on a single PC or networked PCs without a midrange server, the dividing line between small and large systems has blurred. Therefore, we are including all software packages in one directory.

Buying decisions begin with an analysis of how a maintenance organization operates today and what its strategy is for the future. These systems can help organizations implement their strategy to decrease downtime and increase the utilization of their resources, and can be viewed as a communication tool to help make better decisions.

Using these approaches, maintenance personnel can access information and work orders in a number of ways—dedicated terminals and PCs, or mobile Palm-type personal digital assistants (PDAs) and handheld computers running Windows CE. Other wireless and radio frequency devices to access information are also at hand. Developments including e-commerce, supply chain integration, the Internet, and wireless technologies that first were implemented in larger plants also are benefiting smaller and midsize plants.

There has been an expansion in the number of companies that are offering an application service provider (ASP) option to their programs. Users pay a monthly per-seat fee to access the software through an Internet-enabled workstation. The ASP stores the program and the data on its server. Users always have access to the most current version of the program. This delivery method eliminates the need for on-site hardware infrastructure, system administration, and associated costs at the user’s end. Another variation of this method lets users access the program through the Internet but the data resides in their own plants.

To meet the needs of the increasing number of companies that recognize the benefits of electronic transactions, some software suppliers provide web-enabled systems that support e-procurement within their own program or allow users to integrate their EAM or CMMS system with other vendor software. Another growing area is connectivity with programs having the ability to integrate with other plant ERP business applications and production automation systems. MT

Continue Reading →


12:57 am
July 2, 2003
Print Friendly

Checking Best Practices for Preventive Maintenance

Example best practices and the questions you need to ask to determine if your plant is using them.

Visiting plants in different corners of the world, we often are asked: “What are the current best practices for preventive maintenance (PM)?” We usually answer that we define preventive maintenance using 95 key elements. We also point out, to some people’s dismay, that there is no single silver bullet for improving PM, but rather many combined efforts will be required to eventually yield results.

Here are a few key elements that have been extracted from our program of Current Best Practices (CBP) along with test questions and best practice (BP) examples to help you gauge how well your plant practices measure up.

Do you have a definition for preventive maintenance?
Interview test: Ask people in maintenance and operations to define what is included in preventive maintenance.
BP example: We have a definition of preventive maintenance that is documented, understood, and well communicated across our plant.

Having a definition of preventive maintenance is important for good communication in meetings, improvement efforts, and training seminars. For example, are detailed cleaning, balancing, and alignment part of preventive maintenance? Is operator inspection part of PM? Are operating practices part of PM?

We have often attended meetings or interviews where we are told a plant is continuously working on improving preventive maintenance. When we ask for the plant’s definition of PM, we notice that there are as many definitions of PM as there are people. How can we expect to improve PM if we are not clear on what PM really is? We define PM as essential care and condition monitoring (PM/ECCM) as shown in Fig. 1. Perhaps you can use the definition in your plant.

Do you know how satisfactory PM is done today?
Test: Ask the plant manager, maintenance manager, and operations manager for the PM improvement plan. If there is one, is it specific with timelines? For example: “Lubrication storage improvement complete by September 2003.”
BP: Plant management is aware of strengths and weaknesses of the PM program. The plant therefore has specific plans and timelines in place for improvement actions.

The experience of Ian Farrell, maintenance manager at UPM-Kymmene in Scotland, illustrates the awareness factor. Farrell, whose company has deployed the CBP education and training assessment in several plants in the United Kingdom, expressed the experience of interviewing people in the plants in a presentation at a recent maintenance conference.

“ We interviewed people in the plants to get a good idea of how well PM is done. When initially asking a person how well PM is done in the plant, the first answer is ‘Yes, we do this all the time.’ After some more discussion and specific questions around PM, the interviewee changes the statement to ‘Well, we probably do this most of the time.’ After more small talk and several cups of coffee and more explanations around PM the interviewee states, ‘I know we definitely, sometimes do it.’

“ The questions become more specific and the interviewee downgrades the statement to ‘I think we do it.’ Time passes and questions around, for example, alignment standards, condition monitoring routes, and operator involvement make the interviewee think of what good PM really is, and the statement is changed to ‘Somebody told me we did it.’ When we finally have defined what best practices in preventive maintenance are and there is a stack of coffee mugs, the person muses, ‘We used to do it all the time.’”

By first defining what PM is, and then educating and training people in the current state of their actual PM performance, the groundwork for improvement is laid.

Do you have an alignment standard, and is it followed?
Test: Ask for an alignment standard and check quality of standard. Go look at equipment for signs of good or poor alignment.
BP: There is a well-documented alignment standard. More importantly, the standard is followed.

In a world-class reliability and maintenance organization, all alignments are done to 0.002 in. (0.05 mm) for equipment running below 3600 rpm and 0.001 in. (0.025 mm) for equipment running above 3600 rpm. There is a well-defined alignment standard explaining how to set up, clean, check for pipe strain, check for soft foot, etc.

Take a tour of your plant. If alignment is done well there are jacking bolts (push bolts) installed on all motors, gears, and other equipment of significance. Bases and foundations are in good condition and no more than four shims are used under the motor feet (Fig. 2). Overall vibration level is low in the plant (0.1 in./sec unfiltered average). As a tracking indicator, see if alignment records are kept for each alignment job.

Do you have a lubrication standard, and it is followed?
Test: The standard should include storage, handling, filtering, and cleanliness of lubricants. Visually check cleanliness of storage areas and handling.
BP: There is a well-documented lubrication standard. More importantly, the standard is followed.

The cleanliness standard for each piece of equipment should match the clearances in the equipment’s lubricated surfaces. For example, a hydraulic unit may need to be filtered down to 3 microns (200 beta) and a gearbox to 12 microns (75 beta).

In order to reach the right cleanliness levels of lubricants, oil and grease have to be stored, handled, and filtered correctly. Few people know that new oil usually is delivered at around 40 microns cleanliness level, which means that oil going into equipment with fine clearances should be filtered.

Are inspections (condition monitoring) done where it is cost effective to do so?
Test: Go through inspection lists, check for level of detail, and make sure the route is actually completed.
BP: There are inspection routes for all mechanical, electrical, and instrumentation equipment (where it is cost effective to have inspections).

In a top-notch plant, inspections are documented and completed according to schedule. The plant is using an inspection list or, even better, a handheld computer. The list or handheld computer describes exactly what to do for each inspection. The inspections are a combination of measuring condition and subjective (look, listen, feel, smell) inspections.

Most inspections are completed while equipment is operating because we do not want to waste valuable shutdown/offline time on inspections that could be done on the run. Inspections can usually be done better when equipment is operating. For example, a pump cannot really be inspected well when it is down since there are no vibration, no operating pressures, and no seal water flow.

To see if your plant is performing according to world-class reliability and maintenance standards, take an inspection list, or handheld computer (if you do not have inspection lists, it is time to develop them), and walk the route. For example, check the following:

• Do we have condition monitoring routes covering all necessary inspections?
• Do we use simple inspection tools such as a stroboscope, infrared thermometer, vibration pen, industrial stethoscope, bright flashlights (500,000 candela), and inspection mirrors?
• Can we inspect couplings, belts, and chains on the run, or do guards make it impossible (Fig. 3)?
• Are inspections being done? Are oil glasses clean enough to see oil levels, are base bolts clean enough to check tightness,etc.?
• Are people educated and trained in basic inspection techniques?

Is detailed cleaning of equipment done well?
Test: Take a walk in your plant and visually check the cleanliness and condition of the equipment.
BP: Detailed cleaning of equipment is done consistently. Dirty areas are redesigned in order to protect equipment from contamination.

Detailed cleaning can be checked easily. For example, a clean hydraulic unit can be inspected for leaks in about 10 sec by taking a quick look at the pan underneath the unit (Fig. 4). A dirty hydraulic unit would take 20-30 min to check for leaks.

Is an ultrasonic or vibration monitor used when greasing bearings?
Test: Check lubricator’s equipment.
BP: Vibration or ultrasonic levels (or other method) are checked while greasing in order to apply the correct amount of grease.

Greasing is done by measuring ultrasonic or vibration levels while applying grease to the bearing. It is almost impossible to know how much grease is applied to a bearing without a measurement. The measurement tools indicate to us when the grease hits the bearings and monitor the vibration or ultrasonic levels as grease is squeezed into the bearing. Over and under greasing can be avoided by using the right tools. An alternate method is to use a volume meter, assuming the required grease volume for the bearing is known.

Although just a sample of the 95 points we use to evaluate plant performance, these example tests and best practices demonstrate the methodology by which one can build a system for discussing performance levels. MT

Tor Idhammar is partner and vice president of IDCON, Inc., reliability and maintenance consultants, 7200 Falls of Neuse Rd., Suite 200, Raleigh, NC 27615-5384; telephone (919) 847-8764

Preventive Maintenance/Essential Care and Condition Monitoring


Fig. 1. Essential Care (EC) includes preventing failure from occurring, with tasks such as detailed cleaning, lubrication, alignment, balancing, operating procedures, adjustments, and installation procedures. Fixed Time Maintenance (FTM) is all replacements that are done on a fixed schedule regardless of condition, e.g., programmed replacements and overhauls. Condition monitoring (CM) is all inspections from simple subjective look, listen, feel, smell inspections done by operators or crafts people to objective vibration analysis, oil sampling, ultrasonic leak detection, pressure checks, current readings, etc.
back to article

0703idfig2a 0703idfig2b

Fig. 2. Left: A motor has not been aligned properly. We can clearly see the beat marks from the sledgehammer, the motor is missing push bolts, and there are too many shims under the feet. Right: A motor with the correct set up to enable good alignment.
back to article


Fig. 3. This equipment guard allows for on-the-run visual inspection.
back to article

0703idfig4a 0703idfig4b

Fig. 4. The hydraulic unit on the left would take 20-30 min to inspect and identify leaks, while the hydraulic unit on the right can be inspected for leaks in about 10 sec by looking for oil in the pan underneath the unit.
back to article

Continue Reading →


12:48 am
July 2, 2003
Print Friendly

Backlog Interpretation and Analysis

Last in a series of articles discussing the management of backlog

The first article in this series, “Essential Elements of Backlog Measurement,” defined backlog as the classification of work that, for whatever reason, has not been completed. The second article, “Measuring Backlog,” presented various methods of measuring backlog in weeks.

However, two questions remain. What does backlog mean? And, how can it be used?

There are any number of meanings, real and implied, relating to backlog. Primarily it is used as a scheduling tool for workload management and balancing. It also can be used to justify decisions concerning changes in organizational work force and budgeting.

Maintaining backlog validity
Before using backlog for any of these purposes, its validity should be ensured. Backlog requires regular maintenance to prove its validity. This maintenance includes sifting through the work to determine if jobs are feasible future work, are duplicated, require additional planning or materials, or require changes in their priority, and if the backlog provides sufficient future required work.

If work exists in the backlog that is not valid, it should either be adjusted to make it valid, such as changing the status or priority of individual jobs, or be canceled and removed from the backlog.

Determining healthy backlog
It is important to have an idea of what a healthy amount of backlog for an organization is. Opinions vary, but it is generally accepted that two to four weeks of backlog is a healthy figure of future work. This provides a good amount of scheduling capability to the organization. Also, it provides for balanced workloads as well as good facility management, where most, if not all, of the required work is completed for maintaining the facility.

If the backlog consistently falls below the two-week minimum, there is not enough work available to provide good scheduling of the labor force and leaves the organization open to running out of work to be done.

At first it might seem that this is a good thing. But if a time comes when all personnel cannot be used, money is being spent without a productive result. If this occurs often, labor resources may require adjustment.

Likewise, if the backlog is consistently above the healthy limit, the organization has more work than can be reasonably handled by the present work force. If the backlog validity is maintained, the effect of excessive backlog is higher overtime use and higher cancellation of valid or required work.

This is cause for concern on many points. First, higher overtime use generally means that the company is paying 50 percent more for those labor hours that are presumably only about 70 percent productive, due mostly to worker fatigue and decreased morale. Second, the cancellation of valid and/or required work is inherently hazardous for the facility. If an organization is so overworked that PMs go unperformed and are cancelled due to lack of resources or, in the case of a daily or weekly PM, duplication of work, then a higher possibility exists that a particular PM will not be performed for long periods, increasing the risk of breakage and subsequent corrective maintenance work.

Managing excessive backlog
There are ways to manage the backlog in this situation. One way would be to evaluate the preventive maintenance program to determine if it is excessive. It could be that too much PM is being performed and some daily work actually should be done weekly and so on. Also, evaluate job plans to determine if any changes are required to the processes involved or if duration estimates require alteration.

This contributes to validity as well. It is important to note that while this is an option to correct excessive backlog and improve validity, it should not subtract from the purpose of the PM program—to prevent a breakage.

Another avenue of managing excessive backlog involves evaluating corrective work in the backlog. As a facility ages, some corrective work will show up in patterns. If these patterns are identified, some care can be taken to more closely examine the cause of the problem and either fix/replace it directly or add it to the PM program, thereby avoiding a costly repair.

The final avenue of managing excessive backlog comes from the acquisition of additional resources. More manpower is a sure way to reduce the effects of excessive backlog. This reduces the strain on workers, budgets, and facilities by reducing the need for high overtime, reducing cancellations of required PM work, and reducing the backlog to a manageable level.

Mentioning that backlog can be used as an element to work force adjustments tends to perk up ears. There will be times when supervisors and managers will have to justify present or requested resources. Backlog, along with other contributing elements, is an important tool for the decision making process.

Any one of these techniques can produce positive results in managing excessive backlog but a combination is more likely the final answer to resolving and avoiding the problem altogether.

Elements of change
When considering a situation involving backlog work and the desire to examine work force requirements, it is important to use a wide base of data. In the example, figures are based on the prior year of data concerning patterns, averages, and summaries.

A quick and dirty method of determining labor resource requirements based on the backlog is to first decide what is an ideal amount of backlog weeks to be maintained in the organization. As stated before, two to four weeks seems to be the ideal range to maintain.

After deciding the backlog weeks, apply this figure to both the average backlog hours (3590 hr in this example) and the technician credit hour capability (34.15 hr/week). See Fig. 1. There are 10 technicians available for work in this shop.

However, there are some key elements in addition to backlog that are required to receive a proper picture of labor resource requirements. Typically, they include workload distribution, overtime usage, cancellations, and workload projections.

Calculating work force requirements involves some of the same operations as applied in the weeks measurement calculation presented in the second article. First, identify projected preventive maintenance hours—for the next year is best.

Projecting PM hours is a simple operation and is accomplished by examining the PM schedule and taking note of frequency, frequency unit, and job plans or labor estimates. Once this information is obtained, find out how many work orders will be performed for this PM schedule.

If, for example, a monthly PM used for lighting inspections has 4 hr in estimated time attached, this PM would generate 12 work orders throughout the year that would total around 48 hr of labor time.

Once the total amount of PM work for the next year has been calculated, the annual shop credit hours capability should be determined (Fig. 2a).

Then, the projected annual workload should be figured. In this calculation, it is important to know what percentage of the expended labor hours for the prior year fit with each work type (Fig. 2b).

Next, by subtracting the annual capabilities from the projected annual workload the amount of the projected labor hours in excess of the shop’s capability can be determined (Fig. 3a).

Finally, the amount of additional labor resources and their effect on the backlog can be calculated (Fig. 3b).

Now it comes to money
Why bother going through all these calculations just to figure the amount of resources a shop would require? Money.

It costs money to employ a labor force to maintain a facility. It also costs money if that facility is neglected through cancellation of required maintenance. The question with this is: what are the options and which is most cost effective?

Option 1 is to continue cancelling work that resources are not available to complete. This is not good because the health of the facility is the primary concern. Option 2 is to continue working with the present work force and absorb the excess work in overtime labor (Fig. 4).

Overtime labor is not especially productive mainly due to worker fatigue and morale compared to regular labor time. It is also very expensive compared to completing the same work with regular time employees.

Option 3 is to add labor resources to the shop. This can accomplish two things. First, the facility can be effectively maintained with less risk of corrective maintenance due to facility failures. Second, the addition of these resources accomplishes the excess work without the need for excessive overtime labor (Fig. 5).

In a world where budget cutbacks often affect a facility’s support group before other “more critical” areas, facility managers are increasingly being told to do more with less. Backlog is the key to success in managing these priorities, sometimes in unconventional ways.

The subject of backlog is complicated. Coupled with other items, backlog can be a guide to operations and capabilities. It is a powerful tool for facility managers that does not give up answers easily or completely unless the manager can fully understand what it is and how it is measured, and can manage it effectively to maintain the validity of the data. MT

Jason Aughenbaugh is a business analyst with Amgen, Inc., MS 21-2-A, One Amgen Center Dr., Thousand Oaks, CA 91320

Fig. 1. Labor resource requirements for average backlog
back to article

Fig. 2a. Projected annual capabilities
back to article

Fig. 2b. Projected annual workload
back to article

Fig. 3a. Projected excess workload
back to article

Fig. 3b. Labor resource requirements
back to article

Fig. 4. Cost of overtime labor for excess work
back to article

Fig. 5
back to article

Continue Reading →


6:51 pm
July 1, 2003
Print Friendly

Start from The Beginning


Robert C. Baldwin, CMRP, Editor

This month we present our annual roundup of enterprise asset management (EAM) and computerized maintenance management system (CMMS) offerings from active suppliers in the industry. We provide some basic information about each, as a starting point for selecting your first system or for moving up to a larger system with more functions.

Now comes the real trick—selecting the system that will provide the most return. That task is not unique to maintenance and reliability.

Robert D. Boyle, a director in Deere & Co.’s Global AgServices Div., had some interesting observations about software selection and implementation in his article on “Unlocking ROI” published in the June 2003 issue of APICS-The Performance Advantage, the publication of APICS–The Educational Society for Resource Management. I was especially struck by the following paragraphs:

“ … an organization should not invest in ERP technology on a standalone basis. Instead, investments in the technology should be part of a holistic approach to effecting organizational improvement, and it should involve a comprehensive review of business goals and metrics, processes and procedures, human resources and training requirements, technology, and so forth.

“ For example, merely implementing new ERP technology without aligning affected business processes and procedures and providing staff the proper training is not likely to produce favorable results. The costs and benefits outlined in the business case associated with such a program should be based on this broad perspective on organizational change. Planning and executing comprehensive change initiatives, as opposed to more narrowly defined ERP projects, is the way for an organization to get the most out of the time, dollars, and other resources it invests in such initiatives.”

Substitute EAM or CMMS for the ERP and it works for maintenance and reliability. And you get a clue as to why so many installations of maintenance software never get used to near capacity and never provide the projected return. Software selection and implementation is undertaken as a project, an end in itself, rather than to support a well-thought-out business process driving toward maintenance excellence.

In many regards, selecting the software is one of the last steps in the process of improving your maintenance and reliability operations. The first step is to develop a strategy for how best to do maintenance and reliability to support enterprise objectives.

Once you have the plan, selecting the right software will be much more straightforward. MT


Continue Reading →


6:49 pm
July 1, 2003
Print Friendly

The Science of Profit Centered Maintenance

In the past few years, there has been a growing focus on the business side of maintenance. In a few circles, the terms “profit centered” and “profit centric maintenance” have been coined to suggest a frame of reference slanted toward maximizing revenue through improved plant reliability while managing maintenance costs. I suggest to you that, as professionals, we must not only embrace these concepts but also expand our tool kits to apply these concepts as a management science.

In my career as a maintenance technologist and manager, I have encountered some companies who treat the maintaining function as a prime driver of company revenues and profits. I have worked with many more who attempt to manage for an acceptable level of equipment failure with minimum maintenance costs.

In each situation, there is a prevailing management philosophy with anecdotal evidence and logic behind that chosen philosophy. The science driving decisions is limited.

In the best situations, to justify the cost of improvements to reduce expense or increase reliability, performance targets are established based on benchmarks or intuition. The value of those targets is quantified to create a business justification for the changes. In the worst scenarios, changes are proposed but implemented with minimal success because of weak justification and promotion relegated to philosophical arguments and war stories. In addition, the behavioral science we apply in our change processes is frequently antiquated.

Even at professional conferences, I see many presenters, both practitioners and consultants, seemingly preaching a preferred approach to maintenance management through an almost endless array of parables. While these sessions are often entertaining, the religions occasionally conflict and provide little management science to optimize plant profitability.

In my opinion, our profession’s focus in the past several years on best practices and benchmarks has done a great deal to align our vocabularies and give us a qualitative frame of reference to compare operations and maintenance processes. I believe it is now time we advance our expertise and professionalism with an integrated set of quantitative management models and situational standards that will promote real optimization.

Like a chemical engineer who can define and control the interrelationships among chemicals and process variables to maximize product yield and throughput, we must more clearly identify and quantify the linkages among maintenance processes to optimize plant reliability, capital productivity, and work force effectiveness.

Although there are many mathematical tools to help the maintenance professional, they are not well known nor do they deal effectively with many of our more common challenges. For example, we need to be able to map a predictive relationship between planning, schedule compliance, and craft skills with true technician productivity. We need a generalized quantitative, statistical understanding of the impact that equipment condition has on reliability, and plant availability specifically. We need to agree on a few common derivatives of the EOQ (economic order quantity) model to set optimal spares levels instead of trial and error or hunches.

We clearly need more robust approaches to forecasting the economic value of work process changes including the likely rate of change after factoring in behavior management issues. Similarly, we need an overall change management model that incorporates the latest practices in industrial psychology.

These are but a few of the quantitative tools we need to move our profession from experiential, anecdotal, and sometimes superstition to a management science.

It is the job of professional organizations and educational institutions to help take us to the next level. But they will not do so without pressure and support. I challenge professional practitioners and maintenance managers to be proactive, to contact the professional organizations, educational institutions, and information providers that they work with, and push for action that will make our profession more profit centered through the use of quantitative decision models and management science. Or, let me know your thoughts and needs and I will do what I can to get them into the hands of appropriate leaders. MT

Jim Humphries, vice president of performance technology for Fluor Enterprises, Greenville, SC, has spent over 25 years in operations and maintenance. He is a licensed professional engineer (P.E.) in South Carolina and Virginia, a certified plant engineer (AIPE), and a certified systems integrator (AIIE).

Continue Reading →


6:46 pm
July 1, 2003
Print Friendly

Listening to the Internet

There are just not enough hours in a day to read all the maintenance information you can find on the Internet.

At alone, you can find at least 5 years of collective wisdom captured in articles, editorials, and directories. You can bookmark them, save them to a file, print them, and even e-mail them. As you read this column, even more information is being added.

In an effort to leverage the multimedia capabilities of the Internet and to provide an alternative to reading, a new site has been launched., is a streaming web-based talk radio show.

You now can hear leading experts such as Joel Levitt, Ricky Smith, and Darren Clark talk about maintenance 24 hours per day, on demand, at your convenience. Download the free Live365 player (155 K) for the easiest set up.

The advantage of streaming audio files rather than providing a direct file download is that the stream starts to play as soon as enough information has been downloaded and the rest of the large file downloads in the background while you listen. The format of the audio file is MP3, a common audio format for the web due to its small file size.

Other sites that offer listening material over the Internet include with over 18,000 audio books and audio programs in MP3 format. You can even listen to magazines such as Fast Company, Scientific American, Technology Review, Science News, and the Harvard Business Review.

Audio Books for Free offers classic fiction for the most part but a search of nonfiction revealed the classic book of strategy Art of War by Sun Tzu and other interesting titles. I wonder if Sun Tzu could have imagined his words being downloaded via MP3 2400 years after he spoke them?

Another good site is It offers business MP3 files that play on your computer or portable MP3 player.

If you are a music fan, you probably have heard of MP3 download sites such as Napster (now defunct) and These free sites are questionable at best from a legal standpoint and almost all of the free music sites download a terrible spy-ware application that tracks your Internet usage on behalf of e-marketing companies.

Luckily there are more music alternatives now and most are free or low cost. My personal favorites are National Public Radio for news and for music.

Launchcast allows you to build your own custom radio station by rating songs and artists as you listen. You then can send your radio station link to friends who can build their own radio station. You can even let your friends “influence” your radio station if you like their taste in music. You can click on a red X if they play an artist or song you do not like and you will never hear it again. The basic service is free and you can get a commercial-free version for under $40 per year.

Apple Computer offers over 200,000 songs at its new iTunes Music store. You can preview any of the songs for free. When you find a song you want, buy it for just 99 cents.

If you have moved to a new town and want to hear your favorite old radio stations from back home, try Radio-Locator, the most comprehensive radio station search engine on the Internet. It has links to over 10,000 radio station web pages and over 2500 audio streams from radio stations in the U.S. and around the world. The Internet defies distance so your far-away radio stations can be delivered right to your PC desktop.

Don’t have time to read your boss’s latest memo? Now you can download a cool application that will read it to you. You can choose a male or female voice and convert the memo to MP3 for listening during your morning jog or transcontinental air trip. Visit Zero2000 Software for 2nd Speech Center ($39.95).

Try mp3 file

.com/”> for the latest audio software and information about various MP3 players and get plugged in to the audible side of the Internet today to hear what you have been missing.

Internet Tip: Watch Google

As much as I like to use the Google search engine, recent privacy concerns have made me a little more thoughtful when I do use it.

According to, the popular search site keeps track of every search you perform. It identifies you with a unique ID. We are all under heightened alert since 9/11 but I am not sure I want Google (a private company) tracking all of my activity.

The folks at Google-watch may be slightly paranoid, but they got my attention. Try the anonymous proxy that allows you to use Google without being tracked here: MT

Continue Reading →


3:09 pm
July 1, 2003
Print Friendly

Avoiding Flashover in Medium Voltage Switchgear

Corona is a serious issue in metal clad switchgear because of its highly destructive nature—it is the cause of most flashovers in the equipment.

The fundamental cause is an electrical breakdown of air brought on mostly by an insulating deficiency within the switchgear. Corona results from the ionizing of gases because of this high electrical stress. The stress is often brought on by contamination and poor insulation values.

If the situation is not rectified, a flashover is imminent, possibly causing enormous damage and personal injury.

Tests have proven that when a flashover occurs, the temperature exceeds 15,000 deg. Most of the damage is caused by explosions from the buildup of unburned gases within the enclosure. Corona discharge is unpredictable and caution must be used at all times.

Applying current technologies
All electrical thermographers brace themselves for possible surprises when entering electrical equipment. They know they should always expect the unexpected. Prior to opening any equipment for a visual inspection, ultrasound detection and sometimes infrared imaging are necessary.

Because of corona byproducts, nitric acid and carbon produced from the corona discharge, a continuous decay of insulation takes place which makes the situation highly unpredictable.

There are various signs that identify corona. Some methods of nondestructive testing show no indication of corona, but closer visual inspection may show that corona discharge had been present, but is not active (see Fig. 1 and Fig. 2).

In both images, flashover was imminent if the problem was not rectified. Even though the corona was dormant at the time of inspection, it would have started discharging again with the appropriate relative air density, temperature, moisture, and lack of air movement. In this case the deterioration is on the conductor insulation and corrosion on the metal clad switchgear.

In other cases, the corona discharge may be extremely active. Extra caution should be taken as a flashover could occur at any time. Corona also produces airborne ultrasound and ultraviolet light, and, if advanced into arcing, it will sometimes show a thermal signature. This is a perfect opportunity for ultrasonic detection, ultraviolet corona imaging, and infrared thermal imaging to find out where and why the corona is discharging (Fig. 3).

The problem in this case is decay in the horizontal insulators causing electrical stress to initiate the corona. This starts with contamination entering the switchgear, trapping moisture, and then, under the right conditions, producing corona. With nitric acid, ultraviolet light, ozone, and carbon deteriorating the insulation, the corona eventually spreads on its own.

At times the environment changes inside the metal clad compartment when a door is opened or a cover is removed. This on occasion stops the corona discharge which makes it more difficult to verify. Another byproduct of corona is ozone, and the odor can be identified.

Sometimes corona discharge events show up well in a thermal image as shown in Fig. 4.

Active corona in Fig. 4 shows a 4160 V junction box which is very close to flashover on three phases and had to be rebuilt because the corona damage was too extensive. The activity was temporarily stopped with cleaning which kept it energized until a scheduled shutdown.

Use multiple tools
There is no one tool that will verify all corona problems all the time because corona discharge is erratic and unpredictable, especially in the early stages. An ultrasound detector can verify if airborne ultrasound is present from the corona discharge. An infrared camera is used to verify if there is any temperature rise from corona discharge. The ultraviolet corona camera is the ultimate tool for verifying corona discharge.

If the discharge is dormant, then none of this equipment will indicate or verify anything. A visual inspection may indicate there was discharge previously—an indication that it will return.

There are times when corona is quite obvious without using any test equipment. Fig. 5 shows severe corona in a 15 kV interrupter switch. The green material on the buss bar is corrosion from active corona and the black material on the bakelite insulation is carbon residue from active corona. This particular switch was very close to flashover.

Fig. 6 shows the same location after it was cleaned. Corona activity was stopped temporarily to buy time for a scheduled power outage. The problem was the red bakelite divider was tight against the 15 kV buss and was starting to short out. The problem was resolved by cutting the insulating divider a few inches back from the buss.

Corona does not always leave a thermal signature. It creates little or no heat, but as it progresses into arcing and tracking, a thermal signature can sometimes be found. Corona is most intense in the 300-400 nanometer range in the ultraviolet spectrum. When it exceeds the 400 nm range, corona can be seen in the visual spectrum if there is no interference from background light.

Fig. 7 shows different scenarios of active corona at 5 kV. The left and right columns show three different spectra. At the top in the ultraviolet spectrum using the corona camera, corona is evident. In the middle, nothing can be seen in the visible spectrum. At the bottom in the infrared spectrum using thermal imaging, there is no thermal signature where the corona activity is present.

Using infrared imaging, corona imaging, and airborne ultrasound detection, problem areas can be found before they become a health and safety risk and a serious financial burden. Cleaning and maintaining electrical equipment under live conditions has proven to be beneficial for everyone. MT

Chuck Humphrey is general manager, Highvec Canada, Inc., P.O. Box 1421, Timmins, ON P4N 7N2; (705) 268-6011

0703flash1 0703flash2
Figs. 1 and 2. Insulation decay phase to phase on a 5 kV feeder line (left) and air flow in the dust during corona activity on a 4160 V conductor (right).
back to article

Fig. 3. A thermogram showed a temperature rise on the 15 kV buss insulators, verified by the ultraviolet corona images.
back to article

Fig. 4. A 4160 V junction box which is very close to flashover on three phases.
back to article

0703flash5 0703flash6
Figs. 5 and 6. Interrupter switch before (left) and after (right) cleaning to temporarily stop corona activity.
back to article

Fig. 7. Three scenarios of active corona at 5 kV.
back to article

Continue Reading →