The Smart Approach to Predictive Maintenance

Alan R. Bessen, P.E.

Introduction

A growing number of aggregate producers are using Predictive Maintenance (PdM) technologies such as vibration analysis or infrared thermography. Within the scope of a comprehensive maintenance program these technologies can be highly effective. However, by themselves, they bring absolutely no value. It is the actions that result from condition monitoring that provide value.

In an aggregates operation most of this value will come from increasing the reliability and efficiency of equipment. If a plant is running 90 to 95 percent up time, reliability has already been achieved. The primary benefit from PdM technologies will be in maintaining reliability by verifying preventive maintenance, equipment replacement and planned overhaul schedules. If a plant is running at 85 percent up time or less there are 10’s or even 100’s of thousands of dollars available each year in potential cost reductions through improving the reliability of equipment.

Predictive maintenance is one of the four basic levels of maintenance common throughout industry. The four levels are:

· Reactive - “Fix it when it breaks”: This is part of every maintenance philosophy because you can never eliminate or predict all failures. However, relying on this level of maintenance leaves little control over cost of repair and lost production.

· Preventive - “Time-based action”: Visual inspections, filter changes, oil changes, overhauls, etc. are based on hours of operation or tons produced. This is a crucial part of any program and provides opportunity to perform maintenance and repair with minimal production loss.

· Predictive - “Condition-based analysis”: Equipment is monitored during actual operation. Technologies used include vibration analysis, oil analysis, infrared thermography and ultrasonics. Data is collected, trended and used to predict component life. The information is often used to establish Preventive Maintenance schedules.

· Proactive - “Root cause analysis”: Failures are analyzed to define base level cause and effort is made to eliminate future failures by changing design, specification or procedure.

Good Maintenance Starts at the Top

The best maintenance programs include all four maintenance levels and are supported by a corporate-wide commitment to ensure the reliability of equipment. Since maintenance and repair is seldom done by corporate-level executives, program ownership must reside with those who are doing the work. This is particularly true for predictive maintenance since its function is primarily to provide information on the condition of equipment. If nothing is done with the data, it is useless.

Technology can enhance a maintenance program but will never replace good planning and a competent crew. However, the regular application of vibration analysis and thermography demonstrates a commitment that, from my experience, will cause the maintenance attitude within many crews to shift away from a reactive maintenance approach to a preventive / predictive approach. This increases the probability that the necessary actions required to improve equipment reliability are willingly performed.

What is the Technology?

Predictive maintenance is the most technologically advanced element in a maintenance program. It uses microprocessor-based technology to evaluate the condition of operating equipment. Two of the most frequently used methods are vibration analysis and infrared thermography.

Vibration analysis shows bearing faults, misalignments and imbalances in mechanical equipment. Vibration data will not only identify a problem, but can accurately indicate the type and severity of problems in bearings, gear reducers and other rotating equipment. Vibration monitoring equipment can also be used to measure the structural competence and effectiveness of vibrating screens.

Infrared thermography detects heat related faults such as loose connections and imbalanced phases in electrical equipment by measuring and displaying temperature variations.

Information derived from a predictive maintenance evaluation provides current data on equipment condition. The data is trended and used to measure changes in condition, set alarm limits and enhance preventive maintenance services.

Implementing the Technology

The relatively small size of typical aggregates plants, and the normal lack of technologically skilled maintenance personnel make it impractical to initiate a predictive maintenance program entirely in-house. In fact, rapidly changing technology and the high cost of hardware, software and training make the use of a contract service specialist imperative at the beginning of most programs.

Hardware and software for vibration analysis costs approximately $30,000, infrared thermography hardware and software can easily exceed $50,000. The expense of acquiring and maintaining PdM equipment makes it desirable to share the initial purchase cost over multiple locations.

Contracting PdM services will cost between $2,000 and $3,000 per evaluation for a typical 3,000,000 ton per year aggregates plant. A plant already operating at over 90 percent up time may benefit from one or two general evaluations per year. Plants operating between 70 and 85 percent up time may benefit from three or more visits. Annual contract service costs for a typical 3,000,000 ton per year aggregates plant should not exceed $15,000 unless special problems requiring frequent monitoring occur.

To ensure local ownership of the PdM process, it may eventually be desirable to train and equip local maintenance employees to become basic level vibration analysts. A basic level analyst would be capable of acquiring data and deciding whether or not a problem exists. This can be accomplished aggressively in 4 to 6 weeks using alarm set points within the software, an experienced technician for support, and a trainee with a strong mechanical aptitude, some computer competence and a desire to learn.

Predictive Maintenance Examples

The following are typical examples from actual operating quarries illustrating the potential for avoiding lost production and reducing unnecessary repair costs.

Tertiary Screen - The screen’s poor performance was directly affecting quality and capacity. Vibration assessment of the screen’s operation determined that the springs had deteriorated dampening its motion. Replacement of the springs returned the screen to normal operation and allowed management to cancel plans for a planned overhaul at a cost of $25,000.

Dredge - Intermittent electrical trips shut down operations on average of three hours per day for over two weeks. Basic electrical troubleshooting was unable to identify the problem because of its unpredictable nature. Infrared thermography revealed loose connections on multiple control fuses. Repairing the connections solved the problem. Estimated cost of lost production was $30,000.

Primary Crusher - Lack of lubricant caused the west side bearing on a primary impactor rotor at a major quarry to fail during a time of low inventory and peak demand. Some shaft damage occurred but the bearing was changed with hopes of making it through the production season. Immediately following the repair, the east side bearing began running hot. Quarry management began planning a full rotor change out requiring four down days at a production and sales loss of over 50,000 tons. Vibration analysis of the east side bearing showed no damage, high temperature was caused by over greasing. Both bearings were monitored frequently for the balance of the season and the 50,000 ton production / sales loss was avoided. Estimated value of the avoided loss was $80,000.

The Bottom Line

Incorporating new technology into your maintenance program will not magically improve plant performance. Monitoring and recording equipment condition does not make it run better. The key to any successful maintenance program is to focus employees on taking actions and adopting practices that will improve equipment reliability. In this setting vibration analysis and infrared thermography are properly defined as useful tools not solutions themselves.

Predictive Maintenance methods have proven effective in other industries, as we learn to properly apply them in the aggregates industry we will be able to redefine expectations for reliability, improving productivity and increasing profitability.