Printer Friendly

Health maintenance for machines?

Health maintenance for machines?

Health-maintenance programs for employees are the latest practical approach to controlling spiraling health-care costs. They reduce our need for serious medical attention by anticipating and controlling our self-destructive tendencies.

But health-maintenance for machine tools? Yes, the same principles can be applied on the shop floor --for the same reasons and with the same results. Here, based on information from IRD Mechanalysis Inc, Columbus, OH, is one approach to a "shop-floor HMO" for machine tools.

Three choices

There are three health-maintenance approaches to machine tools:

Breakdown maintenance. In this time-honored system, machines simply run until they fail or perform so poorly that they must be shut down because they are producing 100% scrap. Failure can be catastrophic, and at the very least, untimely; and the necessary replacement parts, tooling, or manpower are seldom readily available.

Scheduled maintenance. Like regular checkups, this program shuts down critical machine tools after specified periods of operation. The machines are then partially or completed disassembled for inspection and replacement of worn parts. This tack has the disadvantages that total dismantling is expensive and time-consuming, and the proper time interval is difficult to predict. Also, healthy machines may be needlessly repaired, degraded by too-frequent disassembly, or reassembled incorrectly.

Predictive maintenance. This approach combines real-time analysis and early detection to pinpoint when specific repairs are actually required. Shutdown can be scheduled at convenient times, and well before damage becomes severe. Repair times can be minimized, and manpower and replacement-part requirements accurately forecast.

Wide application range

Predictive Maintenance (PM) is being applied to a wide range of rotating equipment: machine tools, spindles, electric motors, V-belt drives, gear drives, fans and blowers, turbines, motor-generator sets, and rolling mills. Industries using PM include: metalworking, automotive, steel, tire, marine, mining, and chemical processing.

The key variable being monitored is vibration. All rotating equipment vibrates, and the degree of vibration (although often indiscernible to the human eye, ear, or touch) is a good indication of equipment condition. It can reveal component unbalance, wear, misalignment, bearing defects, and bent shafts.

System elements

The Mpulse[TM] system is a modular approach to fully networked hardware and software for PM data collection and analysis. It provides real-time untended acquisition of vibration data and other parameters from machinery throughout the plant.

Data from the shop floor is collected by a host PC from a layered set of hardware modules providing signal conditioning, data acquisition, and operator displays wherever they are needed. It includes:

Sensors. These include accelerometers, velocity probes, displacement probes, current and voltage sensors, and temperature devices permanently mounted on equipment. This can be supplemented by data manually collected with portable instruments.

Sensor personality cards. The sensors are linked to signal-conditioning boards called sensor personality cards that can accommodate up to 16 sensors. These cards include intelligence to detect transducer faults or cabling shorts or breaks.

Sensor stations. Up to four personality cards can be housed in sensor stations that protect them from the plant environment and up-link sensor data via multiplexing over a single, multi-wire cable to condition-surveillance processors.

Condition-surveillance processors (CSP). These are multi-channel real-time analyzers and data collectors located up to 300 ft from sensor stations. They are linked via coaxial cable (Ethernet) to the main workstation (PC), using TCP/IP protocol.

The CSP can measure acceleration, velocity, displacement, spike energy, rpm, DC temperature or pressure, machine on/off condition, and do amplitude-spectrum and frequency-band analysis.

Workstation. The workstation is an industrial MS-DOS personal computer. It contains a database-management system on a hard drive. Signal data flows to it on a regular daily schedule, yet runs independently of the CPSs so that if the PC loses power, data continues to be collected by the network. Once back on-line, the workstation interrogates each CSP to get fresh information.

Data analysis

What do you do with all this data? At the high end are critical machines whose failure would be rapid and catastrophic. These are continuously monitored and can be shut down in a matter of seconds if a serious problem is detected.

The majority of machines, however, are less critical and sampled less frequently. Area alarm panels provide continual surveillance to detect dangerous conditions and sound alarms. On-line analysis can then be used to view real-time signatures and compare them with historical patterns.

Trend analysis continually updates the database for display and on-going analysis. Expert analysis software provides predictive functions, automatically searching for machine problems and using vibrational analysis and bearing-problem expertise to pin-point upcoming maintenance requirements.

Some of the analytical capabilities include:

Acceleration. G-peak measurements of acceleration are typically used where frequencies exceed 60,000 Hz.

Velocity. RMS velocity measurements are used where vibrational frequencies are less than 60,000 Hz.

Displacement. Peak-to-peak displacements are measured under conditions of dynamic stress.

Signature analysis. Vibrational signatures are compared to baseline signatures to detect changes in the mix of frequencies. Fast Fourier Analysis (FFT) algorithms are used for decision-making, including digital techniques to break up the signature into specific samples for comparison.

Spike energy. High-frequency vibrations from rolling elements often have unique spike-like characteristics. Special circuits enhance these signals and suppress lower-frequency interference to identify these conditions.

Time waveforms. Some signals have characteristic amplitude-vs-time patterns, once nonharmonic noise components are filtered out, that can identify abnormal rotation conditions.
COPYRIGHT 1991 Nelson Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1991 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Quality Solutions
Publication:Tooling & Production
Date:Nov 1, 1991
Previous Article:Manufacturing process designed for the global market.
Next Article:Investments result in world-class stamping.

Related Articles
Maintenance: If it ain't broke, keep it from breaking!
Cut costs with predictive maintenance.
Maintenance... the business opportunity of the 1990s.
Maintenance software tailored for injection molders.
Molders benefit from new software for a host of shop needs.
Preventive maintenance: stand alone manufacturing compared with cellular manufacturing.
Smart manufacturing means `fix it before it's broke.'.
Drinking-water quality and issues associated with water vending machines in the city of Los Angeles.
Cincy Machine leads way with Smart Machine plan.
Rebuilds may make sense: when money's tight, buying new equipment isn't the only option.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters