Online vibration spectral data analysis helps mine keep critical equipment running

The ups and downs of monitoring: Canadian mine drum hoist raises the level of remote data analysis.

At a large potash mine in Esterhazy, Saskatchewan, Canada, a drum hoist, 19.5 ft. in diameter, rotating at 60 rpm, brings freshly mined material to the surface by raising and lowering two loading sleds.

The system flags identified faults, alerts the hoist control room via a mimic display, and sends an e-mail to the hoist engineer.

This is the only drum hoist at this mine location, and its performance is vital to the mine’s operation. Azima DLI’s online condition monitoring system is used to ensure the highest level of monitoring of the drum hoist’s condition. The SpriteMax platform continuously captures and diagnoses potential faults from the machine’s vibration spectral data.

Following each sample, the data is run through the SpriteMax integral automated diagnostic system for analysis. Deviations from the expected vibration spectral signature are passed through a number of fault models specific to the machine type and application. The system flags identified faults, alerts the hoist control room via a mimic display and sends an e-mail to the hoist engineer.

To ensure optimal performance from the drum hoist, the only one at this Canadian mine location, the machine’s vibration spectral data is use to capture and diagnose potential faults via an online condition monitoring system.
To ensure optimal performance from the drum hoist, the only one at this Canadian mine location, the machine’s vibration spectral data is use to capture and diagnose potential faults via an online condition monitoring system.

The cycle time for one sled to be raised from the bottom and the other sled to be lowered from the top is roughly 90 seconds. The sleds are loaded and unloaded simultaneously during the next 12 seconds, and then they begin the return cycle.

That 90-second cycle is broken into three sections. The first 25 seconds is the initial acceleration from rest to full speed. The sleds are at full speed for the next 40 seconds and then slow to a stop during the final 25 seconds.

To ensure the current spectra is being compared to the established average spectra — the expected pattern of frequencies and amplitudes — the same speed and load conditions need to be maintained for each test. Based on the 90-second cycle, the only interval that meets that repeatable condition is the 40-second stretch when the sleds are at full speed. A tachometer signal is read multiple times and compared to the previous reading to allow the SpriteMax to determine when the sleds are at steady speed.

The system determines if the hoist was operating in the clockwise or counterclockwise direction. Data acquisition begins after steady speed and rotation have been verified. Because of the slow rotation rate of 60 rpm, a long time waveform (7 seconds at 40,960 Hz) is captured first. This waveform also passes through the Fast Fourier Transform module to generate spectral data of 0 Hz to 500 Hz and 0 Hz to 12,000 Hz frequency spans.In addition to the time waveform and spectral graphs, the unit acquires a demodulated graph with a frequency range of 5,000 Hz to 10,000 Hz. This demodulated test is designed to detect the signal from rolling-element bearings during the early stages of wear and fatigue.

Before commissioning the SpriteMax system, the hoist engineers had to perform a battery of specialized monthly tests using handheld instruments. With SpriteMax as the early warning system tracking the health of the hoist bearings, the hoist engineers can focus their attention on other tasks. When the system alerts them to a change in bearing health, the engineers can observe the data or perform other specialized tests to determine the rate of degradation. They’re able to apply timely and planned corrective action, such as additional lubrication, or schedule repairs to coincide with planned hoist/cable maintenance.

Show Comments
Hide Comments

Join the discussion

We welcome your thoughtful comments.
All comments will display your user name.

Want to participate in the discussion?

Register for free

Log in for complete access.

Comments

No one has commented on this page yet.

RSS feed for comments on this page | RSS feed for all comments