Coal Preparation » General
A previous project developed a method using a portable PC, a triaxial accelerometer and suitable software to determine the stroke and frequency of a vibrating screen and to display the three dimensional motion of the screen on a rotatable and resizable XYZ graph. Screen and panel manufacturers who observed this equipment in operation expressed a strong interest in having a portable device with the same capabilities.
A brief literature search showed that many vibration instruments were available, however the majority were designed to measure bearing vibration or shaft movement for maintenance purposes only. Only two manufacturers SKF (UK) with their Copperhead range and Br?el & Kj?r, Schenck's Condition Monitoring Systems were specifically designed as a screen fault detection system. These systems concentrate on bearing condition monitoring but can also measure the overall screen vibration and can be alarmed for cut-off when the vibration is outside of set ranges. These devices while measuring vibration do not output frequency, stroke or attempt to quantify the screen motion.
This project's objective was to produce a prototype robust portable, hand-held measurement system capable of measuring, displaying and recording the screen motion in terms of frequency, stroke and three-dimensional movement. For reasons of flexibility and ease use it was decided to use off the shelf components for the construction of the accelerometer package and a wireless (Bluetooth?) enabled IPAQ? pocket pc for the data analysis and display.
The unit was tested on CSIRO's pilot banana screen located at the Queensland Centre for Advanced Technologies and at two plants in the Hunter Valley region.
Measurement of screen frequency was carried out between 6Hz and 17Hz and compared to the frequency as determined using a strobe. The instrument measurement of frequency was found to differ from the measurements using the strobe by approximately 0.2% relative at the higher frequencies and approximately 10% relative at the lower frequencies. Inaccuracies in the measurement of the frequency using the strobe were noted as it became increasingly difficult to accurately measure the point at which all movement appeared stopped at low frequency. With this in mind the actual measurement of frequency using the prototype instrument may be more accurate than the data suggests.
The prototype instrument was found to be very sensitive to the measurement of stroke (STD. DEV. 0.2% relative). This degree of sensitivity suggests that the measurement of stroke may be one of the variables capable of indicating mass rates on operating screens. Limited experiments were carried out to determine the capability of the instrument to measure unbalanced screens, as would be the case with a weakened spring or uneven feed distribution. Major differences between normal and out of balance operation were observed with a 80kg weight placed on the front and rear spring support tabs respectively.
During the plant evaluation ten screens were examined for stroke and frequency during operation. These included desliming, product and reject drain and rinse and high frequency screens. The unit could be attached to any part of the screen, a reading taken and the data saved within one minute. No communication problems were encountered within the plant environment.