Submitted to: Transactions of the ASAE
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/12/1999
Publication Date: N/A
Citation: Interpretive Summary: Research in automatically controlling the cotton ginning process has made great strides in recent years. Gin process control allows cotton cleaners and dryers in the gin to be controlled according to trash levels, moisture, and color of incoming seed cotton. Thus, a prescription sequence can be established to produce an optimal product and economic return. In the continual quest to identify new, useful evaluations for process control an improve sensing techniques, a sensor was developed to measure mass flow rate of lint cleaner waste. Lint cleaners are used in the gin to further clean lint of trash after the lint is seperated from cotton seed. Accurate determination of lint cleaner waste (trash) amounts can provide feedback on lint cleaner performance which is useful information for process control. The sensor uses an array of lights above a duct carrying lint cleaner waste and a light bar at the bottom to detect light output. Waste traveling through the duct interrupts the light beam and the amount of waste flow ca then be related to the amount of light blockage. The objective of this study was to evaluate the sensor by comparing its readings with known waste quantities collected in plastic containers. If the sensor was found to work well, equations relating the amount of waste flow to the amount of light blockage could be developed. Results from the tests clearly showed the device to be robust in the ginning environment and capable of providing distinct representations of lint cleaner waste using many combinations of lint cleaners. Equations were then developed so the sensor could give a direct reading of waste traveling through the duct. This device will meet the needs for improved gin process control and provide the information necessary to fine-tune the control process.
Technical Abstract: A sensor was developed to measure mass flow of lint cleaner waste in a small-scale (1 bale/hr) cotton gin to provide additional information on lint cleaner performance for automated gin process control. The device consisted of a light source bank and light bar at opposite sides of the waste duct. The amount of light blocked as detected by the light bar was proportional to mass flow. Experiments were conducted to evaluate the sensor by varying the number of lint cleaners activated. Statistical correlations between sensor output and output from photodiodes configured to detect temporally varying light levels were examined. Curves were developed relating measured mass flow to sensor output, consistent with expected sensor responses. A Weibull-type function was found to best describe the sensor response as a function of experimentally measured mass flow. A re-parameterized Weibull-type function was found to be suitable as sa predictor of mass flow rate from sensor output when the data included only gin motes (zero lint cleaners activated). When the zero lint cleaner treatment was not included in the data set, a linear relationship best predicted mass flowrate. Total mass of particulate was best predicted by a reciprocal log model.