Field Evaluation of Infrared Moisture Meters: A Progress Report

W.Stanley Anthony and Lionel Gervais


Two infrared type moisture meters manufactured by Moisture Systems Corp. were installed in a commercial gin at Burdette, MS, during the 1988 and 1989 ginning seasons In 1988, one meter was installed at the feed control and another at the holding chute between the conveyor-distributor and an extractor-feeder. In 1989, one meter was again installed above the extractor-feeder while the other was initially installed in a special device that extracted lint from a lint flue, analyzed it, and returned the lint to the lint stream. That meter was subsequently moved to the feed control in order to give the ginner more information concerning the moisture content of the cotton as it came into the ginning system. Initial calibration of the meters was achieved by placing samples of known moisture content in front of the moister meters and taking successive readings. The samples were originally conditioned in chambers above salt solutions in order to yield known relative humidities. For dynamic calibration, both meters were installed at the feed control in the Microgin at Stoneville and conditioned, then 100 pound lots of cotton were fed continuously in front of the meters and about 100 samples were removed by hand immediately after the cotton had been viewed by the moisture meters. The seed cotton was ginned and oven moisture tests were used on the lint as the reference for the samples.

The meters were then installed in the commercial gin using specially built support structures. Small changes in the meter calibrations ere make to account for differences in the mounting locations. In 1988, reference samples were taken periodically throughout the 10 weed ginning season, but no attempt was made to change the calibration of the meters. The meters performed will without maintenance, but the data suggested that improvements could be made in the accuracy of the measurements. This appeared to be as a result of the gross differences in the surface density of the sample.

In 1989, one of the infrared moisture meters was replaced by a more advanced version with dual wavelength filters in order to compensate for the density variations (Quadra-Beam model 60). The other meter was the sam one that was used in 1988. Reference oven moisture samples were collected from the cotton during gin processing throughout the ginning season and changes were made to the calibration in order to improve accuracy of the meters. comparison of individual moisture samples during ginning to individual meter yielded moisture variations of ± 1%. Integrating and averaging 10 successive samples coupled with electronic gating to disregard measurements when insufficient cotton was present produced a significant correlation (coefficient of determination) of 73% with typical variations of ± 0.3% for one location while the other produced a correlation of 0.78% with typical deviations of 0.6%.

During the approximate 10-12 week ginning season in which 15,000 to 18,000 bales ginned each year, the meters perform without any maintenance problems. In order to enhance utilization of the infrared type moisture meters in moisture meters in commercial ginning systems, simplified procedures must be developed for initial calibration as well as subsequent verification of the validity of the instruments.

Reprinted from 1990 Proceedings: Beltwide Cotton Production Research Conferences pg. 106
©National Cotton Council, Memphis TN

[Main TOC] | [TOC] | [TOC by Section] | [Search] | [Help]
Previous Page [Previous] [Next] Next Page
Document last modified Sunday, Dec 6 1998