ABSTRACT
Improvement of the measurement of strength in the high volume instrument (HVI) system by adjusting the measured strength for measured moisture content shows promise. In this study, 275 samples were pretreated in five different relative humidities before HVI and resistance moisture testing. The samples represented eight growth regions in the United States. Comparisons of several possible models of the strength including either the sample conditioning relative humidity or the measured moisture as an indicator of sample moisture content showed that the measured moisture was as good at controlling the effect of the pretreatment as was knowledge of which pretreatment was used. Small, but statistically significant differences in the strength variation by region of origin of the sample were noted, but the effects were minimal and were neglected in further analysis. Two selected models were used to predict a "corrected" strength for all of the data based on the measured strength and moisture content. Using the better model resulted in a standard error of prediction of 1.12 g/tex with sample moisture contents in the range 5.7% to 9.2% while the standard deviation of the strength data was estimated to be 1.11 g/tex. Both models yielded excellent prediction of the standard strength in the data used to choose the model coefficients, showing that the moisture correction of strength would allow greatly relaxed conditioning of samples for HVI strength measurement.
|