No abstract

A new method to evaluate the cut resistance of protective glove materials has been developed. The method consists of sliding a straight blade on a sample material and determining the horizontal blade displacement required to cut the material at a given load. A special arrangement allows a load applied on the sample material to be kept constant throughout the test. The effect on the cutting results of a) degradation in blade sharpness, b) blade speed, c) sample holder geometry (semi-circular or flat), and d) the load applied was characterized. The results demonstrated that a) the blade edge degraded even with soft materials such as neoprene; b) the blade speed had a negligible effect on the cutting results; c) the holder's geometry does not have a significant effect on the variability in the cutting results, but the use of a semi-circular holder is recommended; and d) the applied load is a function of the material's cut-resistance. It was found that the blade displacement increases non-linearly with the applied load. The cutting test conditions were set as follows: the blade is 70 mm long and its speed is 150 mm/min; the blade edge must be used only once. To evaluate the resistance of a material to cutting, tests must be performed with at least two different loads. The load required to cut a material to a standardized distance, namely 10 mm, is calculated by interpolating the experimental values. However, in the preferred method, the cut test is performed with three or four different loads and the load required to cut to 10 mm is calculated by a non-linear regression. Results are reproducible with a coefficient of variation (CV) lower than 16% with homogeneous materials. However, higher CV's are obtained with knitted materials such as Kevlar gloves, or steel reinforced materials.

Two methods for evaluating the cut resistance of gloves were compared: one developed at ITF-Lyon (Institut Textile de France) and one developed at the IRSST (Institut de recherche en santé et en sécurité du travail du Québec) in Montreal, Canada. The ITF method uses a circular blade with a pressure of 5 N applied on the blade. The blade speed is sinusoidal with a maximum of 100 mm/s. The value to be measured is the number of cycles to cut the material and is compared with that of a reference material. The IRSST method uses a straight blade and the pressure is applied on the sample holder. Series of tests using at least two different weights must be performed at a constant blade speed. The cut resistance for the IRSST method is measured as the load required to cut the material at a 10-mm blade displacement. Eighteen glove materials of different levels of cut resistance were compared. With uniform materials such as neoprene, no variability is observed with the ITF method, while a coefficient of variation of approximately 10% is observed with the IRSST method. These results may be due to different sensitivities of the test methods. The ranking of cut resistance obtained with both methods can be considered as equivalent and the results are comparable with a correlation coefficient r of 0.89.

A new cut-test apparatus for standardization has been developed. The apparatus is based on a previously developed cut-test method that consists of sliding a blade on a material and determining the distance traveled by the blade to cut through the material when a constant force is applied. To build a machine for standardization, several requirements were considered. One of the most important requirements was that the force applied at the blade/material point of contact must be constant throughout the test. Other requirements were that the test apparatus be easy to use by an untrained person, that the applied force should be known, and that the results should be reproducible. The force is applied at the material/blade point of contact from bottom upwards through a ram-type arm. To satisfy the requirements, a mechanism based on a double Watt system was developed that allows a perfect vertical displacement of the sample and a constant applied force throughout the test with a maximum error of 2×10-2N.

No abstract