| Resumo: | This paper presents the advantages of analysing yarn evenness, based on mass variation, using a capacitive sensor which allows the measurement of 1 mm yarn length. The variation of capacity is due to the dielectric change and is proportional to the quantity of yarn mass between the sensor plates. Available commercial equipment uses a solution of 8 mm with the drawback of decreasing the resolution eight times. This limits the correct detection of Neps and other parameters. This type of fault (Nep), usually has a length between 1 and 4 mm, which can be measured directly with the new sensor. Commercial equipment detects this fault based on estimation. This paper also presents the spectral analysis of the yarn mass. This is important for textile quality control. Two approaches are presented, one based on a FFT model definition of narrow bands to aggregate the harmonics and other based on the Walsh-Hadamard transform. They show the wavelength detection of the sample taken into test, enabling the detection of the most common faults and errors, namely: fibre wavelength, and production faults. Furthermore, with the resolution presented it is possible to measure wavelengths in the range |2 mm, 2 cm|, which is impossible with commercial available solutions. The information obtained in this interval, is of utmost importance as it allows the characterization of the fibres in some types of yarns. The main objective of the project is to increase the quality specifying strict requirements allowing, in the near future, real-time actuation on a ring spinning frame.
|
|---|