Abstract

For driving belt condition monitoring, the main interest is the certification of the capacity to keep their qualities unchanged over a long period of time and secondary to detect the imminence of the catastrophic failure. This paper presents a study on the behaviour detection and description of a flat driving belt health condition, used in a rotary machine electrically driven, particularly a lathe headstock gearbox running idle. It was discovered that in the mechanical power transmitted from electromotor to gearbox via a flat belt some specific sinusoidal components (a fundamental and some harmonics) of variable power are generated. The description of these power components (by values of amplitude, frequency and phase at origin of time) is indirectly detectable in the evolution of the active electrical power absorbed by the drive electromotor. Two arguments are available for this approach. Firstly, there is a reasonable assumption that between the mechanical power and the active electrical power there is an approximated proportionality relationship through the power efficiency. Secondly, the evolution of the active electrical power (or mechanical power as well) is a deterministic signal with a low level of noise. A simple computer assisted procedure of active electrical power signal acquisition and data processing was conceived, the detection was done by computer aided curve-fitting procedures in Matlab applied on active electrical power evolution absorbed by the driving motor in stationary working regimes (the electromotor playing the role of a mechanical power sensor). Mainly two ways of graphic representation have been proposed in order to describe the variable power generated by this flat belt (in time and frequency domains). The behaviour of many other types of belts involved in rotary machines driving can be similarly described.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call