Kaikki aineistot
Lisää
Ethernet-based fieldbuses are increasing popularity in industrial environments. Ethernet technology offers high bitrates, relatively cheap price and it is a well known and widespread technology. This thesis introduces the basics of Ethernet technology and some industrial Ethernet protocols. The aim of the thesis was to make a simple handbook of industrial Ethernet protocols and to examine and compare redundancy properties of the different Ethernet-based fieldbuses. Lower levels of EtherNet/IP are examined thoughtfully and redundant EtherNet/IP implemented to ACS350 frequency converter from ABB.
Relative emission spectra of light-emitting diodes (LEDs) depend on the junction temperature. The high-energy region of the emission spectrum can be modelled with Maxwell-Boltzmann distribution as a function of energy and junction temperature. We show that according to the model and our experiments, the normalized emission spectra at different junction temperatures intersect at a unique energy value. The invariant intersection energy exists for many types of LEDs and can be used to determine the alloy composition of the material. Furthermore, the wavelength determined by the intersection energy can be used as a temperature invariant wavelength reference in spectral measurements.
LED lighting and smart controlling can decrease the consumption of electric power in zero or positive energy buildings. This study demonstrates that cyclic dimming, which appears in smart controlling, e.g. through occupancy sensing, might have an effect on the luminaire lifetimes and failure rates. Two different types of LED luminaires were aged in 30-s cycled dimming and undimmed modes. The manufacturer-specified lifetimes for both luminaire types were 100 000 hours. For one of the luminaire types, four out of five cycled units failed before 30 000 operating hours. For the uncycled luminaires, the lifetimes estimated from the measurements were over 100 000 hours. For the other type, the estimated lifetimes were 75 000 hours and 64 000 hours for uncycled and cycled luminaires, respectively.
Relative emission spectra of LEDs depend on the junction temperature. The high-energy region of the emission spectrum can be modelled with the joint density of states and the Maxwell-Boltzmann distribution as a function of energy and junction temperature. It can be shown that the normalized emission spectra at different junction temperatures intersect at a unique energy value. Thus the wavelength and the relative intensity of the intersection point do not depend on the junction temperature of the LED. The invariant intersection energy exists for all LEDs manufactured using the elements from groups III-V. The wavelength determined by the intersection energy can be used as a temperature invariant wavelength and relative intensity reference in spectral measurements.
We have developed spectral models describing the electroluminescence spectra of AlGaInP and InGaN light-emitting diodes (LEDs) consisting of the Maxwell-Boltzmann distribution and the effective joint density of states. One spectrum at a known temperature for one LED specimen is needed for calibrating the model parameters of each LED type. Then, the model can be used for determining the junction temperature optically from the spectral measurement, because the junction temperature is one of the free parameters. We validated the models using, in total, 53 spectra of three red AlGaInP LED specimens and 72 spectra of three blue InGaN LED specimens measured at various current levels and temperatures between 303 K and 398 K. For all the spectra of red LEDs, the standard deviation between the modelled and measured junction temperatures was only 2.4 K. InGaN LEDs have a more complex effective joint density of states. For the blue LEDs, the corresponding standard deviation was 11.2 K, but it decreased to 3.5 K when each LED specimen was calibrated separately. The method of determining junction temperature was further tested on white InGaN LEDs with luminophore coating and LED lamps. The average standard deviation was 8 K for white InGaN LED types. We have six years of ageing data available for a set of LED lamps and we estimated the junction temperatures of these lamps with respect to their ageing times. It was found that the LEDs operating at higher junction temperatures were frequently more damaged.
We have developed a setup for measuring differential spectral responsivities of unifacial and bifacial solar cells under bias light conditions. The setup uses 30 high-brightness LEDs for generating a quasi-monochromatic light source covering the wavelength range 290–1300 nm. Halogen lamps are used to generate bias-lighting conditions up to the irradiance level of 1000 W/m2. The setup has been fully characterized for spectral irradiances and spatial uniformities of all light sources. Validation measurements carried out using a reference cell of 2 × 2 cm2 area from Fraunhofer ISE demonstrated an agreement better than 2% over the wavelength range of 425–940 nm, with an expanded uncertainty of 2.6%. In the UV and IR regions, the discrepancies are higher but still within estimated uncertainties. The setup is also tested in measuring spectral responsivities of large 15 × 15 cm2 bifacial solar cells. The associated problems are discussed.