The power division behavior of single mode branching waveguides is studied theoretically. The dependence of the power division on the branching angle and on the coupling length is shown. A new qualitative formula relating these important parameters is presented. The calculated results for various power divisions are plotted. Certain power divisions are found to have certain respective limiting branching angles above which the desired power divisions can no longer be obtained.

An algorithm for the real-time measurement of the fetal and maternal RR intervals from a single-lead abdominal signal has been developed on a low power TMS32010 digital signal processor development system. The algorithm is based on a combination and modification of earlier techniques which have been proposed to perform similar processing. The performance of the algorithm is improved by the incorporation of validation procedures which enable the algorithm to continue operation despite temporary data loss during maternal movement. The reliability of the R-wave detection is determined from information tagged to the measured R-R intervals indicating the signal condition at the time of detection. These tags are then used as a basis for error correction and the measurement of the algorithm performance. The performance achieved from tests on active pregnant women shows the possibility of implementing the above system for ambulatory use.

Knowledge of bituminous creep modulus is very important to.fully characterize a premix. Conventional methods to determine it involve elaborate laboratory procedure requiring specialized equipment and trained personnel. Theoretical methods to calculate dynamic creep modulus (E*) have been available but very little comparison between theoretical and laboratory results are available in the literature. A study was carried out to determine the laboratory E* values of a set of field cores. The same cores were then analyzed for their physical properties such as penetration, viscosity and air-void content which were used as input for the determination of the theoretical E*. A comparison was then made between the laboratory and theoretical E* values in graphical plots which showed agreement within a factor two. In many analysis, this is close enough given the inherent variability within a premix. Thus, in cases where laboratory E* could not be performed, a quick estimate could be obtained using the method shown in this study.

Factor analysis was applied to batch shake-flask fermentations for the production of the enzyme cytochrome p-450 by the yeast Saccharomyces cerevisiae NCYC. 754 on a medium based on coconut-water allowing a screening of the experimental variables temperature (T), pH, shaking rate (RPM) and the initial concentrations of glucose (G), yeast extract (YE), peptone (P) and phosphoric acid (PA)supplements. Both the enzyme specific yield (nmol g-1) and the biomass yield (gL-1) were found to be strongly influenced by all the above experimental variables except P. Previous media formulations for the fermentation have shown that peptone is very important in ensuring good growth and enzyme production. The possibility of formulating a coconut-water based medium which needs less peptone supplement would lower medium costs as well as partly solve the wastewater treatment problem of the coconut processing plants.

The enzyme volume yield (nmol L-1) in batch shake-flask fermentations for the production of the enzyme cytochrome p-450 by the yeast Saccharomyces cerevisiae NCYC 754 on a medium based on coconut-water was optimised with respect to the experimental variables temperature (T), shaking rate (RPM) and the initial concentrations of glucose (G), yeast extract (YE) and peptone (P) supplements using a half-replicate 25 factorial and steepest ascent methods to locate the area containing the maximum enzyme volume yield and using a rotatable composite design to find the optimum levels of the experimental variables at the point of maximum yield. It was found that for maximum enzyme volume yield the levels of the variables are as follows: T = 26.8° C, RPM 159 rpm, G = 139.18 gL-1, YE = 17.04 gL-1 and P 6.95 gL-1. Experiments conducted with these optimised levels of variables gave an average enzyme volume yield of 372.62 nmol L-1 with a biomass yield of 16.08 gL-1. The optimised level of P at 695 gL-1 is very much lower than the previously optimised level of P at 11.90 gL-1 in a medium not based on coconut-water while the optimised levels of the other variables remain almost the same. Peptone costs RM158.00 kg-1 at current market price and the lowering of the optimum level of peptone in a fermentation medium based on coconut-water confirms the usefulness of coconutwater as a base for a fermentation medium in this particular fermentation.

The enzyme volume yield (nmol L-1) in batch 5L stirred tank fermentations for the production of the enzyme cytochrome p-450 by the yeast Saccharomyces cerevisiae NCYC. 754 on a medium based on coconut water was optimised with respect to the experimental variables temperature (T), pH and stirring rate (RPM), which were all under constant set point control, using the 23factorial experiments and steepest ascent methods to locate the area containing the maximum enzyme volume yield and using a rotatable composite design to find the optimum levels of the experimental variables at the point of maximum yield. It was found that for maximum enzyme volume yield the levels of the variables are as follows: T = 26.88° C, pH = 5.12 and RPM = 253 rpm. Experiments conducted with these optimised levels of variables gave an average enzyme volume yield of 637.36 nmol L-1. This represents an improvement of 71% compared to the maximum yield in shake-flask fermentation of 372.62 nmol L-1.

The use of coconut-water as a cheap fermentation medium for the production of the enzyme cytochrome p-450 by the yeast Saccharomyces cerevisiae N.C.Y.C 754 was investigated in a 5L stirred tank fermenter by building state variable models and generating optimum control time-profiles for the control variables temperature (T), pH and agitation rate (RPM) Using Pontryagin’s Continuous Maximum Principle. The theoretical maximum enzyme yield that can be achieved under optimum timeprofiled control of the three variables is 694.87 nmol L-1 in 30.5 hours of fermentation. Except for T, the continuous optimum time-profiles of the variables do not differ significantly from that of set point control. Experiments conducted with the three variables controlled along these time-profiles gave an average yield of 633.70 nmol L-1 in 28 hours of fermentation.