Mechatronic development of dispensing modules - optimize design on robustness, tolerance and interfaces

Abstract

Almost everyone would have interacted with the ubiquitous, yet easily forgotten automated teller machine, the ATM. These sophisticated machines perform numerous tasks effortlessly and have become historically significant and pervasive because of a transformative function, cash dispensing. The cash dispenser is a complicated piece of technology which comprises of many modules and each module is made up of a number of sub-assemblies and parts. All the parts and sub-assemblies should come together to deliver the functionality of dispensing cash as part of an automated telling machine with at most efficiency and reliability with no scope for an error. Thus it becomes highly important for manufacturers to completely understand the functional requirements of such a product. Critical parameters based on the top-down approach are identified and key characteristics delivering different functionalities are studied. Based on these studies, parts contributing to key characteristics have been noted and their influence on achieving the required functionality has been examined. On the contrary knowing the limits of the design is perhaps equally important. A systematic analysis of variation of tolerances of these critical parts contributing in achieving key characteristics is carried out to know the limits of the cash dispensing module. Tolerance analysis using the worst case method and statistical analysis using root-sumsquare methods have been compared to give a conservative as well as a lenient perspective. Also for each key characteristic, a Monte Carlo simulation is carried out for one hundred thousand random samples. The transfer equation of each key characteristic was formulated using constituent dimensions and was used in the simulation. This gives a much more realistic figure on variation of tolerances as compared to worst case and root-sum-square method. Dr Daniel Whitney’s datum flow chain, a concept of conveying the intent and architecture in the form of a iagram, also was used to know the sensitivities of specific dimensions of a part that are included in the functionality delivery chain. With this study, we have relevant and strong data to back a design and know that a particular design is good enough to meet manufacturers’ quality standards and customer’s requirements. In case there are problems encountered during manufacturing or later on during actual usage, the root cause of the problem can be easily arrived at, saving time and cost. The tolerances of all dimensions are studied and their impact to the fmal assembly is documented, which can be altered by removing unnecessary tight tolerances which increase cost of manufacture and tighten loose tolerances which had resulted in malfunction of the product. Thus design to cost is controlled benefitting the manufacturer.