Font Size: a A A

The Research On The Application Of DOE In The Process Optimization And Circuit Optimal Design

Posted on:2010-06-12Degree:MasterType:Thesis
Country:ChinaCandidate:L LiFull Text:PDF
GTID:2178360275997648Subject:Microelectronics and Solid State Electronics
Abstract/Summary:PDF Full Text Request
In this paper, the statistical characterization and optimization are completed for the Screen Printing process in the Thick-film Resistors process using DOE(Design of Experiment) technique. In the circuit optimization, the Radio Frequency Amplifier, as a example, is optimized based on the experimental design technique. And the comparison and analysis is made between the optimum results optimized by the DOE technique and the"optimization module"in PSPICE. At the same time, some issues existed in the simulation process are analyzed and resolved. The main works of this paper include the following aspects:For the Screen Printing process in the Thick Film Resistors process, the statistical characterization model is established with experimental data using Response Surface Design, based on which this process equipment is optimized. The optimization finds the optimal combination of factors, with which the discreteness of resistance thickness is minimized while the thickness meeting the target range. The discreteness of the resistance of thick-film resistor products is reduced consequently.Now it is difficult for many EDA softwares(such as PSPICE) to find the optimal solution by their own module and the optimizing time is too long. Considering these limitations, this article conbines DOE technique and EDA software and makes the metamodel of the RF amplifier established. And the circuit performance is optimized based on the metamodel, which shows that the optimization result based on DOE technique is far superior than the one based on PSPICE optimization module.In the current circuit simulation trials, each process requires people to complete three steps: modifying components'parameters, implementing circuit simulation, and obtaining circuit performance parameters. With a little bit more complicated circuit, the number of the simulation trial is often a dozen times, or even more than a hundred times. It is heavy workload and error-prone. In order to solve this problem, an interface to call PSPISE software is written using C Sharp language, which implements the automation of this process. Thus the work of the circuit optimization based on DOE technique is reduced greatly, and the application of DOE technique in the circuit is more feasible.
Keywords/Search Tags:Design of Experiment, Thick-film Resistor, Statistical Characterization, Circuit Metamodel, PSPICE Calling
PDF Full Text Request
Related items