Return to search

DISTRIBUTED PROCESSING APPROACHES TO REAL-TIME DATA COLLECTION: AN APPLICATION IN NMR SPECTROSCOPY

Traditional approaches to the computerization of laboratory instrumentation result in a one-to-one correspondence between computer and instrument. This becomes ineffective in a multiple instrument laboratory. An alternative approach is one in which each instrument has a dedicated data acquisition and control computer, with several such computers linked through a laboratory data network to a centralized data management computer. The development of such an approach is the subject of this dissertation. The problem is first defined in the context of a Nuclear Magnetic Resonance (NMR) laboratory, followed by a brief history of computers and a discussion of the ways in which computers have been applied in NMR. / After this, an introduction to the concepts essential to computer networking is presented, continuing with a discussion of the development of a local area network that utilizes industry-standard IEEE-488 protocols and incorporates differential line transmission for long-distance error-free operation. A Z80 microcomputer-based network interface is described which facilitates the easy connection of virtually any host computer to the network. Specific circuitry is also presented to interfaces S-100 (IEEE-696) microcomputers to the network. / Following discussion of the laboratory data network, the dissertation provides discussion of a microcomputer which was designed to operate as a stand-alone data acquisition and experimental control system, as well as a node in the network discussed earlier. This microcomputer utilizes several Z80 microprocessors and is based on S-100 circuit cards. Operator interaction is handled via a keyboard and a high resolution video graphics monitor. All software is based in Read Only Memory, providing a low-cost stand-alone system for a variety of laboratory applications. All data acquisition and experimental control is performed with sophisticated custom-built circuitry. / The remainder of the dissertation concentrates on the development of the custom circuits for data acquisition and experimental control. A subsystem is discussed that allows the interfacing of the host microcomputer to as many as 255 independent control processors via a single twisted-pair cable: useful in low-speed control and monitoring applications. A second subsystem handles data collection and time-critical instrument control. This subsystem, called the DAS, or Data Acquisition Subsystem, is based on modular design approaches, and consists of from two to five S-100 circuit cards. A minimal system is one Analog to Digital Converter card and a Signal Averager card. The AD circuit has on-board controls for analog gain, low-pass filtering, and digital resolution. The Signal Averager has three on-board counters to provide all of the control and timing signals for data collection. A hardware averager is also provided to allow the coaddition of repetitive signals. The Signal Averager is designed to support an optional Direct Memory Access Sequencer for high-speed signal averaging. A second AD card can be added to this subsystem, providing simultaneous two-channel data collection. The DAS optionally includes a Programmable Timer/Pulse Controller circuit which executes sophisticated timing programs of up to 16 steps with 50 nanosecond time resolution and maximum step length in excess of 15 hours. Precise experimental control is obtained through any of 24 programmable outputs that can be activated on any step. / Examples of the use of this system in NMR data collection are included. / Source: Dissertation Abstracts International, Volume: 42-06, Section: B, page: 2359. / Thesis (Ph.D.)--The Florida State University, 1981.

Identiferoai:union.ndltd.org:fsu.edu/oai:fsu.digital.flvc.org:fsu_74565
ContributorsTERPSTRA, DANIEL KEVIN., Florida State University
Source SetsFlorida State University
Detected LanguageEnglish
TypeText
Format388 p.
RightsOn campus use only.
RelationDissertation Abstracts International

Page generated in 0.0012 seconds