Use of failure modes and effects analysis in design of the tracker system for the HET wide-field upgrade

Hayes, Richard; Beets, Tim; Beno, J.H.; Booth, John A.; Cornell, Mark E.; Good, John M.; Heisler, J T; Hill, Gary J.; Kriel, Herman; Penney, C.E.; Rafal, Marc D.; Savage, Richard; Soukup, Ian; Worthington, Michael S.; Zierer, Joseph

doi:10.1117/12.925500

Cited by 2 publications

(1 citation statement)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is only when the tracker is operating outside of the safe parameters that the HFC is called into action. This device addresses the majority of potential hardware and software faults identified during the Failure Modes and Effects Analysis (FMEA) 13 by forcing a safe shutdown of the system when a fault conditions is detected.…”

Section: Hardware Fault Controller Design and Implementationmentioning

confidence: 99%

HETDEX tracker control system design and implementation

et al. 2012

Self Cite

View full text Add to dashboard Cite

To enable the Hobby-Eberly Telescope Dark Energy Experiment, The University of Texas at Austin Center for Electromechanics and McDonald Observatory developed a precision tracker and control system -an 18,000 kg robot to position a 3,100 kg payload within 10 microns of a desired dynamic track. Performance requirements to meet science needs and safety requirements that emerged from detailed Failure Modes and Effects Analysis resulted in a system of 13 precision controlled actuators and 100 additional analog and digital devices (primarily sensors and safety limit switches). Due to this complexity, demanding accuracy requirements, and stringent safety requirements, two independent control systems were developed. First, a versatile and easily configurable centralized control system that links with modeling and simulation tools during the hardware and software design process was deemed essential for normal operation including motion control. A second, parallel, control system, the Hardware Fault Controller (HFC) provides independent monitoring and fault control through a dedicated microcontroller to force a safe, controlled shutdown of the entire system in the event a fault is detected. Motion controls were developed in a Matlab-Simulink simulation environment, and coupled with dSPACE controller hardware. The dSPACE real-time operating system collects sensor information; motor commands are transmitted over a PROFIBUS network to servo amplifiers and drive motor status is received over the same network. To interface the dSPACE controller directly to absolute Heidenhain sensors with EnDat 2.2 protocol, a custom communication board was developed. This paper covers details of operational control software, the HFC, algorithms, tuning, debugging, testing, and lessons learned.

show abstract

Section: Hardware Fault Controller Design and Implementationmentioning

confidence: 99%