Designing Superconducting Magnet Controls for the Super High Momentum Spectrometer

Brian Mastracci

Office of Science, Science Undergraduate Laboratory Internship Program

Virginia Polytechnic Institute and State University

Blacksburg, Virginia

Thomas Jefferson National Accelerator Facility

Newport News, Virginia

July 30, 2010

Prepared in partial fulfillment of the requirement for the Office of Science, Department of Energy’s Science Undergraduate Laboratory Internship under the direction of Paul Brindza and Steven Lassiter in the Experimental Nuclear Physics division at Thomas Jefferson National Accelerator Facility.

Participant: ______

Signature

Research Advisor: ______

Signature

TABLE OF CONTENTS

ABSTRACT

INTRODUCTION

MATERIALS AND METHODS

RESULTS

DISCUSSION AND CONCLUSIONS

ACKNOWLEDGEMENTS

REFERENCES

FIGURES AND TABLES

ABSTRACT

Designing Superconducting Magnet Controls for the Super High Momentum Spectrometer. BRIAN MASTRACCI (Virginia Polytechnic Institute and State University, Blacksburg, VA, 24061) PAUL BRINDZA, STEVEN LASSITER (Thomas Jefferson National Accelerator Facility, Newport News, VA, 23606).

Hall C of Jefferson Lab is building the Super High Momentum Spectrometer (SHMS), which will be capable of analyzing charged particles scattered by its upgraded 11 GeV electron beam. Particles will be focused into detectors using five superconducting magnets. Each magnet will be monitored and controlled using integrated hardware and software systems to ensure particles are delivered precisely and accurately to the detectors. Electronic data flows from instruments built into the magnet through various hardware devices that transform the data into a signal usable by a Programmable Logic Controller (PLC), where user-designed software continually monitors the status of the magnet. The focus of this project was to design and document the hardware that will be used to control the SHMS superconducting magnets. This was accomplished by revising wiring diagrams for the existing High Momentum Spectrometer (HMS) controls using AutoCAD drafting software. Technical drawings of new magnet designs and hardware manufacturers’ documentation were also used to ensure the proper signal will be delivered to each PLC. The end result was a map of the signals used to control liquid cryogen flow valves and to monitor temperature, strain, and potential difference within the magnets’ superconducting coils. A complete and accurate signal diagram is essential for ease of construction and maintenance of control systems, since poorly designed or incomplete hardware configurations could supply improper PLC input, leading to malfunction of and damage to SHMS components. A complete and well-designed control system optimizes performance of the machine, increasing its value to the study of particle physics.

INTRODUCTION

Jefferson Lab is in the process of increasing the energy of its electron beam accelerator from 6GeV to 12GeV. The equipment in each experimental end station is also being upgraded to handle the increased beam energy. Hall C is building the Super High Momentum Spectrometer (SHMS), which will be capable of analyzing charged particles at momenta created by a beam energy of up to 11GeV [1]. Five superconducting magnets focus scattered particles into detectors. Each magnet is precisely monitored and controlled using integrated hardware and software systems.

A complete control and monitoring system must be designed for safety and functionality of the SHMS. A programmable logic controller (PLC) will process data from sensors built into each magnet using software that must be designed and programmed into the PLC. Electronic data from sensors monitoring temperature, pressure, strain, liquid levels, potential taps, cryogenic flow valve position, vacuum pressure, and helium mass flow need to be amplified, conditioned, and converted to usable PLC input signals [2].

In order to create a complete control system all hardware, including cabling, must be acquired and assembled; technical drawings must be created to document signal paths and show how the components are to be connected; and software must be written that tells the PLC how to process input signals and what signals to output.

The High Momentum Spectrometer (HMS) that already exists in Hall C uses the same style of integrated control system that will be designed for the SHMS. Designing controls for the SHMS will be largely a modification of the HMS controls [2]. Changes will be made to incorporate different types of magnets and different hardware components and configurations. It was determined that modifying the existing technical drawings and software would be more efficient than creating completely new documents.

MATERIALS AND METHODS

HMS wiring diagrams were created using AutoCAD 2D drafting software [3]. AutoCAD was used to edit existing drawings and create the wiring diagrams for the SHMS. RSLogix 5000 software [4]was used to write the programs that allowed the PLCs to control and monitor the HMS magnets. Inputs and outputs, or program tags, were modified for the purpose of testing some components of the SHMS control system.

Creating Wiring Diagrams

The main procedure for creating wiring diagrams was to update labels on HMS wiring diagrams. In some cases lines representing wires needed to be removed and redrawn and hardware representations needed editing to reflect the use of new hardware components. In the case of strain gauge measurement, this was accomplished by erasing the representation for Oxford’s strain gauge measurement devices [5] and instead connecting strain gauges to Dataforth signal conditioners [6] so that the signal could be analyzed by the PLC. Low-voltage conditioners (LVCs) were also updated by removing representation for the old unit and reconnecting wires to a diagram of the new unit. Proper methods for wiring the new components were identified by consulting the manufacturer’s documentation provided with the product.

Assembling Hardware

Installing hardware onto the local control rack was a simple matter of snapping it into place. Wiring together the devices was a majority of the work required to assemble the rack. Wires needed to be cut to the proper length and stripped using wire cutters. Ferrules were applied to the end of each length of wire using a crimper to facilitate connection and removal of cables. Finished cables were connected to the terminals specified in the updated wiring diagrams.

Low-Voltage Conditioner Calibration

LVC signal conditioners required calibrating once they were properly connected. Each LVC will measure the position of a Joule-Thomson (JT)cryogenic flow valve by converting the signal output by a linear voltage differential transducer (LVDT) to a current within the input range of the PLC. An electric linear actuator drive motor, which moves the JT valve, and LVDT were connected to the control rack for sensor calibration. By changing program tag values in RSLogix, the valve was moved to its fully open and fully closed positions. The current and voltage outputs from the LVC were recorded at each position. At the maximum closed position, the signal conditioner offset was adjusted until the dual output read 2V and 4mA. At the maximum open position, the signal conditioner gain was adjusted until the output was 10V and 20mA. Adjusting the offset or gain affected the both the high and low signal output. The process of opening and closing the valve was repeated until the open and closed output signal was within ±0.1V of the desired value. Once the first LVC was calibrated the valve was connected to the next LVC unit so that all six signal conditioners could be calibrated using the same valve assembly.

RESULTS

A set of wiring diagrams for monitoring temperature, strain, potential difference, and JT valve position was produced. Drawing labels were revised to reflect new SHMS temperature sensor names. Potential difference will be measured across different coils of each superconducting magnet. Drawings of potential taps were revised to show which coils will be measured on each of the new magnets (see example in Figure 1). Strain gauge representation was changed to show the use of a full-bridge instead of half-bridge strain gauge. The drawing of the Oxford strain gauge measurement unit was replaced by a series of signal conditioners that will send a signal to the PLC for measurement. Drawings for the JT valve system, including LVDT sensors, LVCs, and electric linear actuator drive motors, were revised to show that the SHMS will use only six JT valves on each magnet. Representation for a different LVC replaced the drawing for an old LVC that was prone to failure (see Figure 2). Calibration data for the new sensors was compiled for use in the software program (see Table 1). All hardware components featured in the upgraded signal diagrams were assembled on a prototype local control panel (see Figure 3).

DISCUSSION/CONCLUSION

The local control rack constructed and documented is a prototype rack for the integrated control system that will be used to operate the SHMS. After the liquid level and pressure, helium mass flow, and vacuum pressure measurement devices have been completed, the rack will be shipped to Michigan State University to be tested with the SHMS Horizontal Bender (HB) magnet. Once the first control rack has been tested and completed, additional racks for each of the four remaining magnets will be constructed, tested, and installed.

ACKNOWLEDGMENTS

This project was completed at Thomas Jefferson National Accelerator Facility between the dates of June 24 and July 30, 2010. Thanks to the United States Department of Energy and Thomas Jefferson National Accelerator Facility for creating, funding, and organizing the Science Undergraduate Laboratory Internship program. Thanks to Lisa Surles-Law and the Jefferson Lab education department for organizing the SULI program at Jefferson Lab. Thanks to Paul Brindza for sponsoring me as a SULI intern. Special thanks to Steven Lassiter for guiding me through completion of this project.

REFERENCES

[1] “Design Solutions Document – Upgrade Hall C,” May 23, 2007 (Updated may 9, 2008).

[2]S. Lassiter and M. Fowler, “SHMS Magnet Controls,” June 27, 2010.

[3]Autodesk, Inc., 111 McInnis Parkway, San Rafael, CA 94903, USA

[4]Rockwell Automation, 1201 South Second Street, Milwaukee, WI 53204, USA

[5]Oxford Instruments plc, Abingdon, Oxfordshire OX13 5QX, UK

[6]Dataforth Corp., 3331 E Hemisphere Loop, Tucson, Arizona (AZ) 85706, USA

FIGURES AND TABLES

Figure 1

Figure 2

Figure 3

Table 1