INTERNATIONAL ACCELERATOR RADIOLOGICAL PROTECTION E-MAIL (IARPE) NEWSLETTER "The Official Publication of the Accelerator Section of the Health Physics Society" (with Contributions from International Correspondents) ====================================================================== October 1992 Vol.1 #9 ====================================================================== OFFICERS ====================================================================== President: Ralph Thomas, LLNL Past President: Wade Patterson, LLNL President-Elect: Geoff Stapleton, SSCL Secretary/Treasurer:Nisy Ipe, SLAC Directors (1 Year): Frank Masse', MIT BATES Paul Neeson, D.O.E. Directors (2 Year): Gerald Fallon, MIT BATES Paula Trinowski Directors (3 Year): Carter Ficklen, CEBAF Jerry Miller, LANL FROM THE EDITOR'S TERMINAL (Nisy Ipe ) ====================================================================== I would like to thank Bob May for serving as the acting editor during my absence. Thanks to Bob's efforts, the newsletter has acquired a "new, more official look". This fall, I would like to introduce a "Letters to the Editor" section, so please feel free to send your contributions directly to me. Please entitle your contribution "Letters to the Editor" and include your name, organization and E-Mail address. Please limit the horizontal length of the letter including blank spaces to 70 characters (this would make my work a lot easier). If you don't know what this means (and it appears that there are some who don't really understand this concept!!!) let me explain. The first sentence in this section has 69 characters (including blank spaces), the second has 66 and the third has 26. I hope this is clear. We are soliciting feature articles for the newsletter, and would appreciate hearing from those who would like to contribute. The position of Editor for the IARPE Newsletter will be open, effective January 1993. If any of you are interested in this pursuing this position, please contact eirher Ralph Thomas or me. There are currently over 100 individuals on the distribution list for the newsletter. ANNOUNCEMENTS ===================================================================== Call for Papers - Accelerator Section, HPS Annual Meeting ---------------------------------------------------------------------- (Geoff Stapleton ) ---------------------------------------------------------------------- This is a call for submission of abstracts of papers to be considered for presentation at the Accelerator Section Meeting of the HPS Annual Meeting at Atlanta, Georgia, USA from July 11 - 15, 1993. Next month (November) the Health Physics Society will include in the HPS Newsletter (not this newsletter) a form for members to make an application to submit an abstract for a paper/presentation at the Annual Meeting of the Health Physics Society. Now it is very important that we in the accelerator section use the meeting to the full and we must try to make the meeting as useful and as interesting as possible. To that end we must try to build a program that takes advantage of our wide international membership and includes information on a wide range of topics. I would urge everyone to review their recent work with a view to making a presentation at the meeting. Please submit an abstract as soon as possible so that we can decide on the amount of time that must be scheduled. It is extremely difficult to reschedule presentations to include papers submitted at the last minute. One suggestion about the program is to include a number of keynote papers from around the world. Of course this would rely on the generosity of overseas presenters paying their own way to the conference and relying on the benefits resulting from collecting together so many international experts with a wealth of knowledge of radiation control at a wide variety of accelerator centers. Members of the section who are not HPS members should let me or the editor know that they are interested in presenting a paper and I will arrange for them to be sent an application form. Accelerator Section Directory (Nisy Ipe ) --------------------------------------------------------------------- About a year ago we had started collecting information from many of you with the intent of compiling a directory of individuals working in radiation protection at accelerator facilities around the world. Some of you are probably wondering as to what became of that venture. Because of the uncertainities associated with handling the membership of individuals in the Accelerator Section, who were not members of the HPS, we had temporarily put the project on hold. Now we are ready to proceed with it and Harry Howe (LLNL) has graciously consented to compile the directory. Since several individuals have had changes in their E-Mail addresses, Harry is now collecting E-Mail addresses for inclusion in the Accelerator Section Directory. If you want to be included in this directory, please send your E-Mail address to Harry at or . Upon receipt, Harry will confirm the establishment of two way communications with a message to you and will include your address in the Section directory database. International Conference on Application of Accelerators in Research --------------------------------------------------------------------- and Industry Wes Dunn ) --------------------------------------------------------------------- An update for those who haven't received previous information: "The Twelfth International Conference on the Application of Accelerators in Research & Industry" will be held November 2-5, 1992 at the University of North Texas, Denton, Texas (approx 35 miles N of Dallas) Chairmen: Jerome L. Duggan, UofNT (817) 565-3252 BITNET:FC66@UNTVAX.BITNET and I.L. Morgan, IDM Inc, Austin. Fee: (before/after 10/1/92) University $170/$200 Industry/Govt $250/$275 There will be one section chaired by the Texas Bureau of Radiation Control that will deal with "Radiation Safety Consieration in Accelerator Production and Processing of Medical Isotopes (2pm Monday, Session BK). This session will feature PET facilities and vendors. NEWS FROM INTERNATIONAL CORRESPONDENTS ====================================================================== News from ESRF (Elke Braeuer ---------------------------------------------------------------------- The ESRF reached its final design current of 100 mA in the month of July. For the first time a safety shutter was opened on a typical ESRF undulator beamline (0,5 Tesla). The hutch structure consists of 8 mm of lead and 2 mm steel panels for the 90 degree direction, 3 mm of lead on the roof and a back wall of 10 cm lead bricks within a 20 degree cone and 5 cm of lead outside this cone. This design proved to be very adequate: dose values between 5 and 50 micro-Sievert/hour were measured at nearly any point around the hutch. The values will certainly decrease with improvement of the vacuum conditions which were still worse than 10E-8 Torr. It was clearly demonstrated, that all dose values originated from high energy photons (up to 6 GeV) due to the gas bremsstrahlung in the ring being very important at the ESRF (with the circumferrence of our storage ring being about 850 m). News from LAMPF (Sara Hoover ) ---------------------------------------------------------------------- It is anticipated that LAMPF operations for 1992 will end on October 9th. It has been a successful operational cycle this year, however by mid- September a decision was made by LAMPF and LANL management to reduce beam currents to the main experimental area by roughly 75% to ensure that LANL stays below the site boundary dose imposed by US Federal Regulation 40 CFR 61, Subpart H of 10 mrem per year. This is necessary because of LAMP's radioactive air emissions which originate from the approximately 800 kW, 800 MeV proton beam operations. Health Physics was kept extremely busy calibrating, qualifying and maintaining the stack monitoring equipment which has received significant upgrades within the last year. A proposed solution to this problem will be the installation of a delay line to delay and decay the radioactive air emissions, which have a spectrum of half lives up to about 20 minutes. This solution is expected to be operational in about 1995. At this point, it is anticipated that LAMPF will begin operations again sometime in May 1993. News from SLAC (Nisy Ipe ) ---------------------------------------------------------------------- Visitors Dr. C.A. Perks, Section Leader of the Dosimetry and Aerosol Sampling Section at Harwell Laboratory, U.K. visited the Radiation Physics Department in September. He gave a seminar on Boron Neutron Capture. Therapy. FFTB Installation of the components of the FFTB beam line is continuing. Most of the magnets have arrived at SLAC from the Former Soviet Union. Currently, the plans are to complete the installation of the beam line components by the spring of 1993. The 55' long iron shield that serves as muon shield and separates the FFTB tunnel from the rest of the BSY is being re-stacked in this down time period to allow for the installation of a 10.75" (OD) laser pipe and a 2" (OD) beam pipe. After the completion of these tasks, SLAC will resume operation of the A-line in early November followed by the continuation of the SLC operation in January, 1993. Sufficient shielding is being added in this down time to allow personnel to have access to sections of the FFTB tunnel while the other beams are passing through the BSY. (Sayed Rokni ) NLCTA Work on plans for the NLCTA (Next Linear Collider Test Accellerator) is entering into its latest stages. A proposal to build this facility might be ready for submission to the DOE as soon as November 12 this year. Originally, the design beam energy was 630 MeV, with hopes that an upgrade of the rf system would enable to reach 1170 MeV in a later stage. These hopes are somewhat stronger now, since 100 MV/m was reached recently in a short test accelerator section. The cost of the NLCTA is estimated to be $12 M. (Vashek Vylet ) End Station A A test run for Experiment E-142 was performed in End Station A from September 16 - October 4. The experiment is designed to study the spin structure functions of the neutron. Polarized electrons scattering on a polarized He-3 target, probe the spin of the nucleons in He-3. The spin structure measurements serve to evaluate two separate sum rules; the Bjorken sum rule and the Ellis-Jafbe sum rule. The data and evaluation of the sum rules are fundamentally important to QCD and to the understanding of nucleon structure. Normal run for experiment E-142 is scheduled to start on November 6, 1992. The beam conditions are: Polarized electron beam, Energy=22.66 GeV, Current=50mA, Rep rate=120 Hz, Pulse width=2 microseconds, Target length=0.004 radiation lengths. (Stan Mao ) News from SSC (Jeff Bull < bull@sscvx1.bitnet>) ---------------------------------------------------------------------- The SSC held the first meeting of its Radiation Control Review/Advisory Group on September 9-10. This group, chaired by Ralph Thomas of LLNL, also includes Don Cossairt from Fermilab, Ken Crook of SLAC, Herbert Dinter of DESY, Hideo Hirayama from KEK, Bob Macek from LAMPF, Lutz Moritz of TRIUMF, and Graham Stevenson of CERN. This group is to meet one to two times a year to review the radiation control program and serve as an advisory group for radiation protection concerns during the design and construction phases of the SSC. Topics covered in this initial meeting included an overview of the collider and booster accelerators, shielding design of the booster accelerators, and the proposed groundwater activation model. A written report from the committee is expected shortly. On October 1, a Radiation Control Group was formed at the SSC as part of the Project Management. This new group was formed to consolidate the various radiation protection efforts at the SSC. This group has the responsibility for radiation shielding calculations as well as operational health physics duties at the SSC. This group is headed by Geoff Stapleton and currently consists of Jeff Bull, Joe Coyne, and Sylvia Revell. Nikolai Mokhov will continue to perform beam energy deposition and shielding calculations while assigned to the Collider machine QUESTIONS? QUESTIONS? QUESTIONS? ====================================================================== Tritium Sampling (David Perry ) ---------------------------------------------------------------------- The Isis spallation neutron source at Rutherford Appleton Laboratory releases some tritium to the atmosphere. Although the amount represents very slight local and environmental hazards, we are required by the regulatory body to measure and control it. At present, we are using water bubblers and are considering converting "elemental" tritium into water in the measuring system, but are concerned that that will exaggerate the hazard. We have tried a proportional counter monitor system, but there are doubts as to whether it is correctly compensating for the relatively large amounts of other nuclides (C-11, N-13 etc.). I would be grateful to hear from anyone who has experience of these problems. In particular: are measurements made "on-line"? Is "elemental" tritium measured separately from tritiated water vapour (or at all, on the grounds that the much lower DAC makes this unnecessary)? If HT (and/or DT) is measured separately, has this been achieved in the presence of relatively large amounts of the other spallation products of air? Finally, has it been possible to convince regulatory bodies, using cost/benefit arguments, that monitoring costs should not be grossly disproportional to the environmental detriment? Copies of response to questions should be sent to the editor so that they can be published in the newsletter. ====================================================================== FEATURE ARTICLE Radiation Security Interlocks by Ken Crook () ====================================================================== 10/13/92 Radiation Security Interlocks How about this for an idea:the next high energy accelerator will be buried so deep underground or surrounded by such massive amounts of shielding, that external radiation will be non-existent, even under the worst-case accident conditions. Furthermore, while we are daydreaming, this accelerator will be so reliable and flexible that it will never require an entry for maintenance or modifications. At the stroke of a pen, we have eliminated the need for interlocked radiation monitors, active electronics to contain beams, and an access control system that permits entry only when areas are safe. Alas, as we all know, the real world is quite different. Even underground accelerators must have entry doors and vent shafts. We can seldom afford the cost or space for massive above-ground shielding. Beams do get missteered and hit points of weak shielding, and entry into tunnels is necessary for repair and modifications. What started out, in the early days of accelerators, as a comparatively simple task of installing a couple of radiation monitors and a gate with a microswitch, has ballooned into a fairly major enterprise for the complex accelerators we have today. For example, at SLAC we have more than 30 independent access zones, and five operating modes, allowing beams to be run in one area while access is permitted in other areas. At the SSC with five different interconnected machines, entry will probably be allowed in one or more machines while beams are running in others. This type of access requirement calls for a carefully conceived access control scheme, with appropriate beam stoppers, beam containment devices and radiation monitors. These systems and the associated issues of reliability, testing and risk analysis are discussed below. Access Control Until fairly recently, access control logic was implemented with relays. There were some exceptions such as TRIUMF, where computer logic has been used for a number of years. Relay systems have the advantage that they can be designed to be fail-safe against problems such as power interruption or cable disconnection. Fail-safe is a basic concept in all designs for safety systems. When applied to relay circuits, the normal operating condition is represented by an energized relay. A normally open contact (which is now in the closed state) provides the logic path. When the coil is de-energized, the resulting open contact breaks the logic circuit and initiates a safety shut-down. Of course, relay contacts can become welded because of short circuits or lightning strikes. This type of failure can remain undetected and may prevent the equipment from performing its intended safety function unless the problem is found by a redundant circuit or during periodic testing. For mechanical devices such as beam stoppers, fail-safe operation includes holding the stopper in the out position by air pressure or with an energized solenoid. Loss of air or electronic power causes the stopper to drop into the beamline under gravity. Redundant circuitry is desirable, and often essential, in systems where safety is critical. In some cases triple redundancy may be justified. At SLAC, the access control logic consists of two identical relay paths. This double redundancy has been found to be satisfactory as long as close control is maintained over the equipment and wiring. Control is achieved by locking all racks and cabinets containing the safety equipment and by requiring written authorization for changes and additions. For safety devices outside locked racks such as beam line safety stoppers, triple redundancy has been used. Three mechanical or magnet stoppers must be in or off before entry is permitted in down-beam areas, and all three drop in or turn off when there is a forced entry. Computer-based access control systems are now becoming more common. CEBAF, the APS and the SSC are all planning to use Programmable Logic Controllers (PLCs) to perform the logic functions and to monitor status signals associated with entry control. PLCs were originally introduced in the 1970s to replace industrial relay control systems. They have several desirable features including ease of use, good immunity to electrical interference, and rugged mechanical and electrical design. They can also handle complex and extensive logic requirements, and futhermore, they can provide automatic documentation of the logic, either in Boolean or ladder logic form. Of course there are some negatives to PLCs. Because they are programmable computers, care must be taken to protect the program from unintended changes or accidental erasures. Programs should be stored in Programmable Read Only Memory (PROM) or in protected Random Access Memory (RAM). In addition, because PLCs include solid state circuits, the failure mode of the logic elements and peripheral drivers is difficult to predict. An output transistor that permanently fails in the conducting state is analagous to and just as potentially dangerous as a welded relay contact. For this reason, redundant PLCs should be used for access control systems. The argument can also be made that the programs in each PLC should be different - preferably written by two programmers who may only wave at each other at the coffee machine, but never speak. While this may be extreme, there is some validity to it. Building the same error into redundant hardware systems is not unknown. Identical software errors are just as easy to introduce. Relay safety systems may be going the way of the dodo bird. Recent electronic engineering graduates regard relay logic as prehistoric. They are reluctant to use relays in new logic systems, not only because they are unfamiliar with relay design techniques, but because professional growth comes with using the latest solid state and computer techniques. We must recognize this, and ensure that every possible safeguard is built into the new computer-based systems. Our peace of mind, if not our safety, depend on it. Beam Containment At SLAC, we use the term beam containment to refer not only to the mechanical devices in beamlines that are used to keep beams contained within the correct beam channel, but also to the electronic devices that aid this process. Some labs use the term "active electronics" instead of "beam containment" electronics. Typically, the electronic system might comprise (a) a beamline transducer such as a current toroid, a secondary emission monitor, a beam position monitor or an ion chamber, (b) an electronic processing module that integrates or counts beam current pulses and (c) a beam shut-off circuit connected to beam stoppers, RF sources or high voltage supplies. These electronic systems can be used for the following: 1. to ensure that the maximum beam power for a particular beamline is not exceeded. 2. to detect missteered beams. 3. to detect beams that have been switched into the wrong beam channel. 4. to protect beam stoppers, dumps and collimators from damage. Usually, a distinction is made between the beam containment electronics that protect mechanical devices, which, if damaged produce increased radiation in occupied areas, and electronics that protect machine components where damage would not result in increased external radiation. In the former category, because radiation safety is involved, great care must be taken in the design and implementation of the system. Self-test provisions should be built into the design, such as housekeeping (or keep-alive) pulses through toroid windings, and convenient test buttons should be provided so that each redundant path can be fully exercised. Each module should shut off the beam in redundant ways. At SLAC, three independent shut-off paths are used:the first path inserts a fast stopper, the second turns off the gun triggers and the third removes the klystron accelerate-time triggers and sets the entire linac to standby timing. Each beamline must have at least three independent systems to keep the beam power below the maximum allowed power for the line. Protection is also provided for mechanical beamline devices that have power ratings below the maximum allowed power. At least two independent protection systems are required. Typically these are (a) one or more interlocked ion chambers (B) interlocked temperature sensors and (c) an interlocked Burn-Through Monitor (BTM). The BTM is a pressurized chamber that ruptures on over-temperature. For a typical experimental run at SLAC, as many as fifty beam containment protection devices might be required - five current toroids, twenty-five protection ion chambers, seven temperature interlocks, three magnet current interlocks and ten flow switches. The decision to require a specific protection device is made by the Radiation Physicist in charge of the particular machine or experimental area, assisted by the Accelerator Department Safety Office. These safety requirements are recorded on the Beam Authorization Sheet, which lists all of the radiation safety requirements for the experimental run. All of the protection devices are checked either daily or weekly,in accordance with a formal procedure. In addition to the beam containment devices, 50 to 150 machine protection interlocks may also be required. These are checked less frequently and are not subject to the same close control as the radiation safety interlocks. Radiation Monitors At SLAC, radiation monitors are called Beam Shut-off Ion Chambers (BSOICs). They are normally placed in occupied areas outside shielding, or in tunnel areas where access is permitted while beams are running in adjacent beamlines. Trip levels are typically set to 10-20 mrem/h in tunnels and 50-100 mrem/h in outside areas. When radiation levels exceed the trip setting, beamline stoppers are inserted. Control room operators receive an early warning signal at 10-20% of the trip level, and an alarm indicating BSOIC fail. The fail signal is generated when the processed signal from the internal source drops below a preset level. Reliability It is essential that safety systems are reliable, not only because an unsafe failure puts workers at risk, but also because safe failures reduce experimental beam time. Safety system designers are beginning to address this problem. At a recent workshop at the APS (Ref. 1), an attempt was made to assign reliability or Mean Time Between Failure (MTBF) numbers to the safety systems to achieve an overall machine reliability of 99% for 48 hours. This translates into a safety system MTBF of 200 days. The sub-system and component MTBFs are of course, much greater. Whether this kind of reliability can be achieved in modern large accelerators remains to be seen. Testing Frequent testing is an important element in discovering unsafe interlock failures such as the welded contacts or shorted transistors discussed above. Complete retesting of access control systems should be done every six months. Testing of devices such as emergency shut-down switches and door interlocks should be done more often, such as during every open access period. For beam containment electronics at SLAC, daily and weekly tests are required because of concerns about radiation damage to transducers and cables in tunnels and because the cable plant and beam shut-off paths are not completely protected. Probabilistic Risk Analysis While the concept of risk analysis is common in the nuclear industry, it is only now being considered in the accelerator community. LAMF hosted a workshop (Ref. 2) in December 1991 to evaluate the credit that might be taken for active electronic protection systems. If one can be convinced through analysis or testing that an electronic safety system has a sufficiently low probability of failure (10**-6 or lower, for example), then this system could substitute for some element of the passive shielding protection. Until the accelerator community can develop good reliability data for electronic safety systems, it would seem prudent that the electronics should supplement but not replace passive protection, especially if the accident case results in a significant radiation exposure. (Ref. 1) APS Reliability Workshop, January 1992. (Ref. 2) Workshop on the Use of Instrumentation and Probabilistic Safety Criteria for Prompt Radiation Protection at LAMPF, December 1991. Other References NCRP Report #88 "Radiation Alarms and Access Control Systems" dated December 1986 has useful information on access control systems and reliability. SLAC #327, "Health Physics Manual of Good Practices for Accelerator Facilities" dated April 1988 contains some of the information presented in this note. Finally the D.O.E. order 5480. ACC (draft) " Safety of Accelerator Facilities" and its "Guidance" section dated 4/7/92, contain a very good presentation of current ideas about interlocks and accelerator safety systems. Ken Crook is an electronic engineer who has spent the last 25 years doing instrumentation control and safety system designs, at SLAC. He has served on numerous review committees and is currently the Safety Officer in the Accelerator Department. CLOSING THOUGHTS ====================================================================== " The worst bankrupt is the one who has lost enthusiasm." The editor remembers reading this quote somewhere, and apologises for being unable to remember who to attribute it to.