To: RFQ/PET Distribution From: Ralph J. Pasquinelli Subject: Meeting Notes 8/16/96 Lab G The accelerator tree has been installed and surveyed to a 0.030 inch tolerance. The relay racks and cable tray that rest above the tree should be installed starting next week. Anode cables for the RF systems have been pulled and are being terminated. All relay racks have been labeled with permanent labels. NOTE that these names should be used for cable designations. See Wes Mueller if there is any discrepancy. Dave Peterson will find interested parties and come up with a designation for locations of devices on the accelerator tree. Glenn Waver needs to inspect the as built dimensions of the accelerator and update the drawings. Ion Source Chuck Schmidt reports the ion source rack has been modified and reinforced to withstand the rigors of shipping. Installation of the electronics is being held up while the cable tray and insulated conduit are installed between the rack and the ion source. Beam Optics/Magnets/Solenoid Iouri Terechkine indicated that the first alpha magnet has been magnetically measured and satisfies the specification. The second magnet is being assembled and should be ready for measurement in two weeks. It was decided to keep the alpha magnets at Technical support until the vacuum chambers arrive at the beginning of September. Assembly will take place in IB2. The power supply for Q5 is capable of the 250 amps required. The magnet has been successfully tested to 300 amps. Q1 and Q2 have been assembled and tested. Procurement for Q3 components is in progress. Poles and yokes are due next week. Procurement for trim quads started this week. Del Larson is putting the finishing touches on the report of the end to end beam optics calculations. RFQ After some delay in the FNAL procurement department, the order for the 425 MHz RFQ B vanes have been placed with Danlin Products. Due to this delay, the partial shipment of vanes for RFQ's A and C will go to the plating firm separately. It is hoped to have the first set of vanes here at Fermilab in early September. Steve Ringler and Ding Sun will be setting up a tuning test stand in the Lab G portakamps. Steve Ringler will be making a trip to Danlin to inspect progress on RFQ vane fabrication. The 212 MHz RFQ has been repackaged in the new vacuum vessel and is leak tight. Charge Stripper Kris Anderson reports that the charge stripper is now 75% complete with the design stage. Vacuum housing detail should be complete within two weeks. Nozzle design will be pursued next. Expected delivery of the stripper cell is end of September. Dave Peterson has begun the design of the driver electronics for the fuel injectors. The chassis will contain integral power supplies. The use of two triggers will be investigated to allow remote pulse width control. Diagnostics No update on diagnostics hardware, but a meeting was held in the last few weeks to more closely understand the operational aspects of controlling the beam for chemistry. The meeting was attended by Ken Krohn and Jeanne Link as well as FNAL and SAIC staff. The minutes of this meeting have been distributed by Bob Webber. Low Level RF/High Level RF Ralph Pasquinelli brought up the topic of software performing protection for hardware and personnel. A case in point is the protection of the RFQ. Historically at FNAL, when the control system software crashes for what ever reason, the accelerator continues to run with the last instructions it received. This could prove to be disastrous for protecting the RFQ. We have already seen that we have sufficient RF power to damage the RFQ vanes in the absence of beam. The project can not afford to lose such a critical device as there are no spares. As such, all critical protection devices must be implemented with hardware loops. Software can be used in parallel as it may have value as a diagnostic. In the case of the RFQ, Dave Peterson, Pete Seifrid and Brian Chase should device a plan to protect the RFQ with a reverse trip diode that also includes a bias on the diode. This technique has been successfully employed in Pbar for the protection of TWT's. The bias allows detection of a disconnected or malfunctioning cable. Anyone that is now relying on software for critical protection needs to re-evaluate and redesign. Brian Chase will be installing the LLRF system and preparing it for use in the 212 MHz RFQ test the first week of September. The RFQ will be tested without the ion source, but with as much of the actual surrounding hardware as possible, i.e. gate valve, solenoid. The intention is to understand the x-ray shielding effectiveness of the new vacuum vessel. Nancy Grossman should be prepared to make measurements of x-rays for the test. Joe Dey and Kermit Carlson are checking out the anode power supplies for all the RF systems. Each RF systems will be tested into a dummy load before being hooked up to the RFQ's. A temporary hook up of bottled nitrogen will be used for the tests while the "tank farm" is being designed. Time should also be spent eliminating excessive nitrogen use by plugging leaks. Wes Mueller and Pete Seifrid have begun the connecting of the RF signals and diodes within the RF cabinets. Interlocks The following entry are the minutes from a meeting concerning interlocks. This was not widely distributed, so it is reproduced here. This meeting was intended to deal primarily with requirements and mechanisms for "soft" off-states and permits for RF power and accelerated beam. These permits and mechanisms are mostly related to operator utility, machine operation, and equipment protection. Some points related to personnel safety issues were discussed. Bob Webber opened with a list of questions to set the scope of items to be addressed. These included: by what means does an operator simply inhibit beam if there is no need for beam for a short time, what handle does the low level RF have in order to inhibit beam when a cavity is sparking, how does LLRF get informed that a cycle is expected to accelerate beam or not so as to deal with beam loading etc., by what mechanism does a vacuum valve (including the fast valve) inhibit beam if necessary when closed, how might we make good use of the IRM "beam permit" outputs (to kill beam when important parameters are out of tolerance), and what mechanism inhibits beam when so requested by the radiation monitoring system? Mike Shea then presented a straw-man block diagram depicting a PLC serving as a signal summation and logic block. The PLC drives the control point of an electronic pulse shifter capable of altering clock event generation to mis-time the ion source pulse relative to the accelerating RF pulse. Anticipated inputs to the PLC include the IRM "beam permit" signals, an operator console hardware beam switch, vacuum status, etc. There was discussion of the PLC response time and requirements which would force a particular response. Dave Peterson described PLC response times in the 1 - 10 millisecond range. We could identify no requirement for response to inhibit beam on a shorter time scale. Elliott McCrory urged that any beam permit system be designed so as to not preclude operation with different "kinds" of beam, e.g., 10 MeV target beam vs. 1 MeV studies beam. It was determined that this could be achieved satisfactorily by avoiding hard "wire-or'ed" signals upstream of the PLC summation and logic block wherever possible. This machine is not expected to require fast, automatic switching between different operating modes. Easily accessible, spare inputs to the summation point should also be provided and pre-programmed into the logic to permit quick attachment of as yet undefined logic controls and permit signals. The proposed model was adopted by consensus as the way we should proceed for "soft" beam permits. Mike Shea's group will design and provide the pulse shifter electronics and an operator's hardware beam switch for the control console. Dave Peterson will implement the "soft" beam permit PLC. The decision whether to include this functionality in the existing vacuum or power supply PLC or to use a dedicated PLC for this purpose was left up to Dave. The three beam control points utilized by either the "soft" beam permit system or the personnel safety system were identified as 1) the ion source pulse shifter, 2) the pre-target Faraday Cup, and 3) the LEBT vacuum valve upstream of the 212 MHz RFQ. The pulse shifter is used as described above and, when sufficient, is the preferred control point for any beam permit. The Faraday Cup (at this point in the design) serves as the radiation safety system critical device for target vault accesses. It is also the critical device for the "Chemist's Permit" to allow 10 MeV beam on target. At this point, the LEBT vacuum valve serves as the radiation safety system backup critical device for target vault accesses and as the critical device for the accelerator radiation monitor system. The Low Level RF system will source a beam inhibit signal when it's cavity spark detection circuitry inhibits RF drive to an RFQ and possibly under other conditions yet to be specified. The Low Level RF system will accept an RF Enable input (50 ohm active high TTL) allowing external controllers to inhibit the RF at a low level. At this time, only a vacuum system permit controls this RF Enable. Signals from clock events identifying beam vs. no-beam cycles are required by the Low Level RF system to facilitate RF beam loading compensation. The realization of a hardware pulse shifter led to discussion that the average beam current control (via duty cycle) might be better implemented in hardware than in software as heretofore envisaged. A hardware solution was thought to be more robust. It was agreed to continue with the software loop approach as the first solution. A single bit input to the beam permit PLC would be the mechanism by which this loop would control beam. Vacuum Richard DeHaas and Mike Donavant are building a electronics driver for the thermocouple gauges. Radiation Shielding and Safety Dan Holslin has submitted a design for the target vault here at Fermilab using standard shielding blocks. This design has been reviewed by Ralph Pasquinelli and Al Guthke of the Research Division. The blocks have been reserved and will soon be staged outside of lab G. There may be a problem in that the design has an extra layer of blocks to deal with the cracks between blocks. The design will have to be modified because it currently occupies the space six inches into the anode crowbar racks. Another modification will be the addition of signal penetrations and ventilation for the helium gas that is used to cool some of the targets. Phil Young will bring this to Dan's attention. Van Nguyen of SAIC is working on the HEBT shielding wall design. He will contact Kris Anderson for the company that pours lead into molds. Jerry Bida of BRF should start investigating a source for shielding bricks in the Shreveport area. Delivery by January 1997 is required. Jerry should contact Phil Young and Dan Holslin for details. Phil Young will submit a list of SAIC employees that require film badges and check that all the appropriate training is obtained. Nancy Grossman is the contact. Now that fixed target has started up, there has been indication of radiation effecting the wall flower detectors at lab G, and therefore everybody in the area. The research division is aware of the situation and is monitoring it. Bill Freeman reported that the level appears to be below the 0.14 mrem/hour level. They will continue to monitor levels. It is good practice to wear your badges while at lab G. Target Windows Nothing new to report. Water Systems Bruce Hoffman will get together with Technical Support to clarify who is responsible for water distribution manifolds for magnet cooling water. This question has not been answered since the last meeting and should be addressed immediately. All RF stations have LCW from the RD system. Leaks have been plugged making the systems ready for test. The water skid drawings have been finalized with the vendor. The size of the skids needs to be entered by Glenn Waver on the floor plans. Glenn should see Bruce for details. Bruce Hoffman indicated that a 20% solution of rydlime would not effectively clean out the aluminum RFQ vanes. A 50% solution may be too much. There has not been any evidence of poor flow through the RFQ's, hence they will be flushed with clean water. Flow switches on individual magnets may present a problem. It has been decided to have flow switches on the main headers with klixons on the individual magnets. Power Supplies/Electrical All power supplies have been delivered to Lab G. Mike Donavant has been checking them out as he installs them in the racks. Howie Pfeffer and Bob Oberholtzer are finalizing the AC power requirements so that electricians can begin wiring the rack. Howie indicated that a few details about remote control and mounting of the current regulation loop transducers needs to be finalized. Controls/Software Duane Voy gave a presentation of the current state of the controls software using lab view. Assignments for main systems and subsystems responsibility were reviewed. Duane will be contacting these individuals, or vice versa, if there are any questions. The importance of integrating the accelerator systems via the control system cannot be underestimated. Administrative Updated schedules for the remaining portion of Phase I were handed out. It is clear that in the next 4-8 weeks, we should be running beam once again in Lab G. This will take a considerable amount of effort. Please make every effort to get your work done as neatly and as soon as possible. The contract with BRF should be settled today. It is hoped that construction in the physical plant can begin as soon as Monday August 19. Kris Anderson and Ralph Pasquinelli will make a trip to Shreveport on Tuesday August 27 to finalize construction details. The PET project has now been awarded an experiment number, E-887. The final report of the DOE from the May Review has been received. Copies were mailed to each of the collaborating institutions. Interested parties at FNAL can review the document by contacting Ralph Pasquinelli. The next meeting will be Friday August 30, at 10 AM in the penthouse. RJP