Troubleshooting Status for the ALS In
Contents
1. for each experiment and and a beam loss was observed at the highest speed 1 mm s vs 0 1 mm s When the beam losses occurred some vacuum ac tivity was observed A Prax 11 107 t Torr but it seems to be too small to likely cause the beam losses One important observation we made during this shift is that when the beam losses occurred we could see through the synchrotron radiation monitor that the beam was blow ing up in the vertical direction before recovering its normal shape in the cases of partial beam losses Closing to minimum gap When the VID vacuum be came good enough 10 Torr range 10 Pa we could manage to close it to the minimum gap gradually This im portant step allowed us to dedicate some time to scrubbing and thus improving the VID vacuum However one of the first times we left the VID at the minimum gap we saw sudden pressure spikes happening without beam loss Then we were able to close the VID to the minimum gap directly without experiencing any beam losses except for one particular case The first successful attempts to close the gap were performed while the orbit feedback was on and all the insertion devices where wide open When we performed the same experiment with all the insertion de vices closed we experienced a partial beam loss between the 6 mm and 5 5 mm gaps Magnets T15 Undulators and Wigglers MO6PFP087 We are continuing our efforts to track the instability source s One of them2. html 2 3 Spring 8 http www spring8 or jp en C Steier P Heimann S Marks R Schoenlein W Wan W Wittmer Successful completion of the femtosecond slicing upgrade at the ALS PAC 07 Evans Analytical Group EAG http www eaglabs com 4 335
3. Proceedings of PAC09 Vancouver BC Canada MO6PFP087 TROUBLESHOOTING STATUS FOR THE ALS IN VACUUM INSERTION DEVICE A Madur S Marks S Prestemon D Robin T Scarvie R Schlueter C Steier LBNL Berkeley CA 94720 USA Abstract In 2006 the 30mm period In Vacuum Insertion Device IVID was operational for the femtosecond phenomena beamline at the Advanced Light Source ALS of Lawrence Berkeley National Laboratory Since routine operation be gan a Series of partial or total beam losses as well as coin cident sudden pressure increases within the VID vacuum system have occurred while changing the VID gap This paper reports these observations and describes the investi gations and the repair attempt performed on this insertion device INTRODUCTION The Advanced Light Source ALS at Lawrence Berke ley National Laboratory installed and commissioned its first In Vacuum Insertion Device IVID in 2006 The IVID was designed and built by the Neomax Corp 1 as a col laboration with the Spring8 insertion device group 2 This IVID is dedicated to two femtosecond phenomena beam lines 3 Figure 1 shows a picture of the IVID Figure 1 ALS In Vacuum Insertion Device The IVID is 2 05 m long 30 mm period hybrid planar undulator Permanent magnet iron poles with a mini mum operation gap of 5 5 mm 1 52 T While operating at smaller gaps beam losses total or partial have been as sociated with IVID gap movemen
4. ancouver BC Canada chlorine chromium cobalt copper iridium iron lead magnesium molybdenum nickel oxygen potassium sili con sulfur titanium and zinc The presence of copper and nickel is consistent with our interpretation from the Fig 3 1 e deterioration of the conductive foil After several dis cussions with Spring8 2 and colleagues from the insertion devices community we decided to replace the conductive foils Replacement of the Cu Ni Foils We replaced the conductive foil in collaboration with the device manufacturer Hitachi Neomax 1 during the September 2008 ALS shutdown With the help of three qualified technicians from Japan we managed to remove the IVID from the ALS storage ring replace the conduc tive foils and re install the VID within four weeks The conductive foils replacement also provided an opportunity to directly inspect the IVID The main observation from this work is that the conduc tive foil seemed to be in good shape the scratches were relatively small and to be acceptable they likely occurred during the initial VID assembly The picture of the discol oration around the hole Fig 3 was actually showing an extra layer of nickel During the plating process it some times happens that some nickel material overflows to the copper side Consequently no copper was missing at that location and the presence of nickel on the copper side does not affect the conductive foil quality While inspec
5. could be the dust deposited on the IVID conductive foils We are investigating the possibility to isolate the VID vacuum during maintenance operations by adding a vacuum valve If the dust particle source is lo cated beyond the IVID such a valve may reduce deposition of these particles within the IVID CONCLUSION The list of possible causes for beam instability associ ated with IVID operation is getting shorter but this is still a work in progress Future investigations include a plan to tune the feedback system to make the beam less stable and consequently more sensitive to any instability The purpose is to be able to reproduce the beam losses we experience so that we can increase event statistics and hopefully iden tify their cause s We also plan to use a fast oscilloscope triggered right after a beam loss to determine the speed of propagation of the instability Troubleshooting the IVID to allow it to operate at small gaps has become an unexpectedly long project While we are continuing our investigations we are still learning about its behavior The random nature of the instabilities makes it difficult to methodically study the problem Im proved diagnostics will aid in determining and eliminating the source of losses ACKOWLEDGMENT We would like to thank Spring8 and Hitachi Neomax colleagues for the fruitful collaboration REFERENCES 1 Neomax Hitachi Metals company http www hitachi metals co jp e index
6. ensure the conductive continuity of the image currents This conductive foil is 110 mm wide and covers the mag nets over the whole length 1 5 m The Cu Ni foil was obtained from a 60 um thick sheet of copper plated with nickel SOum The magnetic properties of the nickel are used to ensure adhesion of the foil to the magnets this pro vides the heat conduction path for foil cooling At both ends the foils are clamped to the water cooled transition ta per plates made of OFHC copper Oxygen Free High Con ductivity These taper plates along with the sliding con tacts ensure the transition between the fixed gap vacuum chamber and the undulator varying gap for the continuity of the image currents Figure 2 shows the transition we just described INITIAL DIAGNOSTICS First Inspection In spite of the initial observations beam losses pres sure bursts we lacked instrumentation to properly diagnose these issues During the May 2008 ALS shutdown we in spected the interior of the VID with a borescope connected to a digital video camera The borescope we used is rigid 333 MO6PFP087 A d w eu Ts ar ry ey fg e a Figure 2 Transition parts between vacuum chamber flange and VID gap from left to right Cooling lines are not shown on this picture and is 80 mm long Since we could only access the interior from the viewport windows we inspected the ends region These regions include the transition plates the RF fin
7. gers and the extremities of the undulator i e the conducting foil installed on top of the magnets From the ends we had a global view of the inside of the IVID vacuum chamber From the inspection we determined the following e no discoloration on the sliding contacts they are not overheating e no discoloration on each part of the transition regions e some small scratches order of 1 mm wide and some wrinkles on the conductive foil e an important discoloration around one hole on the conductive foil the foil is designed with pairs of 2 mm diameter holes distributed longitudinally near the outer edges in order to get a flat surface on the foil no trapped volume after its installation e some dust deposition this dust is aligned with the transverse edges of the magnets Figure 3 Pictures from inspection Left discoloration and missing copper around a hole Middle dust deposited along magnet edges Right Line of dust and a scratch on the surface shiny spot The left picture from Fig 3 captured our attention Our interpretation of this picture was that it showed copper peeling off and cracking due to heat induced stress Moreover we collected some samples of the dust in order to analyze it The analysis has been performed by Evans Analytical Group EAG 4 using the Energy Dispersive X ray Spectroscopy method to determine the dust com position The dust materials include aluminum carbon 334 Proceedings of PAC09 V
8. nitial operation Beam losses and pressure bursts were again observed at smaller gaps Although the conductive foil replacement improved our knowledge concerning the state of the IVID it did not re solve the issue of the spurious beam losses and pressure bursts We are dedicating some time during the accelerator physics shifts to continue investigating the probable causes of our problems Accelerator Physics Shifts As the statistics for beam losses events was small and the reproducibility of such events was not evident the goal of the dedicated accelerator physics shifts is to find the condi tions that consistently trigger these events Gap motion We focused our initial attention to the gap motion Most of the events were indeed observed while the IVID gap was changing and only in the closing direction Since we were not able to close the VID to small gaps di rectly lt 9mm consistently without beam loss we tried to find a process that would allow us to reach any desired gap without beam losses We closed the VID gap by small steps 100um from 9 5mm to 5 5mm gaps at a speed of 1 mm s waiting for the vacuum to recover between each step We repeated this experiment twice and both experi ments were conducted at high beam current 500 mA and 440 mA Beam losses from the mA range to the 100 mA range were observed and they occurred at similar gaps for both experiments except for the 5 5 mm gap At this gap the speed was different
9. t To address this issue and to understand this unexpected behavior we started in vestigating all the possible causes This paper reports the observations we performed a review of the conductive foil replacement and the experiments performed during accel erator physics shifts after the conductive foil replacement This work was supported by the Director Office of Science U S Department of Energy under Contract No DE AC02 05CH11231 Magnets T15 Undulators and Wigglers INITIAL OBSERVATIONS Events Description Beam losses as well as pressure bursts were observed while the IVID was operating at small gaps This occurred particularly for currents above 200 mA and up to the max imum ALS current of 500 mA Because of the rapid nature of the observed beam losses it was impossible to distin guish between the cause and effect relationship of beam loss and pressure spike Below are listed the causes we considered mechanical obstacles damaged conductive foils misbehaving RF fingers bad contacts overheating conducting foil vibrations during mo tion misalignment bad thermal conduction e release of particles due to friction between two mov ing surfaces in contact bellows cooling parts of the transition pieces Components Likely Related to Beam Loss The magnetic structure is located within the vacuum chamber and a 110um thick Cu Ni conductive foil is in stalled over both magnetic structures top and bottom to
10. ting the old foils we also noticed tiny brown spots that were randomly spaced all along the foil It was not possible to tell if they were the result of the orig inal fabrication or if they have been generated by vaporized pieces of dust COMMISSIONING AND ACCELERATOR PHYSICS SHIFT OBSERVATIONS Commissioning After the foil replacement and the installation of the IVID back in the ALS storage ring the VID was again commissioned The result of the commissioning demon strated that the beam dynamics in term of tune lifetime dipole errors and beam size instability were not affected with the attempted repair We checked the VID vertical alignment with the elec tron beam and found an offset of 200 um which is similar to the one measured before the conductive foil replacement An alignment of the VID will be scheduled in the near fu ture in order to have a balanced beam clearance and heat deposition between the top and bottom conductive foils Following re installation of IVID in the ALS storage ring it took more than three months to achieve the desired vacuum performance 1071 Torr range 107 Pa Ther mal outgassing was the result of image current heating and upstream synchrotron radiation dissipating in the conduc tive foils This issue has prevented us from operating at Magnets T15 Undulators and Wigglers Proceedings of PAC09 Vancouver BC Canada small gaps and we had to limit the operational gap to 10 mm for i
Download Pdf Manuals
Related Search