Linac RF Commissioning with the SNS HPRF Systems 4 th CWHAP Workshop May 3, 2006 M. McCarthy, R. Fuja, P. Gurd, T. Hardek, Y. Kang SNS RF Group
The SNS RF Group The mission of the RF Group is to ensure high power rf, precisely controlled in phase and amplitude, is provided to the linac beam in a safe, highly reliable and cost effective manner. RF Engineers - 8 RF Technicians 9 Klystron based RF Systems 92 Gridded tube RF Systems 8 Total peak pulsed RF power available 84.13 MW 2
Operators Linac RF Status Screen 3
Waveguide Installation into Empty Gallery Tanks and Tubes Readied at Test Facility 4
550 kw Klystrons Transported to Gallery CPI THALES 5
Weather Not Always Cooperative HARDEK S HELPING HANDS 6
Source of Low Power Output Discovered in TH-2177 (805 MHz, 550kW klystron) magnet water leak filled up the magnet cylinder with water. 7
HV Oil Tank is Moved into Location using Air Pads HV Tank Final Prep before Oil Filling 8
Installation and Power Testing Complete - 5-5-05 9
Component 2.5 MW Klystron (DTL1) 2.5 MW Klystron (DTL1 - DTL6) 2.5 MW Klystron (DTL3) DTL5 Circulator 5 MW CCL klystrons 5MW waveguide 5 MW Loads SCL 550kW klystron SCRF Xmtr 9 Circulator Problem Water leak into HV oil tank Arcing at klystron output coax when Pout > 350 kw Burned through input cavity when HV was inadvertently applied with no magnet power. Slow accumulative internal water leak Arcing at output. Arcing in waveguide Arcing at window One internal body water leak, one internal magnet water leak. External water leak. Solution Replace section of body water circuit in-situ Factory design error. Installed spacer disk to equalize length of inner and outer coax conductors. Plated input cavity externally to seal leak. Reprogrammed PLC to apply mag power with filament power. Replaced circulator (factory warranty), installed water drains in connecting waveguides. Added SF6 to output and circulator. Air-cooled waveguide. Removed waveguide runs, cleaned flanges of all silicone, reinstalled. Replaced seals with different material. Returned to the factory for repair. Replaced with spare. Repaired on site with brazed part. 10
Klystron Internal Water Leak (into HV Tank) Early production units had QA problem with water fittings. High temperature brazing is potentially damaging to solenoid coils. 11
Modification to Air Cool CCLWaveguides 5 MW klystron will heat the waveguide and increase the chance of arcing. The un-used blower from the 2.5 MW tubes was put to use here. 12
5 MW Circulator Window: Before / After Window was weakened by arcing along bottom of flange Arcing caused by leakage of SF6 13
5MW water load Seal replacement to prevent arcing Magnet Cable junction box added to Klystron HV tanks 14
15
16
Congestion Between Tunnel Wall and Cryomodule 17
Plate Added to Klystron Base to Allow Separation of Tube From Tank Without Removing Socket 18
Socket Holding Plate Under 550 kw Tube 19
402 MHz DTL Window With Low Water Flow 20
Water Passage Connections 21
Some Debris Found in the Water Passage 22
Source of Low Power Output Discovered in TH-2177 (805 MHz, 550kW klystron) magnet water leak filled up the magnet cylinder with water. 23
Leaky Circulator DTL5 Water leak at internal joint Replacement coming in. 24
Leaking Magnet Filled Cylinder Up to Here 25
Broken 2.5 MW Klystron Output Coax-to-WG Connection Symptoms: Arcing seen in air vents at coax to waveguide interface when power exceeded 350 kw peak. Problem: Inner and outer coax unequal in length causing stress on connecting ring. 26
2.5 MW Klystron Output Coax to Waveguide Connection. Gold Ring Connects Inner to Outer Conductor. Inner Conductor 1-2 mm longer than Outer Conductor. Resulting Stress Fracture. 27
Loose Water Guide Pipe Inside 550 kw Water Load 28
Klystron Operating Parameters Nicely Labeled on Name Plate. 29
Technique for Pre-Bending Flexible Waveguide 30
Dedicated Fill Spout Hastens Oil Fill Operations and Eliminates Static Electricity HVCM Interlocks Junction box facilitates quick disconnect to klystron and allows measurement of magnet current and voltage. 31
Much Support Came From LANL 32
RFQ/DTL 2.5 MW Klystrons 33
Most Recent DTL-3 Klystron Failure 34
DTL-3 2.5 MW Klystron with shorted magnet cables. Cables were likely shorted by water leaking into barrel connector. 35
Melted Electrical Tape and Corroded Magnet Connections Short May Have Started By Water Leaking into This Connector 36
Klystron Water/Electrical Connections 37
Power Upgrade Project (PUP) Preliminary Layout Basic Fit 12 Klystrons per HVCM Water Pump Room Water Systems Machine Shop HVCM Repair Area HPRF Tested and Ready Spares with local quick test bench Layout courtesy of Marianne Champion 38
SCRF Power Requirement Comparison 700 600 500 623kW Peak Klystron Output Power Requirements 42 ma PUP cryomodules RF Power (kw) 400 300 26 ma 42 ma 26 ma 200 100 0 1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 96 101 106 111 116 SC Cavities 39
SCRF Cathode Voltage Requirements shows HVCM Must be Modified to Drive 12 klystrons with more than 74 kv Cathode Voltage Requirements HVCM outp ut (V olts) 90000 80000 70000 60000 50000 40000 30000 20000 10000 0 1 7 13 19 25 31 37 43 49 55 61 67 73 79 85 91 97 103 109 115 SC cavity 11 pk (76 kv) PUP Cathode Voltage Baseline Cathode Voltage 9-pack limit: 81kV 12-pack limit: 74 kv 40
Conclusion RF Equipment problems mostly mechanical (QA and water related). Beginning of Beam on Target Operations started 4-28-2006 Change over to operating regime/culture Track failure-mode statistics Ensure tested spares are ready and accessible Planning for PUP RF Equipment layout and acquisitions starting. 41