805%20MHz%20Pillbox%20Cavity

1
805 MHz Pillbox Cavity
(IIT, Chicago, IL, Feb. 7th, 2002)

• Derun Li
• Center for Beam Physics
• Lawrence Berkeley National Laboratory

2
Collaborators

• J. Corlett, A. Ladran, R. MacGill, R. Rimmer,
• J. Wallig, M. Zisman
• Lawrence Berkeley National Laboratory
• Berkeley, CA 94720
• D. Summers, M. Reep
• University of Mississippi, Oxford, MS
• A. Moretti, A. Rowe, Z.B. Qian, V. Wu
• Fermi National Accelerator Laboratory
• Batavia, IL

3
Introduction Review

• Development of the 805 MHz Pillbox cavity
• High shunt impedance and high acceleration
  gradient at order of 30 MV/m
• Allow for testing of Be windows with different
  thickness, coatings and as well as other windows
• Study RF cavity operation issues under the
  influence of strong magnetic fields at both the
  solenoid and gradient modes
• The cavity design and status
• The 805 MHz pillbox cavity design should allow
  for testing of different windows ? demountable
  windows to cover the beam irises (we have five Be
  windows, four Cu windows ...)

4
Cavity Design status
Coupler
Waveguide window
Thermo-couples or view ports
Be (or Cu) windows
Three more view ports on the equator
Pillbox cavity
5
Cavity Design Parameters

• Frequency 805 MHz
• Shunt Impedance
• 38 M?/m (Z0) 32 M?/m (ZT2) Z V0,T2/P
• Quality factor Q0 18,800
• Coupling Constant
• ?c 1.0 at critical coupling ltEgt 30 MV/m
  requires 2 MW peak power, 350 watts average, 52
  watts on windows (Cu, 66 watts for Be) at duty
  factor of 1.8x10-4 (12 FNAL19 us pulse length
  and 15 Hz repetition rate).

6
Manufacturing of the cavity

• The cavity was fabricated at University of
  Mississippi, brazed at Alpha Braze Comp.

7
Cavity and coupler tuning

• The cavity was ready for final tuning for the
  frequency and coupling in June, 2001.
• (June 24-27, 2001 in University of Mississippi)
• Before tuning f 803.198 MHz, ?c 0.12
• After tuning f 805.486 MHz, Qext 12,800
• After final brazing
• Measurements done in November 14, 2001 at LBNL
  before shipping to Lab G at Fermilab

f 804.946 MHz, ?c 1.3, Q0 15, 000
8
Cavity Tuning (continued)
Michael and Daniel assembling the cavity
Cavity halves, coupler and Cu windows
9
Cavity Tuning (Continued)

• Frequency tuning (shortening the gap)
• Coupler tuning (widening coupling slot and
  shortening the transition waveguide)

Coupling slot
10
Coupler Tuning (continued)

• The couple slot angle
• was widened from
• 400 to 500 (two cuts)
• in order to get close
• to critical coupling.
• Measurements of
• the coupling agree
• well with time domain
• MAFIA simulations

11
Vacuum issues

• After final brazing at Alpha-Braze Comp.
• Vacuum leaking was detected at the coupler.
• Cu coating on the big transition waveguide was
  stripped and re-coated at LBNL.

12
Modification to coupler
O-Ring Grove
13
Vacuum issues (continued)

• Old flanges were removed, and welded with new
  ones
• UHV cleaning, RGA scan
• Test setup at LBNL

14
Vacuum issues (continued)

• After many failures, an O-ring at the transition
  of the coupler guard vacuum were employed
• Cold and hot N2 gas was used to flush and bake
  the system
• Before shipping to Lab G at Fermilab
• Leak tight with base pressure (after a few days
  baking) 3x10-8 Torr and a clean RGA scan
• Shipped to Fermilab under pressured N2 gas.
• Lesson learned

DONT MACHINE A BRAZED JOINT ? IT MAY LEAK !
15
Diagnostics

• Three view ports on the equator of the cavity
• Three RF probes (adjusted to about -52 dB gain
  from the standard waveguide port) available for
  E M field measurement
• An optic bore-scope can be used to inspect the
  windows and inner surface of the cavity ( need
  lighting and TV/VCR for viewing and recording)
• Maximum six thermo-couples can be attached to
  monitor temperature distribution on windows
• Two compartments behind the cavity can be used to
  measure window deflection by frequency shift

16
Diagnostics (continued)

• Equipment available
• at Lab G
• x-ray, dark current
• spectrum, forward
• reflected RF power
• and arc detector
• Lab G layout ?

17
Be windows

• Pre-stress study of the Be windows
• Received two more Be windows (10,20 mills)
• Two new copper windows with coatings
• Halogen lamp heating measurements on two failed
  Be windows
• Accelerometer measurements
• Acoustic measurements
• FEA (ANSYS code) modeling
• Magnetic and E M forces on windows

18
Be windows (continued)
Acoustic Measurement
Setup for accelerometer measurement
19
Current Status

• Cavity with Cu windows is now at Lab G
• Leak tight at pressure of 2x10-7 Torr (1/22/02)
• Braces are being made at Fermilab
• Leak checking is underway
• Waiting for vacuum parts to be made for final
  assembling of the vacuum and waveguide system
• LBNL personnel (with test plan) will be at Lab G
  in Fermilab before turning on the RF power

20
Test Plan

• A complete microwave measurement
• Resonant frequency
• Coupling constant
• Q factor
• Verify calibration of the RF probes
• Final assembling of the vacuum/waveguide systems
• Bake with N2 flow at 100 oC for two days
  preferably with RGA scans

21
Test Plan (continued)

• Continue baking at vacuum for two days
• RF conditioning slowly starting at low power
• Push for high gradient
• 2.5 MW peak power ? 34 MV/m accelerating gradient
  and 500 watts at 1.8x10-4 duty factor
• Turn on superconducting magnet in solenoid mode
  to repeat above studies
• Replace two Cu windows with the ones with TiN
  coatings

22
Test Plan (continued)

• Should we test Be windows? NO!
• EHS issues need to be addressed first
• may contaminate the vacuum system
• Future test plan depends on the upcoming test
  results and inputs from MUCOOL collaboration.