HOM Transfer Function Measurements and other topics Larry - - PowerPoint PPT Presentation

hom transfer function measurements and other topics
SMART_READER_LITE
LIVE PREVIEW

HOM Transfer Function Measurements and other topics Larry - - PowerPoint PPT Presentation

HOM Transfer Function Measurements and other topics Larry Doolittle, LBNL ICFA Workshop on High Order Modes in Superconducting Cavities 2014-07-15 Applicability any linac with photoinjector (context is LCLS-II) HOMSC14, July 13-15, 2014


slide-1
SLIDE 1

HOM Transfer Function Measurements and other topics

Larry Doolittle, LBNL ICFA Workshop on High Order Modes in Superconducting Cavities 2014-07-15

slide-2
SLIDE 2

Applicability

  • any linac with photoinjector

(context is LCLS-II)

HOMSC14, July 13-15, 2014 2

slide-3
SLIDE 3

LCLS-II Numbers

Game of subharmonics

  • L-band RF is f0 =1300 MHz
  • Gun cavity is f0/7 = 185.714 MHz
  • APEX construction uses 1 laser pulse per 5 gun cavity cycles, 37.143 MHz

Most short-pulse lasers have rep-rate between 30 MHz and 100 MHz 99% of the documentation for the planned LCLS-II talks about the photon pulses at the photocathode, and by derivation the electron bunches, being regularly spaced at ‘1 MHz’, sometimes with the footnote that this is really 928.6 kHz.

  • One nominal bunch every 40 laser pulses

HOMSC14, July 13-15, 2014 3

slide-4
SLIDE 4

What’s in a laser?

“A chain of 4 pre-amplifiers gets the seed from the oscillator. After the first two pre-amplifiers, the repetition rate is decreased to 1 MHz by a acousto-

  • ptic kicker (AOM), driven by a 2 W 100 MHz pulsed rf signal synchronous

with the oscillator. The AOM’s rise and fall time are at the 20-30 ns level.”

  • The Photocathode Laser System for the APEX High Repetition Rate

Photoinjector D. Filippetto et al., Proceedings of FEL2013

Flexible!

There are many limits to how often the AOM can be fired, but especially at low bunch charge, it is pretty much pulse-on-demand subject to the 26.9 ns granularity, and keeping approximately 1 pulse per microsecond or less over tens of microseconds.

HOMSC14, July 13-15, 2014 4

slide-5
SLIDE 5

Normal

Usual spectrum of ‘1 MHz’ beam, with TE111 mode impedances shown.

5 10 15 20 25 30 35 40 2442 2444 2446 2448 2450 2452 2454

Strength (dB) f (MHz)

HOMSC14, July 13-15, 2014 5

slide-6
SLIDE 6

Abnormal

Same conditions, but jiggle each pulse’s timing ±13 ns to hit 2449.9 MHz

5 10 15 20 25 30 35 40 2442 2444 2446 2448 2450 2452 2454

Strength (dB) f (MHz)

HOMSC14, July 13-15, 2014 6

slide-7
SLIDE 7

Normal

Usual spectrum of ‘1 MHz’ beam, with TE111 mode impedances shown.

  • 40
  • 30
  • 20
  • 10

10 20 30 40 2380 2390 2400 2410 2420 2430 2440 2450 2460

Strength (dB) f (MHz)

HOMSC14, July 13-15, 2014 7

slide-8
SLIDE 8

White phase noise added

Random jitter modulated

  • 40
  • 30
  • 20
  • 10

10 20 30 40 2380 2390 2400 2410 2420 2430 2440 2450 2460

Strength (dB) f (MHz)

HOMSC14, July 13-15, 2014 8

slide-9
SLIDE 9

White amplitude noise added

Random 5% pulse dropouts

  • 40
  • 30
  • 20
  • 10

10 20 30 40 2380 2390 2400 2410 2420 2430 2440 2450 2460

Strength (dB) f (MHz)

HOMSC14, July 13-15, 2014 9

slide-10
SLIDE 10

Transfer function

  • Excite from photoinjector
  • Couple to monopole cavity mode
  • Synchronous RF measurement on HOM probe (or fundamental probe)

Timing system planned for LCLS-II is flexible enough to synchronize this stimulus/response over whole accelerator site

HOMSC14, July 13-15, 2014 10

slide-11
SLIDE 11

Alternative

Modulating the bunch charge, as described by Daniel Hall of Cornell, is perhaps simpler to think about. But a full linac’s longitudinal bunch propagation is sensitive to bunch charge due to short-range wakes. Nothing in the system besides HOMs (and presumably the end-station photon experiments) is sensitive to ±13 ns changes in bunch arrival time.

HOMSC14, July 13-15, 2014 11

slide-12
SLIDE 12

Conclusions

  • Photoinjectors don’t have to be operated only at their nominal setting
  • Pulse-on-demand feature can be used as a programmable frequency

source to excite (or avoid exciting) HOMs at any frequency Thanks for listening!

HOMSC14, July 13-15, 2014 12

slide-13
SLIDE 13

Familiar Block Diagram

Conditioning Signal Conditioning Signal Conditioning Signal

housekeeping and custom functions FPGA

Conditioning Signal

DAC DAC ADC ADC Host CPU or PHY Host I/F Network

HOMSC14, July 13-15, 2014 13

slide-14
SLIDE 14

Familiar Block Diagram

housekeeping and custom functions FPGA DAC DAC ADC ADC Host CPU or PHY LO Clock Host I/F Network

HOMSC14, July 13-15, 2014 14

slide-15
SLIDE 15

Data Acquisition

Antenna 1.7 GHz 20 MHz Antenna 3.9 GHz 20 MHz ADC ADC LO1 LO2 LO 2.8 GHz Antenna ADC 1.7 GHz

HOMSC14, July 13-15, 2014 15

slide-16
SLIDE 16

Data Acquisition - DESY/Bou Habib

0.1 0.2 0.3 0.4 0.5 0.45 0.5 0.55 0.6 0.65 0.7

fs (GHz)

1.3 GHz 1.7 GHz 2.4 GHz

HOMSC14, July 13-15, 2014 16

slide-17
SLIDE 17

Data Acquisition - LCLS-II LLRF compatible

75-177 MHz freq. monitor 1320.0 MHz downconversion LO 20 MHz IF x3 VCO 5100-6800 MHz HOM Antenna 1200-2800 MHz? PLL 188.57 MHz reference ÷16 3960 MHz HOM second LO 3940 MHz 20 MHz 20 MHz IF HOM Antenna 3940 MHz 20 MHz 2-board set, plus

  • ff-board 3954 MHz

cavity filters (20 MHz BW) 1200-2850 MHz HMC586? ADF4106? HOMSC14, July 13-15, 2014 17