1. Accelerator Concepts Workshop
Heterodyne measurement of Coherent Transition Radiation (CTR) from Seeded Self-Modulation (SSM) in AWAKE
Falk Braunmueller, P. Muggli, M. Martyanov, F. Batsch, K. Rieger, A. Caldwell & AWAKE team
27 September 2017
3rd European Advanced
Accelerator Concepts Workshop
Elba, Italy

2. Outline Setup of heterodyne CTR-measurements Measurement principle
Measurement processing
Main result: fCTR = fplasma(nRb)
Further results: Dependence of SSM on Rb-density gradient 2

3. SSM-Diagnostics via CTR
Coherent transition fmodulation (90-280GHz)
preliminary
Modulated
p+-bunch
OTR
OAP-
mirror
CTR
F. Batsch, Poster session 19:30
Courtesy: T.Haubold
preliminary
Coupled into WR90 waveguide 15m transmission

4. SSM-Diagnostics via CTR
Coherent transition fmodulation (90-280GHz)
Frequency:
Heterodyne mixing
Modulated
p+-bunch
OTR
OAP-
mirror
15m transmission
Laser-based
Courtesy: T.Haubold
Waveguide-
based
Amplitude:
Schottky diodes

5. Diagnostic setup End of transmission line
3 Heterodyne receivers for CTR:
Laser-based mixing (last presentation)
WR8 / GHz: Radiometer- systemnew
WR3.4 / GHz: VDI-system from EPFL
 replaced by WR4.3/ GHz system
Can detect 2nd harmonics of fmodulation
Mirror
Beam splitter |
August
June 5

6. Measurement principle
Signal:
Expected signal (fIF=10GHz)
Intermediate
frequency:
fRF ~260GHz
fIF ~10-20GHz
Reference:
fref~270GHz
Mixer
fref from frequency-multiplication of tunable local oscillator
fref = nharm fLO
Also mixing with weaker parasitic reference frequencies
fref = nharm,1 fLO . (nharm,1= nharm +/-1 , …)
Confirm that signal on oscilloscope is from mixing with correct reference frequency:
fix
fIF= | fRF – nharm fLO|  nharm = ∆fIF/∆fLO
setting
measured
to be determined 6

7. CTR-signal from mixer Short signal, close to expected length
Very precise
Strong single-frequency- component (find via spectrogram)
 fIF 7

8. Data-selection Choice of useful data:
Signal level large enough, e.g. > 40mV
Use only ‘prominent peaks’:
Significantly higher than other IF- peaks
(shot-to-shot variation of parameters)
Previously: selection ‘by eye’ 8

9. CTR-analysis Fit fIF vs. fLO to check nharm
 In general, expected nharm =8 / 12 / 24 is confirmed (sometimes ambiguous)
fRF = nharm fLO +/– fIF
Average & standard deviation of fRF
(here: 255.9GHz +/- 1.4GHz)
Unclear if from change of CTR-freq. 9

10. Results of CTR-analysis
Result: fCTR vs. nRB
fCTR = fplasma(nRb)
SSM with fCTR=fplasma as predicted
Rb fully ionized
Good match between fundamental & 2nd harmonics
 proof that correct nharm(fLO) was chosen
Excellent fit result: parameters within 0.3%
Error analysis incomplete
Error bars: Std GHz
vapour
Preliminary result
(95% confidence of fit on mean) 10

11. CTR-amplitude Amplitude increasing with beam-charge ECTR~q, but:
(WR4-system)
(Standard dev.)
Amplitude increasing with beam-charge
ECTR~q, but:
SSM-amplitude affected in nontrivial way
Emission angle & coupling may be affected
Promising for future analysis
Preliminary result 11

12. fCTR-dependence on nRb-gradient
138GHz
10% gradient
10 m 12
Preliminary result
131GHz 12
5% gradient
134.5GHz
Negative Gradient: fCTR=129GHz~const.
10 m
131GHz
Moving average
Evolving interaction over several meters!
Frequency increasing with positive gradient,
but basically constant with negative gradient
 Explanation from SSM?

13. SSM-Dependence on nRb-gradient
No/Small gradient  microbunches reach less far
fRF~fplasma(end)
Preliminary result & possibly varying parameters
Gradient >5%:
Microbunches longer visible after seeding
fRF corresponds more to fplasma(end)
 longer interaction in plasma?
ξ along bunch [a.u.]
Negative Gradient: fCTR=129GHz≈const.
Check with simulations ?!
138GHz
10% gradient
entry
10 m
fplasma
131GHz 13

14. Summary Several successful upgrades of heterodyne CTR setup
Consistent results after data down-selection
Very successful measurement of fCTR=fplasma(nRb), confirming full ionization + character of SSM
Clear correlation between beam charge & signal amplitude
Investigation of self-modulation physics:
fCTR=fplasma(nRb,downstream) for positive nRb-gradient
Longer persisting microbunches
Analysis to be continued
 longer interaction?
Preliminary results 14

15. Thanks for your attention!
Acknowledgement ( GHz-system):
Work supported by Requip, Sinergia and (No: /1), grants of the Swiss National Science Foundation, by the Ecole Polytechnique Federale de Lausanne (EPFL) and by Faculty of Basic Sciences of EPFL.

16. Additional slides

17. Analysis/Measurement To-Do-List
Apply criterion of prominent peak to all points
Analysis of signal amplitude: need to correlate with ‘good shots’ from streak camera & two-screen halo-BTV
Frequency-variations correlated with alignment/ angle of p+-defocusing/… ?
Ratio of signal amplitudes V(2nd harmonics)/V(fundamental) vs. p+ charge
idea: more non-linear  stronger 2nd harmonics? 17

18. Measurement principle
fRF = nharm fLO +/– fIF
=nharm fLO
known
measured
to be determined
fRF ~260GHz
fRF=100GHz
fRF=260GHz
fIF ~10-20GHz
fref~270GHz
Mixer / Schottky diode
Expected signal (fIF=10GHz)
Find nharm by scanning fLO:
nharm = ∆fIF / ∆fLO

19. Heterodyne Measurement
Measure intermediate frequency (IF) between CTR-signal (RF) and known reference
Reference signal from frequency-multiplied tunable local oscillator (LO)
Waveguide Transmission of RF over 15m
Small measurement bandwidth
Good signal efficiency
VDI heterodyne receiver from Swiss Plasma Center (SPC) at EPFL (Lausanne)
~10-20 GHz out
Reference
RF in

20. Waveguide Transmission Line
Detector behind shielding wall
15m of overmoded waveguide WR90 (fundamental mode 8- 12GHz)
E-field
polarization

21. Measurement principle
fRF = nharm fLO +/– fIF
=nharm fLO
fixed fLO
known
measured
to be determined
fRF ~260GHz
fRF=100GHz
fRF=260GHz
fIF ~10-20GHz
fref~270GHz
Mixer / Schottky diode
Expected signal (fIF=10GHz)
nharm fLO = fRF
Single fRF with fixed fLO can give several fIF-signals
 Signal frequency must be constant to within 1-2GHz!