Download Diagnostics for high repetition rate ERL injectors Florian Loehl

Diagnostics for high repetition
rate ERL injectors
Florian Loehl
for the Cornell ERL team
FLS2010 Workshop, Stanford, March 1-5, 2010
Florian Loehl (Cornell University)
Cornell Injector Prototype
deflector
cryomodule
photocathode
DC gun
beam dump
buncher
experimental beam lines
Design parameter:
Nominal bunch charge
77 pC
Bunch repetition rate
1.3 GHz
Beam power
up to 550 kW
Nominal gun voltage
500 kV
SC linac beam energy gain
5 to 15 MeV
Beam current
100 mA at 5 MeV
33 mA at 15 MeV
Bunch length
0.6 mm (rms)
Transverse emittance
< 1 mm-mrad
FLS2010 Workshop, Stanford, March 1-5, 2010
Florian Loehl (Cornell University)
Diagnostics challenges for high
repetition rate ERL injectors
• The dynamic range of many systems needs to be HUGE!
Examples:
– bunch charge:
– duty cycle:
~ fC to 100 pC
~ 10-5 to 1
• High repetition rates of up to 1.3 GHz
• Pulsed operation AND CW operation
• Very high power levels in many systems
– up to 550 kW beam power
– up to 50 kW beam power from the DC gun alone
– Laser power in the IR >100 W
• Similar beam quality similar to state-of-the-art FEL injectors
FLS2010 Workshop, Stanford, March 1-5, 2010
Florian Loehl (Cornell University)
Advantages of high repetition rates:
fast measurements
Emittance measurement system:
Faraday
Cup
corrector pair
“scanner 1”
slit 1
corrector pair
“scanner 2”
slit 2
• No moving mechanical parts
 Allows for very fast measurements
(~ 2 s to 5 s)
Will be used for a parametric
optimization of the injector
FLS2010 Workshop, Stanford, March 1-5, 2010
Florian Loehl (Cornell University)
Difficulties with high repetition rates
Transverse deflecting cavity
Number of cavities
Max transverse kick voltage
Max RF power
Average power
Pulse duration
Max rep. rate
1
200 kV
3.8 kW
200 W
60 µs
1 kHz
time
unstreaked
beam
(leakage)






streaked
beam
Tuner and tuner
mechanism
Input coupler
beam energy
Water cooling
channel
Protrusion
• Very good extinction ratio required in
pulsed operation mode (> 106) for many
integrating measurements
FLS2010 Workshop, Stanford, March 1-5, 2010
Beam pipe
Field probe
Pumping port
Florian Loehl (Cornell University)
Difficulties with high
power levels
beam current (mA)
12
9 mA maximum current
5 MeV, 250 keV gun voltage
10
8
 45 kW total beam power
 2.3 kW only from gun
6
4
2
0
10:00:00 PM
10:30:00 PM
11:00:00 PM
• Main limitations:
- Gun high voltage instabilities at various beam currents
- Laser amplitude instabilities
Heavy beam loading even in the gun!
FLS2010 Workshop, Stanford, March 1-5, 2010
Florian Loehl (Cornell University)
Laser system
Simplified layout:
50 MHz / 1.3 GHz
Yb laser oscillator
timing FB
• Work on increasing the
extinction ratio for pulsed
operation mode by installing
acousto-optical modulator
Pre-amplifier
power FB
main amplifier
• Implementation of additional
feedback loops for laser
power and timing control
AOM
Pockels cell
SHG
beam shaping
FLS2010 Workshop, Stanford, March 1-5, 2010
Florian Loehl (Cornell University)
Gun voltage stabilization
Gun
We implement a beam based
feedback loop based on a time-offlight measurement
pick-up1
Time-of-flight
measurement
Regulation uses uTCA regulation
electronics developed at DESY
pick-up2
e-beam
FLS2010 Workshop, Stanford, March 1-5, 2010
Florian Loehl (Cornell University)
Laser Position Stabilization
• Variable attenuators increase
dynamic range by 4 orders of
magnitude
quadrant
detector 1
variable
attenuator
piezo
mirror
quadrant
detector 2
variable
attenuator
piezo
mirror
FLS2010 Workshop, Stanford, March 1-5, 2010
• At full laser power (20 – 50 W)
the attenuators are operated
close to the damage threshold
(~30 W / cm2)
• Not yet tested over a wide
range of laser operation
conditions
• Dynamic range could be
further increased by polarizer
based attenuators
Florian Loehl (Cornell University)
Beam current stabilization
Gun
Pockels cell
laser beam
• We implement a beam based
feedback loop based on a bunch
charge measurement
• Maximum regulation bandwidth: 3 MHz
Expected monitor resolution:
pick-up
charge
measurement
10 fC
1 pC
100 pC
BW=
100 kHz
0.012%
0.003%
0.003%
BW=
10 MHz
0.85%
0.25%
0.25%
e-beam
FLS2010 Workshop, Stanford, March 1-5, 2010
Florian Loehl (Cornell University)
Transverse beam profile at
high beam currents
‘Flying wire’
• 20 µm carbon wire
• should withstand ~MW beam
• up to 20 m/s wire speed
Not yet commissioned
FLS2010 Workshop, Stanford, March 1-5, 2010
Florian Loehl (Cornell University)
Longitudinal beam properties
at high beam current
THz interferometer ?
Installed but not yet
commissioned
Electro-optic diagnostics?
Resolution limited by low
beam energy (1/γ)
FLS2010 Workshop, Stanford, March 1-5, 2010
Florian Loehl (Cornell University)
transverse
position
How to measure the beam positions
of both beams in an ERL?
accelerating beam
deaccelerating beam
∆t = 0.77 ns
time
beam pick-up
Options:
• direct sampling with several GHz ADCs
single bunch measurement
but: expensive, no high resolution ADCs available
sensitive to ADC timing stability
pick-up signal
amplitude
• Two-frequency down-conversion scheme
FLS2010 Workshop, Stanford, March 1-5, 2010
Florian Loehl (Cornell University)
How to measure the beam positions
of both beams in an ERL?
Two-frequency down-conversion scheme:
~ 2.6 GHz ± fIF
2.6 GHz
~
~
~
~
~
~
1.3 GHz
~
IQ detection
average amplitude for both beams
IQ detection
difference between amplitudes
of both beams
1.3 GHz ± fIF
 phase information only used to
determine sign of difference signal
• If the second beam is not exactly at a phase of 180 deg with respect to
the first beam, this causes a constant offset in the 1.3 GHz signal.
 no concern
• Non-constant phases of the second beam:
Can be corrected for with the phase information of IQ detectors?
Two beam BPMs can be tested in our ERL injector by
splitting and delaying the 1.3 GHz drive laser beam.
FLS2010 Workshop, Stanford, March 1-5, 2010
Florian Loehl (Cornell University)
Conclusion
Special challenges for high repetition rates
• Large dynamic range
• Large power levels
Pulsed mode / low current mode
• Many diagnostic systems can be similar to low current injectors
• All important monitors are available
High current CW mode
• Still need research on
- Transverse beam profile measurements
- Longitudinal beam profile measurements
- Stability !
FLS2010 Workshop, Stanford, March 1-5, 2010
Florian Loehl (Cornell University)