Transcript Slayt 1 - Indico

Notes From SwissFEL
A. AKSOY, A. LATINA
Layout
• There are many things which can not be compressed
–
–
–
–
Bunch compressors
Beam transport
Photon lines
S-band section
• Do we gain enough with high gradient?
Stability
• Amplitude and phase stability of X-band RF modules
is critical;
– SwissFEL X-band linearizer requires
• phase jitter <0.07 deg
• amplitude jitter < 0.1%.
• The jitter is dominated by the driver amplifier and the HV stability
of the modulator.
• the modulators pulse to pulse HV needs to be stabilized to better
than 1e-10.
• We should make calculation as soon as possible not
only for RF but also alignment..
Breakdown
• SwissFEL operating @100Hz has breakdown-rate of
4x10^-8
– which means ~1 BD/h over the entire facility.
– CLIC operating @500Hz has breakdown-rate of 10^-7
which means 50 BD/h.
• Breakdown-rate tolerances are still very tight
• In case of a breakdown, thermal variation can be
such that the system gets detuned, and we will loose
a lot of time for recovering
– We should study for fast tuning (additional module or
controlling the detuning..)
RF issues
• X-band will have high power density at pulse
compressor (500MW for X-band)
– can cause high breakdown-rate.
– Even at C band the breakdown occurring in cavity is less
than breakdown occurring in waveguide
• We should be careful with the length of compressed
RF pulse.
– The filling time of the cavity is already about 100ns.
– For a pulse length of 150 ns might be challenging to
accelerate several bunches.
– We should take into account stucture length
Issues about 500 Hz repetition
• Pulse to pulse RF phase and amplitude stability
problems with the modulators The linear voltage level
needs time for being corrected.
– 100 Hz already gives problems but probably manageable..
• Cooling down, lot of heat dissipation which make the
cavity engineering challenging.
– Needs 4 times more cooling than 100 Hz. More operation cost
(SwissFEL would be 25 MCHF/year)
• Operation S-Band @500Hz would be challenging
• Beam loss @500Hz due to dark current would increase
cost of infrasturcure
• SwissFEL building 160 MCHF for 100 Hz repetition
• Photon detectors @400 Hz can work.
Undulators
• In general high “k” values are better:
– they allow you change the undulator gaps to tune the wave length.
– Photon flux is proportional to K.
• Large K values require big undulator period length and higher
beam energies
• If one wants to tune undulator strength “K”
– Planar standart is prefereble
• Constant wake field from beam bipe
• Easy operation and machining
• .However there exist limitation on period length.
– The wake due to small aperture varies wiht K for in-vacuum undulators
• One needs to make fine tunning for each K
• in this sense going to higher energy and small laser
wavelength, standart undulators has advantage
User Requirements; Hard X-ray
• Running at 15 GeV would give 5 times more photons and
relax the micro-bunch instability by a factor 2
– Undulators would be similar, max 60 meters long each but longer
period
• means easiest fabrication
– Relax a bit the emittance requirements
– Photon lines would be longer to avoid damages on the screen
• ~150 m instead of ~100 m lines
– Single molecule/crystal imaging will need about 20 GeV machine
in order to get 10^34 photon /pulse.
• But at higher energy the incoherent synchrotron radiation starts
deteriorating the electron beam along the undulator line.
– Very high photon density per pulse would make the machine
unique.
• This allows decreasing the size of the samples (easy crystallization)
increasing the chance of getting defect free samples
– Example Interesting physics at 14 KeV photon energy
• “Mössbauer effect”. Very sharp edge of Fe.
Soft x-ray
• Energy would be around 2 GeV
• User requires high average laser power
• Users work with the idea that they don’t need
to re-use the samples)
• Superconductive RF is the solution / MHz rep
rate (take LCLS-II)
• However full CLIC time structure, 150 bunch
/pulse and 500Hz rap-rate would be useful for
soft X-rays
Recommendation from SwissFEL:
• Choose one good application for proposing
FEL. High photon power at one photon pulse.
E.g. Single-Crystal Imaging, which is not
covered by any FEL.