To: J. Cook, D. Helms, W. Skinner
cc: Division (M7), FEL Coordination Group
From: F. Dylla
Subject: IR Demo Project Weekly Report, January 10-15, 1999
Date: January 15, 1999
Management
This week is a continuation of the January shutdown of the FEL
for planned maintenance and installation activities. A summary
of these activities is given below.
Final preparations were made for JLab presentations to the BESAC
Panel Meeting on Novel and Coherent Light Sources next week in
Gaithersburg, MD. In addition to a summary of the JLab FEL, supporting
presentations will be given by Len Feldman (Vanderbilt) Materials
Science, Bill Cooke (CWM) in Atomic Physics, and Kevin Lehmann
(Princeton) in Chemical Physics.
C. Bohn gave a presentation on JLab's FEL to the Directed Energy
Directorate of the Air Force Research Laboratory last Friday,
8 Jan 99.
Secretary of Energy Bill Richardson visited JLab this week and
voiced his strong support for the FEL program.
FEL Installation/Maintenance Activities
The following maintenance items were completed this week:
The new cesiator for the photocathode gun was installed and activated.
The photocathode was also installed in the gun, and the vacuum
bake is underway. In addition, a new residual gas analyzer for
monitoring the constituents of the gun vacuum was received, installed,
and is now operating. The metering circuits in the gun's high-voltage
power supply (HVPS) were modified to read 5 mA as the maximum
current, rather than the previous 4 mA. A test of the HVPS is
planned for next week to determine its ability to produce 5 mA
current into a low-impedance load.
After modifying the 3-way cross at the insertable dump in the
reinjection region into a 4-way cross, we installed a new viewer
there. It will aid in configuring and steering the recirculated
electron beam back into the cryomodule. In addition, the viewer
foil at the exit of the cryomodule was replaced with one having
a 5.75 mm-dia hole in the center. This will allow inserting the
viewer without disturbing the injected beam, thereby providing
a pristine view of the recirculated beam after deceleration through
the cryomodule.
New beam-current-monitor cavities were installed at the exit of
the first recirculation arc and in the reinjection region. They
will be used primarily as aids in setting the path length of the
electron beam around the machine.
A mechanical support was installed for the beam-position monitor
in the energy-recovery (ER) dump line. It had been omitted from
the original design of the beam line. Klixon temperature switches
were installed and wired on the lower part of the ER dump. They
are designed to open at 88 C and are for additional protection
from beam-induced heating.
FEL Commissioning Activities
The activities scheduled for the next two weeks are designed to
relieve stress on the people doing installation/maintenance while
providing for restoration of the "Dec 98" machine by
29 Jan 99. The vault will be open all next week to permit installation/maintenance.
Gun processing will proceed early the following week, and injector/accelerator/FEL
restoration will proceed after that. The schedule is only a projection;
it will be modified, if need be, as turn-on gets closer.
A series of benchtop measurements were made to provide data by
which to formulate plans for circumventing the waveguide IR detector
faults that surfaced in December. The faults are very likely
caused by higher-order modes. Bottom line: A copper screen of
~1 mm mesh size removes all measurable rf response at 19.5 GHz
and transmits ~60% of the IR. Screens are being installed over
all eight of the detectors in the cryomodule, and three gate valves
are also being installed to permit testing alternative solutions
and measurement methods with rapid turn-around. The trip points
will need to be reset during the time-frame of system checks (likely
early the week after next).
R. Li has first results from her coherent-synchrotron-radiation
simulation. She finds that the emittance growth through the first
recirculation arc is sensitive to the energy spread at the wiggler
location (which corresponds to the starting point of her runs).
For an assumed gaussian longitudinal bunch profile with rms parameters
previously measured at the wiggler, she predicts a 34% CSR-induced
emittance growth from wiggler to exit of the first arc. She has
benchmarked her code against analytic calculations pertaining
to the longitudinal CSR force, and is presently doing likewise
for the radial CSR force (the latter being more difficult analytically
than the former). She will also do runs to check for sensitivity
of emittance growth to longitudinal bunch profile to provide guidance
on the steps needed for a careful empirical study.
Fundamental limitations on pushing the FEL power to higher levels
have been further investigated theoretically. The essential concern
is ultimately reaching an rf instability in the energy-recovery
mode that cannot appear in the straight-ahead mode. We have found
empirically that the power output and the gain are extremely and
nonlinearly sensitive to the optical-cavity length in the vicinity
of saturation (which occurs at ~311 W in the straight-ahead mode),
a property that will adversely drive the rf system. This means
it is possible to excite an rf instability whereby the FEL rapidly
switches on and off, i.e., with a frequency ~10 kHz. By contrast,
as one backs the optical-cavity length away from saturation (to,
e.g., ~210 W in the straight-ahead mode), both the gain and power
output become much less sensitive to cavity length and the rf
instability will not appear. The bottom line is that with ~5
mA beam we should be able to get close to 1 kW (nominally 900
W) with the 5 µm optics presently installed, and have some
hope of pushing to the full kW, without getting too close to saturation
and exciting the rf instability.
The prediction of 1 kW with 5 mA assumes that there are no mirror
limitations. The laser is less sensitive to mirror distortion
if we operate in the low-slope region where the small signal gain
is high so we may get lucky, especially if we have one silicon
mirror. If we are close to 1 kW, and the mirrors are not limiting
the performance, we can try to increase the power by (1) getting
the bunch tails incorporated into the main bunch using injector
settings, and (2) increasing the acceptance of the recirculator
beyond its specification and lengthening the cavity until 1 kW
is reached. Finally, it is worth noting that in December, with
limited time to optimize, we achieved the power one predicts theoretically
for a 1.6 mA beam. The most productive path to 1 kW still seems
to be to increase the current and continue optimizing, which is
what we plan to do upon resumption of commissioning.
Regarding the FEL Accelerator Readiness Review (ARR) process, we cannot formally close out the ARR until we are ready to do routine user experiments, for which we are unfunded. Presently we are exclusively doing commissioning tasks, ultimately to include test experiments, for which we are following established EH&S policies and procedures. Today we sent our DOE site office a memo to that effect, namely, stating that lack of operating funds prevents us from reaching the point of being able to support routine user experiments.