FEL Upgrade Project Weekly Brief
September 4-7, 2001

 
Highlights:
The first 100 kW klystron for the upgrade injector was successfully tested to 85 kW this week.
This is more power than is needed to supply the required 10 mA of beam current.  The vendor expects the tube to top out at
120 kW.

Management:
We are spending time carefully analyzing the cost performance data as we prepare for close-out of the Phase 1 project by
Sept. 30.  We are ensuring that the total project costs do not exceed the project budget of $9,029k.  We are looking at the
remaining obligations on Phase 1 accounts to make sure that they can all be costed before the cost close-out on Dec. 1.  We
are expecting the close-out of the August expenses early next week so that these data can be included in our analysis.
Potential cost and schedule variances by WBS element are being analyzed.

The project monthly financial and technical reports for July 2001 were distributed to the DOE and DOD contract monitors
today.

Draft presentations were prepared for the annual DOE/SURA review of Jefferson Labís Science and Technology which is
scheduled for Sept. 24-25th.

A CRADA with STI Optronics for the design and analysis of compact , permanent magnet recirculation magnets for energy
recovered linacs was sent through the lab management chain for sign-off this week.  Preliminary discussions were opened up
with AES, Inc. for negotiation of a CRADA to support an SBIR grant that AES recently secured to begin the development of
a 100 mA injector based on the IR Demo dc photogun.

A draft budget request was prepared for the Commonwealth of Virginia to support the upcoming request (Oct. 12th) for input
for the state 2002-2003 budget.  The Virginia budget allocation over the last 4 years has supported the FEL operations
budget and we look forward to and appreciate continued state support of the FEL program at Jefferson Lab.

WBS 3 (Beam Physics):
Field analysis & certification of the GY performance was completed and documented (JLAB-TN-01-042, at
http://www.jlab.org/~douglas/FELupgrade/technote/JLABTN01042.pdf)

A preliminary energy recovery dump design was developed and sent into engineering.  See:
http://www.jlab.org/~tennant/TN-01-038.doc ; interferences are being resolved and diagnostic layout worked up.

The recent beam transport experiment to test 3-pass operation with the Demo was documented:
http://www.jlab.org/~douglas/FEL/technote/JLABTN01043.pdf , for your viewing pleasure, we attach a single result:

 

WBS 4 (Injector):
Continued work on the shield door test set up.  The cathode stalk subassembly is in brazing. Initial design of the gun stand and
ceramic stack removal cart is done.  The hat flange assembly was completed by the shop. A schedule was set with W&M for
implanting the cathode ball and support tube.

Gun HVPS - Design of the tank and holders for the stacks continue.

WBS 5(SRF):
Completion of the first production 7-cell cavity should occur next week.  (Helium vessel and end groups welded). This
assembly will allow us to determine if HOM coupling will be adequate for the FEL upgrade cryomodule and will also conclude
the development of the assembly procedures (if successful).
 
WBS 6 (RF):
Zone 3 - The Filament and Mod Anode board tests were successfully completed.  Testing of interlocks is progressing.  The 5
watt pre-amplifiers and their power supply are being installed.  High voltage to the klystrons is expected next week.

Zone 4 - Progress was made this week in getting RF Control Modules calibrated and then returned to the zone.

Injector HPA- First Article klystron tests are scheduled for next week.  The first klystron has already successfully operated to
85 kW and they expect to run it above 120 kW.  The factory's HVPS is being modified for additional power.  Low power S
parameter tests were made and plotted for the 100 kW circulators.  This report is being forwarded to the vendor for
suggestions of how to tune the circulators for optimal performance.

Injector HVPS - Design work continues on the upgrade of the HVPS's.
The Main Circuit Breakers, Contactors, SCR Controllers, 12 Pulse Transformers, Rectifier Assemblies, and Inductors have
been specified, ordered, and many have already been received.  Modeling of the power circuit, crowbar, and klystron
continues for best noise and stability performance.

WBS 8 (Instrumentation):
A new combination amplifier/driver board is in the prototype stages.  The testing on the timing/delay board is complete.  It is
installed and working as expected.

The SF6 chassis has been upgraded and the readout is now available in EPICS.  Documentation on this chassis is being
updated as well.

Cables were pulled to four of the five labs for installation of the intercom system.  More cable is on order and the task should
be completed soon.

The AMS patch panels were "cleaned-up" and any cable that was not connected on both ends was removed.

Parts for beamviewers continue to arrive daily.  The mounting bracket for the solenoid/lamp box is out for fabrication and the
box will be out for machining soon.  Assembly should begin before the end of the month.

User lab upgrades continue.  More tools are available to assist users in experimental set-ups.  A video distribution amplifier for
each user lab is also in the fabrication stages.

The new dump current monitor was tested successfully this week.  It resolves 1 microsecond pulsed beam, this is important
for operation of 5 microsecond / 60Hz alignment mode operation. The design is now being finalized and documented.

Drawings completed this week:  F0032 - Solenoid box assembly and fabrication Rev. B; F0033 - Beamviewer Interlock
Box-Version A, assembly details; F0101 Beamviewer Interlock Box-Version B, Fabrication Details; and F0102 -
Beamviewer Interlock Box Mounting Plate, Fabrication Details.

WBS 9 (Transport):
Dipoles
Optical Chicane Dipoles (GW)
o  Proposals were received and evaluated.  Several items needed clarification from the vendors.
    Clarifications due September 10.
Injector Dipoles (DU/DV)
o  The final round of rework on the Small Magnet's (GV) Assembly Drawing is nearly complete.
    They started the final correction of the Large Magnetís (GU) details.
Arc Dipoles (GY, GX, GQ)
o  David Douglas analyzed the magnetic model data for the 180 Dipole (GY) from AES.  The
    result was permission to complete the drawings per the magnetic model.  His specifications
    were met.
o  AES has almost completed their manipulation of the magnetic model of the Reverse Bend
    (GQ).  They had to add a .001 inch step to the edge of the pole, adjust the stepís width and
    add a taper to get uniform field and integral.  They have yet to adjust the gap height and faces
    to get the correct bulk field and effective length.
o  Design continued on backchecking the detail drawings of all three magnets with the comments
    and corrections.  They also started adjusting the DY drafting model (and hence, automatically,
    most of the details) to the dimensions of the successful magnetic model.
Quadrupoles
3 inch quad (QX)
o  The specification for assembly and the assembly drawing were signed and placed in the
    procurement cycle.  Work continues on cores and coils.  Request for bids on assembly went
    out on Sept. 7.
Trim Quad (QT)
o  The 2D detail drawings and statement of work are in final checking Procurement package
    ready to start process on Sept. 10.
Sextupole (SF)
o  Work slowed as the octupole statement of work and quadrupole assembly and a well earned
    vacation took Robin Wineís time. Draft drawings have been started.  Finally back to work on
    the 3D model.
Octupole (OT)
o  DULY Research had results from their magnetic modeling. They found that a mushroom cap
    pole tip of the right shape, placed on the recycled corrector coils gets a 2 to 3 % magnet with
    simple flat truncated ends.  They are continuing their pursuit of the 1% goal by adding field
    clamps, nubbins at the pole tip edge and rounding the pole ends. Analysis shows that the next
    level power supplies used for the Trim Quads (20 A, 50 V) will be necessary for these
    magnets also.  Kevin Jordan will be happy about this.  Not! DULY continues iterations to try
    to achieve 1%.
 Corrector Dipoles (DB, DJ)
o  Milhous Co., in developing their production process, made a corrector coil that looked good on
    their second try but it had a short.  Vendor shipped DB prototype on Sept 7.
o  We received feedback on our concept design of the new mounting bracket for the DB/DJ
    correctors from the plastic molding company.  Those suggestions are being incorporated into a
    final design.
Beam Line and Vacuum
o  Work continued on the definition of the Optical Chicane chambers and continued to the
    mounting stand for the GW magnets. Continues.
o  The bid package drawing set for the girders in the return leg continues in detailing.
o  We continued engineering work on the LCW distribution.  LCW flowrates and manifold
    locations for the dipole have been defined and put on layout.
o  Butch Dillon-Townes is working on finalizing the style of the recirculation dump.  It still is a
    subject requiring further definition and analysis. Received preliminary dump line definitions
    from D. Douglas. Dave feels that we can live with a standard dump design and a well
    diagnosed 4 inch diameter beamline based on his DIMAD studies. George Neil brought up the
    cost/benefit of a larger beamline (say 6 inch) in this area.  We will contine the design
    discussions when we have a preliminary layout from engineering which they have started.
    Confirmed that we have the need for a QJ magnet and investigating the use of the QI
    magnets.
o  Work started on the injector line stands and vacuum chambers.  There was extensive
    discussion of the aperture in the injector line and the x-chamber at this weekís beam transport
    engineering meeting.  Dylla asked for input all relevant concerns such as impedance, beam
    scraping and compatibility with upgrades so that consistent decisions can be made.

WBS 10 (Wiggler):
The final probe calibrations were completed for the wiggler magnet test stand.  The probe is extremely linear with no evidence
of transverse Hall effect at the 1 Gauss level.  The measurement apparatus was installed on the wiggler and some test scans
carried out.  The apparatus as assembled was not centered in the wiggler and so did not have sufficient range at one end.  This
was corrected.  The next few days will be devoted to debugging the measurement apparatus before "real" scans can be done.
Welding was started on the wiggler vacuum chambers.  We may have to machine up new end flanges because there is not
much extra metal at the ends of the chambers for the final machining.  Still expect the chambers to be completed before the
end of the month.

WBS 11 (Optics):
The designers for the optical cavity chambers produced the first set of prints to support the procurement of components, and
we are to place purchase orders for some of them today (9/7).  We are planning an internal review of the design next week,
and after receiving any and all comments/suggestions, will continue and complete the detailing.  We have sent a number of
spare mirrors, procured for the IR Demo, to Veeco (Wyko Div) to determine their surface roughness.

As the cost of the new drive laser is far over our budgeted amount, we are revisiting our planning and developing alternate
plans.

Support for operations/experimenters:

We are testing new hardware, preparing samples, and planning for our next JTO-funded run, scheduled to start on 9/17.  As
part of our preparations, we have been interacting with our colleagues at NRL in order to coordinate their equipment with
theirs.

Operations/Commissioning:
Some machine time was devoted this week to testing of the beam current monitors in the recirculation beam dump.  Fine
tuning of this diagnostic will allow shorter pulse operation without beamloss trips.

Planning continues for the second phase of JTO sponsored experiments later this month and for the start of the last user run in
October.