FEL Gun Committee Meeting
Thursday, 14 Jan 99
Recorder: C. Bohn
Next Meeting
- Date: 21 Jan 99
- Time: 1500-1600
- Place: FEL Facility Break Room
Agenda for Next Meeting
- Discuss HV Processing, Heat Clean Procedure to be used on the new wafer
(Siggins, 15 min).
- Review surface analysis of metallic spatter on previous cathode
(Siggins, Phillips, 15 min).
- Discuss and prioritize low-cost incremental gun upgrades (Biallas, 20 min).
This Week's Attendees
G. Biallas, C. Bohn, F. Dylla, J. Gubeli, K. Jordan, G. Neil, L. Phillips,
T. Siggins, C. Sinclair, R. Walker
Discussion
The nature and source of the metal spatter on the previous cathode entered
as a topic of general discussion. The discussion was skewed and came to no
closure because there is as yet no data. However, C. Sinclair pointed out,
and all realized that, it is important to identify the source, or at least
potential sources, of the impurity before we can know how to correct the
problem. For example, if the problem were to persist, including a load
lock would be fruitless in that the wafer would need to be removed, not
just reprocessed. T. Siggins and L. Phillips will work together prior to
next meeting to gather at least some cursory data. One comment: Two
earlier FEL cathodes failed due to spatter; one got copper on it, and the
other got silver on it, meaning their failure mechanisms apparently
differed.
Sinclair proceeded with a presentation of his ideas concerning a "simple"
load lock (indicating that perhaps there is no such thing). His charts are
attached. The bottom line is that Charlie recommends adding a high-voltage
standoff to bring the load lock to ground. Perhaps the simplest
implementation would be to do a rear-mounted load lock with a new ceramic
supplying the voltage stand-off. A large ring would replace the present
hemispherical structure that receives the HV from the power supply and
provide a link between the new ceramic and the existing HV stack.
Disadvantages are that it would necessitate a long feed for the cathode and
a long load-lock chamber. The key advantage is that no new electrostatic
problems would ensue. By contrast, if one would go for a side-mounted load
lock, one would need to bore a hole in the side of the ball, or modify the
ball in some other pronounced fashion, thereby introducing risky
electrostatics.
Charlie also pointed out that the availability of a load lock would then
motivate looking at other cathode types, perhaps the most attractive of
which would be K_2CsSb, a material that exhibits high quantum efficiency in
the green, possibly long lifetimes, and only needs to be reprocessed
through a load lock for rejuvenation. A study of any new cathode material
would necessitate going through a learning curve, but a load lock should
significantly reduce the e-folding time of the curve.
One manifest implication of Sinclair's presentation is that a load lock
cannot be viewed as an incremental upgrade. Most likely its development
would best transpire by way of constructing and testing an altogether new
gun in the Injector Test Stand. People are invited to come up with simpler
and cheaper alternatives, but to do so would appear (at least to Bohn) to
be difficult.
F. Dylla advised that the definition of "incremental upgrade" is one that
involves no more than ~50 k$ and a few man-weeks of labor. A load lock is
ruled out at once, meaning it cannot be developed in the absence of a new
funding source for the FEL. G. Biallas had previously tabulated low-cost
options for incremental upgrades; he will update it and send it to
Committee members for review and consideration. Further discussion will
take place at next week's meeting per the agenda above.
Closed Action Items
- Biallas: Produce final version of wafer change-out schedule and post in
the FEL Facility. Status: Posted on the daily scheduling whiteboard.
- Dylla: Meet with Sinclair regarding procurement of DLC-coated samples.
Status: Reached
agreement to procure samples, but procurement not yet started.
- Phillips: Assess availability of profilometer. Status: One is
available on the first floor of ARC. It is a good one, with resolution
of a few hundred angstroms. Phillips needs to get
with M. Gupta to obtain permission to use it, which should be permitted
under the ARC contract.
New Action Items
- Sinclair: Prepare procurement package for DLC-coated samples for
Dylla's signature. (due 21 Jan 99)
- Biallas: Update list of incremental upgrades and their respective
payoffs, and e-mail the list to Committee members.
Old Action Items
- Siggins/Biallas: Specify design of experiment to measure cathode
temperature. (due date TBD)
- Dylla: Ascertain feasibility of low-cost "load lock" for
cathode-processing experiments in the
ARC. Status: No progress due to lack of funding. (due date TBD)
- Fugitt: Follow through with Ling-Ling on having her develop a strawman
table of HV experiments
on candidate electrode materials. Status: Discussions are now ongoing.
(due 22 Jan 99)
- Siggins: Begin scanning gun data and, with D. Gilchrist's help,
compiling information into an
electronic logbook. Status: Effort delayed pending V. Bookwalter's
creation of web-site address. (due 21 Jan 99)
****Charts from C. Sinclair Presentation****
1) LOAD LOCK ISSUES FOR THE FEL GUN
I have not been a proponent of load locks for polarized guns,
because they do not solve the underlying problems, but rather "work around" them.
They are very costly in $$, and involve a great deal of skilled staff time.
However, in some situations, the realities require "work arounds" rather than
"solutions" or "understanding". This may be one of these situations. However,
IT IS ESSENTIAL TO CLEARLY UNDERSTAND WHAT PROBLEM(S) A LOAD LOCK IS SUPPOSED TO GET
AROUND OR "SOLVE"
2) WHAT PROBLEMS SHOULD A LOAD-LOCK "RESOLVE"?
- Reduce bakeout turn-around time
- Allow cathode heat cleaning without heating gun innards
- Reduce or eliminate the introduction of cesium into the gun by preparing the
cathode outside the gun proper
- Keep the cathode out of harm's way during HV processing
- Allow rapid cathode replacement, by having a "ready reserve"
- Preserve the processing to HV once it has been established
- ????
3)THINGS TO WATCH OUT FOR
- We need to be sure, as best we can, that adding a load lock
doesn't bring new and comparably unpleasant problems of its own
- We need to be sure that adding a load lock is the best way to
reach a clear solution to the agreed upon problems
- We need to match the style of the load lock to the agreed upon
problems to be solved, to minimize the cost, effort, and time required to implement
4)LOAD-LOCK STYLES
- At High Voltage, always attached
- likely simplest from a vacuum system standpoint
- causes much high voltage complexity, due to the
large volume and surface area at high voltage
- may involve much effort to disassemble and
reassemble the high voltage system after access
- cathode heat cleaning is a design challenge
- ???
- At High Voltage, but removed during gun operation
- reduces complexity of high voltage system
- a fair bit more complex, both mechanically and vacuum-wise
- more time involved to install, remove
- bakeout of a limited volume required
- cathode heat cleaning is a design challenge
- ???
- At Ground Potential, always attached. Two options available:
- Re-work the electrode support and HV feed to the cathode
- lectrostatics issues are complicated, and may
require unpleasant solutions
- requires complete redesign of gun, HV feed....
- cathode heat cleaning easier to solve
- Add HV standoff to bring load lock to ground
- solution demonstrated at NIKHEF
- leaves most of existing system unaffected
- requires simple extension of existing SF6 tank
- probably get away with simple ceramics to extend
back to ground (like on existing lower voltage guns)
- leaves many options open as to what is included
in the ground potential load lock
5) SHOULD WE CHANGE CATHODE TYPE?
- Alkali antimonides are a possibility
- K2CsSb offers high QE in the green (I made a 17% one on
my first try, with Bob Springer's simple apparatus.)
- multiple wafers would not be required - just clean and reform
the cathode on a simple substrate
- lifetime issues clearly uncertain in our gun, but recall
Springer's demonstration at the Sante Fe FEL conference
- Cs3Sb may be an even simpler starting point, with far less
alkali hazard, though lower QE
- CsTe and its variants seem not useful - the laser required for
operation is exceptionally demanding - and unavailable commercially
- EXPECT A NON-TRIVIAL LEARNING CURVE......