Next Meeting
Agenda for Next Meeting
Item Person Responsible Time ---- ------------------ -------- Status of Open Action Items All 10 min Progress Report on Commissioning Fugitt/All 20 min Scheduling Fugitt 10 min New Issues All 5 min Agenda for next week All 5 min
This Week's Attendees
S. Benson, G. Biallas, J. Bisognano, C. Bohn, D. Douglas, F. Dylla, J. Fugitt, A. Grippo, L. Harwood, K. Jordan, R. Legg, R. Li, L. Merminga, P. Piot, Q.-S. Shu, R. Walker, B. Yunn
Closed Action Items
Items of Discussion
The FEL cryounit and cryomodule are being cooled to 4K today. Cooldown to 2K is anticipated to occur during day shift Monday, 2 Feb 98. Checkout of srf and rf in both the cryounit and cryomodule will follow, and it appears that there is sufficient time to complete both before 0800, 5 Feb 98, at which time plans are to resume FEL commissioning with electron beam. Meantime, the FEL injector group has made steady progress in refurbishing the gun. HV processing and heat treatment were completed this week, and cathode preparation commenced this morning.
C. Bohn reiterated the goals of "pre-GEN" (5-17 Feb 98) commissioning:
Bohn (with T. Siggins) will be first up at 0800 5 Feb 98 in anticipation that there will be the normal amount of difficulties to overcome in turning on the machine after a long shutdown. Bohn felt he should deal with these difficulties himself. Legg will follow Bohn's day shifts with swings, and Engwall with owls, in a concerted effort to turn on the injector efficiently.
R. Walker e-mailed a 3-month shift schedule to the FEL commissioners yesterday. Both Bohn and Walker solicited inputs from many people, and discussed the schedule with nearly all (if not all) of the participants as part of the process of putting it together. The overall scheduling philosophy is to treat all participants equally and (by intent) fairly. Consequently, everyone will be rotated through the various shifts. For example, Bohn starts with 3 successive day shifts, followed by 6 days "off", then 3 successive swing shifts, then 6 days off, then 3 successive owl shifts, then 6 days off, then repeats the cycle. Some people will deviate from this pattern over the first few days so they can come in early in the turn-on process and experience a "setup shift." This is intended to provide them with some training, experience, and a sense of overall edification.
Bohn indicated that people can trade shifts when personal difficulties arise, encouraged everyone to communicate their needs as they arise (some of which he already knows about), and offered his personal assistance. At minimum, however, "traders" need to inform the FEL Experiment Coordinator with as much advance notice as possible. J. Fugitt will be FEL EC for the foreseeable future and will carry a cell phone (879-9999). Bohn will likewise carry a cell phone (879-1096).
According to R. Lauze and A. Hutton, sometime soon (probably in mid-Feb) we will be able to enlist the help of CEBAF operators. At first, subject to CEBAF crew chief approval, an operator might join the 2-person commissioning team in the MCC Control Room to assist and gain familiarity with FEL control screens and operations. (Note: Bohn envisions that FEL operation from the MCC will commence on or about 19 Feb 98, coinciding with GEN target irradiation, though if possible it may be tried prior to that.) Eventually we will be able to rely on CEBAF operator support as a matter of routine, always with the understanding that the CEBAF crew chief can pull the operator away for main-machine "emergencies". Since it is not possible to give hard dates for operator availability, we have by necessity developed the shift schedules assuming no CEBAF operator is available. Please be patient and remain calm -- this will eventually work out in some sensible way.
Bohn and Walker (with input from R. Legg) are developing an addendum to the existing OSP for FEL electron-beam operations to enable running the FEL during times when staffing in the MCC is curtailed (see old action items below). Indications are that this might be the case prior to 7 Feb 98, as well as, for example, during future CEBAF maintenance periods. The addendum is being written to cover contingencies over the next three months.
K. Jordan provided an update on the status of the MPS and an overview of its logic. It is being configured per the four average-current-dependent beam modes established long ago. He projected that enough of the MPS will be complete, prior to resumption of beam operations, to permit running 1.1 mA current in the injector before shutdown for GEN target installation (planned for 17 Feb 98). A question arose whether all of the BLMs will need to be running during GEN target processing, seemingly an unnecessary requirement since the average current will nominally be only 2µA. Jordan indicated Mahoney will provide a work-around, probably along the lines of what is done in CEBAF (turn off BLM voltages). Discussions about current settings in the dump-line magnets have been ongoing, and a formal meeting was held this week (on 28 Jan 98) to decide on strategies and inputs. The meeting minutes are attached at the end for the benefit of the interested reader.
R. Li summarized her considerable progress in simulating the beam dynamics attending coherent synchrotron radiation. She has written a two-dimensional self-consistent code that incorporates gaussian macroparticles and has been benchmarking it. First, she successfully benchmarked it against analytic theory for a rigid-line bunch propagating through a simple 90-degree dipole magnet. Next, she compared the code's predictions to those of DESY, which lacks self-consistency but should be accurate when CSR is only a perturbative effect. To do so, she invoked a model problem posed by P. Emma at SLAC which involves a single dipole. Predictions for CSR-induced energy spread were in agreement for this problem, which is one where the DESY code can be validly applied. Rui also modeled the first optical chicane in the FEL, obtaining 2.4% emittance growth for 120 pC bunches. Interestingly, this compares with 3% emittance growth predicted analytically from the rigid-line bunch model accounting for transients (one obtains 10% growth if the steady-state CSR field is used, which we had inferred to be an upper bound). The reason for agreement is that the simulation indicates the line bunch serves as a fair approximation to the actual bunch configuration throughout the chicane. This will not be true, however, in the first recirculation arc of the FEL where the bunch dynamics is quite violent. Rui will be simulating that arc in the near future.
D. Douglas reported that DIMAD, modified to include the steady-state CSR force, shows a 10% CSR-induced emittance growth in the first arc with nominal setup and a 135 pC bunch, a result that can later be compared to simulations. If the code prediction also turns out to be this small, then Douglas will develop arc setups for parametric studies that "force" emittance growth by, for example, keeping the bunch short throughout the arc. Of course, all such predictions will need to be checked by way of direct experimentation in the FEL.
Presently, the basic difficulty with the macroparticle code is that it takes a long time to run, several hours of Cray II CPU time for the aforementioned examples, principally due to the necessary incorporation of causality. Rui has been optimizing the code to reduce run time, and she continues to do so.
New Issues
New Action Items
Old Action Items
Procedures in Work
Emittance Growth from CSR
MEMORANDUM FOR RECORD
Subject: Magnet Interlocks for FEL MPS -- Meeting Minutes
Date: 28 January 1998
From: C. L. Bohn (recorder)
To: Distribution
Meeting Attendees: G. Biallas, C. Bohn, D. Douglas, K. Jordan, K. Mahoney, M. Wiseman, B. Yunn
Meeting Purpose: To specify magnet-interlock strategy and set points for the FEL MPS.
Prior to the meeting Bohn contacted Dave Kehne and obtained from him a document he had authored in April 1996 concerning specifications for the 10 MeV beamline that had been envisioned for the now defunct Injector Test Stand. Kehne designed the injector and energy-recovery dump lines (0G, 1G) for the FEL, and the essential considerations for their designs are described in the document. Bohn began the meeting by distributing Kehne's paper. Biallas distributed a 23 Oct 97 memo from B. Legg that tabulates raster-magnet specifications for the injector and energy-recovery dump lines. M. Wiseman is able to specify the rasters for the straight-ahead dump line (2G), and he has a table of results of burn-through calculations for all three dumps.
Key points of agreement among the meeting participants are delineated in what follows.
The overall philosophy for the MPS is that it must greatly decrease the probability of catastrophic failure due to errant beam, but it cannot set the probability to zero. Rather, it sets the probability "as low as reasonably achievable."
The general strategy for cw setup of the FEL accelerator driver is:
A. For first light:
Given the cw setup strategy and the MPS philosophy, the MPS strategy follows as a corollary: With magnet currents locked, look for changes in currents of the most sensitive magnets only. Plan nominally that ±100 mA (given 10 A trim cards) will trip the MPS. "Trip" means shut off the electron beam.
The "most sensitive magnets" are, in descending order of importance to MPS:
With the MPS strategy thus resolved, K. Jordan needs the current-interlock set points and tolerances (roughly ±10% of the set points) for all of the raster magnets, and he needs MPS time scales for the dumps (roughly 1/10 of the burn-through times). Biallas offered to develop and maintain a raster-magnet spreadsheet, which he and Bohn will compile. Bohn, as owner of the FEL MPS limits, will then transmit the spreadsheet to Jordan before week's end.
Distribution:
S. Benson
G. Biallas
J. Bisognano
D. Douglas
F. Dylla
A. Hofler
A. Hutton
K. Jordan
G. Krafft
R. Legg
K. Mahoney
G. Neil
C. Sinclair
S. Suhring
M. Wiseman
B. Yunn