Next Meeting
Agenda for Next Meeting
This Week's Attendees
J. Bennett, S. Benson, G. Biallas, C. Bohn, I. Campisi, D. Douglas, R. Evans, A. Grippo, R. Li, L. Merminga, G. Neil, P. Piot, C. Reece, M. Shinn, B. Yunn
Closed Action Items
N/A
New Action Items
Old Action Items
Items of Discussion
C. Bohn opened by summarizing the goals and "restrictions" for the next FEL run. Accelerator Division management has sanctioned a month-long summer run principally to acquire data that strengthens the design of the FEL Upgrade. Accordingly, the run will focus primarily on experimental accelerator physics, especially CSR studies, BBU studies, and injector studies. There will also be another attempt at achieving a full kilowatt of cw lasing, and there will be time for key lasing studies, including the commissioning of laser diagnostics, "wiggler" studies and the effects of lasing on the electron-beam phase space, and high-priority (ONLY high- priority!) tests in the User Labs. Injector studies are expected to be invasive, such as moving the displaced solenoid #1 back to its design position and resteering with the "experimental correctors". Thus, the injector studies are chronologically relegated to the last week of the run. Target dates for the run are 12 Jul 99 through 13 Aug 99, i.e., just before the next FEL Conference.
M. Shinn indicated that new sapphire optics (for 3 µm lasing) with hopefully improved coatings might be available for installation in early July but it will be close. Both the 5 µm and 6 µm optics will be mounted and ready for installation. We achieved the 710 W power record with the former, in which case the power limit was the cathode (average current available); it may prove possible to get a full kilowatt with these mirrors and higher (~5 mA) current. The 6 µm optics should prove thermally robust, but they require a machine setup we have not done, i.e., one that delivers ~35 MeV beam. As we get closer to the start date, we will eventually decide on a mirror suite, and plans are to retain that suite for the duration of the run (unless there is some truly high-priority reason to deviate and install another suite).
Shift scheduling will be according to the following scheme, in which Week 1 starts once the gun is producing acceptable beam (meaning at least 60 pC and at least 320 kV):
Week 1:
Weeks 2 and 3:
Week 4:
This meeting then focused on summaries of electron-beam studies done during the last run and plans for BBU experiments. P. Piot summarized injector measurements, D. Douglas summarized transport measurements, and L. Merminga summarized plans for BBU. As a matter of general principle, Piot noted that, according to PARMELA, if the beam is mismatched to the cryomodule entrance, there can be very strong emittance growth, and therefore the injector studies should concentrate on establishing a good, careful match.
Piot indicated first that the wire scanner at the gun exit is working and reveals a very strange beam profile with hints of significant halo at 320 kV. In particular, he saw a signal even when the scanner is retracted, but he could zero this signal by adjusting the field of solenoid #1. No attempt was made during the last run to optimize empirically the solenoid settings for the "new" gun voltage (320 kV) because the machine was "working" and we had very little run time due to a cantankerous gun, and it may be wise to do so sometime during this run.
According to both rf calibration and (very nearly) PARMELA simulation, the calibration of the cryounit cavities is: Cavity #3 MEDM = 9.5 MV/m (canonical setting) implies real gradient of 10.4 MV/m Cavity #4 MEDM = 12.45 MV/m (canonical setting) implies real gradient of 13.1 MV/m. The energy at the injector exit is 9.56 MeV kinetic as inferred from the injection-chicane setpoint of 11080 G-cm, and so the total energy is 10.1 MeV. Multislit #1 consistently gives a normalized rms horizontal emittance of about 6 mm-mrad, and the vertical emittance appears to be better (too good to measure with the mask). We have yet to be able to do a measurement with the diagnostics near the cryomodule entrance (Multislit #2 and Happek #1). Happek #1 was relocated to the wiggler where it is working fine (meaning it produces useable signal). Perhaps this means that in its original location the CTR transport is misaligned, and/or the low-energy beam produces too little CTR, and/or the bunch length in the injection line is much larger than anticipated. Injector test plans will include provisions for assessing these possibilities.
Piot has been able to get excellent agreement between simulated and measured time-of-flight measurements in the injection line. Agreement comes given a buncher gradient of 1.85 MEDM, a value that does NOT produce the least energy spread as viewed on ITV0F06. He has rough agreement between PARMELA-predicted and measured spot size at ITV0F06 versus buncher gradient.
Douglas reviewed his difference-orbit measurements at 48 MeV and showed how he iterated to achieve good agreement between the design and measured orbits in the recirculation loop. It turns out that even better agreement would manifest itself were the actual beam energy about 1.5% lower than he thought it was. S. Benson pointed out that the FEL wavelength usually corresponded to about 1% lower energy (48.2 MeV versus Douglas' presumed value of 48.5 MeV), an observation that works in the right direction. He has some indication that IPM2F09 is misaligned, as well as some hints that the button BPMs aren't resolving well.
Douglas advised that the proper quadrupole excitation curves be installed in software during this down. It will be done (see New Action Items above). Save/Restore files will be created as modified versions of the existing key BURT files to reflect the new magnet strengths.
L. Merminga summarized the most recent plans for BBU studies. To obtain an experimental indication of the BBU threshold current in the machine within the short time available, plans are to infer it from measurements of beam-transfer functions in the recirculating mode. Specifically, for a given average current, the beam displacement will be modulated both vertically and horizontally using stripline BPMs already installed in the injection line (Merminga is in the process of calculating the required modulation amplitude). The modulation will then be detected using the field probe in one (or more) of the srf cavities in the cryomodule to monitor the strength of a key higher-order deflecting mode. The amplitude of the signal registered on a network analyzer is the basis of the inference of the threshold current. The required electronics for the kicker and pickup are available; however, there is one complication. The coupling of the srf probes, as they are presently set up, are weak (-21 dB at 1.9 GHz). To acquire sufficiently strong signals will probably necessitate installing 3 dB hybrid splitters to enable a direct, relatively unattenuated reading of the cavity probes. Unfortunately, to do so will introduce phase shifts and attenuations that change both the phases and GASKs of the cavities, in which case additional setup time will be required at machine turn-on.
Bohn asked that the splitters not be installed yet so we can turn on unimpeded and quickly go for the kilowatt. Then (at Week 2) the splitters should be installed as needed one at a time, and the respective cavity phases and gradients be set one at a time, so that we are never far from a viable operating point. This strategy should minimize the risk to, and impact on, the run while adding only modestly to the time required for the BBU study.
New Issues
None, other than those mentioned above.