Notes from the BPM Pelimenary Design Review.
Febuary 27, 2001, Los Alamos, NM.
Committee members: R. Webber (FNAL/chairman), J. Crisp (FNAL), R. Johnson (SLAC).
Non-LANL guests: Craig Dawson (BNL), Dave Purcell (ORNL), Saeed Assadi (ORNL), Ken Reece (ORNL), Tom Shea (ORNL).
Labs attending via video conference: LBL, BNL.
Quality control will be very important during fabrication and installation of the pickups, especially for the case of the DTL pickups. We should make a list of quality control steps and test procedures. Examples of good tests include high potting the connectors and cables inside the DTL stem, and high current tests of the same, while wiggling and vibrating the unit. Mechanical stress relief of the cables before they go up the DT stem was also mentioned. Since the DT pickup will be welded into the DT by a weld shop, extra care should be taken to protect the vacuum feedthrough connectors during shipping and welding. It would also be valuable to record the TDR signatures at various steps along the assembly process.
The self calibration feature needs at least about 125 feet of cabling between the pickup and the electronics. The MEBT installation plans call for locating the electronics several tens of feet from the pickups. We need an installation plan for the MEBT that covers this.
It was mentioned that it might be nice for the control system to track the calibration constants over time to check for drifts. This could be made available via Channel Access.
The calibration system is a critical part of the BPM system that needs careful attention.
The accuracy of the BPM system relies on a 2.5 MHz reference signal with a stable amplitude and phase. We need a spec on these parameters.
The time of flight energy measurement method relies on absolute phase measurements between two BPM pickups. Good accuracy is only possible for BPM electronics that use the same Phase Stable Reference Multiplier (PSRF). J. Stovall stated that this will be necessary for at least about 5 locations along the linac. It is good to do this anyway to minimize the number and therefore the cost of the PSRF, estimated to be about $6k each. We need an installation and cabling plan that takes this into account.
The front end signal processing may excessively degrade the signal we deliver to the LLRF folks. We need to take a closer look at this, and spec their signal requirements.
Since BPM signal processing relies on software, it will be important to have good version control for it.
In general, the review committee felt that the BPM design looks sound and we that should proceed rapidly with final design and prototyping.
|
