BLTWG Meeting 11/16/2009
Revision as of 16:10, 24 February 2017 by Marki (Text replacement - "http://argus.phys.uregina.ca/cgi-bin/private" to "https://halldweb.jlab.org/doc-private")
- Time: 11:00 EST
- Place: EVO and ESNET (with telephone bridge)
- Connecting: instructions are here
- Present: Alex S., Richard J., Jim M., Victor T., Dan S., and Bill C.
- Update on the tagger resolution results, including multiple scattering -- Alex
- Status report on microscope electronics -- Igor (away at INT)
- Status report on diamond development -- Richard
- Other progress reports -- others
comments by R. Jones on microscope resolution studies
- What happens when, as we agreed last week, the microscope is moved onto the focal plane and the fixed array and exit window are moved inward be the same amount? This should improve the resolution in the microscope a little, and at the same time reduce the size (and hence cost) of the vacuum chamber.
- So far we have only considered the focus in the dispersion (xz) plane. Where is the focal surface in the y direction? As I remember, the focal surface in the xz plane is mainly due to the dipole field and weakly depends on the quadrupole. However this is not true for the y-focus, which comes about entirely from the quadrupole field. Without the quad, the y-focal length is negative. Dan Sober once showed using TRANSPORT that the y-focal surface is not parallel to the x-focal plane, so the two can be made to cross anywhere in energy simply by dialing the strength of the quadrupole field. In Geant we can do this by placing an imaginary cut in the field map upstream of the dipole and rescaling the field there by some factor of order 1 such that the two focal planes cross in the region of the microscope. Now that we have changed the dipole optics, we need to repeat this exercise to optimize the y-resolution. The latest figures show that with the reduced pole width, the y-focus is no longer optimal.
Status of Microscope Electronics
- Control board: FPGA firmware debugging is complete!
- all functionality there. Robust against Ethernet problems (failing links, traffic etc.)
- inputs/outputs understood, but calibration of DAC may be necessary. Change of its reference voltage may be necessary to remove the systematic shifts present with the current less-than-required supply voltage.
- yield of good boards is poor for now: only one of the three prototype boards has a functional DAC. One more may be restored. It is suspected that the ball grid array DAC chip was not soldered on (reflow processed?) with the proper temperature profile. The DAC manufacturer specifies tight processing requirements and it is not clear the company populating the boards is aware. We are currently talking to them to establish where the problem lies with that procedure and make sure this is not a problem in the future.
- Backplane (control/amplifier board interface and signal feed-through)
- Layout complete
- Final details of printing being worked out (especially w.r.t. the light-sealing)
- Manual assembly (at least for the prototype) is intended to save cost - all components are through-hole.
- Amplifier board: in the process of population. Some issues of component reel format (related to the small size of batch) being worked out right now.
- Low-level software (microscope comm. protocol-aware Ethernet packet communication routines) exists. General, socket-based, Java-based layers being added that will become a proper microscope monitoring suite.
- Full system bench-top test expected by winter (subject to academic schedules).