Jan. 25, 2012 Discussion on CDC FDC alignment
The following are some notes Simon wrote up after a discussion amongst several of the Hall-D staff on how we might calibrate/align the CDC and FDC.
The discussion took place on Jan. 25, 2012.
Here's a summary of my notes. Feel free to add corrections or things that I missed. We had an informal meeting to discuss calibration and alignment of the hall-D drift chambers. As far as the alignment is concerned, the CDC as a whole and the FDC as a whole would be tied to the rest of the detector via the survey targets. The relative positions of the ends of the wires with respect to each other in the CDC will be very well established before installation inside the solenoid; therefore the task becomes to determine the offsets and rotation angles of the CDC volume with respect to the beam line. The surveyors will be able to determine these parameters assuming that we provide enough targets with appropriate lines of sight. We would then use cosmic rays (using BCAL wedges above and below the CDC to form a trigger) to verify (or refine?) this alignment. For the alignment of the FDC, we would again begin with survey, but the task as a bit more complicated because we have to determine the rotation angles of adjacent wire planes with respect to each other and need to determine offset parameters in x and y for each plane. The angles of most cosmic rays are not conducive to calibrating the FDC, so we will need to take beam data with the magnetic field off (so as to provide straight-line tracks). There was some discussion about how to provide these tracks and what particles would be appropriate. We considered Compton scattering and pair production, for example. A different target other than liquid hydrogen, such a thin foil, may be needed for this purpose and the target may have to be pulled upstream. The FCAL could be used as part of the trigger and as a particle identification tool. We expect the incident beam photon energy would have to be fairly low for this purpose and the beam intensity would also have to be low. There was also some discussion regarding how we know where the beam line is: a downstream fiber detector a la the one PrimEx used would be very useful and we have the pair spectrometer and the active collimator to give us some kind of handle on the upstream side. As far as software is concerned a straight-line track fitter will need to be developed. We will also probably need to write code to be able to manipulate the raw survey data into something that makes sense for our detector. The time-to-distance relationships for the CDC and the FDC will need to be measured. This can be done with uniform illumination of the cells but a reasonable start time needs to be established. This could be done with beam data assuming that we can get enough tracks heading through the CDC. The start time could be established by the start counter. For very forward-going tracks some other detector will be needed. We need to do a relative gain calibration between the CDC channels. There was some hope that we would be able to use the test signal from the FADC-125 to do a preliminary calibration, but it is not clear that this actually works. In any case we will need to establish a program to accomplish the gain calibration and write the appropriate software. We also need a means to calibrate the CDC amplitudes for the purpose of dE/dx determination. We think we should be able to use minimum ionizing particles (from the cosmic ray spectrum) to give us points near the pedestal and can take advantage of the different path lengths through the straw to probe a range of energy depositions. Signals from protons would potentially be useful, but we would need beam for that. We considered whether or not we need to worry about the non-linearity of response for large pulse heights. We need to calibrate the strip-to-strip relative gains in the FDC. We already have a first version of this calibration that takes advantage of the fact that the facing cathode planes are measuring induced signals from a single avalanche in a given event. We need to map out the inefficient regions of the FDC (near the wire-deadened region and close to circuit boards). This can be done with beam data with the magnetic field on. One thing we did not discuss in the meeting is the correction in the FDC for the deflection of the avalanche position along the wire due to the Lorentz force. An initial correction can be provided (or I should say is already provided) from parametrizing the results from Garfield simulations, but we need to consider how well this can be determined from the real data. Simon