Present:
Chris Allgower, Andy Bacher, Chris Lavelle, Hermann Nann, John
Olmsted, Tom Rinckel, and Ed Stephenson
MAGNETIC SHIELD TESTS
Before making a commitment to a large project, Tom Rinckel decided to test what would happen with a smaller steel plate (27 inches high) put between the Pb-glass and the present shield. Tom found a scrap steel plate (27 x 48 x 1/4 inches) and inserted it on the beam right side. Subsequent tests by Chris Allgower with the 6-degree manget off and then on showed that this plate cut gain losses from "complete" in some cases to less than 25%. The evaluation was made by fitting Gaussian peaks to the cosmic ray spectra. Variations of a few percent were noted in the centroids. It was decided that such a modest loss could be easily accommodated with equally modest gain adjustments.
Next, this plate was trimmed and put on the beam left side for a similar test. In this case, there is less room, and the threaded rod that holds the clamp to the top of the Pb-glass stack was removed. This test was less successful, but it appeared that all PMTs could be recovered. At the same time, two smaller 1/8-inch plates were tried on beam right. These were also less successful, but satisfactory.
We decided to try improving the size of the plates on beam right, and to go with this scheme for the future. These additional plates are connected to the tables that carry the Pb-glass.
Pb-GLASS TIMING
One of the pieces of information not yet used in the analysis is the timing of Pb-glass hits.
John Olmsted reported that he found a tight correlation between this time for 3He+pi0 data (last proton run) and the time of flight. This originates from the shift in start timing (from delta-E2 in this case) with 3He velocity. The correlation can be easily taken out with a simple linear correction. The slope of the correlation is consistent with what one expects based on the time-of-flight shifts and the calibration of the relevant TFCs (about 16 channels/ns for the Pb- glass). John then obtained an "offset" for each Pb-glass channel and used this to align all timing peaks at channel 1250 to make gating easier.
The analysis scheme was set up so that a straightforward shift is made to apply this to 4He+pi0. John showed missing mass spectra gated on scintillator cuts and correct timing. For various small Pb-glass multiplicities, these gave reasonable missing mass spectra with a few dozen events.
A discussion ensued about how many events we should expect from the last three days of running. To the nearest order of magnitude, the integrated luminosity was probably in excess of 10^36/cm^2. Thus a 1 picobarn cross section would be evident as 1 event. A 4He+pi0 cross section of 10 picobarns and a 4He+gamma+gamma cross section of 16 picobarns would be consistent with what we are getting.
[The outstanding software tools still needed are (1) Pb-glass energy summing over clusters, (2) Pb-glass angles, and (3) traceback to the target.]
SCHEDULE
The new schedule is out, and it gives the following dates to CSB:
REFERENCE CROSS SECTIONS
Ed discussed the data base for p+d cross sections. The nearest data sets are from Postma and Wilson at 146 MeV [Phys. Rev. 121, 1229 (1961)] and Sekiguchi et al. at 135 MeV [private communication]. A useful summary of the early data is found in Bunker et al. [Nucl. Phys. A 113, 461 (1968)]. From Bunker, it is clear that the cross section is falling at all angles as the energy goes up. The Sekiguchi (and earlier Sakamoto) data from RIKEN violate this trend, being too small, in particular at forward angles. For the moment, we will set the RIKEN data aside and look at smooth trends to get an estimate that applies to our energy.
Ed had on file Faddeev calculations from the Bochum group at 97, 108, 120, and 135 MeV. Our energy is roughly 114 MeV. From these, smooth trends emerge. But the calculations are for n+d, and do not include the interference with the Coulomb amplitude at forward angles. This correction is small at the angles where the forward scintillators are now positioned. At larger angles, the Postma and Wilson data seem to fit well with the trend of the Faddeev calculations. So it is possible to make an estimate with an error perhaps as small as 10%.
At the larger angles, the Faddeev calculations are known to be too small because of the neglect of (at least) three-body force effects. Here it is harder to estimate the cross section at 114 MeV because the three-body effect decreases with energy at a rate that is not well determined. Using the Sekiguchi/Sakamoto data would reduce the correction for three-body substantially. So while an estimate is possible, scaling errors could be as large as 25% (most likely downward).
[The RIKEN group reports that data they measured at 100 MeV will not be released because of problems with beam integration in the Faraday cup. They have no comment on the problems with the trend of the data with energy.]
SMALL ITEMS