GEn01 analysis meeting May 2, 2002 =================================== attended by: Oscar, Paul in Charlottesville Hongguo, Donal, Frank, Nikolai, Glen at JLab *) Target polarization: Paul worked on trying to fit the baselines to a low ordered polynomial, but found that there was significant higher ordered structure. He is not optimistic that this approach can be used to smooth out the baseline data. He also looked into the 17% change of the TE constant for the one calibration. He could not find any obvious problems. *) Target volume: Paul did a numerical integral and found that the average length of the helium between the cell wall and the nose piece wall considering the raster distribution is 5.19 mm. This can be compared to 4.99 for naive integration which does not include the raster location. Note that we have not included a realistic raster distribution.... rather it was assumed to be a uniform raster. We need to check if the radius of the nose piece is ID or OD. *) Pion production: Hongguo looked at carbon data. He stills sees missing strength in the dip when comparing MC to data. He also tried to increase the amount of Al in the target, but said that it had the wrong W dependence to fill the hole. He has not done the target core test, but the suspicion is that we are missing some reaction mechanism. *) Monte Carlo and Data differences: Hongguo reviewed our 98 results. He believes that additional pion production strength cannot fill the hole. He also found that when he does a cut hsdelta>-3%, then most of the differences in W for Monte Carlo and Data go away when one looks at the full range of cuts. He does, however, see a 7 MeV shift in the W peak between MC and data. There was a difference in the ndet angles between MC and data that resulted in a 2 cm shift, which brings the difference to 1 cm. 1 cm is the best we can do to align the detector anyways, so we declare victory and move on. Hongguo will change E, E' and theta individually in the MC to see how much of a change will be necessary to explain the 7 MeV shift. Also, Hongguo needs to change the beam energy from 2.332 to 2.330, although this is different from the Arenhoevel grid. Donal suggested that Hongguo and Frank but together a short write-up about our the missing strength. We can then distribute the write-up to collaborators to let them think about it. I was also suggested that Hongguo look at the different HMS momentum settings to see if he observes anything interesting. Oscar suggested that there may be target field orientation issue. This might explain many of the strange things we see, such as the hsdelta cut getting rid of the missing strength as well as some of Nikolai problems. *) Butcher: Frank presented his first extensive look at asymmetries from the second pass. The asymmetries are not charge asymmetry corrected, which are significant. For the record, I have copied some of Frank's observations about these results from an email he had distributed earlier: ------------ Observations: At least in the case of the neutron, most of the cuts result in little change from the established standard cuts, most notably the target cuts. Everything else is certainly well covered by statistics, including the raster dependence. In the theta_np bins, the proton data are very severely affected by the neutron data: where the n is statistically significant, the p is less so and also has a much more positive value. Tentative conclusions: We are misidentifying a considerable number of neutrons as protons because there's enough noise hits in the paddles. Maybe for subsequent replays (these and Q2=1 data) we should re-examine the criteria for matching a paddle hit to a track? ---------------- Frank would like a list of runs when we change the cell and stick change. Paul (or Marko with Paul's prompting) will work on it. Hongguo will give Frank a new list of cuts for the Butcher. *) Packing Fraction: Nikolai's code is working on Frell now, and it is about 10 times faster. He believes that there is a problem with the HMS momentum settings. He showed plots showing good agreement between Data and MC for the first momentum setting for dry-thin carbon. There was okay agreement for the wet-thin carbon run at the same momentum. He also showed plots of hse which agree only if the momentums of the second and third settings are shifted by 10 and 20 MeV compared to what is used in the engine. We do not have a solid explanation for this. The target field size and orientation may play a role in this. The target field was 5.003. Survey report is that on July 30, the target field was 29.64 degrees with respect to the beam. Our first video conference went okay. We agreed that it would be better to run a VNC setup next time to share plots. Whether we try video or just voice is not clear. It will probably depend more on which rooms are available than anything else.