User name A. Deur
Log entry time 15:50:49 on July21,2003
Entry number 105889
keyword=Problematic pressure curve points out rate dependent cuts
Here the W spectrum for N2 run at 2.134 GeV, 6 degree (fig 1) and the N2 pressure curve (fig2). The W spectrum looks fine while the pressure curve is problematic, which seems to come from rate dependent cuts.
In Fig 1, we can see the domination of the elastic peak, however, we see clearly the N2 excited states. The first one is the shoulder on the elastic tail at about W(-nitrogen mass)=0.011 GeV, The 2nd state is the clear bump at W= 0.017, 3rd is at W=0.022 and there is probably a 4th one at 0.026. These are at the expected positions if we compare, for ex, to a Nitrogen W spectrum from spin duality or g2 where the excited states are relatively big.
For the N2 pressure curves, I plot 3 curves: Pressure curves from
*trigger without any cut
*with cut on the elastic peak+pid detector+reconstructed quantities and corrected for lifetime and the fact that the prescaler 2 are different from run to run.
The pressure curve with scaler and no cut look very nice. However, the one with elastic cut+reconstructed quantities cut shows a non-linearity between the yield and the pressure. Further inverstigations shows that the problem is mainly in the reconstructed quantity cuts (phi, theta and yt). I initialy used the same cuts as for the elastic asymmetry analysis and then I made them loose but this did not improve, the non-linearity persists. An explanation could come from the fact that many events are misreconstructed and I cut them away even with loose cuts. It turns out that the percentage of misreconstructed events varies with the (non-prescaled) rate because the relative amount of multitrack events (which tend to be misreconstructed) increases with the rate. Hence my cuts are rate dependents, qed. It turns out that if I make the problem worst by adding a cut selecting only 1-track events, then I get an even larger non-linearity so I am pretty sure that this explanation account for most or maybe all of the non-linearity. However I cannot test further this hypothesis as long as Espace does not have efficient subroutines to deal with multitrack reconstruction.