Analyze harps near target

1. Assuming that you are in the g0superharp directory on cdaqh1
2. Type: make_harp_file directory_name run_number
3. Type: use cernlib/2000
4. Type: paw
5. Depending if beam was unrastered or rastered. If the beam was unrastered and one wants to fit with a Gaussian then add fit=yes when execute plot_harp.kumac in next step. If the beam was rastered then execute with fit=no.
6. In paw++ Executive window type: exec plot_harp harpname=IHA3H00_runnum fit=yes or no
7. A plot of the raw harp scan is shown. Click with left mouse to make gates around each peak. Click the right mouse to quit. The first peak is the x1 wire, the second is the y wire and the third is the x2 wire. The program has offsets for each wire ( for spring 2003 these were determined by comparison to the BPMs). For each of the following plots, the offset for that wire is subtracted from the raw data and for the X1 and X2 the sign is reversed to agree the BPM sign conventions.
8. Then the x1 plot appears. If not doing a fit the left click at the lower left, center and upper left of the peak (click with right mouse to quit). The kumac will return the difference between the first and last position and the center position. For a fit make a click in the flat region below the peak, at the left base of the peak, at the top peak ( the y position is used as the starting height), at the right base of the peak and in the flat region above the peak . The fit is a constant background plus a Gaussian. Unfortunately fitting with PAW can sometimes have problems, and you just have do it again. If you get a negative number for sigma you can just take the absolute value since PAW uses the square and is insensitive to the sign. All units are mm.
9. Then the y plot appears. Do as for x1 plot.
10. Then the x2 plot appears. Do as for x1 plot.