Notes on the MTest ntuples

Doug Jensen  has developed a program to generate sets of ntuples for MTest running.

The ntuples contain assorted data from the Accelerator Lumberjack data logging system. There are some beam parameters, intensities and the downstream wire chamber data. The ntuple is part of a standard paw/hbook histogram file. There are several histograms as well as the ntuple. This is a draft – as the effort is evolving.

One ntuple is generated for each day of running. Typically, MTest runs for 12 to 14 hours a day. MTest gets beam once per minute, so that there are, in 14 hours, a maximum of 840 samples. The actual number varies, as there are gaps in the beam delivery, the running time may be extended or shortened, and the time between spills is not exactly 1 minute ( see histograms 11 and 12 in the ntuple ) Histogram 13 currently histograms the accelerator beam to be delivered to Mtest – Beam21. The title of the histogram includes the name of the directory that contains the raw data – so the ntuple contains some internal documentation.

The ntuples are PAW based ntuples. Conversion to ROOT ntuples should be straightforward using the ROOT package routine h2root. (But it is not included in all installations – for example is on neither may Mac or PC.)

Some of the items in the ntuple are:
Time seconds – time stamp time/3600 = hour.fraction
sample sample number for a set
Beam21 accelerator beam for MTest
scaler1 number of counts to MTest mt6sc1

There are additional notes at the end of Appendix A which is a listing of the ntuple. A significant part of the ntuple is the data that describes the beam at the wire chambers in the beam line. There are three chambers in the downstream part of MTest, the last chamber being in MT6.

For each chamber, for example 61 (there are also 62 and 63 and there may be MW1SEM)

mt61_p1 position of peak – in wires hist 21
mt61_h1 height of the peak hist 20
nt61_s1 sigma of peak ( forced ? 0 ) hist 22
mt61_a1 area of the peak / beam hist 24
mt61_b1 background hist 25
mt61_c1 chisq ( not a true chicq, because errors not well defined ) hist 23


There are each of these variables with a 2, e.g. mt61_h2 p1 corresponds to the horizontal, p2 to the vertical. For each plane, the histogram number is increased by 10, so histogram numbers 30..35 correspond to the vertical in 61, 40..45 to horizontal 62 etc

These numbers are based on a minuit fit to a gaussian plus a constant background.

The file contains histograms for all of the wire chamber plots. Also included are sample plots from each of the chambers. These plots have histogram numbers of the for ij0nnn where nn is the sample number, i = {1,2,3,4} corresponding to the 3 MTest downstream multiwires, or mw1sem, . j = {2,4} for {horizontal, vertical}

Histograms 11 and 12 plot the time difference between samples, histogram 13 has the name of the directory where the data is saved, so the histogram file becomes self-documenting. Currently, Beam21 ( the internal beam intensity ) is plotted.

This is a project recently under way. If thre are questions or suggestions, please contact

Several references may be of use: