Notes on how to carry on Andrea’s work:

I am working through the photomultiplier tubes (PMs) located in the large wooden box under the whiteboard to find ~70 of them with the best operating characteristics. “Best” is of course a qualitative term; I am trying to gather sufficient data on the tubes to determine a quantitative “figure of merit” on which to base a decision on whether a tube is good or bad.
The process of gathering the data on each tube is more complicated than lengthy; for an experienced user of CAMAC, KMax, and ROOT, the time it takes to gather and examine the data will be greatly reduced!
Hopefully I won’t forget any steps. The stages of the procedure are
(1) installing tubes in the test stand, (2) preliminary examination of their output under high voltage, (3) recording the tubes’ signal response to the LED at 5-6 levels of high voltage, (4) moving the data to Linux and using ROOT to analyze it. My program of developing software to do all the analysis work is not totally complete, so it should be possible to continue refining the ROOT scripts.

(1) Installing tubes in the test stand:

The test stand is inside the large square wooden box on the table in room 434. There is a wooden structure that looks like a wine rack in the near right corner, which holds 9 PMs for testing. In all that follows it is important to remember that the center tube is for reference only; it should remain in the same position at all times, for all tests. If there are still tubes loaded into the wine rack, they can be removed by disconnecting the signal and high voltage (HV) cables from the rear of each tube, and lifting the wine rack straight up out of the box. Carry the wine rack and tubes over to the table next to the box, and remove the 8 testing tubes through the front of the rack (the front, back, right, left sides are labeled). Cover each tube’s photocathode (the glass window) with one of the squares of black felt from the small cardboard box, and place them in the tall gray cabinet with the other tubes. If you get to the end of a row, cover the tubes with a large piece of felt and make another row on top of them. It is easiest to place them in the cabinet in the order of their numbering 1 - 8, which is written in pencil on the white label of each tube.

To reload the wine rack with fresh tubes, it’s important to remember that we need to be able to tell which tube goes with which set of measurements, so you have to write down the ID number of each tube. This is done in 2 ways: each tube has an old CERN ID number written on it in pen, and I am also writing a unique number on each in pencil to make the marks distinguishable. The CERN number is usually of the form #.##, or 5.32 for example. Start a new entry in the log book and write down the CERN numbers for each of the 8 new tubes you take from the box. Also write (on each tube) an identifier with your initials, the group number, and position number where you are installing each tube, i.e. “AMS 8.4” would mean that Andrea had used this tube in group 8 and in position 4 in the rack. You are starting with group 8. Remember to leave the same center (reference) tube in its place. Take the wine rack over to the light-tight box and lower it into place. This is the hardest part; there are marks on the box floor to show where to put the rack, but after it is in place you can align it better by sighting along the top tube and pointing it straight toward the LED in the opposite corner.

(2) Preliminary examination of output

Connect the HV and signal cables to the back of each tube; each cable is labeled with the tube number to which it should be connected. Now CLOSE THE BOX! This is critical whenever the HV is on.
Turn on the HV supply and put 1200 V on each tube. The HV channels correspond somewhat to the tube numbers, as follows:

Tube #	HV channel
C	0
1	1
2	2
3	3
4	4
5	5
6	12 (channel 6 may not be flaky, you can check it out)
7	7
8	13

Now put the scope in Average mode (in the Acquire menu) and look at the output from each tube. You should see a smooth signal with a (negative) height of about 60 mV or so. Do this for each of the 9 tubes. If you see any anomalous signals, look more carefully in other scope modes or get help, but if they still look bad, reject the tube and place it on the lower shelf of the cabinet with the other rejects. Replace it with another tube from the box

(3) Recording the tubes’ responses to the LED at high voltage

Once you are confident that the 8 tubes in the box are ready to be tested, close the box again and throw the black felt “blanket” over the left side where all the cables come through (to keep out more light from the room). Plug all the signal cables from the tubes into the ADC in the CAMAC crate - the tube number and ADC inputs all match except for Tube C (the reference) which is plugged into ADC input 11. The Gate input should already be set unless someone has changed it, if so, they should have left a note explaining the change.
Turn on the CAMAC crate by pushing the black button under the yellow light (it is just under slot 11 of the crate). It’s important to do this before you use the computer! You should hear a loud fan come on. Now warm up the Mac terminal and open “Andrea’s Test Instrument” from the desktop (it may already be open). The control panel for the KMax instrument should appear along with 9 empty histograms and a “Report” window. Wait a few minutes before proceeding to let the HV output and tube signals stabilize.
To get all the tube data measured and stored, iterate the following procedure:
Click the “Go” arrow on the control panel. The “Counts” window should show a rapidly growing number of ADC triggers received. Wait until this number reaches about 5000 and click Stop. Each of the tube should have a histogram of its measurements that looks like an upside-down parabola.
Save the Histograms as Text via the File Menu into a directory marked with the group number. The text files should be named by the voltage level followed by the tube number, e.g. 1200Ref.Text or 12004.Text. Make the file Space delimited, 1 column.
Click on the histogram and type OpenApple-I, which will bring up the statistics for it. Write down the Centroid and FWHM for each histogram.
Do these 3 things for HV levels from 1200 to 1600 V. You should accumulate 45 text histograms in the directory!

(4) Getting the useful information

Moving the stored text histograms to the Linux terminal is nontrivial; actually I haven’t succeeded in doing it yet. Until we can do this, it will be necessary to type a text file containing the data which the analysis program will read in. What I usually do is copy a file with the right format (for example g7all) and change the numbers. The analysis macro that I wrote requires the following format:

Name of group and date
Overall label for the x data (usually HV)
Group N, Tube C label
1200. ADC-Centroid ADC-FWHM at 1200 V
1300. ..... at 1300 V etc.
...........
1600. 3452.2 253.6
Group N, Tube 1 label
tube 1 data in 3 columns
.
.
.
Group N, Tube 8 label
Tube 8 data in 3 columns
eof

I think the data file should have 65 lines. Write it into a directory where you have a copy of my macro called “readall2.cpp”.
Open a ROOT session by typing >root or >root -l at the command line. Then execute the following commands:

Root> .L readall2.cpp
Root> getdata()
What file? <enter your filename here> Root> getdata()
What file? <enter the same filename again>

The first time that getdata() is run, it should produce a plot of data points corresponding to the logarithm of each tube’s ADC peak at each voltage, and a histogram of the slopes of each fit line. Unfortunately there is something wrong and only one real fit is performed. That’s why you have to call getdata() again. After calling getdata twice, the end of the ROOT output should have a list of the fit line parameters for each tube. Write down all the tube names, slopes and slope errors from the output.
You’re done! I may be reached by email ( asharp@uiuc.edu ) if there is a huge problem, and Steve and Brendan can help with the ROOT scripting. They are free to update and fix the macro if they like!