O.M. Signals, Peak to Valley Studies
IceCube 2007 Home
Here is a new R7081-10 P.M.T. P/V Analysis
Introduction
We set up an oscilloscope to generate histograms of the peak size of O.M. signals (a function of the scope) with the trigger was set low enough to let through noise signals. The histograms were expected to have an initial noise peak and then a second peak representing a one photon event.
Equipment:
- Optical Module
- Hamamatsu BB-7085, with a R59 12-02 PMT
- Oscilloscope
- Ortec 584 Constant-Fraction Discriminator
Setup
The O.M. was inside the dark box without a scintillator plate. Its signal was split with a T. One part went to the CFD, and the other went to the oscilloscope The CFD was set to a 5 mV leading edge trigger level, so that it would let through a great deal of noise as well as legitimate signals. The scope was set to trigger on a timing signal from the CFD. The scope would then look at the signal from the O.M. and measure its peak height, peak area, and average the signal. It would then generate histograms of the peak height.
Note: The CFD was used to trigger instead of the scope because the scope was incapable of triggering at less than about 25% of the base mV/div scale it was set at. Since this experiment was done, new updates of the scopes software have been downloaded, and it is now capable of triggering at any level on any scale. If the experiment was repeated it would be unnecessary to use the CFD. It's fortunate that we did use it this time though. It was discovered that the CFD input socket is not at 50 ohm unless the CF delay sockets were also terminated in 50 ohms. The CFD was causing a definite reflection. This means that the counting statistics done on the P.M.T. R7081-10 earlier are probably wrong, although the general trends should still be the same.
Results
Since the input voltage needed for a gain of 10^8 has not been determined for BB-7085, a wide range was used. Reasonably well defined double peaks were found for the range of 1400V to 1800V (shown below). They have peak to valley values ranging from 2.0 to 3.0. An unexpected result was that at high input voltages a definite pre-signal is apparent. The CFD consistently saw two events at before the real signal. It usually saw three triggerable events after the signal as well (Click here). This is not really a problem but it is good to know.
IceCube 2007 Home
