APPENDIX A: Electron Gun

Fig. 20. Electron gun, primary dimensions [inches].

 

Fig. 21. Electron gun, cross-sectional dimensions (V1 and V2 only) [inches].

 

TOP

Appendix B:  SIMION

      SIMION is a program, that, in an earlier incarnation for DOS, was used in the development of the gun by Kopp and Fishpaw.  Kurt Wick purchased the latest version for this project.  The gun was built to scale as well as could be managed.  The program uses “GU” or general units, which are not converted to inches or millimeters until the visualization stage.  Since there was no clear relation given between these units and inches, it was difficult to know if the gun really was to scale.  In particular, the very narrow holes in the end plates for V1 and V2 and the mid-plate in V2, although having clearly different diameters in the physical gun, have the same diameter in the model.

Fig. 22. Beam focusing visualization.

The program did not have a high enough resolution for a measurement in thousandths of an inch to be used.  As such, all predictions made by the program were questionable.

      Mostly the program was used as a visual aid in the working of the gun.  It allowed me to see the effect of a higher or lower V2 with fixed V1 and V3.  The program also was useful in the design and initial testing of the deflection system.  The general order of magnitude of voltages needed to move a 3 KeV beam of electrons was determined relatively easily.

      Figure 22 shows a lengthwise cross-section of the gun in SIMION.  This view lets one see the effects of the V2 potential on the focusing of the beam.  Notice that the beam’s diameter narrows in V3.  In practice, the narrowing would be centered over the surface of the dielectric.

Fig. 23. 3D view of the potentials in and around the gun.

TOP

Appendix C:  Vacuum System

      The Leybold-Heraus Trivac D8B vane pump has the following operating characteristics (supplied by manufacturer’s website):

Nominal pumping speed 50 / 60 Hz

      (m³ x h^-1).................................... 9.7 / 11.6

Pumping speed 50 / 60 Hz

      (m³ x h^-1).................................... 8.5 / 10.2

Ultimate partial pressure without gas ballast (Torr)       0.75 x 10^-4

Ultimate total pressure without gas ballast

      (Torr)....................................... <1.5 x 10^-3

Ultimate total pressure with gas ballast

      (Torr)....................................... <3.8 x 10^-3

Water vapor tolerance (Torr).................... 18.8

Noise level to DIN 45 635, without/with gas ballast (dB[A])      50 /52

Admissible ambient temperature (°C).. 12 – 40

Motor rating (HP).................................... 0.50

Nominal speed 50 / 60Hz (rpm)... 1500 / 1800

Weight (kg).............................................. 21.2

 

      The Leybold-Heraus Turbotronik NT 150/360 has the following operating characteristics (supplied by manufacturer’s website):

 Main connection, 60Hz (V)...................... 120

Max. power consumption during run up

      (VA)................................................... 750

Power consumption during normal operation, approx. (VA)    130

Max. output voltage / run-up current

      (V / A)..................................... 3 x 45 / 5.5

Overload current limit (A).......................... 3.5

Permissible ambient temperature (°C). 0 to +40

Weight (kg)................................................ 8.5

 

TOP

Appendix D:  Deflection

      The electron beam that comes out of the gun is aimed both downwards and to the right.  To correct for this deviation, two sets of plates are used to create electric fields in both the x-axis and the y-axis (z being the direction of motion).  In addition to the DC deflection voltage, the vertically aligned plates have an AC component that is used to “chop” sections of the beam out, thus giving the pulsed electron beam needed to improve the intensity of the Cerenkov radiation.  A pulsed beam also gives a pulsed radiation signal, thus making it easier to extract the signal from noise in the frequency domain.

      Some trials were conducted in SIMION to see if plates as small as were being considered could indeed deflect the beam enough.  Based on these constraints, the plates would have to be brought to a few hundred volts to work. 

Fig. 24. Deflection “cage” views (left, front, top)

 

      The four deflection plates (see Fig. 25) were designed to be held in a “cage” or sorts (see Fig. 24).  This cylindrical assembly was designed to fit snuggly in an elbow (an elbow is a 5” cylinder with a third port at a right angle coming out of the center of the pipe, see Fig. 26) that would be installed in front of the gun.  Only two of the plates would need to connected to exterior power supplies, as the other two would need to be grounded to the chamber.  Therefore only a single two-plug BNC connection plate would be needed. 

Fig. 25. Deflection plate dimensions (plate is 1/16 inch thick, as are the diameters of the holes) [inches].

Fig. 26. Extension to the gun placement, with deflector / chopper plates positioned down-beam of the gun.

 

Thus far I have milled all four plates and begun the assembly of the “cage.”  Due to welding issues, however, the cage was not completed (one of the four lengthwise supports could not be welded to the end ring with the support legs in place).  Once this problem is tackled, however, the deflection system can be operational in a very short time frame.  All the raw materials are available and the majority of the specialty parts have been prepared.  The only missing pieces are the power supplies and cables to set the plate potentials.  To oscillate the voltages at high frequency and at above normal voltages (the function generators can only go up to 10 VPP) a transformer will be used to step up the voltages.

 

TOP

Appendix E:  Sensor

      During the rapid pressure changes at the outset of pumping and when the air valve is opened at the end of a run, the diode voltage displayed a peculiar and sudden shift (see Fig. 27).  This is believed to be due to a rapid charge buildup on the diode, inducing some capacitance between it and the walls of the chamber.  Several trials with a resistor in parallel with the diode failed to offer any significant improvement.

      The slow oscillation back to equilibrium is present whether or not the multimeter is attached; trials were performed to test this hypothesis and sloping appeared whether the diode voltage was being monitored or not.  It also should be noted that the oscillation about equilibrium was not noted until late in the project, since no trial had ever been run without interruption for the nearly three hours it takes for the voltage to relax after pumping (see Fig. 28).  Even longer runs would be necessary to fully map the process.  Collecting data for the diode behavior after rapid repressurization is a bit easier, since the experiment can be left running unsupervised (while the pumps are active, they must be monitored continuously).  Again, more data is needed for a better understanding of the phenomena being observed.

Fig. 27. Voltage variation during a complete pump cycle.  Note that the rapid increase in voltage occurred during the pumping and the rapid drop was when the air valve was opened.

Fig. 28. An extended trial.  Note that the final voltage is less than the noise level present before pumping.

 

TOP

Appendix F:  Data

      For the following data analysis, only 92 of the 100 seconds from each subset were analyzed.  This cut out the first 5 and last 3 data points form each set.  These points were removed since they cover the reaction time of the power supplies.  It was noted during the trial that the voltage source did not change its output in less than one second, and therefore the shift in voltages affected several data points.  To be safe, these points were ignored in the individual analysis.  They are, however, present in the complete data analysis conducted earlier

Fig. 29. Data Subset 1: Diode: voltage [V] vs. time [s]

Fig. 30. Data Subset 1: FFT: power [watts] vs. frequency [Hz]

Fig. 31. Data Subset 1: Phase: phase angle [radians] vs. frequency [Hz]

Fig. 32. Data Subset 2: Diode: voltage [V] vs. time [s]

Fig. 33. Data Subset 2: FFT: power [watts] vs. frequency [Hz]

Fig. 34. Data Subset 2: Phase: phase angle [radians] vs. frequency [Hz]

Fig. 35. Data Subset 3: Diode: voltage [V] vs. time [s]

Fig. 36. Data Subset 3: FFT: power [watts] vs. frequency [Hz]

Fig. 37. Data Subset 3: Phase: phase angle [radians] vs. frequency [Hz]

Fig. 38. Data Subset 4: Diode: voltage [V] vs. time [s]

Fig. 39. Data Subset 4: FFT: power [watts] vs. frequency [Hz]

Fig. 40. Data Subset 4: Phase: phase angle [radians] vs. frequency [Hz]