Preliminary Simulation

• The geometry below was simulated with CST, where the return loss was monitored and the level and the resonant frequency of the signal was recorded.

• The return loss is very sensitive to the separation between the waveguide wall and the antenna, the level being below -30 dB at 15 GHz over a 29 micron separation range, below -40 dB this range is only 9 microns.

Design Investigation

• Two modified feedthrough designs were investigated: an antenna with a bead on the tip ,left, and a "stepped" antenna of two different radii, right.

• The bead design was chosen to be investigated further as it had a slightly larger acceptable separation range. The stepped design also seemed to be quite sensitive to the step parameters.

Bead investigation

• The radius and length of the bead were altered and the separation ranges were measured.

• Bead Length was 1 mm.
• For the -20 dB threshold, a bead radius of 1.65 mm gives the largest separation, for -30 dB, 1.55 mm gives the largest range.
• 1.55 mm bead radius was chosen because 1.65 mm is too close to the outer coaxial radius of 1.75 mm.

• Bead radius was 1.55 mm.
• The separation range increases with decreasing bead length.
• Due to an error a bead length of 0.5 mm was chosen to be studied further, even though 0.3 mm gives a better separation range.
• However, it was later discovered in ACE3P that decreasing the bead length increased the simulated loaded Q. 0.5 mm turned out to be a fair compromise between Q factor and separation range: ClicRedesignSim.

• A bead of radius 1.55 mm and length 0.5 mm at -20, -30 and -40 dB thresholds gives separation ranges of 204, 63 and 20 um respectively.

Coaxial Tolerance

• The offset and angle of the inner coaxial component relative to the outer were altered to investigate the tolerances on the coaxial.

• Increasing the offset of the inner coaxial from the center of the outer coaxial does not have an appreciable change on separation ranges at -20 and -30 dB threshold levels.
• There is a large change in the peak frequency, >200 MHz, which accounts for the change in the -40 dB separation range. At larger offsets this frequency change will be great enough to affect the other threshold values as well.

• In the Z direction (the long dimension of the waveguide) a negative tilt (towards the cavity) has a greater affect on the frequency and therefor the separation ranges.
• However, over a ±0.5 degree range, the change in the allowed range is still small.
• In the Y direction (dimension parallel with the beam pipe) the negative tilt (towards the waveguide center) again affects the frequency more.
• The ranges are affected less by tilts in this axis.