Reference Cavity

• The dimensions of the first copper prototype were known to be incorrect. Predicting a frequency drop of ~330 MHz and a loaded Q drop of 160.
• The BPM was clamped together and the feedthroughs also clamped without the vacuum seals so the separation was increased by 0.45 mm.
• The VNA was connected to the two ports of the reference cavity where S parameters could be measured.

Monopole Mode

• The measured resonant frequency is 300 MHz below the desired 15 GHz.
• The feedthrough antennae were not well concentrically aligned and frequency was seen to vary on repeated repositioning of the feedthrough.
• Simulations need to be performed to see if this effect is reproduced.
• The Q values can be determined by plotting a smith chart.

• The unloaded Q factor is much smaller than expected, bringing the loaded Q down.
• This could be caused by uneven clamping or not clamping strong enough or by excessive surface roughness.
• The frequency and Q factor was seen to vary with antenna offset, so the effect of this offset needs to be investigated.

Feedthrough Offset Variation

• Simulations were performed with the feedthrough offset ±1 degrees the along the beam pipe direction.
• Corresponding to a antenna tip movement range of ± 0.26 mm within the cavity.
• Positive values correspond to offsets towards the flat surface of the cavity, negative values correspond to offsets towards the reentrant part of the cavity.

• The frequency change depending on the antenna offset is ~700MHz/mm.
• The Q factor change is not enough to explain the low loaded Q factor of the position cavity.

Dipole Mode

Position Cavity

• The BPM was clamped as above and all four ports connected to the VNA.
• A drop in frequency from 15 MHz was predicted from simulation.

• The frequencies are very close to the expected value of 14.985 GHz, the Q factors are three times larger.
• The coupling here is too weak to rely on the Q factors calculated from the Smith chart but it does offer clues to the source of the high loaded Q.

• The external Q is much higher than the design value of ~600, possibly caused by the incorrect coupler.

Position Cavity with Reference Cavity Antenna

• The position cavity was simulated with the feedthrough antenna from the reference cavity.

• In the simulation the external Q is much larger than expected, driving the loaded Q up.
• The value of the loaded Q is closer to that measured from the transmission curves.
• This is encouraging evidence that using the correct feedthroughs and vacuum seal will results in Q's close to the design values.

Cross Coupling

• All maximum cross couplings are below -30 dB.