345 GHz Single Ended ‘barney’ Rx, Data analyses Jacob … · 345 GHz Single Ended ‘barney’ Rx, Data analyses Jacob W. Kooi 6/23/2006 ... we derive a front end receiver noise

345 GHz Single Ended ‘barney’ Rx, Data analyses Jacob W. Kooi 6/23/2006

Fig. 1 Instrument sensitivity in Hilo and the CSO. The red dot data is at the CSO.

Fig. 2 IV, Y-factor and Phot/Pcold curves. Optimal Y-factor at 2.08 mV. The ambient temperature was 276K. The M3 Hot/Cold data was measured at M3 with a 4” square LN2 dipped eccosorb pad. Rsg/Rn ratio: 12.7, Rp=63 Ohm, Rn5.37 Ohm., RnA~7.6, Area 0.71um2 (DeviceB6D21C3). R295K=175 Ohm.

Fig. 3 Optimal sensitivity at 2.08 mV. Conditions are in the graph. The M3 Hot/Cold data was measured at M3 with a 4” square LN2 Pad. Ambient temperature was 276K Using the intersection line technique (Blundell/Feldman et al.) we derive a front end receiver noise contribution of 15K+- 2K.. Fig. 4 Shot noise deduced IF Noise temperature (4.45K at 2.08mV) and DSB Mixer gain. This is with a B-field Current of 29mA. With less magnetic field (~20 mA) the mixer gain will improve slightly to about unity! The ambient temperature was 276K.

Fig. 5. DSB Mixer noise temperature. There is a 10K additional Optics loss when going from the cryostat exit to M3. 7K can be attributed to Ohmic loss in the mirrors. The remainder is in the air (100% humidity at the time of the measurement) and possibly some reduction due to not catching all the sidelobes/cross polarization signals that naturally occur when doing a measurement directly at the dewar exit. All noise temperatures have been calculated using the Callen-Welton formulism with half a photon noise in the loads. Tmix is 1.47 * hf/k, where the quantum noise limit at 345 GHz equals 16.6K Fig. 6. Given a Rsg/Rn ratio of the AlN barrier SIS junction of 14, it is reasonable to assume the presence of multiple Andreev Relflection current (Dieleman et al.). From fitting to the data, we find that of the of the measured current, 80% is contributed as tunnel current and 20% due to M.A.R. The effect on Tmix has not yet been calculated, but can be assumed quite small.

Fig. 7 Derived mixer gain and Fourier Transform Spectrometer measurement of the instrument. The waveguide Cutoff frequency is 250 GHz. The RF bandwidth is limited by the corrugated feedhorn, which is designed to work from 280-420 GHz.

Fig. 8 Calculated Sideband Ratio from the Fourier Transform Spectrometer measurements.

Fig. 9 IF Noise Contribution. From the intersecting line technique the RF optical loss is 15-20K over the frequency range: 280-420 GHz.

Fig. 10 Allan Variance on the telescope for a 4 GHz IF BW. In a 1.5 MHz AOS channel BW, this equates to about 51 seconds total power stability time.

280 GHz Fig. 11a Left: Ph, Pc with Tambient=17C Trec and Timx. I sis = 90 uA. Fig. 11b Left: Phot vs LO Power(Isis) I/V curve vs LO Power. Fig 11c Trec vs B-field. 30 mA appears optimal (blue curve) I/V curve vs B-field

Fig. 11d Left: Mixer Gain vs LO pump power (85-90 uA is Optimal) Fig 11e Mixer Gain vs Magnet field, 30 mA appears about optimal

316 GHz

Fig. 12a Left: Ph, Pc with Tambient=17C Trec and Timx. I sis = 90 uA.

Fig. 12b Left: Phot vs LO Power(Isis) I/V curve vs LO Power.

Fig 12c Trec vs B-field. 30 mA appears optimal (blue curve) I/V curve vs B-field

Fig. 12d Left: Mixer Gain vs LO pump power (85-90 uA is Optimal)

Fig 12e Mixer Gain vs Magnet field, 30 mA appears about optimal

345 GHz

Fig. 13a Left: Ph, Pc with Tambient=17C Trec and Timx. I sis = 90 uA.

Fig. 13b Left: Phot vs LO Power(Isis) I/V curve vs LO Power.

Fig 13c Trec vs B-field. 30 mA appears optimal (blue curve) I/V curve vs B-field

Fig. 13d Left: Mixer Gain vs LO pump power (85-90 uA is Optimal)

Fig 13e Mixer Gain vs Magnet field, 30 mA appears about optimal

384 GHz

Fig. 14a Left: Ph, Pc with Tambient=17C Trec and Timx. I sis = 90 uA.

Fig. 14b Left: Phot vs LO Power(Isis) /V curve vs LO Power. Fig 14c Trec vs B-field. 30 mA appears optimal (blue curve) I/V curve vs B-field Note 2nd Josephson step with SQUID interaction.

Fig. 14d Left: Mixer Gain vs LO pump power (85-90 uA is Optimal)

Fig 14e Mixer Gain vs Magnet field, 30 mA appears about optimal

424 GHz

Fig. 15a Left: Ph, Pc with Tambient=17C Trec and Timx. I sis = 90 uA.

Fig. 15b Left: Phot vs LO Power(Isis) /V curve vs LO Power

Fig 15c Trec vs B-field. 30 mA appears optimal (blue curve) I/V curve vs B-field Note 2nd Josephson step with SQUID interaction.

Fig. 15d Left: Mixer Gain vs LO pump power (85-90 uA is Optimal)

Fig 15e Mixer Gain vs Magnet field, 30 mA appears about optima

Barney Optics

Fig. 16 Measured and fitted beam Profile on CSO secondary mirror.. Fig. 17 Calculated Optics parameters based on measured Secondary Edgetaper.

Fig 18. ‘Barney’ on the telescope. Measured beam motion on M3 with telescope at Zenith to horizon is negligible.

Fig. 19 Checking beam alignment secondary (Fig. 10)

Fig 20. Mixer inside Cryostat

Fig. 21 Mixer block and optics

Fig. 22 Cryostat.

Fig 23 SIS junction in Groove. 50 um Quartz, twin AlN junctions. Device: B6D21C3. R295K = 175 Ohm, Rn=5.37 Ohm

Fig` 24 Assembled mixer block. IF 4-8 GHz.