Memorandum. Introduction. List of Figures. To: E. Bryerton K. Crady G. Ediss N. Horner A. R. Kerr D. Koller G. Lauria S.-K. Pan K. Saini D.

Transcription

1 Memorandum To: E. Bryerton K. Crady G. Ediss N. Horner A. R. Kerr D. Koller G. Lauria S.-K. Pan K. Saini D. Thacker cc: From: J. Webber J. Effland R. Groves Date: Subject: Gain vs. LO Power of SIS Mixer-Preamps for ALMA Band 6 Introduction Gain, noise temperature and total power variation of ALMA mixers as a function of LO power is required to calculate how LO amplitude noise degrades receiver stability. This memo provides measured gain, receiver noise temperature, and total power vs. LO levels for ALMA Band 6 mixers [1]. Although unbalanced sidebandseparating mixers are planned for the Band 6 receivers, measured data for balanced mixer-preamps are also included here for comparison. List of Figures FIGURE 1: EQUIPMENT SETUP FOR GAIN VS. LO POWER OF SINGLE-ENDED MIXER-PREAMP FOR ALMA BAND FIGURE 2: GAIN AND NOISE TEMPERATURES VS. IF, SINGLE-ENDED MIXER-PREAMP...6 FIGURE 3: GAIN VS. LO POWER, 2 GHZ, SINGLE-ENDED MIXER-PREAMP...7 FIGURE 4: RECEIVER NOISE TEMPERATURE VS. LO POWER, 2 GHZ, SINGLE-ENDED MIXER-PREAMP...8 FIGURE 5: IF NOISE POWER VS. LO POWER, RF HOT LOAD, 2 GHZ, SINGLE-ENDED MIXER-PREAMP...9 FIGURE 6: GAIN VS. LO POWER, 260 GHZ, SINGLE-ENDED MIXER-PREAMP...10 FIGURE 7: RECEIVER NOISE TEMPERATURE VS. LO POWER, 260 GHZ, SINGLE-ENDED MIXER-PREAMP...11 FIGURE 8: GAIN AND NOISE TEMPERATURES VS. IF, BALANCED MIXER-PREAMP...12 FIGURE 9: GAIN VS. LO POWER, 2 GHZ, BALANCED MIXER-PREAMP...13 FIGURE 10: RECEIVER NOISE TEMPERATURE VS. LO POWER, 2 GHZ, BALANCED MIXER-PREAMP...14 FIGURE 11: IF NOISE POWER VS. LO POWER, RF HOT LOAD, 2 GHZ, BALANCED MIXER-PREAMP...15 FIGURE 12: GAIN VS. LO POWER, 260 GHZ, BALANCED MIXER-PREAMP...16 FIGURE 13: RECEIVER NOISE TEMPERATURE VS. LO POWER, 260 GHZ, BALANCED MIXER-PREAMP...17 FIGURE 14: THEORETICAL DEPENDENCE OF GAIN AND NOISE TEMPERATURE ON LO AMPLITUDE USING THE QUASI FIVE- FREQUENCY METHOD AND ASSUMING A 50-OHM IF LOAD [2]...18 File: \\CVFILER\cv-cdl-sis\MeasSys\Data\GainVsLO\memo1.doc Page 1 of 18

2 Equipment Setup The equipment configuration used to measure the gain and noise temperature of the mixer-preamps is shown in Figure 1. Although Figure 1 shows the setup to measure single-ended mixer-preamps, the setup for balanced mixer-preamps is almost the same but includes another mixer bias supply to power the second component mixer. LO power was measured at the output of the tripler using an Anritsu ML-83A power meter with MP84B1 head. All power levels quoted in this memo are referenced to the output of the LO plate and do not include waveguide and coupling losses between the LO plate and mixer. Mixer and preamp gains are found using the usual P/ T equation by taking the ratio of measured noise powers to temperatures when the input is connected to hot and cold loads. The gain of the IF system is obtained by connecting its input to a hot and cold load using the 6-way switch in the Dewar: G IF = 1 P kb T hif hif P T cif cif where G IF P hif P cif is the IF system gain, is the power output from the IF system when its input is connected to a hot load, is the power output from the IF system when its input is connected to a cold load, T hif is the radiometric temperature of the hot load connected to the IF input, T cif is the radiometric temperature of the cold load connected to the IF input. In a similar fashion, overall system gain is found from a P/ T equation when power levels are measured with the RF input of the mixer switched to hot and cold loads using the chopper wheel. Gain of just the mixer-preamp is the overall system gain normalized by the IF system gain, which becomes: G MXR PREAMP P = P hrf hif P P crf cif T T hif hrf T T cif crf where G MXR-PREAMP is the gain of the mixer-preamp, and P hrf is the power output from the receiver when its input is connected to a hot load, P crf is the power output from the receiver when its input is connected to a cold load, P hif is the power output from the IF system when its input is connected to a hot load, P cif is the power output from the IF system when its input is connected to a cold load, T hrf is the radiometric temperature of the hot load connected to the receiver input, T crf is the radiometric temperature of the cold load connected to the receiver input, T hif is the radiometric temperature of the hot load connected to the IF input, and T cif is the radiometric temperature of the cold load connected to the IF input. Printed: 13 February 04 Page 2 of 18

3 The db attenuator installed between the mixer-preamp and the IF switch helps maintain nearly the same noise power levels when measuring RF and IF system noise with hot and cold loads. This attenuator was measured on a network analyzer and its value must be added to the measured mixer-preamp gain. Graphs Figure 2 through Figure 7 show results for a single-ended mixer-preamp. Similar data were measured last spring for balanced mixer-preamp, and those data are shown in Figure 8 through Figure 13. For the single-ended mixer-preamp operating at optimum LO power levels, gain and receiver noise temperature as a function of IF are shown in Figure 2 for two different LO frequencies. At an LO frequency of 2 GHz, noise temperatures span K to 50K as shown in Figure 2, compared to 45K to 55K quoted in Reference 1 for the same mixer-preamp topology. Measured noise temperatures at 260 GHz in Figure 2 span 42K to 70K, compared with 49K to 61K in Reference 1. Measured mixer-preamp gains in Figure 2 are within measurement error of being the same for both LO frequencies and range from 37 db to db. This is essentially the same as the limited gain data presented in Reference 1. Irregularities are evident in some of the measured points in the mixer-preamp gain and receiver noise temperature data shown in Figure 3 and Figure 4. The irregularities probably result from system gain changes because the data were measured over several days time and the LO powers were not changed monotonically. Single-ended mixer noise power in the 60 MHz passband of the IF filter is plotted in Figure 5 as a function of LO power at a LO frequency of 2 GHz. The power level at each IF is adjusted using a programmable attenuator on the warm IF plate to maintain a single power meter range when the chopper switches between hot and cold loads. This means that the attenuator value must be added to the IF power reading to find the power level assuming a constant IF attenuation. Balanced mixer-preamp data is presented for completeness, although the baseline plan for ALMA Band 6 cartridge uses sideband separating, single ended mixers. The balanced data were measured at 2 GHz and 260 GHz LO frequencies, but through an oversight, the singleended data were measured with LO frequencies of 2 GHz and 260 GHz. The balanced data, which were measured six months earlier, spans a much narrower range of LO levels than the single-ended data. A much larger LO power range is spanned during measurement of the single-ended data to show more clearly the optimum noise temperature and optimum mixer-preamp gain. Gain and receiver noise temperature as a function of IF for a balanced mixer-preamp operating at optimum LO power levels is shown in Figure 8. Balanced mixer noise power in the passband of the IF filter is plotted in Figure 11 as a function of LO power at a LO frequency of 2 GHz. Theoretical Sensitivity to LO Power Variation Analysis of the mixer using the quasi five-frequency method [2], and assuming a 50-ohm IF load, gives the dependence of gain and noise temperature on LO amplitude shown in Figure 14. Note that the range of LO amplitude shown in the figure is approximately 3 db i.e., ± 1.5 db about the point of minimum noise temperature. Printed: 13 February 04 Page 3 of 18

4 Acknowledgements The authors wish to thank A. R. Kerr for his many useful suggestions, comments, and his theoretical sensitivity contribution. References 1 E. F. Lauria, A. R. Kerr, M. W. Pospieszalski, S.-K. Pan, J. E. Effland, and A. W. Lichtenberger, "A 0-0 GHz SIS Mixer- Preamplifier with 8 GHz IF Bandwidth," 01 IEEE International Microwave Symposium Digest, pp , Available as ALMA Memo 378 at 2 A. R. Kerr, S.-K. Pan, and S. Withington, "Embedding Impedance Approximations in the Analysis of SIS Mixers," IEEE Trans. Microwave Theory Tech., vol. 41, no. 4, pp , April This paper was originally presented at the Third International Symposium on Space Terahertz Technology, March Printed: 13 February 04 Page 4 of 18

5 Chopper Wheel Dewar Single-Ended Mixer/Preamp IF db Port 4 Warm IF Plate F = GHz 0.5 db 0-63 db G = db NF = 5.3 db 18 db Cold Load Hot Load Heater G = 23 db NF = 2.5 db 3 db YIG FILTER BW = 60 MHz 3 db Cold Load Hot Load LO Rack Anritsu Power Meter ML81A LO Plate Pwr Hd Anritsu MP717A Wave Meter Hitachi X3 Tripler Millitech FTT J4-3RS00 Square Law Detector HP 436A Power Meter Spectrum Analyzer HP 8484A Gunn Osc Carlstrom H55 Coax SW Controller Preamp Bias Supply Lambda Power Supply DC Out 10V In TRG 5 Mixer Bias Supply Mixer 6-Wire Bias Detail Ij + Ij - 10K Ω 10K Ω 07 JEE NEW ON CVFILER, ADDED MXR/AMP BOX 06 JEE CHANGED BAL MIXER TO SINGLE-ENDED 05 JEE ADDED AMP AND 2 ATTENS IN DEWAR 04 JEE REMOVED 3RD AMP & 7DB PAD-WARM IF 03 JEE REMOVED COLD IF, ADDED 0.5 DB PAD 02 JEE ADDED COLD IF AMP 01 JEE INITIAL S + Ibias 1K Ω 50 Ω 5 Ω 10K Ω Mixer Diodes S - Bias Supply Vj+ Vj- Color Key RF IF Bias Figure 1: Equipment Setup for Gain vs. LO Power of Single-Ended Mixer-Preamp for ALMA Band 6 Printed: 13 February 04 Page 5 of 18

6 Single-Ended Mixer/Preamp Gain and Noise Temps. vs. IF L586A-2-F6-2-B3-371C-01/IF4-12P.04 Measured Noise Temp, Flo = dbm Noise Temp, Flo = dbm Gain, Flo = dbm Gain, Flo = dbm Receiver DSB Noise Temperature (K) Mixer/Preamp Gain (db) IF Frequency (GHz) Data File: "\\CVFILER\cv-cdl-sis\MeasSys\Data\GainVsLO\SingleEnded.xls" Sheet: "GainvsIF260" Figure 2: Gain and Noise Temperatures vs. IF, Single-Ended Mixer-Preamp Printed: 13 February 04 Page 6 of 18

7 Single-Ended Mixer/Preamp Gain vs. LO Power L586A-2-F6-2-B3-371C-01/IF4-12P.04 Flo = 2 GHz Measured Mixer/Preamp Gain (db) Power at LO Multiplier Output (dbm) Data File: "\\CVFILER\cv-cdl-sis\MeasSys\Data\GainVsLO\SingleEnded.xls" Sheet: "GainvsLOpwr2" Figure 3: Gain vs. LO Power, 2 GHz, Single-Ended Mixer-Preamp Printed: 13 February 04 Page 7 of 18

8 Rx Noise Temperature vs. LO Power Using Single-Ended Mixer/Preamp L586A-2-F6-2-B3-371C-01/IF4-12P.04 Flo = 2 GHz Measured Receiver DSB Noise Temp (K) Power at LO Multiplier Output (dbm) Data File: "\\CVFILER\cv-cdl-sis\MeasSys\Data\GainVsLO\SingleEnded.xls" Sheet: "TempvsLOpwr2" Figure 4: Receiver Noise Temperature vs. LO Power, 2 GHz, Single-Ended Mixer-Preamp Printed: 13 February 04 Page 8 of 18

9 Single-Ended Mixer/Preamp IF Power vs. LO Power L586A-2-F6-2-B3-371C-01/IF4-12P.04 Flo = 2 GHz Measured IF Total Noise Power (dbm) Power at LO Multiplier Output (dbm) Data File: "\\CVFILER\cv-cdl-sis\MeasSys\Data\GainVsLO\SingleEnded.xls" Sheet: "IFPwrvsLOpwr2" Figure 5: IF Noise Power vs. LO Power, RF Hot Load, 2 GHz, Single-Ended Mixer-Preamp Printed: 13 February 04 Page 9 of 18

10 Single-Ended Mixer/Preamp Gain vs LO Power L586A-2-F6-2-B3-371C-01/IF4-12P.04 Flo = 260 GHz Measured Mixer/Preamp Gain (db) Power at LO Multiplier Output (dbm) Data File: "\\CVFILER\cv-cdl-sis\MeasSys\Data\GainVsLO\SingleEnded.xls" Sheet: "GainvsLOpwr260" Figure 6: Gain vs. LO Power, 260 GHz, Single-Ended Mixer-Preamp Printed: 13 February 04 Page 10 of 18

11 Rx Noise Temperature vs. LO Power Using Single-Ended Mixer/Preamp L586A-2-F6-2-B3-371C-01/IF4-12P.04 Flo = 260 GHz Measured Receiver DSB Noise Temp (K) Power at LO Multiplier Output (dbm) Data File: "\\CVFILER\cv-cdl-sis\MeasSys\Data\GainVsLO\SingleEnded.xls" Sheet: "TempvsLOpwr260" Figure 7: Receiver Noise Temperature vs. LO Power, 260 GHz, Single-Ended Mixer-Preamp Printed: 13 February 04 Page 11 of 18

12 Balanced Mixer/Preamp Gain and Noise Temps. vs. IF L586A-2-F6-2-B3-371C-01/IF4-12P.04 Measured Noise Temp, Flo = dbm Noise Temp, Flo = dbm Gain, Flo = dbm Gain, Flo = dbm Receiver DSB Noise Temperature (K) Mixer/Preamp Gain (db) IF Frequency (GHz) Data File: "\\CVFILER\cv-cdl-sis\MeasSys\Data\GainVsLO\Balanced.xls" Sheet: "GainvsIF260" Figure 8: Gain and Noise Temperatures vs. IF, Balanced Mixer-Preamp Printed: 13 February 04 Page 12 of 18

13 Balanced Mixer/Preamp Gain vs LO Power BBIA371-A-UVAVIII-L811B-1-0-C2-6-BMIF4-12P.02 Flo = 2 GHz Measured Mixer/Preamp Gain (db) Power at Mulitplier Output (dbm) Data File: "\\CVFILER\cv-cdl-sis\MeasSys\Data\GainVsLO\Balanced.xls" Sheet: "GainvsLOpwr2" Figure 9: Gain vs. LO Power, 2 GHz, Balanced Mixer-Preamp Printed: 13 February 04 Page 13 of 18

14 Rx Noise Temperature vs. LO Power Using Balanced Mixer/Preamp BBIA371-A-UVAVIII-L811B-1-0-C2-6-BMIF4-12P.02 Flo = 2 GHz Measured Receiver DSB Noise Temperature (K) Power at Mulitplier Output (dbm) Data File: "\\CVFILER\cv-cdl-sis\MeasSys\Data\GainVsLO\Balanced.xls" Sheet: "TRvsLOpwr2" Figure 10: Receiver Noise Temperature vs. LO Power, 2 GHz, Balanced Mixer-Preamp Printed: 13 February 04 Page 14 of 18

15 Balanced Mixer/Preamp IF Power vs. LO Power BBIA371-A-UVAVIII-L811B-1-0-C2-6-BMIF4-12P.02 Flo = 2 GHz Measured IF Total Noise Power (dbm) LO Power at Mulitplier Output (dbm) Data File: "\\CVFILER\cv-cdl-sis\MeasSys\Data\GainVsLO\Balanced.xls" Sheet: "IFPwrvsLOpwr2" Figure 11: IF Noise Power vs. LO Power, RF Hot Load, 2 GHz, Balanced Mixer-Preamp Printed: 13 February 04 Page 15 of 18

16 Balanced Mixer/Preamp Gain vs LO Power BBIA371-A-UVAVIII-L811B-1-0-C2-6-BMIF4-12P.02 Flo = 260 GHz Measured Mixer/Preamp Gain (db) Power at Mulitplier Output (dbm) Data File: "\\CVFILER\cv-cdl-sis\MeasSys\Data\GainVsLO\Balanced.xls" Sheet: "GainvsLOpwr260" Figure 12: Gain vs. LO Power, 260 GHz, Balanced Mixer-Preamp Printed: 13 February 04 Page 16 of 18

17 Rx Noise Temperature vs. LO Power Using Balanced Mixer/Preamp BBIA371-A-UVAVIII-L811B-1-0-C2-6-BMIF4-12P.02 Flo = 260 GHz Measured Receiver DSB Noise Temperature (K) Power at Mulitplier Output (dbm) Data File: "\\CVFILER\cv-cdl-sis\MeasSys\Data\GainVsLO\Balanced.xls" Sheet: "TRvsLOpwr260" Figure 13: Receiver Noise Temperature vs. LO Power, 260 GHz, Balanced Mixer-Preamp Printed: 13 February 04 Page 17 of 18

18 Figure 14: Theoretical dependence of gain and noise temperature on LO amplitude using the quasi five-frequency method and assuming a 50-ohm IF load [2] Printed: 13 February 04 Page 18 of 18