ALMA Memo December 2001 rev. 5 April Saturation by Noise and CW Signals in SIS Mixers

Transcription

1 Presented at the 00 International Symposium on Space THz Teccnology, Cambridge, MA, March 00 ALMA Memo December 001 rev. 5 April 00 Saturation by Noise and CW Signals SIS Mixers A. R. Kerr National Radio Astronomy Observatory Charlottesville, VA 903, USA ABSTRACT In ALMA Memo 31, Plambeck pots out that saturation (ga compression) is likely to be a significant factor limitg the calibration accuracy of ALMA observations. In this memo, saturation by broadband noise and CW signals is analyzed for representative SIS receivers operatg at different frequencies. Many SIS mixers current use are expected to exhibit a significant degree of ga compression when connected to a room-temperature source. Previous analyses of saturation SIS mixers have applied only to CW signals. To analyze saturation by noise, the statistics of the output voltage are derived from those of the put signal. A sgle constant, applicable to all SIS mixers, is determed experimentally by fittg the predicted CW ga compression curve to measured data. Keywords: Superconductor-Insulator-Superconductor mixers, saturation, ga compression, dynamic range. 1. INTRODUCTION Saturation (ga compression) can be a serious problem SIS mixers not designed with appropriate put power levels md. Plambeck pots out [1] that ga compression is likely to be a significant factor limitg the calibration accuracy of ALMA observations. SIS receivers have been reported with substantial ga compression at source temperatures as low as 50 K (1.4 db compression []) and 300 K (1.7 db compression [1]), while others have been capable of observg the sun (~6000 K) with only 0.6 db ga compression [3]. Previous analyses of saturation SIS mixers have considered only saturation by CW (susoidal) put signals. For amplifiers, it is well known that a given degree of saturation occurs at a lower put power level with noise than with a CW signal, and similar behavior is expected SIS mixers. This memo describes a method for analyzg saturation by noise or CW signals SIS mixers, and gives generalized saturation curves and representative results for mixers at several frequencies. The established phenomenological theory of saturation by CW signals SIS mixers is reviewed Section, and shown to agree quite well with experimental data, even at ga compression levels as high as 3 db. When saturation is caused by broadband noise, a different approach is required, as described Section 3, which allows the statistics of the output noise to be deduced from those of the put noise. The conclusion, not surprisgly, is that small and moderate degrees of ga compression are produced by noise powers several db lower than the CW power required to produce the same ga compression. Many SIS mixers current use are expected to exhibit a significant degree of ga compression when connected to a room-temperature source. It is noted that the learity of a partly saturated SIS receiver to a small CW test signal can lead to the erroneous conclusion that the SIS mixer is not saturated by room-temperature put noise.. ANALYSIS OF SATURATION BY A CW SIGNAL For an SIS mixer with a well developed quantum response, the small signal power ga is a function of the bias voltage; this is evident Fig. 1. Smith and Richards [4] argued that the large signal power ga could be considered as an average value of the small signal ga over an IF cycle.

2 Fig. 1. I-V characteristics and output power vs. bias voltage curves for a 70 GHz SIS receiver. From [5]. The small signal power ga can be expanded as a Taylor series about the bias voltage V 0 [6]: '' Gv ( ) GV vv G V / (1) If a CW put signal produces an output voltage v = V 0 + V IF s(ω IF t), the time-averaged ga is then IF G G V G 0 '' 0 / 4 VIF The IF power delivered to load R L is PIF PG, from which it follows that [7] R '' L G G 0 P 1 P sig sat 1. (), (3) where Psat / G0RL. Because the degree of saturation of an SIS mixer depends only on the magnitude of the IF output voltage relative to the width of a photon step, i.e., (G R L P ) ½ relative to Nhf/e, where N is the number of junctions series and f is the LO frequency, P sat can be written the form hf N P K C N f sat sat sat, (4) e GR 0 L GR 0 L where K sat and C sat are constant for all SIS mixers. K sat has been evaluated [6] from experimental data which gives K sat = 0.10, correspondg to C sat = 1.7e-30. Figs. and 3 show the results of saturation measurements on two SIS mixers usg CW signals [7]. The saturation curve accordg to eq. (3) (the solid le) is seen to fit the measured data well. The value of P sat is chosen to give the best fit to the data. --

3 Fig.. Measured CW saturation for a -junction Pb/Ox/Nb SIS mixer at 115 GHz, from [7]. The solid curve is from eq. (3) with the value of P sat chosen to give the best fit to the measured data. Fig. 3. Measured CW saturation for a 4-junction Nb/Al-AlOx/Nb SIS mixer at 115 GHz, from [7]. The solid curve is from eq. (3) with the value of P sat chosen to give the best fit to the measured data. -3-

4 Table I shows the values of C sat obtaed from measurements of the CW saturation power of four different SIS receivers. It is clear that the values of C sat are reasonably close to the value 1.7e-30 above. Table II shows, for the same four SIS receivers, the CW signal power required to produce a ga compression of 1 db and 1%. Table I C sat evaluated for four SIS mixers LO GHz Type N R n L SSB db R L P sat nw C sat Pan et al. [8] 115 Pb-Nb E-30 Feldman et al. [7] 115 Pb-Nb E-30 Feldman et al. [7] 115 Nb E-30 Tong et al. [5] 70 Nb E-30 Table II CW saturation powers for the mixers Table I LO GHz N L SSB db R L P 1dB nw P 1% nw Pan et al. [8] Feldman et al. [7] Feldman et al. [7] Tong et al. [5] ANALYSIS OF SATURATION BY BROADBAND NOISE The previous section considered saturation caused by a susoidal RF put signal. When the RF put is noise, different saturation characteristics are expected. Because of the statistical nature of noise, the stantaneous amplitude of the signal at times exceeds the peak value of a CW signal havg the same power, so ga compression begs at a lower power level than with a CW signal. To analyze ga compression by noise, it is convenient to consider the voltage ga of an SIS mixer rather than the power ga which was the focus the previous section. If the probability density function of the put noise voltage is known, and also the nonlear voltage characteristic of the ga compression mechanism, a new probability density function can be computed which characterizes the compressed output signal. In this section, this approach is applied to saturation by noise and CW signals the latter case givg results close agreement with those the previous section. The SIS mixer is characterized Fig. 4 as a non-saturatg mixer series with a saturatg IF element. The non-saturatg mixer has ga and impedance characteristics, at all put power levels, identical to those of the actual mixer when operated under small signal conditions. The saturatg element has an put impedance equal to the IF load impedance, R L, an output impedance equal to that of the mixer, Z m, and unity small-signal voltage ga when connected to the IF load. Hence, under small-signal and moderately large-signal operation, the circuit of Fig. 4 is distguishable from the real mixer. -4-

5 Fig. 4. Representation of the SIS mixer as a non-saturatg mixer series with a saturatg IF element. The non-saturatg mixer has ga and impedance characteristics identical to those of the actual mixer when operated under small signal conditions. The saturatg IF element has unity small signal ga, an put impedance equal to the IF load impedance, R L, and an output impedance equal to that of the mixer, Z m. The saturatg element Fig. 4 is characterized by its nonlear voltage characteristic v out (t) = f(v (t)), measured with the mixer and IF load place. Note that v and v out are the stantaneous values, not amplitudes, of the IF voltages at the put and output of the saturatg IF element, as dicated Fig. 4. When the mixer is biased near the middle of a photon step at the small-signal ga maximum, v out is approximated by an odd function of v, and the differential ga of the saturatg element is an even function of v which can be written, to first order, as 1. (5) A dv out 1 dv 1 Av 3 Sce the degree of saturation of an SIS mixer depends only on the magnitude of the IF output voltage relative to the width Nhf/e of a photon step, it is convenient to use normalized put and output voltages: e V and. (6) Nhf v e Vout Nhf v out Then, (5), dvout dvout 1 AV ( ), (7) dv dv 1 CV 3 where C 3 is a constant, dependent of N and f, for all SIS mixers. The stantaneous (large signal) normalized output voltage is then V dv 1 V t out () CV t, (8) 1 CV 3 arctan 3 0 () t C3 and the stantaneous large-signal voltage ga is v out Vout arctancv 3 t A t. (9) LS () v V CV () t When saturation is caused by broadband noise, eq. (9) allows the probability density function of the output noise to be computed from that of the put noise. The probability density of a Gaussian noise signal of mean square voltage σ is given by 3 1 Other even functions could be chosen to approximate A(v ), for example, the truncated Taylor expansion (1 + Bv ), which would give a good approximation to saturation the real mixer for small values of v. It was found that the form of eq. (5), 1/(1 + Bv ), gave good agreement with measured data even at levels of ga compression as high as 3 db. -5-

6 1 v pv () exp. (10) For the circuit of Fig. 4, if σ is the RMS voltage at the put of the nonlear element whose put impedance is R L, then the power delivered to the nonlear element P = G 0 P sig = σ /R L. The output power delivered to the IF load R L is P out GP where v out (v ) is given by eq. (9). Therefore, G G 0 sig 1 1 vout p( v ) dv L R out v p( v ) dv, (11). (1) Normalizg voltages to Nhf/e as eq. (6), and defg the normalized RMS voltage at the put of the saturatg e element as S, eqs. (8), (10) and (1) give Nhf G G 1 V 3 CV 3 C S arctan exp S 0 3 dv. (13) e e Sce S GP 0 sig RL, eq. (13) allows G/G 0 to be computed as a function of P sig, G 0, R L, f, Nhf Nhf and N. If the put noise is from a broadband source with noise temperature T sig at the put of the receiver, then P sig = kt sig B 1, the factor accountg for the noise received both upper and lower sidebands of the receiver. Determation of the constant C 3 is described the next section, and the appropriate value of the effective IF bandwidth B 1 is discussed Section 4. Determation of C 3 To determe the constant C 3, saturation by a CW signal was analyzed usg (9), and C 3 adjusted to fit the experimental results given the previous section. A CW RF put signal of power P sig delivers power P 0 G P element Fig. 4. If the voltage at the put of the saturatg IF element is 0 sig L P G P a / R sig to the put resistance R L of the saturatg v () t as( t), then. (14) The stantaneous output power delivered to the IF load is P () t v ()/ t R, which has an average value over the IF period 1 1 IF Pout GPsig vout t dt R () L IF 0. (15) From eqs. (14) and (15): G 1 IF vout t dt. G0 a () IF 0 (16) Usg normalized voltages, as eq. (6), eq. (8) gives G 1 IF CA 3 IFtdt, G AC arctan s 0 (17) 0 3 IF out out L IF -6-

7 where A e Nhf a is the normalized amplitude of v. Eq. (17) describes saturation by a CW signal any SIS e mixer. Sce A RLG0 Psig, eq. (17) allows G/G 0 to be computed as a function of P sig, G 0, R L, f, Nhf and N. To determe the constant C 3, saturation by a CW signal was analyzed usg (17), and C 3 adjusted to fit the experimental results given the previous section. A good fit to the measured data at low saturation levels is obtaed with C 3 = 3.3. Fig. 5 shows the agreement between eq. (3) with P sat = 3.5 nw, which closely fits the measured saturation data for the 115 GHz 4-junction mixer, and the saturation characteristic given by eq. (9) with C 3 = 3.3. In the above approach, the degree of ga compression was determed from the mean power delivered to the IF load (eqs.(15)-(17)). It could also be determed usg the fundamental Fourier component of the power delivered to the IF load. We found that these two methods of calculation had negligible difference for ga compression levels up to ~50%. Fig. 5. The graph shows the close agreement between eq. (17) with C 3 = 3.3 and Eq. (3) with P sat = 3.5 nw. Eq. (3) was shown to be an excellent fit to the measured ga compression curve for a 115 GHz 4 junction SIS mixer [7] (see Fig. 3). Fig. 6. Saturation by noise and CW signals. For noise, S is given by eq. (18), and for CW, A / is given by eq. (19). The curve labeled CW[7] is computed from eq. (3) as [7]. Comparison of Saturation by Noise and CW A comparison of saturation by broadband noise and CW signals of equal powers is possible usg eqs. (13) and (17). For noise, the signal power ( both sidebands) P sig is related to S by sig RL e S GP 0, (18) Nhf and for a CW signal P sig is related to A by A e GP 0 sig RL, (19) Nhf where N is the number of SIS junctions series, G 0 is the small-signal ga of the mixer, and R L is the IF load resistance. Fig. 6 shows G/G 0 as a function of S (for noise) and A / (for CW). Also shown for comparison is the saturation curve computed from eq. (3) as [7], which has been shown to fit experimental data well the range 0.5 G/G 0 1 (see Figs. and 3). -7-

8 Examples To illustrate the degree to which ga compression is likely to affect SIS receiver calibration when usg a room-temperature calibration load, we make the followg assumptions: (i) The mixer put bandwidth is B 1 each sideband, with B 1 equal to 0% of the LO frequency. Roughly, this corresponds to a receiver which downconverts all frequencies with the full RF waveguide band, half the LSB and half the USB, to an extended IF band from 0 to B 1 Hz. (ii) The IF load on the mixer is taken as R L = 50 ohms over the extended IF band 0 < f < B 1, an approximation which will be discussed below. An IF impedance of 50 ohms is used because this is the nomal put impedance of many low noise IF preamplifiers, cludg the present NRAO 41 GHz HFET preamplifier [9]. We assume the mixer and preamplifier are connected directly together with no terveng impedance transformer. (iii) The small-signal ga has a constant value G 0 over the full put band 0.8 f LO < f < 1. f LO. Under these assumptions, Fig. 7 shows the expected ga compression caused by a room-temperature calibration load for SIS mixers at four LO frequencies. The ga compression produced by a CW put signal of the same power as the noise put is shown for comparison. It is seen from Fig. 7 that mixer receivers satisfyg assumptions (i)-(iii), with small signal ga G 0 = 6 db SSB and a sgle SIS junction, can be expected to have a ga compression due to a room-temperature source of 16% at 115 GHz and 5% at 460 GHz. With four junctions, the ga compression is 1.4% at 115 GHz and 0.4% at 460 GHz. Fig. 7. Curves of ga compression caused by a 300 K source (solid red les), as a function of small-signal mixer ga, G 0, for SIS mixers at four frequencies. The parameter N is the number of junctions series. In all cases: (i) the put noise bandwidth B 1 each sideband is equal to 0% of the LO frequency, (ii) the IF load impedance is 50 ohms over the extended IF band 0 < f IF < B 1, and (iii) the small-signal ga is constant over 0.8 f LO < f sig < 1. f LO. Shown for comparison is the ga compression by a CW signal of the same power (dotted black le). -8-

9 4. Discussion It is clear from the simulations Section 3 that noise and CW signals of equal power produce different degrees of saturation SIS mixers. Furthermore, it is apparent that a room-temperature source may cause significant ga compression many practical SIS receivers, a phenomenon which affects the accuracy of astronomical observations made with an SIS receiver calibrated usg a room-temperature source. Bandwidth and IF Load Impedance The noise saturation results Fig. 7 are based on the followg major assumptions: (i) the put noise bandwidth B 1 each sideband is equal to 0% of the LO frequency, (ii) the IF load impedance R L = 50 ohms over the frequency range 0 to B 1, and (iii) the small-signal ga G 0 is flat over 0.8 f LO < f sig < 1. f LO. The frequency range 0.8 f LO < f sig < 1. f LO corresponds approximately to a standard waveguide band when f LO is at the band center. A typical broadband waveguide to TEM mode transducer has poor couplg outside the waveguide band, and thus acts to some extent as a band-limitg filter. However, as saturation SIS mixers is primarily determed by the magnitude of the output voltage, the behavior of the embeddg impedance Z e (f IF ) over 0 < f IF < B 1 seen by the SIS junction(s) is the major determant of the saturation behavior. In the analysis, it has been assumed that Z e = R L at all frequencies, whereas reality Z e depends on the RF choke circuit of the mixer, any circuit elements between the mixer and IF amplifier, and the put impedance of the IF amplifier itself. The importance of Z e over this extended IF range has been poted out [10]. Improvg the Dynamic Range Are there ways to improve the dynamic range of an SIS receiver? It is clear, from eqs. (3) and (4) the case of CW signals, and eqs. (6) and (9) the case of noise, that the dynamic range of an SIS mixer can be creased by: (i) creasg the number N of junctions series, (ii) reducg the transducer ga G 0, or (iii) reducg the IF load resistance R L (which also affects G 0 ). If saturation is caused by broadband noise, a bandpass or lowpass IF filter immediately followg the SIS mixer could reduce the RMS IF noise voltage at the mixer output if the filter were designed to have low impedance at all frequencies above the desired IF band. This has been suggested [10] but is difficult to implement unless the filter is corporated to the SIS mixer chip. In [11], a GHz SIS mixer is described which Z e < 50 ohms over 0 to 150 GHz when the IF amplifier has an impedance of 50 ohms over that band ( reality, of course, the impedance of the IF amplifier is unlikely be 50 ohms at frequencies far from the nomal IF band). Fig. 8 shows Z e (f) for that mixer, cludg the effects of the RF choke, junction capacitance, and all the RF circuit elements. The decrease of Z e with frequency reduces the IF voltage at the mixer output due to a broadband noise put, and thereby reduces the degree of saturation compared with a mixer which the IF embeddg impedance is largely the high impedance region of the chart. Fig. 8. Embeddg impedance Z e (f) seen by the SIS junctions a low-parasitic GHz SIS mixer [11]. Z e (f) cludes the RF choke and all the RF circuit elements, and assumes that the put impedance of the IF amplifier is 50 ohms at all frequencies. Z e is plotted on a 50-ohm Smith chart over the frequency range GHz with markers every 10 GHz. -9-

10 In prciple, the dynamic range of an SIS mixer can be creased by creasg the junction area by a factor M (> 1) while reducg the embeddg impedance at all frequencies by the same factor M; for example, the larger junction could be connected to the same embeddg circuit through an ideal transformer with impedance ratio M at all frequencies. A given degree of ga compression occurs both the origal mixer and the modified mixer when each has the same IF voltage at the junction. However, the modified mixer has an IF load impedance (seen from the junction) lower by factor M than the origal mixer, so the IF power delivered to the load is larger by factor M, and the RF put power is therefore also larger by a factor M. Measurement of Ga Compression usg a CW Signal the Presence of Broadband Noise It is of terest to consider the response of an SIS mixer, partly saturated by broadband noise, to a small CW test signal. The test signal produces a response with a small-signal ga which is a function of the total put power (test signal + noise). The ga is not substantially affected by the test signal as long as its power is small compared with the noise put power, and the small-signal response is then lear. Only when the test signal power approaches that of the broadband put noise power RF bandwidth B 1 (B 1 is the effective IF noise bandwidth as defed above, and the factor accounts for the two RF sidebands a DSB mixer) will it contribute substantially to saturation, which will be evident from the nonlearity of the curve of test signal output power vs test signal put power. It is sometimes assumed that a lear curve of test signal output power vs test signal put power implies that a receiver is not saturated by broadband noise accompanyg the test signal. This is the assumption [5], for a receiver with a room temperature put load which no ga compression was measured at low CW powers. It was concluded that "...the receiver is still highly lear when subjected to radiation from an ambient load..." If one applies the noise saturation analysis of Section 3 to the mixer used [5], assumg an extended IF noise bandwidth B 1 of 15 GHz, a room temperature black body source is predicted to produce ~5% ga compression this is consistent with the other data given [5]. To determe the degree of ga compression produced by broadband put noise, a small CW test signal can be used as an dicator. The small-signal ga curve (test signal output power vs test signal put power) is measured first, the presence of the high level broadband noise source (e.g., a room temperature load), and a convenient signal level the lear region of this curve is chosen. Then, the high level noise source is replaced with a low level noise source (e.g., a liquid nitrogen load) and the test signal output level re-measured. Any ga compression caused by the high level noise source will be dicated by an crease the test signal output when the high level noise source is replaced with the low level noise source. Balanced and Sideband Separatg Mixers In balanced mixers and sideband-separatg mixers, the put power is divided equally between two unit (double-sideband) mixers. The power required to produce a given degree of ga compression is therefore twice that required to produce the same ga compression a sgle unit mixer connected to the same signal and IF embeddg impedances. In the case of a balanced sideband-separatg mixer, the put power is divided between four unit mixers, so the saturation power is four times that of the unit mixer. -10-

11 ACKNOWLEDGMENTS The author thanks Shg-Kuo Pan of NRAO and Charles Cunngham of HIA for their valuable discussions on saturation SIS mixers and comments on the manuscript. REFERENCES [1 ] R. L. Plambeck, "Receiver amplitude calibration for ALMA," ALMA Memo 31, August 7, 000. Available at [ ] L. R. D'Addario, "An SIS Mixer for GHz with Ga and Wide Bandwidth," Int. J. Infrared Millimeter Waves, vol. 5, no. 11, pp , [3] E. S. Palmer, T. Dame, private communication, November 001. The mixer used two SIS junctions series and was tuned for SSB operation usg its two waveguide tuners. [4] A. D. Smith and P. L. Richards, "Analytic solutions to SIS quantum mixer theory," J. Appl. Phys., vol. 53, no. 5, pp , May 198. [5] C.-Y. E. Tong, R. Blundell, S. Pae, D. C. Papa, J. Kawamura, J. Stern, and H. G. LeDuc, "Design and Characterization of a GHz Fixed-Tuned Superconductor-Insulator-Insulator Receiver," IEEE Trans. Microwave Theory Tech., vol. MTT-44, no. 9, pp , Sept [6] M. J. Feldman and L. R. D'Addario, "Saturation of the SIS direct detector and the SIS mixer," IEEE Trans. Magnetics, vol. MAG-3, no., pp , March [7] M. J. Feldman, S.-K. Pan, and A.R. Kerr, "Saturation of the SIS mixer," International Superconductivity Electronics Conference Digest, Tokyo, pp. 90-9, Aug [8] S.-K. Pan, A. R. Kerr, M. J. Feldman, A. Klesasser, J. Stasiak, R. L. Sandstrom, and W. J. Gallagher, "An GHz SIS receiver usg ductively shunted edge-junctions," IEEE Trans. Microwave Theory Tech., vol. MTT- 37, no. 3, pp , March [9] E. F. Lauria, A. R. Kerr, M. W. Pospieszalski, S.-K. Pan, J. E. Effland, and A. W. Lichtenberger, "A GHz SIS Mixer-Preamplifier with 8 GHz IF Bandwidth," 001 IEEE International Microwave Symposium Digest, pp , May 001. Available as ALMA Memo 378 at [10] L. R. D'Addario, "Saturation of the SIS mixer by out-of-band signals," IEEE Trans. Microwave Theory Tech., vol. MTT-6, pp , no. 6, June [11] A. R. Kerr, S.-K. Pan, A. W. Lichtenberger and H. H. Huang, "A Tunerless SIS mixer for GHz with low output capacitance and ductance," Proceedgs of the Nth International Symposium on Space Terahertz Technology, pp , March Available as ALMA Memo 05 at