noise sources comparition around the europe - IRA
Contents
1. AROUND THE EUROPE e FINAL REPORT ASTRON N lt 74 Fraunhofer Author Revised by Date Sergio Mariotti A Orfei J D Gallego Oct 9 2010 240 AJ N NO circ avg cor Te circ avg cor Te 220 MEDIAN MAD NO circ MEDIAN MAD YES circ 200 x EN 180 T 160 Y tt 140 Ne A N N v 100 16 18 20 22 University of Cantabria HP346CK01 NO circ avg cor Te 240 circ avg cor Te MEDIAN MAD NO_circ_ MEDIAN MAD YES circ 220 200 180 160 140 120 100 16 18 20 22 24 26 University of Cantabria HP346CK01 10dB Attenuator 10 NOISE SOURCES COMPARISON AROUND THE EUROPE FINAL REPORT Fraunhofer Author Revised by Date Sergio Mariotti A Orfei J D Gallego Oct 9 2010 240 NO circ avg cor Te circ avg cor Te MEDIAN MAD NO circ MEDIAN MAD YES circ 220 200 180 160 140 120 100 16 18 20 22 24 26 University of Manchester SNS N4002A 2NO circ avg cor Te 240 circ avg cor Te MEDIAN MAD _ _ MEDIAN MAD N 2202. 4 A brief description of the setup including the equipment p n S Name and contact of the person that really have made the measurement 6 The tracking number of the shipment 7 Some additional in formations if necessary or needed T H s p FM starts from here Inter laboratory Noise Source Comparition User Manual July 2009 Fig 3 The ferrite circulator in front of the LNA Expected Results Once all partner will have performed the measurements the data will be collected and plotted together on the same graphic The plot will be sent to all partners Postal address Email s mariotti a ira inaf it Mr Sergio Mariotti INAF IRA INAF Ist di Radioastronomia www ira inaf it Radiotelescopio di Medicina v Fiorentina 3508 B Phone 39 051 6399356 39 051 6965829 I 40059 MEDICINA BO ITALY Mobile 39 328 6248631 Fax 39 051 6965810
3. 200 A 160 140 120 100 16 18 20 22 24 26 10 Centro Astronomico de Yebes HP346C 10dB Attenuator 11 NOISE SOURCES COMPARISON AROUND THE EUROPE FINAL REPORT Author Sergio Mariotti Date Oct 9 2010 Revised by A Orfei J D Gallego 240 T 220 200 NO circ avg cor Te YES circ avg Te MEDIAN MAD NO circ MEDIAN MAD YES circ 180 160 140 120 100 16 18 20 22 24 26 11 Centro Astronomico de Yebes NC346KA 10dB Attenuator 240 220 NO circ avg cor Te YES circ avg cor Te MEDIAN MAD NO circ MEDIAN MAD YES circ 200 180 160 140 120 100 16 18 20 22 24 26 12 Centro Astronomico de Yebes SNS N4002A 12 NOISE SOURCES COMPARISON AROUND THE EUROPE m FINAL REPORT ASTRO N 4 Fraunhofer Author Revised by Date C AY Sergio Mariotti A Orfei J D Gallego Oct ot 2010 240 220 200 NO circ avg cor Te circ avg cor Te MEDIAN MAD NO circ MEDIAN MAD YES circ 13 Astron HP346A 240 220 NO circ avg cor Te circ avg cor Te 200 MEDIAN MAD NO circ
4. Fraunhofer IAF SNS N4002A Fraunhofer IAF SNS N4000A 260 240 220 200 Noise Temperature Te K 16 0 16 5 17 0 17 5 18 0 Frequency GHz 4 Noise Temperature of same LNA measured by different labs YES input circulator 16 18 GHz only Even if the different colours may help the trace identification the whole traces are not the best way to compare different measures because is confusing A good way to compare each measurement is a plot of every measurement compared with the mean of the measurements As discussed before the median value rather the average better describe the most probable mean NOISE SOURCES COMPARISON AROUND THE EUROPE FINAL REPORT S ASTRO N Fraunhofer Author Revised by Date Sergio Mariotti A Orfei J D Gallego Oct 9 2010 NO circ avg cor Te 240 circ avg cor Te MEDIAN MAD NO circ MEDIAN MAD YES circ 220 200 4 180 160 140 120 100 16 18 20 22 24 26 5 INAF IRA HP 346C 10 dB attenuator 240 circ avg cor Te circ avg cor Te 220 MEDIAN MAD NO circ MEDIAN MAD YES circ 200 180 160 140 120 100 16 18 20 22 24 26 INAF IRA HP346A NOISE SOURCES COMPARISON INAF MANCHESTER
5. MEDIAN MAD YES circ 180 160 140 120 100 16 18 20 22 24 26 14 Astron SNS N4000A 13 NOISE SOURCES COMPARISON AROUND THE EUROPE 222 FINAL REPORT ASTRO N 2 Fraunhofer Author Revised by Date Sergio Mariotti A Orfei J D Gallego Oct 972010 240 2NO circ avg cor Te L Idm circ avg cor Te 220 i 7 MEDIAN MAD NO circ REA MEDIAN MAD YES circ 200 f s 180 160 140 120 100 16 18 20 22 24 26 15 Fraunhofer SNS N4002A 240 220 200 NO circ avg cor Te YES circ avg cor Te MEDIAN MAD NO circ MEDIAN MAD YES circ 180 160 140 120 100 16 18 20 22 24 26 16 Fraunhofer IAF SNS N4000 14 NOISE SOURCES COMPARISON UC ina e Jr Olfa 1824 AROUND THE EUROPE FINAL REPORT ASTRON lt Fraunhofer Author Revised by Date eT Sergio Mariotti A Orfei J D Gallego Oct 92010 Trace behaviour similarities and noise generators summary In the following table the behaviour of the compared noise generators and the uncertainty labelled by the factory are summarized NOISE Freq Similar Uncertainty HP 346C K01 20 07 2009 16 265 0 12 HP 346C K01
6. correct it by using the algorithm provided ACKNOWLEDGEMENTS Thanks to all very skilled and motivated participants especially to Eduardo Artal and Juan Luis Cano Uni of Cantabria Spain Juan Daniel Gallego CAY Spain Mohamed Missous and Shahzad Arshad Uni of Manchester U K Eric Van Der Waal Astron NL Hermann Massler and Beatriz Aja Fraunhofer IAF Germany Federico Perini INAF IRA Italy 19 4 EI UC ar oia NOISE SOURCES COMPARISON AROUND THE EUROPE asss FINAL REPORT ASTRO N 4 Fraunhofer Author Revised by Date C AY Sergio Mariotti A Orfei J D Gallego Oct 92010 BIBLIOGRAPHY 1 2 3 4 5 6 7 8 9 10 N Kuhn Curing a Subtle but Significant Cause of Noise Figure Error Microwave Journal June 1984 AN 57 2 Noise Figure Measurement Accuracy The Y Factor hp Agilent Technologies year 2001 http contact tm agilent com data static eng tmo Notes interactive an NFU Cal uncert xls W C Daywitt Radiometer equation and analysis of systematic errors for the NIST automated radiometers NIST Technical Note 1327 march 1989 M De Dominicis Strumentazione e Metodologie per la Modellistica di Rumore di Dispositivi Attivi ad Alta Frequenza PhD Thesis Uni Tor Vergata February 2004 in Italian AN 57 1 Fundamentals of RF and Microwave Noise Figure Measurements hp Agilent Technologies July 1983 No
7. limit hv k while the associated uncertainty isn t affected by any reduction It isn t rare a case we quote the noise temperature of an LNA as Te 10 20 The causes that generates and propagates the Uncertainty are well described and discussed in 1 and 2 Here s a summary of that causes and typical values for the travelling LNA described later measured by a typical mod 346C Noise Source 3 Uncertainty related to the LNA Noise Figure strongly dependent on U ENR Uncertainty related to Instrument Noise Figure Uncertainty related to LNA Gain Measurement Uncertainty related to ENR Overall RSS Uncertainty on Noise Figure 1 o 0 25 In this case the NF of the LNA is NF 1 7 dB 0 25 dB or 139 26 K 16 It s evident that the strongest cause of U Te is the U ENR The most obvious questions is how the U ENR can be reduced Before to answer it s necessary give the answer to another question What is the minimum theoretical limit of U ENR The answer may be found in 4 and in 5 and may be summarised as follow ENR is determined by the help of a radiometric chain At the input a relays switch sequentially three arms of the chain the cold standard the hot standard and the unknown noise source U ENR is the resultant propagation of many causes of uncertainty but the most significant are The uncertainty of the temperature of the cold noise source U Tcold The uncertainty of the spar of th
8. over the 18 GHz boundary In this way even the laboratories that are arranged up to 18 GHz may takes advantage of the comparison NOISE SOURCES COMPARISON UC 4 MANCHESTER Fela 1824 AROUND THE EUROPE FINAL REPORT ASTRO N 4 Fraunhofer Author Revised by Date C AY Sergio Mariotti A Orfei J D Gallego Oct 982010 Reflection coefficient and impedance matching The LNA has been impedance matched as well as possible but it cannot be regarded as a well matched device As higher is the product of the reflection coefficient pNS pLNA as higher is the ripple of the Te curve over freguency Two measurements are reguested to the laboratories one placing a selected low VSWR ferrite circulator in front of the LNA and the other one without that circulator All the measurements made with the circulator will be compared all together as well as the measurements performed without the circulator Comparing the two obtained measurement traces it s possible distinguish if the ripple is mainly due to mismatch errors or due to ENR differences Drift Since a drift on gain and noise temperature may occur during the measurement session a measurement of the thru is requested both at the beginning and at the ending of the measurement session If an important drift will be seen the Te data file will be corrected Room temperature The handling of the Noise Source and LNA as well as the solar light c
9. two reasons e A metrological mistake The calibration of a Noise Source by the knowledge of Te of LNA is same as the calibration of a pocket rule by comparing to the length of a table desk while the right way is the contraire e A mathematical mistake It isn t a rigorous method it s just s an approximation The reason is because for a rigorous calculation it should be necessary to know and take into account for the noise temperature of the Noise Figure Meter the Calibration data while those data are unknown However since the LNA gain is high enough the Noise Temperature of the Noise Figure Meter is strongly masked so accepting a small error the Noise contribution of the noise figure analyzer may be neglected In any case even if it s an approximation the method is converging so the error will be reduced and there is no risk of overcorrection But if the difference of ENR is suspected to be very large a corrective action is perhaps better than do nothing So if the preceding simplification is accepted the actual measured Noise Temperature is Th Y Tc 7 1 The similar equation form may be written for the value we would read the corrected Noise Temperature 7 T h Y Tc Y Te T e Q Now let perform the ratio T e Th Y Tc Te T h Y Tc 3 Solving for the corrected hot temperature T h Th Y Tc Th Y Tc 4 17 NOISE SOURCES COMPARISON UC MANC
10. 10dB 20 07 2009 16 265 m 012 INAF IRA HP 346C 10 dB 03 09 2009 16 265 m cm 0 06 INAF IRA HP 346A 17 08 2009 16 18 4 042 SNS N4002 30 10 2009 16 265 o 0 06 HP 346C 10dB 30 11 2009 16 26 5 m cm 012 Yebes NC346KA 10dB 30 11 2009 16265 70102 Yebes SNS N4002 30 11 2009 16 25 o 006 Astron 4 20 042010 16 188 4 02 Astron SNSNA4000 20 04 2010 16 18 06 Fraunhofer IAF SNS N4000 25 05 2010 16 18 9 006 Fraunhofer IAF SNSN4002 25 052010 16265 o 440 06 Where 0 K01 stands for 1GHz 50GHz 2 4mm coax connector Uncertainty reduced by in house calibration 9 KA stands for 10MHz 40GHz 2 92mm coax connector the behaviour of the traces is similar among themselves the behaviour ofthe traces is similar among themselves CM the trace is closer to median value It has been looked for the most similar curves In the following two graphics are plotted le closest curves because the number of similar families of curves is two 15 j MANCHEST fel EX AROUND THE EUROPE FINAL REPORT NOISE SOURCES COMPARISON Revised by Date A Orfei J D Gallego S ASTRON 22 Fraunhofer n CAY mm w Sergio Mariotti Oct 9 2010 As it may be seen in the following charts the measurements performed by laboratories that involves the mo
11. For all data set both the Noise and Gain drifts were negligible Since the data set have been collected at different physical temperatures a correction has been applied in order to normalize the actual Noise Temperature The experimental found law alk 0 5 K C has been applied ATamb Two families of Cartesian graphic have been plotted the case without input isolator NO CIRCULATOR and the case with the isolator in front of the LNA YES CIRCULATOR As it may be seen the spread is very large Also it s difficult to understand the whole traces A selective analysis is much more indicative Three kind of analysis has been performed e The effect of the mismatch loss is analyzed and discussed e the actual Noise Temperature trace is compared to the mean value calculated at every frequency along the points coming from each Noise Generator e Traces performing similar behaviour has been looked for grouped analyzed and discussed What kind of mean The most known method to estimate the mean value of a number of samples is the calculation of the arithmetic average accomplished by the standard deviation An alternative method widely used at NIST is the computation of the median accomplished by the MAD 11 For uniformly and normal distributed samples the average and median are coincident as well as the standard deviation and the MAD But in the case that few samples fall far outside the distribution law the median bett
12. HESTER Fela 1824 AROUND THE EUROPE FINAL REPORT ASTRO N 4 Fraunhofer Author Revised by Date C AY Sergio Mariotti A Orfei J D Gallego Oct 982010 Finally the new corrected ENR value is 5 Or if the operator prefers the advantages of the small correction for 7 To even if formally different by the ENR definition we get T h Tc ENR 10 log 290 5 bis Where Te nput Noise Temperature of the LNA the actual Noise Temperature Th Hot Noise Temperature generated by the Noise Generator the actual value To 290K The standard temperature Y The Y factor T e Input Noise Temperature of LNA the wanted value T h Hot Noise Temperature generated by the Noise Generator the wanted value ENR The Excess Noise Ratio the actual value ENR The Excess Noise Ratio the wanted value ND Note that the variables Te and should not only be considered alone but also their ratio e has a physical meaning The knowledge of the variable Y is reguired but probably was not recorded at the time of measurement It doesn t matter because the Y factor may be calculated offline by solving the 1 for Y and combining the 5 or the 5bis ENR Th To To 10 9 6 ENR _ Te To To 10 Y 7 In order to help users who want correct their ENR table a MS Excel file is provided The user s
13. ISTITUTO NAZIONALE DI ASTROFISICA ITUTO DI RADIOASTRONOMIA v Gobetti 101 40129 BOLOGNA ITALY INAF RAPPORTO TECNICO W IST Noise source calibration with a travelling amplifier Background The overall uncertainty in Noise Figure measurements is dominated by the uncertainty in the amount of noise generated by the noise source Excess Noise Ratio ENR and uncertainty of the ENR For more than 30 years the most accurate noise generators in the market had an uncertainty 0 12 dB Leading brands were Noise Com HP Agilent Micronetics In 2003 work between INAF IRA and Uni Rome Tor Vergata UTV reduced the uncertainty to 0 06 dB at frequencies up to 26 5 GHz A theoretical study between INAF IRA and UTV showed that the uncertainty cannot be less than 0 055 dB Note that an uncertainty of the ENR of 0 055dB becomes an uncertainty of 0 09 dB on the noise figure i e 7 at ambient temperature In 2004 Agilent offered a very accurate noise generator 0 06 dB up to 18 GHz Comparison of the most accurate noise sources around the Europe the comparison of the most accurate noise generators cannot further improve the intrinsic accuracy But a comparison campaign will reveal if each laboratory is inside or outside the uncertainty boundaries Measurements closer to the overall mean build trust for that particular noise generator and test system Conversely measurements far from the mean indicate that correc
14. Yebes SNS N4002 ASTRON 346A ASTRON SNS N4000A Fraunhofer IAF SNS N4002A Fraunhofer IAF SNS N4000A 2 Noise Temperature of same LNA measured by different labs NO_input_circulator YES_input_circulator Noise Temp of the same LNA measured by different labs 280 260 240 220 200 180 Noise Temperature Te K 160 140 Frequency GHz 346C 10dB IRA 346A U Cantabria 346CK01 U Cantabria 346CK01 10dB U Manchester N4002A Yebes 346C 10dB Yebes 346KA 10dB Yebes SNS N4002 ASTRON 346A ASTRON SNS 4000 Fraunhofer IAF SNS N4002A Fraunhofer IAF SNS N4000A 3 Noise Temperature of same LNA measured by different labs YES input circulator NOISE SOURCES COMPARISON AROUND THE EUROPE FINAL REPORT 4 O MANCHESTER S ASTRON Fraunhofer C AY rm IAF Author Revised by Date Sergio Mariotti A Orfei J D Gallego Oct 9 2010 Noise Temp of the same LNA measured by different labs YES input circulator 16 18 GHz only 280 IRA 346C 10dB IRA 346A U Cantabria 346CK01 U Cantabria 346CK01 10dB U Manchester N4002A Yebes 346C 10dB Yebes 346KA 10dB Yebes SNS N4002 ASTRON 346A ASTRON SNS N4000A
15. ached the value of 0 055 dB cannot be considered a small value Note that realistically a U ENR 0 055 dB once propagated produces a U NF 0 09 dB 7 K at 1 o or 14K at 2 o 3 What action can be taken in order to further improve the uncertainty As suggested by 8 and 9 the only possible action is a comparison The inter laboratory comparison by itself doesn t further reduces the uncertainty but it enhances the differences and should suggest some corrective actions especially for the values farther from the mean NOISE SOURCES COMPARISON AROUND THE EUROPE Uc O MANCHESTER FINAL REPORT ASTRO N Fraunhofer Author Revised by Date C AY Sergio Mariotti A Orfei J D Gallego Oct 982010 COMPARISON The inter laboratory comparison is not a re calibration By itself doesn t change the ENR of the compared Noise Sources The differences in readings occurring at each laboratory will not be cleared Only once this document has been read each laboratory will have the freedom to apply corrections to ENR In any case each correction should be applied with great precaution and cautiousness The canonical way to do a comparison is a comparison of a travelling standard An Input Noise Temperature Standard an LNA with a standard Noise Temperature doesn t exist Te of LNA is every time dependent on ambient temperature bias supply and even on the reflection coefficient o
16. d NC346C 10 dB attenuator are closer among themselves Also measurements performed by laboratories that involves the SNS N4002 are among themselves But the two families are different among themselves It may be suspected that the differences may be imputed to differences in the original calibration procedures at Agilent factory 250 IRA 3460 10dB 240 Uni Cantabria 346CK01 10dB 230 Yebes_346C 10dB 220 AVG selected over 346C s 210 200 si n 00r r 190 180 170 160 150 16 18 20 22 24 26 17 Measurements performed with Noise Generators mod 346C 10 dB Attenuator by different labs 250 U Manchester SNS N4002A Yebes SNS N4002A Fraunhofer IAF SNS N4002 240 230 KS AVG selected over SNS N4002A s 16 18 20 22 24 26 18 Measurements performed with Noise Generators SNS N4002A by different labs 16 NOISE SOURCES COMPARISON AROUND THE EUROPE UC SKS lt FINAL REPORT ASTRO N 4 Fraunhofer Author Revised by Date C AY Sergio Mariotti A Orfei J D Gallego Oct 92010 ENR TABLE CORRECTION For laboratories like to correct their ENR table in order to reduce the distance from the average or median a method is described It s important to write in an explicit way that this operation is conceptually a mistake for
17. e Source Comparition O fe User Manual July 2009 10 11 12 13 14 15 16 17 Fig 1 Screwing the coax connector Fig 2 The thermometer sensor is installed If You are using hp8970 or hp8971 or Maury Eaton Read the temperature on the digital thermometer convert in K if necessary put it into the for hp8970 type 6 0 SF and writes to me by email If You are using the new Agilent NFA read both the digital thermometer and the actual Cold Temperature of NFA and writes to me by email both Perform the measurement without ferrite circulator at the input of LNA Record the file as no circulator Disconnect the Noise Source insert the ferrite circulator s n 004 see Fig 3 wait 5 minutes check once again the cold temperature and if necessary set once again into NFA Perform the measurement with ferrite circulator at the input of LNA Record the file as circulator disconnect the LNA and connect the Noise Source to NFM Wait 5 minutes and perform the 4 and last measurement This is a OK check in order to know the drift during the procedure Record the fileas final The measurement is finished Put the LNA ferrite circulators and all the materials into the cardboard box and ship back Send anemail to s mariotti ira inaf it containing the following 1 The four text files 2 The value of Cold Temperature 3 Type of conversion SSB or DSB
18. e arms of radiometric chain used for comparison U sij uc nar eife Kn NOISE SOURCES COMPARISON AROUND THE EUROPE m FINAL REPORT ASTRO N lt Fraunhofer Author Revised by Date SUME Sergio Mariotti A Orfei J D Gallego Oct 9 2010 Given the most accurate data values obtainable from a commercial Network Analyzer and a Standard Cold Load both operated by a very skilled operator a Noise Generator may be calibrated with U ENR 0 055 dB 1 o at 18 GHz For more than 20 years the most accurate Noise Generators sold by hp Agilent the mod HP346 had U ENR quoted 0 15 dB 1 o In the hp Agilent documents for the mod HP346 has been wrote that the vector method to reduce the U ENR was known but not applied 6 the uncertainty was quoted at 1 o instead the more common adopted value 2 o 7 In Year 2003 a join workgroup of INAF Institute of Radioastronomy and University of Tor Vergata has lowered the uncertainty of their own Noise Generators by application of the most accurate vector method known 5 The estimated uncertainty reached was U ENR 0 055 dB 1 o at 18 GHz 1 In house Noise Source calibration with secondary standard In year 2004 Agilent started to sell a more accurate Noise Generator the model series SNS N400x the related Uncertainty was quoted U ENR 0 06dB 16 Despite of the uncertainty has been strongly reduced and the theoretical limit has been almost re
19. er filters away the farther samples Even the MAD rather than the standard deviation better filters the rare and farther samples So the median method is more robust than the average one Uc O Ira MANCHESTER DE CANTAS NOISE SOURCES COMPARISON AROUND THE EUROPE FINAL REPORT S ASTRON Fraunhofer CAY tar Sergio Mariotti Revised by A Orfei J D Gallego Date Oct 9 2010 RESULTS Impedance Mismatch effects A mismatch error should generates a ripple on the frequency domain representation of the trace shows more ripple than the trace The mismatch error is as large as the trace NO CIRCULATOR YES CIRCULATOR The mismatch effect has been summarized on the next table NOISE Mismatch error Freq ORGANISATION GENERATOR difference on range the ripple 4 292 Uc ffe MANCHESTER NOISE SOURCES COMPARISON AROUND THE EUROPE FINAL REPORT e ASTRON 22 Fraunhofer n C AY IAF Sergio Mariotti Revised by Date A Orfei J D Gallego Oct 982010 Charts Noise Temp of the same LNA measured by different labs NO_input_circulator 240 IRA 346C 10dB IRA 346A 220 200 180 160 140 Noise Temperature Te K 120 100 Frequency GHz U Cantabria 346CK01 U Cantabria 346CK01 10dB U Manchester N4002A Yebes 346C 10dB Yebes 346KA 10dB
20. f the source But if the travelling LNA will be measured at known ambient temperature biased by a very stable power supply and close on well matched load its Te can be considerate stable enough around the travel The comparison is more accurate as smaller is Te of the LNA So a cryogenic comparison in principle should be preferred to a room temperature comparison 10 But practically speaking a balancing of the Pros amp Cons must be done Room temperature LNA comparison Man power required week s hours Cryo LNA comparison Accuracy best high Sensitivity to how the setup is arranged more less Sensitivity to U ENR less more Easy to do less more Probability to find candidate laboratories less more Due to that reasons a room temperature comparison has been preferred Frequencies Often the frequency boundaries comes at 18 GHz 26 5 GHz 40 GHz 50 GHz 67 GHz The best seller Noise Generators covers the very popular bandwidth 10 MHz 18 GHz Of course higher frequencies are also very important for radio astronomy It has been chosen a frequency range 16 26 5 GHz for the following reasons e The high end of the bandwidth say 18 or 26 5GHz is probably affected by greater uncertainty or errors than the low end of the bandwidth A comparison is probably more meaningful at the high end of the bandwidth e 16 26 5GHz has been chosen because step
21. hould input frequency by frequency the actual ENR the actual Te and the spreadsheet will calculate the ENR corrected that produce the median Te 18 uc nar eife Kn NOISE SOURCES COMPARISON AROUND THE EUROPE FINAL REPORT AST 3ON Z Fraunhofer Author Revised by Date SAt Sergio Mariotti A Orfei J D Gallego Oct 92010 CONCLUSIONS The Noise Temperature measurement of a travelling LNA has shown strong differences among laboratories Assuming as the major cause of uncertainty on Noise Temperature the uncertainty of ENR the spread of the data is much larger than the expected Many of the Noise Sources were lowest uncertainty type so there is no room for improvements involving the factory The only way to further reduce the uncertainty of the ENR is the calibration with a cryo load that acts as a primary or secondary metrological standard This is especially thru for the millimetre wave frequencies where the uncertainty is greater Because the cryo loads are very expensive 50000 100000 a join or consortium is suggested for the purchase as well as a sharing in the use Perhaps the best way would be a common European Centre for Noise Source Calibration that involve metrological standard cold loads as shown in the following pictures Noise Com Cold Load B Maury Microwave Cold Load In any case each laboratories that believe its ENR is wrong has the freedom to
22. ise Sources 10MHz to 26 5GHz Hewlett Packard Technical Data Jan 1989 http www arftg org s parameter meas html J Randa and others International Comparison of Thermal Noise Temperature Measurements at 2 4 and 12 GHz e IEEE T I M APRIL 1999 J D Gallego J L Cano Estimation of Uncertainty in Noise Measurements Using Monte Carlo Analysis Radionet FP7 Workshop Meeting June 2009 http www radionet eu org fp7wiki lib exe fetch php media na engineering ew gallego final pdf 11 http en wikipedia org wiki Median absolute deviation 20 INAF Inter laboratory Noise Source Comparition User Manual July 2009 Inside the cardboard box Quantity Description 1 Power supply 1 LNA 2 Ka Band ferrite circulators with 50 Ohm load 1 Digital thermometer 1 paper tape 1 This manual Please take care the cardboard box it will be used for ship back to INAF IRA Frequencies The optimal frequency range is 16 00 GHz 26 50 GHz The steps must be 50 or 100 MHz wide and includes the integers of GHz i e 16 00 16 05 16 10 26 50 Please do not acquire strange frequency numbers like 16 000 16 0525 16 105 If Your own equipments do not allow the 16 00 26 50 GHz frequency range You may perform the measurement in a limited portion of the bandwidth 1 e 16 00 18 00 GHz or 18 00 24 00GHz Operations The operations are very
23. oming trough a window as well as the vicinity of an heater or air conditioning changes the physical temperature of the LNA and consequently it s Te For this reason a proper shield form heat cold sources is required as well a period of time long enough to reach the thermal equilibrium 5 minutes Aging check At INAF IRA the LNA has been measured both at the beginning and one year later at the end of the campaign Differences on Te was smaller than 2 K except for frequencies greater than 26 3 GHz where the difference was slowly increasing with the frequency up to 10K NOISE SOURCES COMPARISON UC ra ms Fela 1824 AROUND THE EUROPE FINAL REPORT ASTRO N 4 Fraunhofer Author Revised by Date C AY Sergio Mariotti A Orfei J D Gallego Oct 982010 DATA PROCESSING The data files coming from each one laboratory have been analyzed Not all data set are organized on the reguested way Some data are 50 MHz spaced while other are 100 MHz spaced One data set is 500 MHz spaced This difference has involved a small complication on the processing but does not affects the comparison Most data set contains Noise Temperature and Insertion Gain while two data set contains NF dB only This small lack does not affect the comparison Some data set do not include the initial and final zero traces It has been assumed that the drift was negligible Where possible the drift has been evaluated
24. simple and intuitive But the quality of the result will depend even on the skill of the operator The measurement require only one person One skilled person is better than two or more non skilled people The estimated time needed for the measurement is 1 3 hours additionally up to 3 hours are required for the logistics There is no need to known the principle of operation and the motivations but if You are interested You may download the presentation on http www radionet eu org fp7wiki doku php id na engineering ew lstew then right click on Practical aspects on noise figure measurements and Save destination 4 Inter laboratory Noise Source Comparition Ie User Manual July 2009 The procedure l Turn on the Noise Figure Meter NFM Set LO power if required and the central frequency i e 22 GHz Connect the Noise Source and let it chopping switching ON OFF quickly Open the cardboard box put the power supply LNA thermometer and ferrite circulators on the table Connect LNA to Power Supply then Power Supply to the mains The mains must be grounded properly Wait the warm up time typically 30 min for hp amp 9704 B mode 1 1 Eaton Maury and Agilent NFA If You are using hp8971B mode 1 5 at least 2 hours of warm up are needed The temperature of environment and the room should be as constant as possible keep away from direct sunlight far from air conditioner air flo
25. tive action such re calibration may be necessary Organisation of the job each contributing Institute Laboratory have performed the measurement of a travelling LNA using their own noise source and related instruments The measurement has being performed in the most accurate and precise way at room temperature for simplicity in this phase of the test programme Each Institute has sent the data file to INAF IRA for the comparison The project is started in July 2009 and finished in September 2010 PARTNERS Contact Name INAF IRA Italy S Mariotti F Perini CAY Spain J D Gallego Uni Canarbia Spain E Artal J L Cano Uni Manchester U K M Missous S Arshad Astron Netherlands E Wan Der Waal Fraunhofer IAF Germany H Massler B Aja others labs in the future NOISE SOURCES COMPARISON UC 1 MANCHESTER Fela 1824 AROUND THE EUROPE FINAL REPORT ASTRON 4 Fraunhofer Author Revised by Date SUME Sergio Mariotti A Orfei J D Gallego Oct 92010 NOISE SOURCES COMPARISON AROUND THE EUROPE FINAL REPORT INTRODUCTION It s well known that the correct way to furnish an experimental measure is to provide both the observed value and its associated uncertainty For the low noise amplifiers the improvement of the technology and the cryogenic cooling has produced a decreasing of the input Noise Temperature down to few times the theoretical quantum
26. w etc etc Also the LNA should be far from the heated air flow coming out from the and or other equipments Visual check of all coaxial connectors They must be clean free or dust burrs also check visually the integrity of coaxial cables Pay attention when connect the circulators they are expensive devices Be gently with the connectors If a circulator drops into the floor or has been mechanically shocked please write it into the email report Calibrate the NFM Use the ferrite circulator S N 006 as front end of Your NFM The calibration plane is now at the input of the ferrite circulator see Fig 1 Set an adequate averaging not too long time please Wait at least 5 minutes and check the calibration Redo the calibration once again if far from 0 K Record fileas initial Insert the LNA by gently screwing the K connectors as usual see Fig 1 Put the thermometer sensor as close as possible to the end part of the Noise Source body OUT connector side Use the paper tape to firmly connect the sensor see Fig 2 Paper tape is more thermal isolator than other tapes also it can be peeled away quickly note even in the case You are using an interseries coaxial transition i e 7mm to SMA or N to SMA the thermometer sensor should be placed closer as possible to the end part of the Noise Source body Wait 5 minutes for the thermalizzation Your finger has been heated the LNA 1 Inter laboratory Nois
Download Pdf Manuals
Related Search