Noise Figure Measurements
Feb 24, 2009 Account Manager : Peter Caputo Presented by : Ernie Jackson Agenda • Overview of Noise Figure • Noise Figure Measurement Techniques • Accuracy Limitations • PNA-X’s Unique Approach
Gain
So/N o DUT
Si/N i
Noise Figure Basics Page 2 Noise Contributors
Thermal Noise: (otherwise known as Johnson noise) is the kinetic energy of a body of particles as a result of its finite temperature
Ptherm =kTB
Shot Noise: caused by the quantized and random nature of current flow
Flicker Noise: (or 1/f noise) is a low frequency phenomenon where the noise power follows a 1/f ααα characteristic
Noise Figure Basics Page 3 Why do we measure Noise Figure? Example...
Transmitter: ERP = ERP + 55 dBm +55 dBm Path Losses -200 dB Rx Ant. Gain 60 dB dB Power to Rx -85 dBm -200 ses: Los Receiver: Path
Noise Floor@290K -174 dBm/Hz C/N= 4 dB Noise in 100 MHz BW +80 dB Receiver NF +5 dB Rx Sensitivity -89 dBm
Choices to increase Margin by 3dB Receiver NF: 5dB 1. Double transmitter power Bandwidth: 100MHz Power to Antenna: +40dBm 2. Increase gain of antennas by 3dB Antenna Gain: +60dB Frequency: 12GHz 3. Lower the receiver noise figure by 3dB Antenna Gain: +15dB
Noise Figure Basics Page 4 The Definition of Noise Factor or Noise Figure in Linear Terms
• Noise Factor is a figure of merit that relates the Signal to Noise ratio of the output to the Signal to Noise ratio of the input. • Most basic definition was defined by Friis in the 1940’s. S S S N N in out N F = in S G , Na N out
Reference: Harald Trap Friis: 1944: Proceedings of the IRE
Noise Figure Basics Page 5 What is Noise Figure ?
Noise Out Noise in
Measurement bandwidth=25MHz a) C/N at amplifier input b) C/N at amplifier output Nin Nout
Noise Figure Basics Page 6 Definition of Noise Figure
S S Nout N in N out Noise N a Added
G , Na GNin
Sout f1 f2 Gain = G = Nout = Na + GNin Sin N + GN S Nout a in F = = Noise Factor N GNin GNin F = in S N + GN NF (dB) =10log a in N Noise Figure out GNin
Noise Figure Basics Page 7 Noise Voltage
Standard Equation for Noise Voltage produced by a Resistor
e2 = 4kTBR
k = Boltzman's Constant = .1 38×10−23 Joules/°K is T is absolute temperature ( °K) is B is bandwidth (Hz) e is rms voltage
Reference: JB Johnson/Nyquist: 1928: Bell Labs
Noise Figure Basics Page 8 Noise Power at Standard Temperature
Thermal Load k = 1.38 x 10 23 joule / k R - j X T = Temperature (K) R+jX L L B = Bandwidth (Hz)
Available Noise Power Delivered to a Conjugate Load,
Pav = kTB
-21 At 290K P av = 4 x 10 W/Hz = -174dBm / Hz
In deep space kT = -198dBm/Hz
Noise Figure Basics Page 9 Noise Power is Linear with Temperature
Nin Nout = Na + GNin
= Na + GkTs B G , Na Z s @ Ts
Nout Slope = kGB N2
N1
Na Output Noise Noise Output Power Ts T1 T2 Temperature of Source Impedance
Noise Figure Basics Page 10 An Alternative Way to Describe Noise Figure:
Effective Input Noise Temperature ( T e)
N = N + GN ∑∑∑ out a in = GkTe B + GkTs B G , Na = 0 = GkB(Te +Ts ) Z s @ Ts Z s @ Te N out Slope = kGB
T = (F −1)T e s N
and a Output Noise Noise Output Power
Ts Te +Ts TemperatureT of Source F = e Impedance Ts Noise Figure Basics Page 11 Te or NF: which should I use?
••UseUse either either - - they they are are completely completely interchangeable interchangeable ••TypicallyTypically NF NF for for terrestrial terrestrial and and Te Te for for space space ••NFNF referenced referenced to to 290K 290K - - not not appropriate appropriate in in space space ••IfIf Te Te used used in in terrestrial terrestrial systems, systems, the the temperatures temperatures cancan be be large large (10dB=2610K) (10dB=2610K) ••TeTe is is easier easier to to characterize characterize graphically graphically
Noise Figure Basics Page 12 Friis or Cascade Equation
• Here we see what contribution a second amplifier has on the overall noise factor.
N = N + G N + G G N Nin out a2 1 a1 1 2 in
G = G1G2 G , N G , N 1 a1 2 a2 Nin = kTs B Z s @ Ts Na1 = kTe1B = (F1 −1)kTs B
Na2 Na2 = kTe2 B = (F2 −1)kTs B N F = out
Na1G2 GNin Na1
kT B kT BG kT BG G F2 −1 Nin = s s 1 s 1 2 F12 = F1 + G1
Noise Figure Basics Page 13 Effects of Multiple Stages on Noise Factor
• The cascade equation can be generalized for multiple stages
Nin
G1 , Na1 G2 , Na2 GN , N N 2 Z s @ Ts F −1 F −1 F = F + 2 +L+ N total 1 K G1 G1G2 GN −1 N F −1 F = F + i total 1 ∑ N i=2 G ∏ j−1 j=i
Noise Figure Basics Page 14 Demo of Cascade Equation F2 −1 F = F1 + G1
ABC
Gain (dB) 6.0 12.0 20.0 Gain 4.0 16.0 100.0 Noise Figure(dB) 2.3 3.0 6.0 Noise Factor 1.7 2.0 4.0
0.2 −1 0.4 −1 F = 7.1 + + = .1 70 + .0 25 + .0 047 = .1 997 ABC 4.0 4.0×16.0 NFABC =10log( .1 997) = .3 00 dB 0.4 −1 0.2 −1 F = 7.1 + + = .1 70 + .0 75 + .0 025 = .2 475 ACB 4.0 4.0×100.0 NFACB =10log( .2 475) = .3 93 dB
Noise Figure Basics Page 15 Agenda • Overview of Noise Figure • Noise Figure Measurement Techniques • Accuracy Limitations • PNA-X’s Unique Approach
Gain
So/N o DUT
Si/N i
Noise Figure Basics Page 16 Measuring Noise Figure
Na + Nin .Ga F = Nin . Ga Nin Nout = Na + kTB Ga
Rs Ga , Na
r e
w Slope=kBGa
o
P
t
u tic p ris t te
u c To solve for Na: ara To solve for Na: O h C 1) establish 2 points on curve ee 1) establish 2 points on curve Fr ise oror No 2) establish 1 point and gradient2) establish 1 point and gradient Na
Te Te Source Temperature (K)
Noise Figure Basics Page 17 Hot/Cold Technique
Nin = kB(Th or Tc) Nout = Nh or Nc
Rs
r
e w
o Nh
P
t
u Slope=kBGa
p
t
u O •Physically hot/ cold source•Physically hot/ cold source Nc •avalanche diode•avalanche diode
Na
Te TcSource Temp (K) Th
Noise Figure Basics Page 18
)
o
N (
No = kGB • Ts + Na
r e
Measured w
o Nh P
Slope=kGB Nh t
Y = u p
Nc t u O Nc
Na Source Temp (Ts) Te Tc Th Input (Th − To ) ENR = To Calculated ENR Na Na = kBGTo −1 Te = Y −1 kGB Na + kGBTo ENR Te + To F = = = Te = To(F −1) kGBTo Y −1 To
Noise Figure Basics Page 19 Noise Source - the Avalanche Diode
Matching Pad
Bias Noise Input Output
Excess Noise Ratio, ENR (dB) = 10 Log ( T - 290) 10 h Frequency ENR dB
290 XXX AAA YYY BBB ZZZ CCC . . . .
Noise Figure Basics Page 20 Noise Figure Analyzer
Nout_cold Nh Nin_cold Source variable DUT impedance attenuator Nout_hot Nc Nin_hot
Nh kGB(Te + Th) Th Y Tc Y= ___ = ______Te = ______ (solve for Te) Nc kGB(Te + Tc) Y 1 Te + To F = ______To
Noise Figure Basics Page 21 Calibrating out System NF
Source variable DUT impedance attenuator
F1, Ga1 F2
F12
F2-1 F1 = F 12 - Ga1
F2-1 if F 12 ≈≈≈ then uncertainty increases, so beware if F 1 and G 1 are low Ga1
Noise Figure Basics Page 22 How does the NFA measure gain
Slope=kBG 1G2 Nh_meas
t r en e em w ur o as Slope=kBG 2
P Nh_cal e
t m u UT p nt t D me u e Nc_meas sur O ea n m atio Nc_cal libr Ca
TcSource Temp (K) Th
Noise Figure Basics Page 23 Avoidable Measurement Errors: EM Susceptance
Lights (Esp. Fluorescent)
Noise Source
FM TV * Double shielded Cables for IF Computers (Ordinary Braid if porous) * Shielded HP IB Cables * Enclose All Circuits * Jiggle Connectors (Esp. BNC) * Try snap on ferrite cores on cables, dampen common mode currents * LO Contributes Spurious and Noise
Noise Figure Basics Page 24 Avoidable Measurement Errors: Choose the Appropriate Noise Source
•15dB ENR noise source to measure NF of up •15dB ENR noise source to measure NF of up to 30dB ( eg N4001A, N4002A, 346B/C) to 30dB ( eg N4001A, N4002A, 346B/C) •Use 6 dB ENR for very low noise figure •Use 6 dB ENR for very low noise figure devices to keep noise detector linearity issues devices to keep noise detector linearity issues to a minimum ( eg N4000A, 346A).to a minimum ( eg N4000A, 346A).
•Use a Noise Source with greater internal •Use a Noise Source with greater internal attenuation if the DUT is match sensitive (eg attenuation if the DUT is match sensitive (eg N4000A, 346A).N4000A, 346A).
Noise Figure Basics Page 25 Avoidable Measurement Errors: Accurate Loss Compensation
Cable C1 Noise Coax/WG adapter Source G 1
DUT Cable C2 G 2 Measurement system Coax/WG adapter
Noise Figure Basics Page 26 Avoidable Measurement Errors: Loss Compensation and Calibration
Noise Figure Basics Page 27 Noise Source - the Avalanche Diode
Matching Pad
Bias Noise Input Output
Excess Noise Ratio, ENR (dB) = 10 Log ( T - 290) 10 h Frequency ENR dB
290 XXX AAA YYY BBB ZZZ CCC . . . .
Noise Figure Basics Page 28 AvoidableAvoidable Measurement Measurement Errors: Errors:
TemperatureTemperature Considerations Considerations (Diode) (Diode)
r e
To=296.5 K w
Th=10,000 or 1,500 o Nh
P
t
u
p
t
u O Nc
Na
Te ToTc Th Th’
Source Temp (K) Report Ambient Temperature to Analyzer for Accurate ENR Interpre tations
Noise Figure Basics Page 29 Noise Figure Basics Page 30 Avoidable Measurement Errors: Loss and Temperature Corrections
• A temperature must be entered in Loss Compensation Setup for valid results NFA & PSA Default = 0 degrees K • The NFA and PSA accept S2P files from a Network Analyzer for the Loss Compensation table
Noise Figure Basics Page 31 Avoidable Measurement Errors: Account for Frequency Conversion (DSB Measurement) Mixer
Noise fMeasure f IF Noise Source Meter
f NP NP LO fLO External LO
f IF f IF f IF Freq Freq DSB is recommended when : • The application is DSB • The DUT is broadband • No SSB filter is available • The frequency range make SSB filters impossible/impractical Loss compensation of 3dB (under perfect conditions) needs to be entered to account for doubling of power during measurement
Noise Figure Basics Page 32 Avoidable Measurement Errors: Account for Frequency Conversion (SSB Measurement)
Mixer
Noise SSB fMeasure f IF Noise Source Filter Meter
fLO NP NP External LO fLO
f IF f IF f IF Freq Freq SSB is recommended when : • The application is SSB • The DUT is narrowband If down conversion is part of the measurement system, filters should be included in calibration path and measurement path because the calibration and measurement are performed at the same frequencies.
Noise Figure Basics Page 33 Reduce Measurement System Noise Figure Using a pre-amplifier to improve accuracy
preamplifier Mixer conv loss 9dB Source DUT impedance
Analyzer NF 6dB
variable attenuator DUT G=10dB, NF=1.0dB F12_wanted=4dB
Noise Figure Basics Page 34 MinimiseMinimise ErrorsErrors whichwhich cannotcannot bebe eliminated:eliminated: ReduceReduce MeasurementMeasurement SystemSystem NoiseNoise FigureFigure
F2 −1 • Most important of all F = F1 + G uncertainties 1 • if DUT has low gain/ low NF - F2-1 watch out! F = F - 1 12 G • NF2 ≤≤≤ (NF1 + G1) - 5dB for a1 F2 current -1 accuracy F2 reduced = F preamp + Gpream • work out new system NF with p F2 current -1 preamp G = preamp F2 -F • work out preamp gain reduced preamp • avoid overload
Pin = -174dBm/Hz + ENR(dB) + 10log(BW) + NF1 +G1(dB) + Gpre(dB)
Noise Figure Basics Page 35 ReduceReduce MeasurementMeasurement SystemSystem NoiseNoise FigureFigure (cont)(cont) ErrorError inin 2nd2nd StageStage CorrectionCorrection Uncertainty improvement by adding preamplifier 3. 0 2. ♦♦♦ 5 2. 0 GAIN = 0 dB 5 dB 10 dB 1. 0. 9. 15 8. dB 7 . 6 . (dB) 5 20 . dB 4
RSS UNCERTAINTY . 3 ♦♦♦ . 2
. 1 0. 5. 10. 15. 20. 0 0 0 0 0 SYSTEM NOISE FIGURE, F2 (dB)
Noise Figure Basics Page 36 AdditionalAdditional RecommendationsRecommendations
•• MinimizeMinimize the the noise noise figure figure of of the the test test system system (especially(especially when when measuring measuring low low gain gain DUTs).DUTs). •• ReduceReduce the the magnitude magnitude of of all all mismatches mismatches by by usingusing isolators isolators or or pads pads •• MinimizeMinimize the the number number of of adapters, adapters, and and take take carecare of of them them •• CalibrateCalibrate Noise Noise Source Source ENR ENR values values regularlyregularly and and use use good good pedigree pedigree calibration calibration •• UseUse Averaging Averaging to to Avoid Avoid Display Display Jitter Jitter •• ChooseChoose the the Appropriate Appropriate Bandwidth Bandwidth •• AvoidAvoid DUT DUT non-linearities non-linearities
Noise Figure Basics Page 37 Agenda • Overview of Noise Figure • Noise Figure Measurement Techniques • Accuracy Limitations • PNA-X’s Unique Approach
Gain
So/N o DUT
Si/N i
Noise Figure Basics Page 38 Calculating Unavoidable Uncertainty Available at : www.agilent.com/find/nfu
Noise Figure Basics Page 39 Measurement Uncertainty Case Studies Noise Figure Uncertainty The uncertainty of the overall noise system is related to the noise source (match and ENR uncertainty), device under test (noise figure, gain , i/p and o/p match) and instrument parameters (noise figure uncertainty, gain uncertainty, level of noise figure and match). See http://www.agilent.com/find/nfu
Example 1. @ 1GHz DUT Noise Figure = 1dB DUT Gain = 20dB DUT i/p match = 1.5 DUT o/p match = 1.5 PSA = 0.167 dB 8970B = 0.19 dB; N8973A = 0.167 dB 12% less RSS noise figure uncertainty with NFA or PSA
Example 2. @ 6GHz DUT Noise Figure = 6dB DUT Gain = 15dB DUT i/p match = 1.5 DUT o/p match = 1.5 PSA = 0.186 dB 8970B = 0.327 dB N8973A = 0.186 dB 43% less RSS noise figure uncertainty with NFA or PSA
Example 3. @ 12GHz DUT Noise Figure = 4dB DUT Gain = 20dB DUT i/p match = 1.8 DUT o/p match = 1.8 8970B/8971C = 0.423 dB N8975A = 0.353 dB 16% less RSS noise figure uncertainty with NFA
Example 4. @ 20GHz DUT Noise Figure = 4dB DUT Gain = 20dB DUT i/p match = 1.8 DUT o/p match = 1.8 8970B/8971C = 0.945 dB N8975A = 0.409 dB Noise Figure Basics 57% less RSS noise figure uncertainty with NFA Page 40 Noise Figure Basics Page 41 Noise Figure Basics Page 42 Agilent’s Noise Figure Legacy
8970 8560/90 with NF 340A 1980 1995 1958 85120 1999 Nearly 50 years of Leadership NFA 2000
NEW!NEW! NEW!NEW! PSA with NF 2002 ESA with NF PNA X with 2003 NF MXA, EXA with 2007 NF 2007
Noise Figure Basics Page 43 Agenda • Overview of Noise Figure • Noise Figure Measurement Techniques • Accuracy Limitations • PNA-X’s Unique Approach
Gain
So/N o DUT
Si/N i
Noise Figure Basics Page 44 Source-corrected Noise Figure PNA Series 325 GHz Measurements 110 GHz for the PNA-X 67 GHz PNA-X 50 GHz 40 GHz 20 GHz 13.5 GHz 6 GHz 1010 MHzMHz -- 26.5 26.5 GHzGHz
Noise Figure Basics Page 45 Single Connection, Multiple Measurements • Easily switch between measurements:
• One signal source • CW S-parameters • Pulsed S-parameters • Gain compression • AM-to-PM conversion • Harmonics • Two signal sources • Intermodulation distortion
• Hot-S22 • Phase versus drive • True-mode stimulus • Conversion loss/gain • Noise figure
Noise Figure Basics Page 46 Noise Figure Definition Noise figure is defined in terms of SNR degradation:
(S i/N i) (N o) F = = (noise factor) (S o/N o) (G x N i)
NF = 10 x log (F) (noise figure)
Gain
So/N o DUT
Si/N i
Test system is assumed to be 50 Test system is assumed to be 50
Noise Figure Basics Page 47 PNA-X’s Unique Source-Corrected Technique • PNA-X varies source match around 50 ohms using an ECal module (source-pull technique) • With resulting impedance/noise-figure pairs and vector error terms,
very accurate 50-ohm noise figure (NF 50 ) can be calculated • Each impedance state is measured versus frequency
(Z 1, F 1), (Z 2, F 2), …
Z’s measured during cal F’s measured with DUT
frequency
Noise Figure Basics Page 48 Speed Comparison: PNA-X Versus Y-Factor
Noise Figure Speed Benchmarks 10 MHz - 26.5 GHz
45
40
35 PNA-X is 40% to 900% faster than NFA!
30
25 NFA PSA EXA Seconds 20 PNA X
15
10
10x* 5 6.8x* 1.4x* 4.2x*
0 11 51 101 201 Points * PNA X sweep speed improvement compared to the NFA
Noise Figure Basics Page 49 Noise Figure Uncertainty Example (ATE Setup)
Amplifier: 4.500 Gain = 15 dB Input/output match = 10 dB Y-factor with noise source NF = 3 dB connected to DUT via switch matrix Gamma opt = 0.27 ∠ 0o Fmin = 2.7 dB Rn = 12 – 33 4.000
0.75 dB
3.500 Y-factor with noise source 0.5 dB directly at DUT input
NF NF (dB) PNA-X 0.2 dB
3.000
2.500
2.000 0.5 2.5 4.5 6.5 8.5 10.5 12.5 14.5 16.5 18.5 20.5 22.5 24.5 26.5 GHz * Note: this example ignores drift contribution
Noise Figure Basics Page 50 Uncertainty Breakdown (ATE Setup) Due to imperfect 50 ohm source match
Y factor with ATE network dB Y factor with noise source connected to DUT PNA X with ATE network Jitter
Mismatch Notes:
interaction Gain = 15 dB (4.5 GHz)
s NF = 3 dB Total uncertainty DUT noise/ Γ ENR uncertainty S parameter Input/output match = 10 dB Fmin = 2.8 dB Gopt = 0.27+j0 Rn = 37 Uncertainty contributors Noise source = 346C 97% confidence
Noise Figure Basics Page 51 Noise Figure Uncertainty Example (Wafer Setup) Amplifier: Y-factor with noise source 4.500 Gain = 15 dB Y-factor with noise source connected Input/output match = 10 dB connected to DUT via switch matrix NF = 3 dB to probe via switch matrix Gamma opt = 0.27 ∠ 0o Fmin = 2.7 dB 1.1 dB Rn = 12 – 33 4.000
Y-factor with noise source connected to DUT input Y-factor with noise source 0.75 dB 3.500 connected to probe input PNA-X
0.3 dB
NF (dB) NF PNA-X
3.000
Amplifier: 2.500 Gain = 15 dB Input/output match = 10 dB NF = 3 dB Gamma opt = 0.268 ∠ 0o Fmin = 1.87 dB Rn = 12 – 33 Wave model correlation = 50% 2.000 0.5 2.5 4.5 6.5 8.5 10.5 12.5 14.5 16.5 18.5 20.5 22.5 24.5 26.5 GHz * Note: this example ignores drift contribution
Noise Figure Basics Page 52 Uncertainty Breakdown (Wafer Setup)
On Wafer 15 dB amp with 3 dB Noise figure at 4.5 GHz (Fmin = 2.8 dB, Gopt = 0.27 +j0, Rn = 37.4)
1.8 1.6 1.4 1.2 1 97% Confidence (dB) 0.8 0.6 0.4 Wave 0.2 Y Factor 0 Y Factor PNA XWave Jitter Mismatch Total Uncertainty ENR Uncertainty S parameters
Uncertainty Contributor Noise Parameters
Noise Figure Basics Page 53 Example NF Measurements 5
4
PNA X method using source correction 3
2 Noise Figure (dB) 1 Traditional Y factor technique 0 0 5 10 15 20 25 Frequency (GHz)
Noise Figure Basics Page 54 Airline Demonstration
NF using NFA (Unit 354)
4
3
No airline DUT 2 dB With airline
1 Measurement 1: DUT alone 0 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2 2.1 2.2 2.3 2.4 2.5 2.6 2.7.8 2 2.9 3 Freq (GHz)
NF using PNA-X (Unit 354)
4.00 7.5 mm (length) airline DUT 3.00
No airline 2.00 dB With airline Measurement 2: DUT with airline 1.00
0.00 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2 2.1 2.2 2.3 2.4 2.5 2.6 2.7.8 2 2.9 3 Freq (GHz)
Noise Figure Basics Page 55 NF Comparison in Pseudo ATE environment
Noise source (measurement) NFA
12” cable DUT Loss comp: 4 GHz 0.30 dB 6 GHz – 0.38 dB
12” cable DUT
PNA X cal and measurement planes
Noise Figure Basics Page 56 2-Port PNA-X Options 219, 224, 029
Source 1 LO OUT 2 OUT 1 To OUT 1 OUT 2 Pulse receivers modulator Pulse modulator
+28V J11 J10 J9 J8 J7 J2 J1 Rear
Noise receiverspanel
10 MHz 3 26.5 GHz 3 GHz
R1 A R2
Pulse generators 35 dB B
65 dB 35 dB
65 dB
Source 2 DU Source 2 Receivers Test port Output 1 Output 2 Test port 1 T 2
RF jumpers Mechanical switch
Noise Figure Basics Page 57 Typical Noise Figure of Port Two
Noise Figure Basics Page 58 Noise Channel Parameters
Noise Figure Basics Page 59 Noise Figure Setup
4 through 7
Noise Figure Basics Page 60 Calibration Procedure
• Calibration uses sinusoidal and noise sources, plus cold terminations • Some differences between high and low band calibrations • Calibration sequence for simplest case (insertable) 1. Connect noise source to port 2 • Measure hot and cold noise power • Measure hot and cold match of noise source 2. Connect through (ports 1 and 2) • Measure gain differences between 0, 15, 30 dB stages • Measure load match of noise receivers
• Measure ΓΓΓs values of ECal used as impedance tuner
• Measure receiver noise power with different tuner ΓΓΓs values (mechanical cal only) 3. Connect calibration standards (ports 1 and 2) • Measure normal S-parameter terms
• Measure receiver noise power with different ΓΓΓs values (use ECal or mechanical standards) • Non-insertable cases require extra steps • additional 1-port calibration to account for adapter if noise source is non-insertable • additional S-parameter cal steps for non-insertable DUTs
Noise Figure Basics Page 61 Calibrating On Wafer
1 2
1 port cal
3
2 port cal
Noise Figure Basics Page 62 Comparison Between Gain Settings
Noise Figure Basics Page 63 Measuring Attenuators
20 dB attenuator 40 dB attenuator
Noise Figure Basics Page 64 Interference Beware of unshielded devices, especially near 0.9, 1.8, 2.4, 5.5 GHz!
Noise Figure Basics Page 65 Compression and Damage • Compression in noise receivers: • Wideband noise compresses front-end amplifiers first • Narrowband noise likely to compress ADC before front-end amplifiers • PNA-X will report overload for both cases • Damage level for noise receivers is lower than standard receivers • Front-panel specification is +25 dBm in noise mode • Firmware checks power before switching in 30 dB stage
Detects front end overload
Noise Figure Basics Page 66 Summary • Y-factor method offers reasonable accuracy when noise source is connected directly to DUT • Source-corrected cold-source technique: • Offers best accuracy in all cases • Works in fixtured, on-wafer, and ATE environments • Is 1.4 to 10 times faster than NFA • PNA-X offers highest accuracy as well as speed and convenience of single connection to DUT for a variety of amplifier measurements
Noise Figure Basics Page 67 References • Applications Notes • Fundamentals of RF and Microwave Noise Figure Measurements, AN 57-1, Publication Number 5952-8255E • Noise Figure Measurement Accuracy - The Y-Factor Method, AN 57-2, Publication Number • 10 Hints for Making Successful Noise Figure Measurements, AN 57-3, Publication Number 5980-0288EN
• Web Links http://www.agilent.com/find/nf NFA : http://www.agilent.com/find/nfa PSA : http://www.agilent.com/find/psa NF Uncertainty Calculators: http://www.agilent.com/find/nfu
Noise Figure Basics Page 68