
(J3)
LMH9226
(J2)
LMH9226 RRL EVM
(J4)
ENR Diode
Spectrum Analyzer or
Noise Figure Analyzer
Marki BAL-0010
or Similar
0°
180
°
2:1, Zo = 50
Ÿ
Zo(diff) = 100
Ÿ
3-dB pad
3-dB pad
3-dB pad
3-dB pad
Zo(se) = 50
Ÿ
Frequency (MHz)
Ou
tp
u
t Trace
L
oss
(dB)
2000 2100 2200 2300 2400 2500 2600 2700 2800 2900 3000
-2
-1.5
-1
-0.5
0
Frequency (MHz)
Inp
u
t Trace
L
oss (dB)
2000 2100 2200 2300 2400 2500 2600 2700 2800 2900 3000
-2
-1.5
-1
-0.5
0
Test Setup Diagrams
9
SBOU237 – December 2019
Copyright © 2019, Texas Instruments Incorporated
LMH9226 Evaluation Module
that is representative of the device performance.
and
give typical input and output
trace losses measured on the LMH9226 EVM, respectively.
Figure 9. Input PCB Trace Loss vs Frequency
Figure 10. Output PCB Trace Loss vs Frequency
3.2
Noise Figure Test Setup
Figure 11. Noise Figure Test Setup
Use the following guidelines for Noise Figure (NF) measurement:
1. The traditional Y-factor method can be used for the NF measurement using a Noise Diode and a
spectrum analyzer (or a Noise Figure Analyzer), as
shows.
2. While doing the measurement, take into account any RF cable losses to the EVM board. Any external
input attenuator added for matching will result in proportional NF degradation and must be calibrated
out in the measurement.
3. Also, onboard losses of the input traces at the device input pin must be factored into the NF
measurement.