Noise Figure and Gain measurements were made on two different mixers, using an Agilent NFA Series noise figure analyzer. The measurements were taken with and without a bandpass filter on the output of the LO, shown in Figure 4 and Figure 5. Figure 4 is a direct comparison of two noise figure measurements on an Infineon CMY 210 mixer using the same LO in both cases, with and without the filter present at the output of the LO. The difference between the measurements is seen to be over 10 dB. As you can see from these results, one must be careful when using a multi-purpose signal generator as an LO, because factors such as phase noise, noise floor and spectral purity of the source, can also affect the accuracy of the measurement.
Figure 4. Mixer Noise Figure Measurements
Figure 5 shows the results of the noise figure and gain measurements on an Agilent mixer tested at 1.9 GHz with varying LO power. Again, the measurements are made with and without a filter on the LO signal. The graph compares the results to the simulations. The difference between the simulation and measurements varies from 1-2 dB depending on the LO power.
Figure 5 Agilent 81008 mixer noise figure and gain simulated and measured with and without LO bandpass filter.
A mixer with decent gain and good LO rejection, is less affected by LO leakage and inherent noise from the output of the LO final stage amplifier, than a lossy mixer with poor LO rejection. However, the effect is still there.
We have seen that an untuned LO buffer amplifier can create problems due to out of LO band noise. Next, we will look at untuned RF amplifiers and their impact on mixer noise figure, both in performance and in measurement. Finally, we will focus on the "Gilbert Cell" mixer, with its "built in" untuned RF amplifier.
| Page: | <<Previous | 1 | 2 | 3 | 4 | 5 | 6 | 7 | Next>> |
