Hi all, I am currently simulating a receiver chain consistent of an LNA-Balun-Mixer combination using QPSS and QPNOISE as engines. My input port drives the LNA with a 50 Ohm impedance, while the output port serves as 100-Ohm load impedance to the mixer. So far, the gain of the complete Chain seems to correspond to the expectations (GAIN1+GAIN2+GAIN3) but the noise figure happens to be between around 3 dB higher than expected if I plug the single component gain/noise values into the Friis NF equation. I plot the NF using the "Noise Figure" option (SSB) under the QPNOISE results. I was wondering if this 3 dB difference comes from the 100 Ohm output /50 Ohm input factor due to the single ended to differential conversion? So far I haven't found how the QPNOISE engine calculates the noise figure, but that would be my first guess. The receiver's building blocks are matched among them, 50 Ohm single ended, 100 Ohm for the differential ones, so I would discard at first instance any increase in the noise figure due to mismatch between components. Any document or explanation would be greatly appreciated. Thanks a lot, Erick
↧