Cascading two 50 ohm RF blocks...

Jan 14, 2006 2 Replies

Something has me really stumped: Since cascading two 50 ohm BP filters can lead to big trouble*, wouldn't similar problems occur when connecting a 50 ohm filter to the input of, let's say, a narrowband LNA? Or to the output of a PA? Would the narrowband LNA (or PA) disrupt the filter's response, since the filter will now no longer see a wideband 50 ohm termination at one of its ports? I know I'm missing something here, but I just can't figure out what...



Thanks,



Bill


  • Since the BP filters themselves are expecting to see only a very wideband 50 ohm termination at both in/out ports in order to function as designed -- but when two filters are cascaded the passbands of these

50 ohm filters are the *only* frequencies that actually and truly see
50 ohms; the stopbands of each filter will now not see 50 ohms at all, as these stopbands are now each looking at the very low return losses inherent each other filter's reflective stopbands.

Yes, that's why it's a good idea to isolate each component (filter, LNA, PA) with a resistive pad of 2 or 3 dB. We once had a problem with a 3 W exiciter feeding a 40W PA. Both were in spec. individually but oscillated out of band when connected together. The problem was that the input to the PA, while 50 ohms in band, was essentially open out of band. It reflected enough energy to cause the exicter to oscillate. The solution was to pad the front end of the PA so the exciter would see 50 ohms at all frequencies and then add enough gain to complensate for the loss of the pad.

The exception would be the output of the PA. Our PAs were required to be stable into both a short and an open attached to the other side of a stretchline at all phase angles. We never had a problem putting filters or cavities on the outputs of the PAs.

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required