"Ashhar Farhan" wrote in message
om...
alan,

the polyphase network used by hans
(http://www.hanssummers.com/radio/polyphase/index.htm) is a very good
idea if you _have_ to use passive components. although hans has used
1% tolerance components, you can get away with even lesser tolerence
and use ordinary capacitors with 1% resistors.

Thanks Farhan. I actually used 0.1% resistors, and matched the capacitors
also to 0.1% by adding parallel capacitance. Of course, all that could
change with temperature due to the different parallel capacitances drifting
different amounts.

I believe the passive polyphase network to be superior to the active phase
shifting networks which use op-amps. For any given level of component
tolerance, a passive network will give much better opposite sideband
suppression (I recall seeing 10-20dB reported somewhere but can't provide
references). Provided attention is paid to the values used, the network can
be made lossless which overcomes any concerns about gain distribution and
harming the overall receiver dynamic range.

So the way I tend to put it, is if you _have_ to use active components ...

73 de Hans G0UPL
http://www.HansSummers.com