[CQ-Contest] Bandpass filtering

[S56A]

I was surprised to find K1TTT repeating the same mantra about high impedance 
of W3LPL filters in his free e-book.
I did some quick simulations with AADE and I have to correct myself first - 
parraleling 10 m and 160 m BPF causes 7 db loss as I overlooked low 50 Ohms 
working impedance.  Return loss unfotunately drops to poor 5 dB.  K1TTT case 
of 160/80/40 m BPF lumping does only neglidgeable 1 dB loss.  However, I'd 
preffer triplex RX output in his M/M environment.

QS1R is digital RX with A/D direct conversion of the whole 0-30 MHz 
spectrum.  CW Skimmer then processes 6 contest sub-bands, each 200 kHz wide 
with 500 Hz bandwidth, 2400 channels total.  Our own TX is causing trouble 
on only one of them and should be notched.  ICE protection with high speed 
clipping diodes makes RX matters only worse.  My CW Skimmer RX antenna is 
simple omnidirectional 6 ft vertical with 220 uH grounding for 1,8 MHz 
resonance and JFET source follower.  It is at maximal distance from 
horizontal  TX Yagis and dipoles as near field drops with the third degree 
of distance.

 http://www.sm5bsz.com/dynrange/intermod.htm figure 8 is a rare picture of 
digital RX colapsing after saturation.  I quote him: "Note that a 2 dB 
change, from 1 dB below saturation to 1 dB above saturation causes the third 
order intermodulation product to increase by 61 dB."  His latest posting 
http://www.sm5bsz.com/lir/sim1/sim1.htm ilustrates nicely colocating QRM 
problems.

I guess this subject might be discussed in more details at "Contest 
University" :-)

HNY & 73 de Mario, M/S56A