[CQ-Contest] FW: Coax Stubs for SO2R

Tue Sep 8 11:03:58 EDT 2015

I meant to post my message to Mike (below) on this reflector also.

-----Original Message-----
From: Bill Hider [mailto:n3rr at erols.com] 
Sent: Tuesday, September 08, 2015 10:57 AM
To: 'Mike Smith VE9AA'
Subject: RE: [CQ-Contest] Coax Stubs for SO2R

Mike,

I don't know what you mean: "parallel to my primary antenna switch".  I
assume you mean an A/B switch to test the stub (??)  Any measurements or
insertion of a stub needs to be in the "final" configuration so you can
evaluate the performance of the stub.  You cannot A/B switch a stub in/out
to test it.  The coax cables and relay or other switch  involved in the A/B
switch will change everything.

First of all, you need to set up an interference matrix showing which bands
of each transmitter interferes with which bands on the other receiver AND to
what level (in dB) each interferes.  That means harmonics & sub harmonics.
The determining factor for a SO2R station is whether or not you can work
through the interference, if it's there.  

This will require accurate power measurements which can be made with an
accurate power meter on the transmitter (or amp), by lowering that power
until the interference level is acceptable or zero and filling out the
matrix in dB.  Then moving the power meter to the output of the other
transmitter (or amp) and making the same measurements with the other
transmitter (amp) transmitting.

When you say as you did, for example " I see a signal of roughly S9 or S9+
20/40 on the receiver", that is not accurate enough.
For each band (say, for example the 6 HF contest bands) you need to
determine how much unacceptable interference is there from transmitter A to
rx B, and from transmitter B to receiver A.  You will need to make FIVE
measurements (one/band) from each transmitter - minimum.  That assumes you
are not interested in same-band interference, as I am not.
My SO2R station interference does not include interference due to
transmitting and listening on same the band.

So, start with XMTR A and RX B.  Transmit full power (whatever your full
legal power capability is ). Record that TX A power level as measured on
your transmit power meter.  Listen to the 1st harmonic on RX B.  
Now, reduce your TX A power until the RX B signal is barely heard.
Measure your TX power change you made and calculate that reduction in power
in dB.
That's one entry in your matrix: The amount of interference from TX A on
band 1 to RX B band 2.
You will have to make entries in that matrix for TX A on Band 1 to RX B on
bands 2,3,4,5,6.
Then from TX A band 2 to RX bands 1,3,4,5,6...etc...etc for all of the TX A
bands AND THEN for the TX B bands into RX A  for all 6 bands.

Once you have that matrix, you can then begin to understand your
interstation interference situation.

Then, concentrate your stub design on those bands in the matrix with
unacceptable interference.
Build it/them, install them and run the matrix test all over again.
Build and install additional stubs for the remaining unacceptable
interference situations.
Then run the matrix test all over again.
You get the idea.

3 years ago, I went through this process with George, W2VJN, on email.
George, in Oregon,  gave me measurement scenarios and I, in Maryland, made
the measurements and provided George with the data.  George designed my SO2R
stub filter network and provided a proposal for him to build it for me,
which I accepted.  Over the next two months, George built, tested,
dissembled for shipping, and shipped me my stub filter network band-by-band.
I reassembled each band, installed each band's stub filter in my shack. 
 It turned out I needed two stubs per band to solve my interstation
interference.
George provided a CD with the lab test measurement curves (attenuation on
each band, insertion loss & VSWR) for each of my stubs.
They worked fantastically!  And still do!

And, Mike, at the beginning of my process, before I contacted George, I had
made the identical measurements you made and drew the identical conclusions
as you stated in your original email.

Glad to  help/comment on your matrix as you proceed...post it here and let
us all know how you are doing with it.

73,

Bill, N3RR

-----Original Message-----
From: CQ-Contest [mailto:cq-contest-bounces at contesting.com] On Behalf Of
Mike Smith VE9AA
Sent: Monday, September 07, 2015 9:07 PM
To: cq-contest at contesting.com
Subject: [CQ-Contest] Coax Stubs for SO2R

This is about coax stubs.

Sent this earlier to "Towertalk" then I thought perhaps it was better suited
to contesting.

Reposting, with some edits:

I've seen the K2TR stubs, K1TTT page and have recently bought the FB book by
W2VJN

Today, for a change, I had a little spare time so built 2 stubs, but lets
just concentrate on one.

It was supposed to be for the 40m transmitter (Radio A).  

I will add  Radio A (transmitter) is using a multiband vertical.

Radio B (receiver) is attached to a multiband horizontal dipole (ZS6BKW) in
case it makes any difference in my question.

I see a signal of roughly S9 or S9+ 20/40 on the receiver (varies per band &
harmonic)

I roughed out the length, attached a T and a 2nd antenna switch (a la VA2UP
method) (which I'll only use for stubs), trimmed it with my AA-230pro

and when it's inline (parallel to my primary antenna switch) I see no
difference at all on 20m or 15m on my 2nd receiver when the stub is in or
out of line

NONE.  Like, not even a titch ESP hopeful. (hi)

I presume this is a type 1 stub (shorted 1/4WL type, x velocity factor)

Thinking I had done something wonky, I made another stub from different
coax..same result.

Later I made one up for 20m - same result. No change whatsoever in the
received signal harmonic(s) on the receiver.

Anyone been down this road?

Too close to rig?

Interaction from my T and 2nd antenna switch?  Gremlins? Wrong method using
the AA-230 pro maybe?  I am close to that 23' mark on the 40m one.

de Mike (SO2R hopeful) VE9AA

Mike, Coreen & Corey

Keswick Ridge, NB

_______________________________________________
CQ-Contest mailing list
CQ-Contest at contesting.com
http://lists.contesting.com/mailman/listinfo/cq-contest

-----
No virus found in this message.
Checked by AVG - www.avg.com
Version: 2015.0.6125 / Virus Database: 4409/10595 - Release Date: 09/07/15