[RFI] tips for finding the source of broadband mixing products

[Matt NQ6N]

I'll add a bit more information.  Thanks much for the suggestions so far:


- The rig is a Flex 6600 SDR transmitting 1 Watt.  There is a Morgan
Manufacturing 40m bandpass filter (I measured it as having 28 db of
attenuation at 14 MHz) between the rig and amp (which is in standby) and
also a VA6AM high power 40m bandpass filter (air coil version with claimed
performance of -70 db at 14 MHz) after the amplifier.

- The TX antenna is an inverted vee in the back yard, and the RX antenna is
an inverted L in the front yard.  So only about 50' of physical separation
of omnidirectional antennas, but lots of BPF filtering and low TX power.

- Completely disconnecting the RX antenna weakens the second harmonic
signal quite a lot and makes the note sound like a pure sine wave, but it
could be that I am not hearing the IMD products when the signal is so much
weaker.

- I had a typo in my original message. It was the second harmonic. I'd just
been listening to the third harmonic when I hastily typed my message, and
didn't notice before hitting send.

- I'll be onsite today with a spectrum analyzer, so any tips on things to
measure would be helpful.  Just wanted to update the thread with the above
info.

73,
Matt NQ6N




On Mon, Sep 30, 2019 at 6:35 AM JW via RFI <rfi at contesting.com> wrote:

>  re: "If that happens, there are errors in design and/or construction."
>
> 50 or 60 dB down?
>
> Pshaw.
>
> de AA5CT
>
> PS 50 dB down can STILL be quite receivable WHEN the rig is operated into
> a
> dummy load. It's WHY operators often think their equipment is faulty IF
> they
> don't sample the RF coming out the antenna port but rather just 'listen'
> with
> another rig nearby connected to an antenna or just a piece of wire.
>
> One of the rigs we demonstrated this on was an old Collins tube-type radio
> too ...
>
> These 'things' are easily demonstrable, Jim Brown. Surprised you have
> _not_
> encountered it.
>      On Sunday, September 29, 2019, 10:28:53 PM GMT-5, Jim Brown <
> jim at audiosystemsgroup.com> wrote:
>
>  On 9/29/2019 7:48 PM, JW via RFI wrote:
> > How are you determining all this - you do realize, in close proximity to
> > a transmitting rig, much (albeit low-level) RF comes straight out the
> > power leads
>
> If that happens, there are errors in design and/or construction.
>
> and can 'modulate'/be modulated by the power supply
> > energy that also escapes, including simple rectifier supplies using only
> > diodes?
>
> Ingress/egress is a linear function, and depends strongly on details of
> both design and construction. Nearly all modern equipment fails to
> terminate cable shields and power green wires properly. They SHOULD go
> the the shielding enclosure (chassis), but they nearly always go first
> to the circuit board, THEN eventually find the chassis after wandering
> around return circuitry for a while. This equipment flaw, first
> discovered by a ham working in pro audio, is called "The Pin One
> Problem," because the designated shield contact of the connector
> commonly used for balanced audio circuits is Pin 1.
>
> The method in which equipment is built usually makes it impractical to
> correct these design errors, so the best fix is a serious common mode
> choke on the cable(s) involved. And because the ingress/egress is via
> the green wire or the cable shield, conventional line filters are
> useless UNLESS they are internal, and with their shielding enclosure
> bonding the green wire to the equipment shielding enclosure! They treat
> only the differential voltage and current between phase and neutral,
> phase and ground, and neutral and ground.
>
> The only effect of signal strength is on the strength of the mixing
> products.
> >
> > We demonstrated this at Heathkit several different ways, including using
> > a spectrum analyzer to 'sniff' the stray RF coming back out via the
> radio's
> > power cable WHICH in turn was modulated and showed 120 Hz sidebands ...
>
> So you added an AC line filter with its shielded enclosure bonded to the
> chassis, right? THAT would work.
>
> 73, Jim K9YC
>
> _______________________________________________
> RFI mailing list
> RFI at contesting.com
> http://lists.contesting.com/mailman/listinfo/rfi
>
> _______________________________________________
> RFI mailing list
> RFI at contesting.com
> http://lists.contesting.com/mailman/listinfo/rfi
>