[RFI] Not your typical invisible fence RFI

Tue Feb 1 09:15:12 EST 2022

Opps, I just spotted an error in my text.  I should have said capacitive
reactance and Xc, not inductive reactance and XL (sorry for the error in my
text).  My calculations were correct, but I went brain dead when I typed my
report.  Here is my corrected statement.

I decided we should start with a 0.001 uF cap since this would provide the
following capacitive reactance: Xc at 7 KHz = 22.7 K ohms, Xc at 14 MHz =
11.4 ohms.

Just FYI,
Don

On Mon, Jan 31, 2022 at 5:48 PM Don Kirk <wd8dsb at gmail.com> wrote:

> Here is an interesting update regarding the 20 meters broadband RFI being
> generated by my neighbors Invisible Fence transmitter that's not the
> typical RFI encountered with such systems.
>
> We tried a brand new transmitter and we still had the same problem but at
> a slightly higher frequency (15 MHz versus 14.3 MHz as an example), but the
> frequency is very temperature sensitive (drifts based on temperature)
> so hard to tell how much difference in transmitters was just due to
> temperature.
>
> We configured the transmitter for 10 KHz output versus 7 KHz output
> operation, but still had the same 14 MHz broadband RFI problem.
>
> We tried the Corcom 10VK6 transformer on the output of the transmitter,
> and sure enough it eliminated the broadband RFI on 14 MHz, and a spectrum
> scan from 500 KHz through 30 MHz showed everything looked great (no
> broadband RFI).  Unfortunately the Corcom filter caused the invisible fence
> transmitter to no longer detect a break in the in ground loop of wire, and
> I had already been told that would be the case by the local invisible fence
> technical manager.  This is not what I have encountered with the Innotek
> brand of transmitters which still detects a break in the loop when using
> the Corcom 10VK6 filter.
>
> Also tried two #31 clamp on cores as a common mode choke on the
> transmitter output wires (both wires passing through the same core 3 times
> and added a second core in series with the first one in similar fashion),
> but this had no effect  as the broadband RFI on 14 MHz was still present.
> Note:  Now looking back at the situation I should have also tried a #31
> clamp on core on each wire individually after seeing that the common mode
> choke configuration had no effect.
>
> I then wondered if I could just use a capacitor across the output of the
> transmitter as a very simple low pass filter to "short out" the 14 MHz
> signal while looking like an open at the 7 KHz frequency of operation.  I
> decided we should start with a 0.001 uF cap since this would provide the
> following inductive reactance: XL at 7 KHz = 22.7 K ohms, XL at 14 MHz =
> 11.4 ohms.  Low and behold this worked great in eliminating the 14 MHz RFI
> and a scan from 100 KHz through 30 MHz looked great (no RFI) and my pickup
> antenna was only about 5 feet from the transmitter.  We then tested the
> wire break detection system on the transmitter and it still worked fine.
> We also tested the distance from the in ground wire relative to when the
> dog collar detected the wire, and noticed no changes in the detection
> distance so the Invisible Fence technician was pleased that everything was
> working just as it should.  Note: we used a 100 volt silver mica capacitor
> because that's what I had on hand.  The capacitor is located between the
> output of the Invisible Fence transmitter and the Invisible Fence lightning
> arrestor, so hopefully the 100 volt rating will be adequate.
>
> I'm now going to write up what we did, and then I'm going to have the
> local Invisible Fence office send this info to their parent company (Radio
> Systems Corporation) who is the actual manufacturer of the Invisible Fence
> brand.  I want Radio Systems Corporation to officially approve the use of
> the 0.001 uF capacitor for elimination of the 14 MHz broadband trash that's
> generated by the transmitter.  I also want Radio Systems Corporation to
> look into the design of the transmitter regarding why it's generating the
> 14 MHz broadband trash as it definitely generates harmful interference (The
> RFI is not coming from the wall wart power supply as previously mentioned,
> it's being generated by the transmitter).  I would also like to go back and
> evaluate #31 cores on each transmitter output wire (high impedance on 14
> MHz but not at low frequencies) as this would be a more robust solution
> regarding component failure as a shorted capacitor would cause the output
> signal to bypass the inground loop which would not be detected and the
> homeowners dog would no longer be contained by the system which would be a
> horrible situation.
>
> In the meantime the 0.001 uF cap is in service.  Upfront I made sure that
> Invisible Fence understood the homeowner must not incur any expense related
> to remediation of the RFI, and they fully agreed.
>
> Just FYI,
> Don (wd8dsb)
>
> On Mon, Jan 24, 2022 at 12:37 PM David Eckhardt <davearea51a at gmail.com>
> wrote:
>
>> My wife sez I'm always in a hurry.  Well, she is correct.  A search on
>> the specific number did not yield anything.  And I did not visually
>> search.  As you found, it's there.  Next time I'll slow down a bit.  Thanks
>> for the verification.
>>
>> Dave - WØLEV
>>
>> On Mon, Jan 24, 2022 at 11:59 AM Hare, Ed, W1RFI <w1rfi at arrl.org> wrote:
>>
>>> I didn't have to do anything special, Dave. I just typed in KE3 in the
>>> grantee box and scrolled down the resultant list until I saw the full ID
>>> number.
>>>
>>> It was about 75% down the list if 116 items that KE3 has certificated.
>>> Perhaps you simply overlooked it.
>>>
>>> I always change the default of showing only 10 items to show 500 items
>>> because I don't like to have to keep clicking the "Show Next 10 Rows"
>>> button at the bottom of the screen of results.
>>> ------------------------------
>>> *From:* David Eckhardt <davearea51a at gmail.com>
>>> *Sent:* Sunday, January 23, 2022 7:52 PM
>>> *To:* Hare, Ed, W1RFI <w1rfi at arrl.org>
>>> *Cc:* Don Kirk <wd8dsb at gmail.com>; RFI <rfi at contesting.com>
>>> *Subject:* Re: [RFI] Not your typical invisible fence RFI
>>>
>>> Ed, please let me know why I did not find it under the listing on the
>>> OET for the FCC ID number.  That is usually where I find such things.
>>> Please educate me on how you found it.
>>>
>>> Dave - WØLEV
>>>
>>> On Sun, Jan 23, 2022 at 10:49 PM Hare, Ed, W1RFI <w1rfi at arrl.org> wrote:
>>>
>>> I found the certification record easily.
>>>
>>>
>>>
>>> https://apps.fcc.gov/oetcf/tcb/reports/Tcb731GrantForm.cfm?mode=COPY&RequestTimeout=500&tcb_code=&application_id=rQqvlnXl0jI%2BsFf%2Fvos1Hw%3D%3D&fcc_id=KE3-3002587
>>> FCC - OET TCB Form 731 Grant of Equipment Authorization
>>> <https://apps.fcc.gov/oetcf/tcb/reports/Tcb731GrantForm.cfm?mode=COPY&RequestTimeout=500&tcb_code=&application_id=rQqvlnXl0jI%2BsFf%2Fvos1Hw%3D%3D&fcc_id=KE3-3002587>
>>> OET TCB Form 731 Grant
>>> apps.fcc.gov
>>> It is not a Class B device; it is an intentional emitter.  The rules for
>>> emissions from intentional emitters, when coupled with a flawed
>>> extrapolation method that the FCC defines is rather high by
>>> radiocommunications standards.  The harmonic suppression required in the
>>> rules is essentially 0 dB, with spurious emissions limited to the same
>>> levels as would be intentional emissions on the same frequency.
>>>
>>> Fortunately, the cure worked.  Interestingly, ARRL has received very few
>>> complaints about these, although by the look of the rules and the
>>> certification test report, it looks like interference is likely.
>>>
>>> Ed
>>>
>>>
>>> ------------------------------
>>> *From:* RFI <rfi-bounces+w1rfi=arrl.org at contesting.com> on behalf of
>>> David Eckhardt <davearea51a at gmail.com>
>>> *Sent:* Sunday, January 23, 2022 1:06 PM
>>> *To:* Don Kirk <wd8dsb at gmail.com>
>>> *Cc:* RFI <rfi at contesting.com>
>>> *Subject:* Re: [RFI] Not your typical invisible fence RFI
>>>
>>> There is no record on the FCC OET site for KE3-3002587 which is shown on
>>> the image of the label as the FCC ID number for the wireless transmitter
>>> -
>>> Part 15, Class B, Subpart B, intentional radiators.  However, if I input
>>> only the Grantee Code of KE3, indeed, all kings of pet care products are
>>> shown.  This might be a knockoff with a bogus FCC ID????  Also, as a
>>> Class
>>> B product - home and small office - the FCC logo is required on the
>>> regulatory label.  It is absent.
>>>
>>> Dave - WØLEV
>>>
>>> On Sun, Jan 23, 2022 at 3:23 PM Don Kirk <wd8dsb at gmail.com> wrote:
>>>
>>> > I just uploaded a video onto YouTube that documents wideband RFI on 20
>>> > meters that I tracked down to my neighbors invisible fence
>>> transmitter, and
>>> > it's not the typical approximate 7 or 10 KHz interval RFI encountered
>>> with
>>> > inground invisible fences.  The RFI which is at least 150 KHz wide at
>>> my
>>> > house moves around based on outdoor temperature and on warmer winter
>>> days
>>> > it drops down to slightly below the 20 meter band.
>>> >
>>> > I was hoping the problem was with the wall-wart power supply feeding
>>> the
>>> > transmitter but it was not.  The transmitter has battery backup and
>>> the RFI
>>> > is still present with the wall wart unplugged, and we also tried a
>>> linear
>>> > lab grade power supply in place of the 19.0 Vdc 1 amp wall wart with no
>>> > noticeable difference in the level of RFI at my house.
>>> >
>>> > Unfortunately the transmitter is not part of a DIY system so the
>>> > manufacturer (Radio Systems Corporation) requires that we deal with the
>>> > local "Invisible Fence" Dealer which is our next step.
>>> >
>>> > An interesting note is that when I first tracked the RFI down to my
>>> > neighbors home I wound up near the main circuit breaker box.  We then
>>> tried
>>> > sniffing the circuit breakers with a very small loop to see what
>>> circuit in
>>> > the house the RFI was originating (assuming the RFI was being
>>> conducted on
>>> > house wiring) and this technique typically works great.  We did not
>>> detect
>>> > the RFI on 20 meters on any of the circuit breakers which was
>>> interesting
>>> > because the RFI was strong near the main circuit breaker box.  It
>>> turned
>>> > out that the invisible fence transmitter is just a few feet away from
>>> the
>>> > main circuit breaker box which explained why the signal was strong
>>> near the
>>> > main circuit breaker box.
>>> >
>>> > For those interesting here is a link to the video documenting the RFI:
>>> > https://youtu.be/rHABjR0GZWg
>>> >
>>> > P.S. I have great neighbors that are wonderful to work with on RFI
>>> issues.
>>> >
>>> > 73,
>>> > Don (wd8dsb)
>>> > _______________________________________________
>>> > RFI mailing list
>>> > RFI at contesting.com
>>> > http://lists.contesting.com/mailman/listinfo/rfi
>>> >
>>>
>>>
>>> --
>>> *Dave - WØLEV*
>>> *Just Let Darwin Work*
>>> _______________________________________________
>>> RFI mailing list
>>> RFI at contesting.com
>>> http://lists.contesting.com/mailman/listinfo/rfi
>>>
>>>
>>>
>>> --
>>> *Dave - WØLEV*
>>> *Just Let Darwin Work*
>>>
>>>
>>
>> --
>> *Dave - WØLEV*
>> *Just Let Darwin Work*
>>
>>