[RFI] RF getting into 4-port wirelss router.

[Roger (K8RI)]

Hi Kenith,

BTW I'm 74.

Lots of meat here and far to much to "line itemize"<:-))
Two thing stand out to me although several more appear to be related.

The two things, no, three things that bother me are Knob and Tube,

The problem goes away when the router is disconnected, but the CAT5 is 
still connected to the other computers without causing them grief

The failures appear to be random.

It appears from what you have said that there may be Knob and Tube 
connected to regular wiring.  The biggest problem with Knob and Tube is 
one of polarization.  It's very easy for the polarization to get 
crossed. Most of the time, cross polarization will not cause problems.  
(most of the time), but neutral is essentially ground.  In those old 
houses polarization was more or less, "potluck" .  That makes for an 
unpredictable system whose characteristics depend on what's turned on 
and what's plugged in. This "might" account for the random, or 
unpredictable failures and the problem of tripping GFI breakers that 
went away.  .Did you have a heater plugged into one of those circuits.
Another problem with Knob and tube is it's rare, but circuits sometimes 
actually do get crossed.

This takes me to the system works with the CAT5 unplugged from the 
router, yet still connected to the computers.  You did say that one 
computer area "might" be connected into the old wiring.  Just suppose 
that one of, or both of those computers is connected to an outlet that 
has the wrong polarization.  A networked computer on cross polarization 
"might" be putting 120 VAC on the CAT5 to the router, or the computer to 
the router might be the cross polarized problem.  This would also 
"likely" cause a LOT of RF between the cross polarized computer(s) and 
routers.  The "problem" could also be coming from the router side.

I forgot to ask and you didn't say, but how is the voltage regulation.  
Those old systems were great for a load on one side pulling down the 
voltage and at the same time, driving up the other side.

So we've gained a couple more possibilities. Cross polarization and poor 
voltage regulation which "may" be dependent on what's plugged in or 
turned on.

A test would be to hook a single wire to the neutral in a  plug going to 
a known outlet where hot and neutral are known and proper.and light bulb 
socket.  Using neutral keeps you from holding onto a hot wire. (You 
never know it's hot until you touch a ground)  Make the wire long enough 
to reach anywhere in the house, so hopefully you have a good outlet near 
the center.  Put a female connector on the far end with short lead with 
a probe off the other pin.  plug in a trouble light and check the 
neutral in all the outlets.  Check the hot as well.  You could use a 
volt meter instead of the light, but the light is quick. Bulb lights on 
a neutral and that outlet is the wrong polarization.  You can also turn 
things on and off as well as plugging in appliances.   You might have 
one with a short plugged into a cross polarized outlet.  Plug light into 
outlet and it should light.  Plug into wire and insert probe into 
neutral side.  Should be nothing. Bulb lights = cross polarization,  
Bulb should light with probe in what should be the hot side.

The house I was raised in was 3 old farm houses mated into one 
reasonably sized one. Sounds awkward, but it looked looked like an old 
farm house.  It too was a mix. Part was over a dug out "Michigan 
Basement" with a cistern. Most had an actual basement. My folks put in a 
driven well at roughly 90 feet.  They went to indoor plumbing a bit 
before I arrived.  Much of the house had Knob and tube, but "most of 
that was replaced early on.

Back then there were NO regulations or inspectors.  If it didn't kill 
you or burn down, it must be good.
"The Panel" was a big board with one, two, and four screw in fuse 
blocks, a couple of two cartridge fuse holders and a knife switch with 2 
cartridge fuses.  Seems like there was one single cartridge fuse holder 
hanging on a couple of wires.  All open!  The entire barn was powered 
out of that...somewhere<LOL>  Wish I had a picture of that "panel"

The wiring in those old home can be like a can of worms

73 & good luck!

Roger (K8RI)


On 1/14/2015 11:51 AM, Kenneth G. Gordon wrote:
> On 14 Jan 2015 at 0:34, Roger (K8RI) wrote:
>
>> On 1/12/2015 11:49 PM, KD7JYK DM09 wrote:
>>> :I suppose if you're an expert at everything, there's no reason to try to :
>>> help.
>> I know of no one who is an expert on everything. Actually I know a
>> number of knowledgeable people, but no experts.
> I'm with you on this one (among others), Roger. I'm certainly no expert,
> although I do know "some" things. ;-)
>
>> I'll repeat what I said before (as did several others) and ask some
>> questions.  The remedies of RFI suppression have been pretty well
>> covered.  Jim's tutorial does an excellent job of showing how to
>> suppress RF that gets in the line.
> Yes. I have downloaded a portion of what Jim has on his site, and what little
> of it I have so far been able to read is very informative and valuable.
>
>> *If these remedies have not worked, it's time to stop and look for
>> reasons!*
> Indeed. I have not yet been able to implement many of Jim's (and other
> excellent suggestions I have received here), but will as soon as I can.
>
> I DID remove that 30+ feet of #8 un-connected "ground" wire
> yesterday...whcih was "a job" in itself. The stuff is very stiff and wound
> around all sorts of other wiring. I rolled it up with the idea of using it to
> connect to the 8' ground rods I presently have in place. I intend to also add
> other 8' ground rods as soon as the ground unfreezes here.
>
>> Is the problem really RF?  If RF, it almost has to be common mode. There
>> has to be a reason why this particular setup is so prone to problems.
>> It's not just RF getting in, it's RF (Or Something. It could be AC, or
>> DC as well when the equipment is powered up)  at a magnitude that is
>> frying the equipment. It again, (if nothing mentioned previously has
>> been overlooked)
> Ha! Are you kidding? Of course something has been overlooked! :-)
>
> That's why I posted my original question(s) to the forum! Like anyone else,
> although I have had years of experience in various fields of investigation, I
> certainly do not remember everything and must be reminded of some thing I
> either forgot, or never ran into.
>
>>   almost has to be something we'd not expect as the
>> regular approaches have not worked.  Jim's air core resonant coil should
>> be effective at removing a specific narrow band of frequencies. Get too
>> many, or too few turns and they don't do much.  (Old terms and thinking
>> die hard) while the chokes should suppress a wide range of
>> frequencies.   They are wide band, but switching  power supply
>> frequencies take more turns than the ham bands.  OTOH the Power supplies
>> can have frequencies well up into the ham bands
> Indeed!
>
>> Remember the RF getting into the house wiring and tripping GFIs. Might
>> this be a key?
>> Knob and Tube wiring in some areas.
> Actually, there is very little K&T wiring left, and that which IS left is impossible
> to get to without a major dismantling of parts of the house. As far as I know
> (!) the only K&T wiring left is a short run of lighting in the attic, and one or two
> wall plugs in the living room. (Hmmm...I'd better check those out...)
>
> FYI, the house, a 2 story one, was built in the late 1800s, moved into its
> present location in the 1930s, a basement was then installed under the
> house (certain basement walls are over 3 feet thick), and electricity was
> added about that time (remember the REA?), then the top of the house was
> cut off since 2 story houses weren't "fashionable" at that time.
>
> I did a major re-wiring of the house shortly after we moved into it, upgrading
> all the wiring I could reach without ripping out walls or ceilings, bypassing
> most of what I couldn't get to.
>
>> IF this is RF getting into the CAT5 why aren't the failures almost
>> immediate? (as has been noted previously)
> Good question, which I have not been able to figure out. One significant
> factor (which I have never been able to figure out either) is that the
> "symptoms" change with the seasons. I suspect that the main reason is that
> as the ground dries out, the ground resistance rises drastically, and during
> the wet seasons, it falls drastically. Soil here is mostly hard clay, which when
> it gets saturated is almost like some sort of thick soup, and when it is dry is
> like concrete.
>
>> Loads connected and disconnected?
>>
>> Are the failures temperature, or weather related.
> Seasonal: see immediately above.
>
>> Knob and tube.  I lived in a house like that many years ago. Normally,
>> "Knob and Tube" is only two wires with no ground wire so the outlets
>> would not have a proper ground even if they do have a ground.  Knob and
>> Tube use only two pin sockets..  IE, they won't work with GFIs and
>> should never be part of a circuit that has GFIs. IE an extension.
>> Running a separate ground wire to outlets on knob and tube is begging
>> for problems and can be dangerous.  A house with ham gear and computers
>> should have the K&T replaced and removed as soon as possible.
> Done...at least as much as I was able to reach. As I said, the only bit of it left,
> as far as I know, is a short bit of lighting in the top floor (what's left of the top
> floor where the bedrooms are), which consists of two lights, one in each
> upstairs bedroom.
>
> However, now that I am reminded of it, there MAY be another bit which feeds
> the bathroom... and the main computer room....which may be a big part of
> the problem...although as I remember it, it is a run of #14 TH, installed by
> those geology students I mentioned earlier. At least it terminates at the panel
> as a #14TH...
>
> It is absolutely impossible to get to without a major de-construction of the
> house.
>
>> What is different about this setup besides the Knob and Tube?  IIRC he
>> has a vertical that's fairly close.  Do any of the radials come close to
>> the CAT5?
> No.
>
>> Is the equipment "earthed" at each end?
> Yes, AFAIK... I suppose I should check that...
>
>   Are the coax and
>> radials "earthed" at the base of the vertical.
> Yes.
>
>>   If so, are all
>> connections secure?
> Yes.
>
>> Is the CAT5 a resonant length?
> The run to the basement shack MAY be...it is around 62 feet long...a 50'
> section, then a dumb-switch, then a 12' section.
>
>    Is there something
>> common between the rig and network?
> Not that I know of.
>
>> I keep thinking Knob and Tube.
>>
>> If removing the CAT5 eliminates the problem, disconnect it and measure
>> voltages to ground while transmitting.
> Good idea.
>
>>   Borrow an O-scope
> I've got three good scopes, two Teks and an HP.
>
>> or spectrum
>> analyzer,
> Don't have a working one. Wish I did. I have a non-working Tek or HP which
> I got from the University of Idaho's EE dept, but it needs work...
>
>> but expect abnormally high voltages.  If you are not familiar,
>> borrow the owner of the equipment too.  That way you shouldn't be
>> responsible.  A quick and check would be to disconnect the CAT5 and put
>> an NE51 between it and ground to see if it really is RF of substantial
>> voltage.
> Ah. Good idea.
>
>> It would help to know the layout of the entire "wired" network.  I lieu
>> of a drawing a good description might help
> OK. I'll work on that...
>
>> Are the other computers still connected when the router/switch are
>> disconnected?
> Yes.
>
>> Does your local ham club have an RFI group,
> No.
>
>> or individuals well versed
>> in hunting down problems?
> Only two of us... ;-)
>
>>   If available make generous use of local brain
>> power.
> Well, on that note, all hams in our small town and for a substantial distance
> around it have been plagued with a terrible amount of RF noise for so long
> most of them think it is normal! It is only since I and two other hams realized
> that the noise level was unusually high that we have attempted to do
> anything about it. Due to the fact that none of us has had any real time to
> devote to it, progress has been very slow...but some progress has been
> achieved.
>
> In my case, where, on 80 meters, my "normal" noise level is a constant 10dB
> over S-9, rising to 20dB over S-9 much of the time, we recently found a large
> neon sign at a local pharmacy about 3 blocks from my home that when
> switched off, reduces my noise level by 15 dB. Unfortunately, the owners
> have not yet seen fit to find someone capable and knowledgeable enough to
> permanently cure it, and I am reluctant to tackle the job since I don't carry
> that specialized insurance needed for such a job.
>
>> A knowledgeable person, or persons present can do a lot more, quickly
>> than we can here.
> Indeed. Yet this forum has proven to be extremely valuable to me before this
> event.
>
>> I will relate a problem I had a while back,  I had fried two relatively
>> expensive duoband (VHF/UHF) verticals which in turn took out some
>> expensive finals. The problem turned out to be a combination of a power
>> supply and grounding problem which all agreed shouldn't happen.  I'm
>> still not sure about it, but with all grounds in place, its been working
>> for some time.
>>
>> Knob and Tube...That is the main difference between this house and most
>> with ham stations and computers.
>> BTW At the risk of dating myself, when I was a kid, Knob and Tube was
>> not uncommon.
> For me either, and I'm not averse to dating myself: I am 72, and was first
> licensed in 1956. :-)
>
> Thanks for the information and help, Roger.
>
> vy 73,
>
> Ken W7EKB
>
>


---
This email has been checked for viruses by Avast antivirus software.
http://www.avast.com