coaxial cable: gradual deterioration of performance
Wrightsaerials
times more signal than it should. The losses are always greater at higher
frequencies. The problem usually arises when the cable is old, and sometimes I
have been able to track a gradual increase in loss over a period of years.
Sometimes the explanation is obvious — the cable has taken up large amounts
of moisture via unsealed ends. But often there is no obvious moisture in the
cable. I wonder if cable takes up moisture through the outer sheath, because
sometimes the dielectric is slightly discoloured. Any thoughts, ideas,
suggestions?
Bill Wright
Peter Bertini
jacket?
Pete
"Wrightsaerials" <wrights...@aol.com> wrote in message
news:20001216151522...@ng-bk1.aol.com...
> Working on communal TV systems I often find 75ohm coax that loses five or
more
> times more signal than it should. The losses are always greater at higher
> frequencies. The problem usually arises when the cable is old, and
sometimes I
> have been able to track a gradual increase in loss over a period of years.
> Sometimes the explanation is obvious - the cable has taken up large
MatJake
As a lineman for CATV company. My knowledge has been moisture enters through
the connectors. Failure to use silicone, bi-seal tape, shink boots around ETE
and amplifier housings. This corrodes the center conductor, whereby different
freqs. are effected (skin effect etc.). This is for hardline and is the same
for subscriber drops. In a subscriber drop once water has a chance to get in
the braid it acts like a wick and can draw water like a pipe if it has been up
long enough.
Bill losses will always be greater the higher the frequency that is being
utilized.
If you want to keep water out, Always use a silicone based non-hardening
compound. Coat lightly on the threads of gorund block, splitter etc., never
ever put silicone on the inside of the fitting as this compound is somewhat of
an insulator and you could effectively null all signals out, by getting some on
the center conductor. Never ever use RTV since it hardens and you may one day
need to get to the fitting.
Keep splices to a minimum the less the better. If you have to always use
bi-seal material to cover the splice, at the least use electrical tape- I
always found electrical tape just allows the water to be held in to corrode the
splice but if you use a silicone compound on the threads of the barrel
connector and on the fittings themselves it should work ok.
Bill I don't know what climate you have to work with or if you are working with
hardline or RG-11 or 6 but if you have any more questions feel free to email. I
just gave a few suggestions off the top of my head but I could give more
specific answers if you give me a better idea of what you are working with in
the way of equipment etc.
Regards,
To reply remove the ns from address,
Roy Lewallen
over a fairly wide range, and one from a piece of new cable. This might
give some clues as to whether the loss is due to corrosion of the
conductors or contamination of the dielectric, and knowing that might
help point to the mechanism.
Roy Lewallen, W7EL
Tom W8JI
<comm...@erols.com> wrote:
>could it be contamination of the braid from chemicals leaching out of the
>jacket?
>
>Pete
Hi Pete,
I called Belden about that, because a popular ham-myth is that
non-contaminating cables slowly increase loss from chemicals leaching
out of the jacket.
After transferring me to several people, I reached a "real" engineer.
She said non-contaminating means the cable jacket is stabilized and
remains flexible and intact over a period of years.
She said as far as she has ever seen ALL loss is due to corrosion of
the conductors, unless the cable is physically damaged. She also said
all corrosion was due to external contaminants getting into the cable,
with water being a major problem. She said they don't put "nasty
chemicals" inside the cable since the chemicals would have the
greatest effect when the cable was actually being produced and
everything was hot.
I recently measured some 20 year old RG-17 that I bought surplus. The
ends were well sealed, and the cable was stored indoors. The loss
actually measures slightly below factory spec for that cable, despite
warnings from people that the cable would be "junk" because of it's
age.
Everything I've been able to see or learn from reliable sources seems
to indicate moisture ingress in improperly sealed cables is the major
problem.
73 Tom
Reg Edwards
range, 50 kHz to 100 Mhz, is the ONLY way of separating
copper from dielectric losses.
Plot the results on graph paper as attenuation versus
attenuation divided by squareroot of frequency. Accurate
measurements are needed. Power-out versus power-in using
your so-called SWR meter or Bird wattmeter is a waste of
time.
Comparisons with manufacturers' spec figures mean nothing
until a cable can already be seen to be dropping to pieces.
Comparison with a known good length offers greater
reliability.
However, strange as it may seem, measurement of attenuation
is the most insensitive method of looking for defects in
cables. Copper oxide has hardly any electrical skin effect
on copper. It just reduces the effective diameter of the
conductor. Loss noticeably increases only when the braid is
severely corroded and is about to fall apart.
Cutting 95% of the wires in the braid, even all at one
point, has no measureable effect on attenuation of a long
length.
Polyethylene is a very inert substance. Underground and
undersea polyethylene insulated telephone cables laid in the
1940's and 50's are still as good as new except where chewed
by rats.
Faults are 99% due to physical damage to the sheath and at
the terminations. And of course, due to defective
workmanship at joints.
The foregoing remarks do NOT apply to the RG-58 junk sold to
amateurs which has aluminised cigarette paper purporting to
be an outer conductor.
---
Reg, G4FGQ
***********************************
Roy Lewallen <w7...@eznec.com> wrote in message
news:3A3BF79F...@eznec.com...
> It would be really interesting to see a plot of
attenuation vs frequency
> over a fairly wide range, and one from a piece of new
cable. This might
> give some clues as to whether the loss is due to corrosion
of the
> conductors or contamination of the dielectric, and knowing
that might
> help point to the mechanism.
>
> Roy Lewallen, W7EL
>
> Wrightsaerials wrote:
> >
> > Working on communal TV systems I often find 75ohm coax
that loses five or more
> > times more signal than it should. The losses are always
greater at higher
> > frequencies. The problem usually arises when the cable
is old, and sometimes I
> > have been able to track a gradual increase in loss over
a period of years.
> > Sometimes the explanation is obvious â?" the cable has
John Burrell
A few years ago I had to fit an antenna system with Andrews Heliax feeders
to some equipment we had constructed . I seem to recall a warning in the
'assembly instructions' for the N type fittings regarding use of silicon
grease on the threads. I cannot now remember the details but I guess there
must have been a sound reason behind this. The grease is ok for use inside
the plug or socket but must presumably have some detrimental effect when it
gets between two connectors.
Can someone check this out - and if I'm right perhaps somebody can explain
the reasoning behind it. After all it seems sensible to spread the stuff
everywhere to try and waterproof joints but perhaps it has an insulating
effect also?
73, John, G8OZH
MatJake <mat...@aol.comns> wrote in message
news:20001216180917...@ng-fx1.aol.com...
Tom W8JI
<g4fgq...@btinternet.com> wrote:
>However, strange as it may seem, measurement of attenuation
>is the most insensitive method of looking for defects in
>cables. Copper oxide has hardly any electrical skin effect
>on copper. It just reduces the effective diameter of the
>conductor. Loss noticeably increases only when the braid is
>severely corroded and is about to fall apart.
The problem isn't cutting the wires, the problem is when they corode
or oxidize at the overlays where the braid weaves in and out.
Skin effect forces the current near the inside of the braid, and every
time a braid wire exits the current tries to flow through the lossy
poor connection to the next weave that is the inside.
>Cutting 95% of the wires in the braid, even all at one
>point, has no measureable effect on attenuation of a long
>length.
Yes, but tarnishing the braid a slight amount, or removing the
pressure on the weaves greatly increases loss. Fresh braid from RG-8
cable has more loss than a solid #14 wire at 30 Mhz when the pressure
is removed, such as when the braid is used in a tank circuit as a
connection.
PA's will typically melt the braid of RG-8 cable when used as a tank
lead, while #14 wire in the same spot will run cool.
73 Tom
MatJake
Mat,
A few years ago I had to fit an antenna system with Andrews Heliax feeders
to some equipment we had constructed . I seem to recall a warning in the
'assembly instructions' for the N type fittings regarding use of silicon
grease on the threads. I cannot now remember the details but I guess there
must have been a sound reason behind this. The grease is ok for use inside
the plug or socket but must presumably have some detrimental effect when it
gets between two connectors.
Can someone check this out - and if I'm right perhaps somebody can explain
the reasoning behind it. After all it seems sensible to spread the stuff
everywhere to try and waterproof joints but perhaps it has an insulating
effect also?
73, John, G8OZH
Hi John,
You are correct the silicone compound does have an insulating effect. That is
why you do not want to get it on the center conductors or any thing that
requires an electrical connection. I quite honestly do not know the exact
specifics for the N type connectors but I would gather it to be the same.
However someone may have the exact answer. I do know on Cable TV hardline
connectors you need to coat the threads of the female connector which you are
mating. Basically all mechanical connections should have silicon compound, but
a little goes a long way so no need to trowl it on the connectors.
I hope this helps. I do not really post that much but usually take in all the
good info I have received and quite honestly have received some great
information.
I am by no means the guru on this subject but I do hope I have been of some
help.
Regards,
Matt
Reg Edwards
> The problem isn't cutting the wires, the problem is when
they corode
> or oxidize at the overlays where the braid weaves in and
out.
>
> Skin effect forces the current near the inside of the
braid, and every
> time a braid wire exits the current tries to flow through
the lossy
> poor connection to the next weave that is the inside.
>
>
> tarnishing the braid a slight amount, or removing the
> pressure on the weaves greatly increases loss. Fresh braid
from RG-8
> cable has more loss than a solid #14 wire at 30 Mhz when
the pressure
> is removed, such as when the braid is used in a tank
circuit as a
> connection.
>
> PA's will typically melt the braid of RG-8 cable when used
as a tank
> lead, while #14 wire in the same spot will run cool.
I agree tarnishing of the braid will have a greater effect
on outer conductor resistance than if the outer conductor
was solid copper. However, the attenuation of coax is mostly
due to the resistance of the solid inner conductor. It is
this fact which allows such shoddy outer conductors to be
used while still maintaining attenuation figures within
specifications.
For 50-ohm coax the ratio of inner resistance to outer
resistance is 3.5 to 1 which is also the ratio of the outer
to inner diameters. For 75 ohm coax the ratio is
considerably higher.
Terrible things can happen to coax cables (they seldom do)
and the very, very last parameter to be affected is dB per
100 feet.
But, regardless of dB per 100 feet, if the stuff looks bad
then to the junk pile with it !
---
Reg
Tom W8JI
<g4fgq...@btinternet.com> wrote:
ill run cool.
>====================
>I agree tarnishing of the braid will have a greater effect
>on outer conductor resistance than if the outer conductor
>was solid copper. However, the attenuation of coax is mostly
>due to the resistance of the solid inner conductor. It is
>this fact which allows such shoddy outer conductors to be
>used while still maintaining attenuation figures within
>specifications.
>
>For 50-ohm coax the ratio of inner resistance to outer
>resistance is 3.5 to 1 which is also the ratio of the outer
>to inner diameters. For 75 ohm coax the ratio is
>considerably higher.
I disagree with that. Clean braiding has three to four times the RF
resistance of a smooth conductor.
73 Tom
Gray Frierson Haertig
Interestingly, I know of one commercial FM broadcast antenna
manufacturer that specifies that the the inner cup on the N connectors
it uses on it's lower power antennas be flooded with petroleum jelly.
This raised alarm bells with me, but I followed the manufacturer's
instructions on the installations under my supervision and have yet to
see a failure. Some of the installations are close to ten years old.
I suspect that the contact pressure on center pin connectors is high
enough that the contact is made regardless of the petroleum jelly. In
fact, since the petroleum jelly does an excellent job of excluding air
and water, the contacts may well stay in better shape than if the joint
were made up dry.
I routinely smear a little of the silicone grease supplied with Heliax
connectors on the threads of Type N connectors. It makes the joint
thread up easier. I have not had any problem with this practice and
have used it for at least 20 years.
Gray
--
Telecommunications Engineering
Gray Frierson Haertig & Assoc.
820 North River Street, Suite 100
Portland, Oregon 97227
503-282-2989
503-282-3181 FAX
g...@haertig.com
Roger Halstead
N833R World's Oldest Debonair? s# CD-2
"MatJake" <mat...@aol.comns> wrote in message
news:20001216180917...@ng-fx1.aol.com...
Two points here.
Flooding connectors with DC-4, or DC-5 compound is a regular practice (for
some users) up through at least the 440 band.
Getting to a fitting that has been potted in most of the RTVs is relatively
easy.
I normally figure fittings are disposable as they are about the cheapest
thing in the system.
The danger with most of the RTVs is that they are corrosive.
Just take a whif. If they smell like acetic acid there is a reason. It's
because they have acetic acid in them. Out in the open it will evaporate
within a few days and should be harmless. In a closed space it can eat the
foil right off a circuit board.
I believe most of the manufacturers make non acetic versions.
BTW, most of the RTVs are water proof, but the compounds are not completely
moisture proof. Although practically speaking they would be for the life of
most installations.
> Keep splices to a minimum the less the better. If you have to always use
> bi-seal material to cover the splice, at the least use electrical tape- I
Try liquid electrical tape.
Between the "coax seal" and liquid electrical tape you should be able to
keep most anything out. I usually use two, or three coats of the liquid
tape and after a suitable cure, give it a good wrapping of regular
electrical tape for protection.
I haven't had any leak yet. Well, there was the one that had lightening
blow off not only the tape, but the metal plating from the connectors...but
that was an exception.
Roger (K8RI)
Roger Halstead
--
Roger Halstead (K8RI) www.RogerHalstead.com
N833R World's Oldest Debonair? s# CD-2
"Tom W8JI" <2w...@contesting.com> wrote in message
news:3a3c72b5...@news.akorn.net...
> On Sat, 16 Dec 2000 15:33:15 -0500, "Peter Bertini"
> <comm...@erols.com> wrote:
>
> >could it be contamination of the braid from chemicals leaching out of the
> >jacket?
> >
> >Pete
>
> Hi Pete,
>
> I called Belden about that, because a popular ham-myth is that
> non-contaminating cables slowly increase loss from chemicals leaching
> out of the jacket.
>
> After transferring me to several people, I reached a "real" engineer.
> She said non-contaminating means the cable jacket is stabilized and
> remains flexible and intact over a period of years.
>
She should have seen the RG-8 and RG-8X I had over in Breckenridge.
I don't remember who manufactured it, but the jacket was a silver color,
copper braid, foam dielectric, and stranded inner conductor.
The stuff was great to work with as it was very flexible.
After being out in the sun for only three years it looked like it had caught
something fatal related to Mange.
The cable had actually become sticky to touch and smelled like a cross
between saran and vinyl. It discolored the braid and the loss had gone
quite high.
*THAT* was the only cable I ever saw do that.
I think it was called "extra flex", or something like that. I might still
have a short jumper around here although I don't recall if it has any
printing on it.
Roger (K8RI)
Roger Halstead
N833R World's Oldest Debonair? s# CD-2
"MatJake" <mat...@aol.comns> wrote in message
news:20001217124613...@ng-fx1.aol.com...
>
> Mat,
>
> A few years ago I had to fit an antenna system with Andrews Heliax feeders
> to some equipment we had constructed . I seem to recall a warning in the
> 'assembly instructions' for the N type fittings regarding use of silicon
> grease on the threads. I cannot now remember the details but I guess there
> must have been a sound reason behind this. The grease is ok for use inside
> the plug or socket but must presumably have some detrimental effect when
it
> gets between two connectors.
> Can someone check this out - and if I'm right perhaps somebody can explain
> the reasoning behind it. After all it seems sensible to spread the stuff
> everywhere to try and waterproof joints but perhaps it has an insulating
> effect also?
>
> 73, John, G8OZH
>
> Hi John,
>
> You are correct the silicone compound does have an insulating effect. That
is
> why you do not want to get it on the center conductors or any thing that
> requires an electrical connection.
Both DC-4 and DC-5 were "designed" to go on electrical contacts
Admitedly it was a tad lower in frequency, but they used in in WWII to keep
the mags from arcing over at high altitude.
I use it up through 440 and have not had any problems with it yet (over the
last 30 years)
.
>I quite honestly do not know the exact
> specifics for the N type connectors but I would gather it to be the same.
> However someone may have the exact answer. I do know on Cable TV hardline
> connectors you need to coat the threads of the female connector which you
are
> mating. Basically all mechanical connections should have silicon compound,
but
> a little goes a long way so no need to trowl it on the connectors.
Put it on a connector and it seems to have a mind of it's own as it seems to
crawl into just about every spot on the connector.
Roger (K8RI)
J. Mc Laughlin
It can be worse. Study has shown that CU braided coax is slightly
non-linear (produces harmonics) and that tendency is increased with
even slight corrosion. RG214 is better - the AU seems to help. 73
Mac
Tom W8JI wrote:
>
>
> The problem isn't cutting the wires, the problem is when they corode
> or oxidize at the overlays where the braid weaves in and out.
>
> Skin effect forces the current near the inside of the braid, and every
> time a braid wire exits the current tries to flow through the lossy
> poor connection to the next weave that is the inside.
>
>
--
J. Mc Laughlin n8...@tir.com Michigan ~
Dick Carroll
Roger Halstead wrote:
>
> She should have seen the RG-8 and RG-8X I had over in Breckenridge.
>
> I don't remember who manufactured it, but the jacket was a silver color,
> copper braid, foam dielectric, and stranded inner conductor.
>
> The stuff was great to work with as it was very flexible.
>
> After being out in the sun for only three years it looked like it had caught
> something fatal related to Mange.
>
> The cable had actually become sticky to touch and smelled like a cross
> between saran and vinyl. It discolored the braid and the loss had gone
> quite high.
>
A ham I know asked me if that stuff would work on two meters and I told
him,
"Sure, but I wouldn't bet on how much RF you'll get out the far end of
it". He was using something over 100 feet to his antenna. So he stuck a
wattmeter out at the far end, at the antenna, and it was delivering 5
watts out of the 45 he was putting inyo the coax from the rig.
Dick
Tom W8JI
wrote:
>Dear Group:
> It can be worse. Study has shown that CU braided coax is slightly
>non-linear (produces harmonics) and that tendency is increased with
>even slight corrosion. RG214 is better - the AU seems to help. 73
>Mac
Any pressure contact joint can cause less than perfect linear
operation. That's true no matter what the materials are.
The level of that problem is insignificant in coaxial lines compared
to the loss problem.
J. Mc Laughlin
-----As with so many things, it depends. With many co-located
transmitters and receivers, the IM produced can be the limiting
factor. I recall that on a flat-top, solid line had to be used to
overcome IM produced in the braided coax. It was thought that the
problem was in the radios, but during a refit with extra filtering in
the radio (but still using coax to the load) the IM would not abate
independently of the filtering! 73 Mac N8TT
P.S. I intended to say Ag (silver) not Au (gold) for RG214.
Tom W8JI
wrote:
>
>-----As with so many things, it depends. With many co-located
>transmitters and receivers, the IM produced can be the limiting
>factor. I recall that on a flat-top, solid line had to be used to
>overcome IM produced in the braided coax. It was thought that the
>problem was in the radios, but during a refit with extra filtering in
>the radio (but still using coax to the load) the IM would not abate
>independently of the filtering! 73 Mac N8TT
I never consider subjective data for anything more than it is, and
opinion.
I'd be interested in published measurements of just how bad IMD can
be.
As a perrsonal aside, I find it EXTEREMELY unlikely that the cable was
a major contributor (let alone a noticable contributor) to IMD in a
system like that if the other cable was not defective. Often times
site people accidentally "fix" a problem (I've seen this hundreds of
times over the years) by blindly changing things and disturbing
everything in the system, and then wrongly conclude they "found" the
problem.
73 Tom