Richard Clark wrote:
On Mon, 1 Feb 2010 08:04:08 +0000, Ian White GM3SEK
wrote:

There seems to be two different meanings of "foil" in this discussion.

Hi Ian,

You don't offer another meaning, simply different examples.


The purpose of the posting was to identify and distinguish those two
very different meanings of "foil".


Most of the criticism seems to have been about "foil" made from
aluminized plastic. I'd agree this is very dubious because the effective
thickness of metal is unknown, especially in low-cost cables. The
presence of a so-called drain wire is also an indication that it's
difficult to make direct contact with the metal in the shield.

Actually, the drain wire is not specifically needed for termination,
but having said that, it is needed for termination - in a practical
sense. The drain wire does not run the length of the cable simply to
provide a handy length of wire available at any arbitrary point of
cut. The drain wire is "so-called" because it serves as a current
drain. It is a necessary component to the electrical design much as
the "so-called" drain lead on an FET is.

The foil has an atrocious conductivity for any significant length. If
it were to be relied upon alone, you could as easily assign it the
name of distributed resistor instead of shield (and yet even a
distributed resistor would satisfy some purpose of shielding).

The drain wire insures that this significant length of atrocious
conductivity is no greater than half the circumference of the inner
insulated wire. At this length, the foil path resistance is a quite
suitably low resistance.

The sense of drain, is electrostatic drain. If the term appears to be
"so-called" it is by purpose and historical application.


Very well, let me re-phrase: the presence of a so-called drain wire can
be taken as an indication that the metalized plastic shield has poor
electrical conductivity and is not suitable for RF applications.


However, "foil" can also mean a thin but solid metal sheet. When applied
as an overlapping wrap of 360deg, this kind of "foil" has close to
perfect shielding properties at HF and above. Its main weakness is that
the metal can tear if the cable is bent too sharply, and the main
purpose of the braided copper cover is to bridge any resulting gaps.
Both copper and aluminium foil-covered cables are available, and copper
will obviously provide a more reliable contact between a connector and
the shield.

Every cable has what is called its minimum turn radius. In use, this
can be violated and the physical and electrical properties can become
compromised. This is not a fault of design.

Manufacturers are fully entitled to specify a minimum bending radius.
What's important here is the *result* of bending the cable at a
progressively decreasing radius. A braided shield will slip and stretch
to relieve the stresses, and will often survive quite excessive bending
without breakage of strands; it will then recover leaving relatively
little disturbance.

In contrast, a foil shield has a very sharp failure threshold, beyond
which it will be torn apart; see below.

That a user can put a cable to misfortune is not remarkable insight,
but attributing the tear in this foil to becoming a great misfortune
seems to be hysterical as that tear is drawing down the shield
coverage from 100% to 99.9999999% except at one specific and
distraught bend where it might actually reduce it to 96% (the native
coverage of the woven shield that embraces it) for an eighth inch. It
is very hard to imagine a situation where this local discontinuity
serves to bring down an entire system when it is a design redundancy.
The user having violated the minimum radius rule should be more
concerned with the inner wire migration through insulation and causing
a short - a vastly higher probability of an issue of greater concern.

Those are two separate problems. The "issue of greater concern" is the
simply the one that happens first; but without detailed knowledge of
each specific installation it's impossible to predict which one that may
be.


Most Hams are quite aware of that consequence, and it alone (if
nothing other) motivates them to observe the minimum bend radius
prohibition.

Those Hams who are not aware of this consequence lead a superstitious
existence where failure arrives by the fault of some mysterious and
elaborate agency:

I have heard these stories of torn foil for years. And yet each and
every one of them has been testimonial, not research based in their
having been the cause of misfortune. Evidence would demand that the
entire length of jacket and woven shield be stripped off the cable in
some form of ritual much like an autopsy.

Here is that story. The cable in question was semi-airspaced with a
shield made from solid copper foil in a 360deg wrap, overlaid by
open-weave copper braid. Having experienced problems with fluctuating
VSWR in a rotor loop, I removed that entire section of cable - and yes,
indeed I did 'autopsy' it.


much like an autopsy. That operation alone is suggestive of general
destruction, a self fulfilling prophecy once you get down to the
fragile foil layer.

Rubbish.

The cable jacket was carefully removed by slitting along its length and
gently peeling it off. In the two sections close to where the rotor loop
had been anchored, the foil shield had been torn circumferentially into
several isolated segments, each a few inches long. The overlying braid
was not broken, and was only slightly disturbed by the surgeon's knife.

Such was the objective evidence.

My deductions were that most of the repeated bending of the rotor loop
had been concentrated into those two sections. As for the VSWR
fluctuations, it seemed that the outer braid had not made sufficiently
good contact to bridge over the breaks in the foil when the antenna was
being rotated. I considered both the observed VSWR problem and the
implied shielding problem to be important because the system was
carrying 1kW at 432MHz.

I accept that these problems were entirely due to my poor installation
technique. I now try to distribute the bending more evenly along the
entire length of any rotor loop, but it isn't easy. Therefore I prefer
to use cables that have some tolerance of excessive bending if it should
occur.


This level of examination is something only a producer would embark
upon, and once they discovered a systemic failure, they would resolve
it (cynics can chime in here with their chorus of "no they wouldn't").

No, they wouldn't. They would simply state that this type of cable was
not designed for repeated flexing at close to the minimum bend radius. I
fully accept that; what I don't like is the drastic mode of failure in
which the foil tears completely apart.

In particular, I don't like the type of cable in which the foil shield
is solidly bonded to the underlying PE, because there is no possibility
of 'slip' to relieve the bending stresses.

In practice, hams have to use whatever is most cost-effective and there
is no doubt that solid copper foil has excellent EM shielding
properties, so long as that shield remains undisturbed. For a rotor
loop, one has to balance the risk of tearing the shield against the
disadvantages of splicing in a section of more flexible and tolerant
cable such as RG213.

More modern low-loss cables have both the solid metal foil shield and a
heavier cover of braid to act as backup.


A Ham would look at a kink in a cable, open it up, discover torn foil,
and it would be immediate proof of the problem. Simply fill in the
blank of what that problem is, and add that to the list of ills that
proceeds from using foil shielded cable.

None of us was talking about a severe "kink", only about moderately
excessive bending.


Although I only had that one experience of failure (and didn't let it
happen again), I did take the trouble to find out what had caused it.
Several other hams have related similar experiences with those kinds of
foil shielded cable. I still use them where low loss is important, but
treat them much more carefully than braid shielded cables like RG213.


--

73 from Ian GM3SEK
http://www.ifwtech.co.uk/g3sek