Moisture: The Enemy
There have been reports of TCAS II
installations showing signs of RF cable deterioration
after only a few years of operation. The cause of these
problems has been traced to moisture either in the coaxial
cable at the antenna connector or inside the connector
itself.
 Moisture
trapped inside a connector can produce effects ranging
from unnoticed to serious, depending on the demands
of the system. But be assured the cable assembly
will be affected if moisture enters. The shields
and conductor can suffer corrosion, especially if cleaning
is less than meticulous when the connector is applied.
However, even if corrosion were not to develop, moisture
present in the connector or dielectric can change the
characteristic impedance of the cable. Where there is
a change in impedance, there is an increase in signal
reflection (VSWR). As a result, a sensitive system may
become unreliable or, worse, fail.
Before making a blanket indictment
of the connectors, cables, cable assembly techniques
and/or installation practices, let's examine the factors
that can create this problem.
Moisture can become trapped inside
the antenna connector as a result of repeated exposure
to significant swings in temperature and atmospheric
pressure in the presence of significant relative humidity.
(See Figure 1.) In commercial
air transport jet aircraft (which account for most TCAS
II installations), air temperatures at the skin can
cycle from as much as, say, +120°F on a Phoenix
ramp to -60°F at 35,000 feet. Pressure may change
from about 30 inches of mercury at sea level to about
7 inches (a 75% drop) at 35,000 feet; cabin pressurization,
however, can reduce the effect to about a 15% drop.
Such temperature changes can precipitate
moisture in the air, and a drop in cabin pressure can
create a partial vacuum in and around the connector.
During descent, though the warming of the skin tends
to return the moisture to the air, some of it can be
drawn into the connector as the pressure returns to
higher levels. All this happens differentially; conditions
inside the connector will not change as quickly as those
around it. Over time, it will equalize again, but there
is no way of knowing what “over time” might
be.
At the risk of downplaying the advantages
of the newer, softer, more porous dielectric materials,
it appears these, too, may be implicated in this. Their
lower loss, smaller diameter, and lighter weight are
important elements in selecting cables for airborne
applications, but there is a price to pay. It has to
do with the porosity of the dielectric. In effect, the
layers of thin Teflon® tape wound around the center
conductor resemble a roll of corrugated cardboard. You
can picture the spaces for "stuff" to get
into. [An exaggerated analogy, of course.]
Solid-dielectric cables go back long
before TCAS and might be considered the solution, but
it must be noted that they are hardly immune to the
effects of moisture and the interior of the connector
is still a trap for moisture. A solid dielectric is
less absorbent, but the corrosion risks to the conductor
and shields are the same. And there are other drawbacks:
solid dielectric cables can add pounds to an installation,
and will occupy more space because of the larger diameter
necessary to meet loss requirements. They are stiffer
and will have greater bend radius requirements and so
may require other routing considerations in installation.
These problems often dictate the choice of the higher-performance
construction.
Now, at least theoretically, whatever
moisture goes in can also get out. Given the conditions
(and freedom) to dry, the problem could be only temporary.
Right?
Actually, trapped in the tiny space
of a connector, a return to dry-as-the-day-it- was-made
is simply impractical — even more unlikely considering
the fact that the water in the dielectric isn't just
sitting there waiting to evaporate or be poured out.
It may have wicked its way back into the cable by capillary
action. Removal by dry heat in a vacuum may be possible,
but even if it worked it is (1) tedious, (2) requires
special equipment, and (3) calls for undoing and rebuilding
the connection at the affected end. All this on top
of the R&R of the installed cables in the first
place!
So it becomes evident that problem
prevention is vital: stop the free flow of moisture
into the connector.
ARINC Characteristic 735 doesn't specifically
call for weather-sealed antenna connectors. In fact,
when it was adopted, the risks of moisture in advanced,
lightweight coax were probably not much of a consideration.
(These issues are not mentioned in the Characteristic.)
Weather-sealed connectors are available,
but no one knows their long-term ability to block the
passage of damaging moisture. However, we do know that
weather-sealing helps. The objective here is to prevent
moisture from oozing into the space inside the connector
where it can get to the exposed end of the cable.
Nothing plastic will seal completely
against water molecules in the air, but weather seals
will stop the condensed water. A layer of dual-wall
(hot-melt glue on the inside) heat-shrink tubing over
the cable and fixed end of the weather-sealed connector
adds strain relief, but is not adequate moisture protection
by itself.
A hermetic seal is the ideal
since it blocks the passage of the smallest molecule.
It is common in situations such as ceramic-package IC's,
light bulbs, etc. where there must be no chance of the
outside atmosphere getting at the sensitive interior.
However, a true hermetic seal is accomplished only with
the flow of metal to glass or ceramic -- obviously a
high- temperature operation, but not practical in cables,
or most other things for that matter.
PIC’s Role
From the beginning, PIC has provided
weather-sealed antenna connectors on the coax assemblies
we make for TCAS II, TCAS I, Mode S, GPS, SATCOM, MLS,
and other RF applications. Also the added protection
of dual-wall shrink tube is standard on all PIC cable
assemblies. These two measures of sealing provide maximum
practical protection against water absorption in coaxial
antenna connectors.
The concerns of the commercial aviation
industry — especially all TCAS II users —
over the effects of moisture in antenna cables are real.
Some suggest that correction and prevention of future
system problems will get constant attention until there
are more answers.
Fact is, connections in all kinds
in aircraft are susceptible to environmental contamination.
Problem connections include those with the following
parameters:
-
RF above 100 MHz
-
System-defined loss budgets
-
Connections at the antenna
-
Non-pressurized cabin location
Of these, it is self-evident that there
is greater proximity to wide temperature excursions
at the antenna. Coupled with moisture in the atmosphere
(as opposed to flowing water, which would suggest a
leak at or around the antenna) there will be absorption
— and not necessarily just a little. One major
airframe manufacturer reports that moisture can wick
deeply into expanded-tape dielectrics (typical of low-loss
coaxial cables) as much as four feet from the connector.
(See “In Defense of Low-loss
Cable” at the end of this paper.)
While loss effects may be unnoticeable
at such a distance, it stands to reason that the longer
the span of wet-dielectric cable, the greater the attenuation.
This contamination by what could be classified as distilled
water (in the air, remember) changes the dielectric
constant materially, affecting impedance of the cable.
As a result, the VSWR is increased and loss suffers.
 Any
cable in systems where a minimum loss is specified is
at risk, especially those produced with a smaller amount
of "leeway" than others, and some system parameters
put a limit on that. See Table
1.
Serious Implications
There have been TCAS and Mode S failures
attributable to moisture in the antenna cables. To the
extent this is the case the potential consequences are
far-reaching.
For one thing, the major role for Mode
S transponders is not to make reports to TCAS-equipped
neighboring aircraft but to transpond to ATC radar for
safe management of the airways. This applies to Mode
A and Mode C transponders as well.
Another complication arises in that,
while some Mode S transponders are set up to notify
the crew of failure, some are not. Dual transponders
may be unable to overcome such failure since it occurs
in the one common element of the system: the antenna
cable.
So something as simple as an
unprotected coaxial connection can cause a transponder
to go silent without notice. Notwithstanding our confidence
in our air traffic control system, it adds to the burden
to deal with aircraft which fail to supply helpful,
if not vital, data.
Repairs
Type C connectors, used on Mode S antennas,
are more susceptible to the passage of moisture than
TNCs which are used on TCAS antennas. The pressure on
the mate gasket is likely at fault; the Type C has a
bayonet-like fitting, while TNCs are screwed onto their
mating connectors with the potential (and probability)
of a tighter fit against the gasket. See Figure
2. Both, however, have been implicated in moisture
problems.
 Next,
if the pathway for moisture at the cable entry point
is unsealed, it may as well be wide open. Some might
argue that if it is so open, it will expel moisture-laden
air just as readily as it takes it in, leaving the connection
in environmental equilibrium. However, there is never
complete reversal of such events; an absorbent cable
dielectric is always going to be "a little thirsty"
when moisture is present. This typifies the principle
of capillary action.
If a section of wet cable can be completely
dried, which is next to impossible, or removed (How
much do you cut off?) it can be re-terminated and deserves
to be sealed as thoroughly as possible.
Some Assumptions
The following may make this a little
less formidable, but bear in mind they are assumptions.
It might be assumed that the worst
of the absorption is in the first few inches of the
cable. This length is going to be longer in low-loss
cables than in solid-dielectric RG-type cables. Secondly,
it might be assumed that cutting and re-terminating
the cable with a new, properly sealed connector (not
just wrapped in dual-wall shrink-tubing) will solve
the problem. However, assurance of restored cable performance
comes only with appropriate testing for loss, VSWR,
delay, and phase angle (if required).
If the removed length is to more certainly
eliminate "all" of the entrapped moisture,
it may shorten the cable too much. Add another piece?
Not a good idea. Now you start compromising.
Here is one final reason to bite
the bullet and replace the cables altogether: with a
repaired cable there is still the uncertainty of trapped
moisture which can in time enhance the corrosion process
and produce yet another mode of failure. Does it buy
time? Yes. Does is overcome the problem? Only maybe.
Prevention — Now
Existing installations may or may not
be a ticking time-bomb, but so far there is no pattern
to suggest that any particular locations or aircraft
are less at risk than others.
Nonetheless, taking immediate steps
to add protection to mated connectors may at least delay
the onset of moisture-caused cable/system failure.
Simply disconnecting the connector
from its mate, air-drying both, and re-connecting it
with a dual-wall heat shrink sleeve over the entire
connection, including several inches of the cable, is
truly inadequate. Moisture remaining inside the cable
and connector are not only ignored; they become trapped.
Actual removal of the connector from
the cable and re-termination after drying may produce
acceptable results temporarily, but it is a poor second
to making a complete cable replacement. Needless to
say, replacement connectors should be of weatherproof
design. (Figure 3 illustrates
typical weatherproofing.)
Field Repairs
 Because
the major airframe manufacturers frown upon —
forbid, actually — the use of standard heat guns
or other such potential fire sources aboard fueled aircraft,
other sealing measures are called for when seals cannot
be applied outside the aircraft.
One recommendation involves wrapping
the connection with a fluorosilicone gel-like tape —
such as PICWrap™. This forms
an excellent seal and is readily and cleanly removable
for inspection or repair. Further, it is a strictly
hand-wrapping operation requiring no heat, thus avoiding
the risk of fire.
Final Analysis
While there is certainly more experience
to be gained, what has been learned so far is :
-
Moisture absorption problems are
worse in cables with wrapped dielectrics than those
with solid dielectrics.
-
Moisture absorption is exacerbated
by the wide pressure and temperature excursions
common to aircraft.
-
One effect of moisture is to reduce
the dielectric constant, changing impedance, worsening
VSWR, and increasing loss.
-
Maximum seal potential is realized
when using connectors with built-in seals and dual-wall
heat shrink tubing or other appropriate sealing
method.
For certain, new installations ought
to include connectors designed to be inherently weatherproof
— at both the cable end and the mating end —
with gasketing designed in. The additional surrounding
protection should be incorporated at the time of cable
assembly , and wrapping the entire mated connection
at the time of installation adds yet one more layer
of protection.
In Defense of Low-loss Cable
In recent years, advances in manufacturing
process for coaxial cable dielectrics have brought us
more air — specifically, microscopic spaces incorporated
into the Teflon® tape which is wound in an overlapping
spiral around the center conductor.
The wound-tape configuration and open
air "cells" lead to capillary action, like
a sponge in some respects.
While this is, in effect, more apt
to draw and trap water molecules, it does result in
improvements in cable performance: notably, lower loss,
higher velocity of propagation, and lighter weight —
to name a few.
These advances far outweigh the risk
of moisture contamination, if steps are taken to create
a suitable barrier to moisture exposure — or better,
if used in a controlled environment.
In aircraft, the environment
is almost hostile. Yet the measures covered in this
paper can and do reduce the ill effects of moisture
on critical parameters.
Back
to Top
Download
a printable PDF of this Technical Paper
|
