From: StarAerospace@aol.com
Message-ID: <70.cc5c438.287b72c8@aol.com>
Date: Mon, 9 Jul 2001 16:49:12 EDT
Subject: Re: Electrical connectors and solder
To: lancair.list@olsusa.com
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit
Reply-To: lancair.list@olsusa.com

          <<<<<<<<<<<<<<<<--->>>>>>>>>>>>>>>>
          <<  Lancair Builders' Mail List  >>
          <<<<<<<<<<<<<<<<--->>>>>>>>>>>>>>>> 
>>
<< Solder is technically unacceptable for aircraft usage due to the fatigue 
problem >>

>From my years in avionics (military, GA, commercial carrier) and over 300 
full panel retrofits from '78 to '87 as well as my time on CH-53s,  B737s and 
BAe 146s in both repair and overhaul, I did far more soldering than crimping. 
 Solder unacceptable?  Without strain relief, perhaps.  With strain relief, 
solder connections are used in a large percentage of connectors on your 
avionics and instruments.  Some of these are mil spec, some are not, all are 
either mil spec or FAA certificated and all depend on correct and adequate 
strain relief.  It is only in the last two decades that crimp pins have fully 
displaced the solder cup pins in mil spec connectors for light wire gauges.  
They are far from fool proof, and STILL require correct and adequate strain 
relief.  I have lost count of the number of failed crimp connections I have 
repaired and they far outnumbered the failed solder connections both as a 
percentage and overall.  Many of those crimp connections that I repaired were 
properly installed, they failed anyway.  I may have seen a failed solder 
connection when it was properly strain relieved, but I don't remember it.

If you use crimp connections, you must use the correct racheting crimp tool, 
the correct pin size for the wire, strip the insulation to the exact 
dimensions specified, then provide strain relief external to the pin.  Why 
all of this concern?

First, toss that smash-jaw-home-depot-flymart set of pliers in the trash.  
The ones that have broad crimping jaws will only match well to one or two 
sizes of pins and you have no quality control vs. over and under crimping.  
Under crimping has the obvious disadvantage of poor contact and possible wire 
pullout.  Over crimping is far more common (a little crimp is good, more must 
be better, right?) and causes far more problems.  An over crimped connection 
can cause stress and failure of the pin itself, bending of the pin which 
prevents proper seating in the connector, and (worst of all) cutting of the 
wire.  The last exhibits an intermittent connection;  they are oh-so-fun to 
troubleshoot.  The "blade jaw" pliers are even worse, since an over crimp 
literally cuts the wire inside the connection.  The worst case scenario is a 
correct crimp on the insulation area and an over crimp on the conductor.  
Guaranteed intermittent and it won't come loose!

Second, too small of a pin requires you to either crimp the wire on the 
insulation (strain relief) section of the pin, or cut a few strands out to 
make the wire gauge fit the pin.  High resistance and poor strain relief are 
the result.  Your life depends on the correct contact area of the crimp 
connector making good contact with the correct amount of wire;  too little 
and high resistance and heat are the result.  The failure may not show up for 
a year or two and Murphy's Law stipulates that it will be in a totally 
inaccessible area of the airframe and the connector will be nicely melted 
when you find it, ruining many good connections and other systems that were 
properly installed. <LOL>  Too large of a pin and you have to over crimp just 
to keep the wire from falling out (see previous paragraph).  Please do not 
stuff more small gauge wire into a large pin to make a small wire crimp 
properly.

Third, crimp pins and splices have two crimp areas:  a small diameter for the 
conductor and a larger one for the insulation.  There is a VERY narrow range 
of stripping dimensions that put the correct amount of wire into the 
conductor area and leave enough insulation to be properly crimped in the 
second (strain relief) area.  Under strip and you deprive the conductor area 
of the required contact (see previous paragraph on high resistance).  Over 
strip and the strain relief area has to crimp on wire instead of insulation.  
If there is motion that the strain relief was supposed to take care of, this 
motion is now focused on an under crimped hunk of wire instead of a correctly 
crimped area of insulation.  The only bright side is that when it fails it 
may fall off and be readily visible upon inspection.  Soldering a crimp 
connector pin or lug melts the insulation that should be providing strain 
relief and forces a stress riser where the solder ends as it wicks up the 
wire.  I have seen a lot of these fail.  If you are going to use a crimp 
connector, crimp it properly and trust it.  If it is in a dirty area, adding 
a piece of heat shrink or self vulcanizing tape insulation is perfectly 
acceptable since their effect is OUTSIDE the connection.  Solder was never 
planned for by the engineer who designed and certificated that crimp.

Last, the strain relief of the insulation area crimp to the wire insulation 
is NOT capable of very much motion or fatigue resistance.  Let's face it, 
we're looking at a very thin metal sleeve pressed up against some PVC or PTFE 
insulation;  bend it and it cuts.  The connection must have external strain 
relief.  For a pin connector this comes in the form of the back shell and 
clamp provided.  The wires should be bundled and clamped such that any 
tension on the bundle pulls only on the clamp and none of the wires from the 
clamp to the pins are under strain.  Especially in homebuilts, I can't count 
the number of connectors I see every year with no back shell ("too much 
trouble" is the most common excuse...)  On a wiring harness, the strain must 
be distributed to the other wires in the bundle;  and that is a whole can of 
worms on its own.

<< But in any event, do not splice where the wires can flex. Anchor both 
sides of the splice down, eliminate all motion at the splice, or expect a 
failure, sooner or later.>>

Sound advice.  A crimp splice in a wire bundle will abrade any other wire it 
comes in contact with.  For that matter, so will wires that are crossed 
rather than combed (all wires in a bundle running parallel).  This issue is 
nonlinear with respect to the frequency of vibration in the airframe;  i.e., 
it becomes critical in turbine aircraft.  All Walter users out there need to 
be more diligent towards their wiring workmanship than the piston users.

Solder is great stuff.  It takes a discontinuous surface of stranded wire and 
forms a continuous conductor to the properly designed-for-solder connector 
pin or another wire.  It is almost impossible to have higher resistance in a 
good solder connection than in the wire itself.  Heat shrink or self 
vulcanizing tape insulation work well for coverage and minimal strain relief 
IF care has been taken to eliminate all angle from the connection and sharp 
edges from the solder and wire.  In connectors, I highly prefer mil spec 
soldered pins;  though I rank the skill required to rework these without over 
temping them right up there with TIG welding.  If used correctly, mil spec 
crimp pin connectors properly stripped, crimped with a calibrated racheting 
crimp tool, and installed with correct back shell strain relief are the 
fastest and most reliable connection available.  The trouble is, almost no 
one does them correctly.

On a butt splice in a wiring harness, there is no contest.  Proper solder 
joints beat crimp splices for piston aircraft.  Since neither is acceptable 
for high frequency due to the rubbing problem on adjacent wires, what do we 
do if we have to repair a wire on a turbine aircraft?  We replace the whole 
wire, end to end.  We sometimes got away with doing a splice and covering it 
with enough self vulcanizing tape to prevent adjacent wire abrasion.  
Remember all of this before you bury your wiring in inaccessible and/or 
uninspectable locations.  The V-22 program is having real fun with this right 
now.

Eric Ahlstrom
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
LML website:   http://www.olsusa.com/mkaye/maillist.html
LML Builders' Bookstore:   http://www.buildersbooks.com/lancair

Please send your photos and drawings to marvkaye@olsusa.com.
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>