1AIRCRAFT CONNECTOR BONDING RESISTANCE TESTS AND MATERIALS ANALYSIS

Co-authors

John A. Ziegenhagen (Retired)University of Dayton Research Institute, 300 College Park, Dayton, OH 45469-0137, USA

James C. HierholzerUniversity of Dayton Research Institute, 300 College Park, Dayton, OH 45469-0137, USA

Edward L. WhiteAir Force Research Laboratory, Materials Directorate, AFRL/MLSA

2179 12th St, WPAFB, OH 45433-7718, USA

2ABSTRACT

This paper discusses aircraft electrical connectorbonding resistance tests conducted by the ElectronicsFailure Analysis Laboratory, Materials IntegrityBranch, Air Force Research Laboratory(AFRL/MLSA) at Wright-Patterson Air Force Base,Dayton, Ohio. The electrical bonding resistancebetween a MIL-C-38999 cadmium and nickel-platedaluminum connector and mating surface has beenfound to increase over time. Indications suggest thegrowth of an oxide film on the faying surfaces beingresponsible for the increase in resistance. The initialevaluation attempted to determine the causes of thisphenomenon by conducting connector bondingresistance tests and materials analyses usingconventional cadmium-plated connectors. Time-based bonding resistance tests were then conductedusing connectors plated with nickel-boron (Nybron)and Ion Vapor Deposited (IVD) aluminum. Resultswere compared to the data obtained usingconventional cadmium-plated connectors. At a laterdate, additional bonding resistance tests wereconducted. Connectors made of bronze-aluminumalloy, stainless steel alloy, and titanium alloy werecompared with conventional aluminum alloyconnectors plated with cadmium and nickel. Testswere conducted both at room ambient conditions andin a salt fog environment. IVD aluminum-plated andtitanium alloy connectors show promise for providingeffective corrosion control with sufficiently lowbonding resistance. Two workshops were conductedat Wright-Patterson Air Force Base to addressconnector bonding issues. Also presented areresistance measurements obtained from actual F-16and F-15 aircraft in storage at Davis-Monthan AirForce Base. These measurements were madebetween various connectors and the associatedairframe components to which they were mated.

INTRODUCTION

Electrical connectors in aircraft come in awide variety of shapes, sizes, configurations, andmaterials. There may be between 500 and 2000 ofthem on a modern aircraft. Connectors are typicallyconstructed of aluminum, titanium, stainless steel, orcomposite materials. Cadmium plating is usuallyapplied to connectors used in Air Force weaponsystems because it is an excellent corrosion inhibitorand has relatively high electrical conductivity. Achromate conversion coating is applied to achieveadditional corrosion protection. The MIL-C-38999connector is a common type found on aircraft andconsists of a chromate conversion coating over

cadmium plating which, in turn, is plated over nickel.The substrate is typically aluminum.

To insure uniform electrical shielding, thewhole perimeter of the connector shell must beelectrically connected to the component or aircraftframe member. The connection path must bebetween the connected members and not thefasteners, and must maintain a low resistance valueover an extended amount of time. MilitarySpecification MIL-C-38999 requires a maximumshell-to-shell bond resistance of 2.5 milliohms for thecadmium-plated (Class W) connector. However, anAir Force laboratory study concluded that bondedcadmium and aluminum surfaces allow the formationof an insulating oxide layer that causes electricalbond deterioration. Bond deterioration is notacceptable in an electronic grounding path because itcan lead to poor electromagneticinterference/electromagnetic pulse (EMI/EMP)protection and, therefore, poses flight safety risks.

DISCUSSION

Connector Bonding Measurements on StoredAircraft

As part of an aging aircraft study, it wasdesired to measure the connector bonding resistanceof connectors and their points of attachment that havebeen in place on aircraft over a long period of time.This was done to determine the effect of long-termusage and storage on the electrical bonding ofconnectors to components and aircraft structuremembers. Access was gained to two F-15s and oneF-16 aircraft in storage at the Aerospace Maintenanceand Regeneration Center (AMARC) at Davis-Monthan Air Force Base. The aircraft selected werelargely intact and had been prepared for long-termstorage in the desert. Figure 1 shows one of theaircraft selected, an F-15, with its armament baycover unlatched.

Figure 1 F-15 at Davis-Monthan AFB

3Tail numbers indicated the F-15 aircraftwere manufactured in 1973 and 1976 and the F-16was manufactured in 1982. The F-15s were stationedin Hawaii, and the F-16 was based in Florida. Accesswas gained to several areas of the interior of theaircraft through landing gear wells, access panels,etc. A variety of bulkhead connectors were accessed,as well as some connected to avionics. Most of theconnectors were of circular configuration (figure 2).

Figure 2 Connectors inside F-15 aircraft

Resistance measurements were madebetween the connector shells and the bulkheads oravionics using the four-point probe system connectedto a Keithley digital multimeter. Nineconnector/bulkhead assemblies were measured onone F-15 aircraft, and four from another F-15. It wasattempted to take measurements from connectors inapproximately the same location on both aircraft.Eight connector/bulkhead assemblies and threeconnector/avionics assemblies were measured on anF-16 aircraft. Resistance measurements variedwidely, and ranged from 3.0 milliohms on aconnector mated to a bulkhead located inside a wheelwell of an F-15 aircraft to 8400 milliohms (8.4 ohms)on a connector attached to a bulkhead located insidean F-15 aircraft underneath the cockpit. Thereappeared to be no correlation between bondingresistance measurements and the age or location ofthe connector on the aircraft.

Cadmium-Plated Connector Bonding ResistanceTests and Materials Analysis

A project was initiated at AFRL/MLSA todetermine the cause of increasing bonding resistancebetween cadmium-plated electrical connectors andtheir mating surfaces over time. Military qualifiedbulkhead-style connectors and chromated aluminummating plates machined to accept them were obtainedfor this evaluation (Figure 3).

Figure 3 Test connector and mounting plate

Cross sections of the connector platingsystem were examined in the scanning electronmicroscope (SEM) using energy dispersivespectroscopy (EDS). This was done to determine thethickness and elemental composition of the platingsand base metal for both the connectors and mountingplates. A polished cross-section of the connectorflange was prepared to show the disposition andthickness of the various platings and base metal(figure 4).

Chromate

Cadmium

Nickel

Aluminum

Figure 4 SEM micrograph of polished cross sectionof cadmium-plated aluminum showing plating layersand thicknesses

Secondary ion mass spectrometry (SIMS), a surfaceanalysis technique, was used to determine theelemental composition of the surface of theconnectors and mating plates. This is a much moresensitive analytical technique than EDS.

Three sets of connector and test panelassemblies were prepared. The connectors werebonded to the panels with screws and nuts electricallyinsulated from the rest of the assembly with Teflontape and Nylon washers. Two assemblies wereprepared with their screws tightened to 18 inch-ounces and the other assembly tightened to 96inch-ounces of torque (field value). One assembly

4from each torque value was tested under roomambient conditions. The other assembly prepared atthe lower torque value was tested under a pureoxygen atmosphere in a glass jar (figure 5).

Figure 5 Three assemblies under test

The assemblies were bonded together for 55days. Electrical resistance was measured daily usingthe four-point probe test method (figure 6).

0.154 mVDC

0.10000 ADC

Voltmeter

Ammeter

D.C. Power Supply

Switch

Four Point Probe Method

+

+

+ _

_

_

Figure 6 Four point probe resistance test

Resistance values climbed steadily over timeon all three samples, exceeding 7 milliohms and 3.5milliohms for the lower and higher torqued samplesin air, respectively. The assembly in pure oxygenexceeded 250 milliohms at the end of the 55 dayperiod. The addition of oxygen increased the rateand final bonding resistance value. Application ofadditional pressure and maintaining a constantpressure at the faying surfaces reduced the rate ofbond resistance increase and final value.

Surface analysis of connector and aluminumplate surfaces before and after exposure tests revealedthe presence of a large number of elements thatreadily form oxides. The examination of bondingsurfaces also shows that only a small area is actuallyin intimate contact. This appears to be a function ofthe surface disparities and distortion that can occur

by fastening the connector at the corners or edges.The increase in bonding resistance is most likely dueto the small area physically making contact betweenthe connector and aluminum plate surfaces and thepresence of various oxides at these contact points.AFRL/MLSA report WL/MLS 97-069 discusses thisproject in detail.

Electrical bond deterioration is a drivingforce behind the decision to replace cadmium inconnector manufacture. Another reason for replacingcadmium deals with the cadmium plating process. Ituses hazardous chemicals and generates hazardouswaste. The Environmental Protection Agency (EPA)has targeted 17 hazardous substances for reduction orelimination because of the quantities used, toxicity,persistence, and mobility. Cadmium is one of thechemicals on this list. The Air Force is currentlyeliminating the use of these EPA-17 chemicals toreduce exposure of workers to these materials, reducemaintenance costs, and to meet increasingly stringentpollution control requirements.

Nybron Plating

Other coating systems were examined thatmay be less susceptible to oxide formation whilemaintaining a stable and acceptable bond resistancelevel between faying surfaces. AFRL/MLSAsuggested plating the connector surfaces with anickel-boron material, trade-named Nybron.AFRL/MLSA had found this plating to be effectivein several structural applications and believed its lowelectrical resistance would also make it an effectiveconnector plating. An investigation was initiated tocompare changes in bonding resistance betweenNybron-plated and cadmium-plated connectors andtheir mating surfaces over time.

Several connectors were plated with anickel-boron (0.42 percent boron), or Nybron,plating. A fracture face cross section of a Nybron-plated connector was examined in the SEM (figure7).

Figure 7 Nybron plating fracture face

5A Nybron-plated connector and aconventional cadmium-plated connector were matedto chromate-coated aluminum plates. Care was takento isolate the mounting screws from the assembly, asin the previous experiments. The screws weretightened to 96 inch-ounces torque.

Both assemblies were tested at roomambient conditions for 55 days. Bonding resistanceof the assembly using the cadmium- plated connectorincreased steadily over time. In contrast, the bondingresistance of the Nybron-plated connector assemblyremained less than 0.5 milliohm. AFRL/MLSAreport WL/MLS 97-095 discusses this project indetail. Nickel-boron plating may be considered whenthere is a long-term, low electrical bond resistancerequirement. Environmental concerns must be takeninto account, however, due to the presence of theelement boron in the plating.

Ion Vapor Deposited Aluminum Plating

In view of this, AFRL/MLSA proposed theuse of connectors plated with ion vapor depositedaluminum (AIVD). This material has distinctadvantages over cadmium and nickel plating onaluminum alloy connectors. AIVD plating providesthe necessary conductivity and corrosion protection,while at the same time being environmentallyfriendly. Both the AIVD plating and the applicationprocess are nonpolluting. The plating is purealuminum and is applied electrically; no chemicalbaths are used.

It was decided to compare bondingresistance changes of cadmium-plated and Nybron-plated connectors to AIVD-plated connectors andtheir mating surfaces over time. Several rectangularconnectors and machined aluminum plates wereAIVD-plated and conversion coated. A polishedcross section of a connector so treated was examinedin the SEM (figure 8).

Figure 8 AIVD connector plating cross section

The one-to-two micron thick chromateconversion coating can be seen at the top of the crosssection. Below this was a striated layer about 12microns thick. It is assumed this is the area that hadbeen compressed from the shot peening procedure.This seals the AIVD coating and is generally done bythe plating specialist as a corrosion preventionmeasure. A cross-sectional fracture face of thisconnector was also prepared (figure 9). Note the topseven microns of plating has a somewhat differentappearance than the remainder. Also note thecolumnar grain structure, which is typical of theAIVD plating process.

Figure 9 AIVD plating fracture face

Several sets of connector and test panelassemblies were prepared using the same bondingprocedures as with the Nybron plating testing. TwoAIVD assemblies were tested at room ambientconditions for a period of 60 days. Another AIVDassembly was subjected to a salt fog exposure in achamber adjusted according to ASTM B-117. Thechamber utilized a continuous salt fog spray. Theexposure time was 500 hours. Also included in the

6salt spray test were cadmium-plated and Nybron-plated connector assemblies for comparison.

Resistance values for the AIVD assembliestested in room ambient conditions were very lowinitially, and did not increase significantly during thetest period. They remained well within theacceptable limit of 2.5 milliohms. Bondingresistance of the IVD aluminum and cadmium-platedsamples in the salt fog rose gradually, exceeding 20and 30 milliohms, respectively, by the end of the test.Bonding resistance of the Nybron-plated connectorassembly climbed very rapidly in the salt fogchamber, exceeding 140 milliohms after 500 hours.The amount of corrosion and deposits increaseddramatically on the Nybron test article. Incontrast, the other two assemblies did not shownearly as much of an increase in corrosion productsover time (figures 10, 11, and 12).

Figure 10 Nybron-plated assembly after salt fog exposure

Figure 11 AIVD-plated connector assembly aftersalt fog exposure

Figure 12 Cadmium-plated connector assembly after salt fog exposure

These experiments indicated the IVDaluminum-plated connectors show promise to provideeffective corrosion control, as well as sufficiently lowbonding resistance. They also indicated the need todevelop methods for measuring and controllingconversion coating thickness and composition. Thisis because electrical bonding appears to be highlydependent on coating thickness, especially theconversion coating. Ion vapor deposited aluminumwas evaluated in AFRL/MLSA report 98-75.

Bronze-Aluminum, Stainless Steel, and TitaniumAlloys An effort was made to find connectormaterials which would give good electricalcharacteristics without the need for platings. Flange-mounted circular connectors made of bronze-aluminum alloy and stainless steel were obtainedfrom the Royal Air Force and J-Tech (Tustin,California), respectively. A jam-nut mountedtitanium alloy circular connector was obtained fromAllesandro Automatic (Los Angeles, California).Bonding resistance tests were conducted comparingthese with conventional cadmium and nickel-platedaluminum alloy connectors. Tests were conducted atroom ambient conditions and in a salt-fogenvironment. Resistance measurements during roomambient tests were acquired with an in-house deviseddata acquisition system that provided continuous 24-hour sampling and storage without the assistance ofan operator (figure 13). This apparatus is shownschematically in figure 14.

7Figure 13 Data acquisition and control interface

To AnalogInput on I/OBoard

Relay

Four Point Probe Data Acquisition System

_

Current Source

From I/O BoardDigital Output

_

+

Power Supply

1k Ohms

1k Ohms

100k Ohms

100k Ohms

500 Ohms5 Ohms40 Watts

Screws electricallyisolated from plate andconnector

Sleeve Washerfor electricalisolation

Screw

+

24 VDC

+ _1 Amp

Figure 14 Schematic representation of test fixture

Room Ambient Tests

After cleaning, the connectors were mated toaluminum mounting plates which had beenconversion coated with alodine chromate conversioncoating. The flanged connectors were fastened withstainless steel screws and nuts to the mounting plate(figure 15). The titanium connector was fastened witha jam nut (Figure 16). To prevent the fasteners fromcontributing to the conductivity of the assemblies, thescrews were wrapped with Teflon tape and fiberwashers were used under the nuts. A fiber washerwas also used under the jam nut for the titaniumconnector. The outside surface of the connector waswrapped in Teflon tape to isolate it from themounting plate. The screws for the three flangedconnectors were torqued to 96 inch-ounces with atorque wrench. The jam nut on the titaniumconnector was tightened to 30-36 inch-pounds. Thisis the recommended torque value for a size 9connector shell. Initial resistance readings weremade using a four-point probe measurement system(AEMC Micro-ohmmeter, Model 5600). This testwas conducted at room ambient conditions. Thetemperature and humidity was respectively 72F and

the 38% RH. Resistance measurements during roomambient tests were acquired with an in-house dataacquisition system that provided continuous 24-hoursampling and storage without the intervention of anoperator (figure 13).

Figure 15 Bronze-aluminum connector mounted

Figure 16 Titanium alloy connector mounted

Initial measurements were made bysampling every hour for the first 24 hours.Subsequent measurements were taken every fourhours thereafter. Previous experience has shown thebonding resistance changes rapidly in the first 24hours. Data were acquired continuously for a periodof two months. Of the four connector materialstested, only the cadmium-plated aluminum typeexceeded the specified resistance limit of 2.5milliohms during the test (Charts 1 and 2). It failedin the first 24 hours of testing. The remaining threeconnector assemblies increased slightly in the first 24hours and then remained nearly constant throughoutthe test (Chart 3).

8In Air

0

1

2

3

4

5

6

7

0 2 4 6 8 10 12 14 16 18 20 22 24

H o u r s

Res

ista

nce

in M

illio

hms

Bronze AluminumTitanium Al loy

S ta in less S tee lCadmium Plated Aluminum

Chart 1 Bonding resistance at room ambientconditions of the four connector materials (first 24hours)

Cadmium Plated Aluminum

0

5

1 0

1 5

2 0

2 5

3 0

3 5

28-F

eb

29-F

eb

1-M

ar

2-M

ar

3-M

ar

4-M

ar

5-M

ar

6-M

ar

7-M

ar

8-M

ar

9-M

ar

10-M

ar

11-M

ar

12-M

ar

Date

Res

ista

nce

in M

illio

hms

Chart 2 Bonding resistance of the cadmium-platedaluminum connector at room ambient conditions

60 Days

0.0000

0.1000

0.2000

0.3000

0.4000

0.5000

0.6000

0.7000

0.8000

28-F

eb

3-M

ar

7-M

ar

11-M

ar

15-M

ar

19-M

ar

23-M

ar

27-M

ar

31-M

ar

4-A

pr

8-A

pr

12-A

pr

16-A

pr

20-A

pr

24-A

pr

28-A

pr

Date

Res

ista

nce

in M

illi

oh

ms

Bronze Aluminum

Titanium Alloy

Stainless Steel

Chart 3 Bonding resistance of bronze-aluminum,titanium alloy, and stainless steel at room ambientconditions for 60 days

Salt Fog Tests

A second set of four connectors were testedin a salt fog environment, according to ASTM B-117-97, (Standard Practice for Operating Salt Spray (Fog)

Apparatus). Upon removal from the salt fogenvironment, the connector assemblies were gentlyrinsed in warm water and a soft acid brush was usedto clean the salt deposits from the surface. Theassemblies were then allowed to dry at room ambientfor one hour. The assemblies bonding resistance wasmeasured with the data acquisition system describedpreviously.

Initially, the connector assemblies wereremoved from the salt fog environment and testedevery 24 hours for the first 4 days. Thereafter, theassemblies were removed from test every two daysfor measurements. This process was continued forfour weeks or until the connector reached a resistanceabove the failure limit of 2.5 milliohms. Notedduring this test was the heavy buildup of whitecolored deposits around the mating areas between theconnector and mounting plate for each of theconnector assemblies. Of the four connectors tested,only the titanium alloy connector was able tomaintain a resistance value below the 2.5 milliohmsfailure limit for the four-week period. The cadmium-plated aluminum connector failed in just one day; thebronze-aluminum connector lasted two days, whilethe stainless steel connector failed after five days(Chart 4).

Salt Fog

0.000

5.000

10.000

15.000

20.000

25.000

6/5/

00

6/7/

00

6/9/

00

6/11

/00

6/13

/00

6/15

/00

6/17

/00

6/19

/00

6/21

/00

6/23

/00

6/25

/00

6/27

/00

6/29

/00

7/1/

00

7/3/

00

7/5/

00

Date

Res

ista

nce

in M

illio

hms

Titanium AlloyCadmium Plated AluminumBronze Aluminum

Stainless Steel

Chart 4 Bonding resistance of the four connectormaterials in a salt fog environment

All connector assemblies tested in the saltfog environment were found to have a heavy buildupof white colored deposits between the faying ormating areas of the connector and mounting plate.Elemental analysis determined these deposits to besodium and chloride ions as expected from the saltfog spray. Deposits were more pronounced on thefaying areas of the cadmium-plated aluminum andbronze-aluminum connector assemblies (figures 17).The titanium alloy, as well as the stainless steelconnector assemblies, exhibited less salt fog depositson these faying areas than the other connectors(figure 18).

9Faying areas

Figure 17 Heavy deposits from salt fog on fayingareas of bronze-aluminum connector and mountingplate

Faying areas

Figure 18 Titanium alloy connector and mountingplate with lighter deposits on faying areas from saltfog

WorkshopsTwo workshops were conducted at Wright-

Patterson Air Force Base in 1996 and 1997 to addressthe issues of connector bonding. Representativesfrom various U.S. government agencies,manufacturers, and universities were in attendance.The Royal Air Force and the Defense Evaluation andResearch Agency of the United Kingdom were alsorepresented.

Both workshops featured presentations by anumber of the attendees. Topics included bondingresistance tests and materials analyses of connectorsutilizing cadmium plating and possible replacementplating materials. These included AIVD, nickel-boron, thermal sprays, and other novel platingsystems. Other issues were discussed, includingsurface treatments, conductive gels and gaskets, andconnector fastening procedures.

A number of candidate solutions wereproposed. These included mechanically removing

the plating from the mating surfaces, followed by theimmediate application of a conversion coating andassembly. A non-conductive sealant would then beapplied. Also suggested was the use of conductivegrease to be applied in a controlled factoryenvironment, and conductive sealants, particles, andgaskets. Alternative finishes and materials wereproposed, such as Ion vapor deposited (IVD)aluminum, nickel-boron alloys (Nybron), titanium,and nickel-plated composite connectors. Methods toincrease faying surface contact were also suggested,along with using the fasteners as part of theconductor path.

As a result of the discussions by those inattendance and the topics presented it was agreedchromate conversion coatings are fraught withproblems and need to be replaced, if for no otherreason, for environmental concerns. It wasrecognized conductivity is presently obtainedprimarily through the fasteners and that the fayingsurfaces are, in reality, very small. It is desirable tohave increased intimate contact area. This can beaccomplished by roughening the mating surfaces tobreak through the oxide layer. Sealers should beused to reduce oxygen exposure at the bond interface.

Conclusions

Cadmium-plated aluminum connectorsmounted to aluminum plates exceeded 2.5 milliohmsbonding resistance on their faying surfaces both inthe room ambient and salt fog tests in approximately24 hours. The bronze-aluminum, titanium alloy, andstainless steel assemblies, after showing a minorincrease in bonding resistance initially, were able tomaintain a steady resistance value far below the 2.5milliohms maximum limit during the room ambienttest. The bronze-aluminum and stainless steelconnector assemblies bonding resistance exceeded2.5 milliohms during the salt fog test in two and fivedays respectively. Only the titanium alloy connectormaintained a bonding resistance below the 2.5milliohms limit. The titanium alloy may havemaintained a low bonding resistance due to a bettermetal to metal seal between connector and plate.This may be due to the jam nut fastening to themounting plate.

Elemental examination of surfaces show theincrease in bonding resistance is likely due to thesmall area physically making contact between thecadmium-plated aluminum connector and aluminummounting plate. Increased resistance during the saltfog test was due to the salt fog penetrating the poormetal-to-metal seal between the connector andmounting plate for the bronze-aluminum, cadmium-

10

plated aluminum and stainless steel connectorassemblies.

System requirements appear to be themotivating force for resolving connector bondingissues. Designs and analyses are based on the 2.5milliohm maximum connector bonding resistancerequirement, for example. Bond measurements arenot regularly or periodically verified; whenmeasurements are made, however, bonds often failthe test. There is universal concern for meeting EMIrequirements and for replacing platings and coatingsthat may not be available in the future.Environmental requirements are becoming morestringent, resulting in fewer acceptable materials. Itwas mutually agreed a team concept was needed tosolve problems. This team would include thecustomers, suppliers, scientists, and engineers.

Consideration should be given toredesigning the connectors and their method offastening and bonding. The use of additionalfasteners and/or increasing the torque on the presentfasteners may improve the opportunity for gas-tightseals between the connector and mounting plate.Bonding tests should be conducted using jam nuts forfastening and bonding connectors made of materialsother than titanium. The jam nut method of fasteningappeared to effect a better electrical bond than thescrews in the corners.

11

BIBLIOGRAPHY

Connector Bonding Resistance Tests and Materials Analysis , Evaluation Report No. WL/MLS 97-069, 13 August 1997

Nybron-Plated Electrical Connector Bonding Resistance Test, Evaluation Report No. WL/MLS 97-095, 27 October 1997

Connector Bonding Resistance Tests Using Ion Vapor Deposited Aluminum Plating , EvaluationReport No. AFRL/MLSA 98-75, 20 April 1998

Connector Bonding Resistance Tests and Materials Analysis, Evaluation Report No. AFRL/MLSA00-107, 23 January 2001

1996 & 1997 Electrical Connector Workshop Notes , sponsored by AFRL/MLSA and ASC/ENAE,compact disc format, available from AFRL/MLSA WPAFB, OH 45433