ARCING FLAMMABILITY FACTORS

Michael F. Walz Aging Electrical Systems Program Manager Cesar A. Gomez Aging Electrical Systems Engineer Airport and Aircraft Safety Research and Development Division William J. Hughes Technical Center Atlantic City International Airport

I. ABSTRACT
Aircraft safety relies in part on the design, integration, installation, maintenance, and proper performance of the electrical wiring interconnect system (EWIS) throughout an aircrafts operational life. Aircraft are designed with redundant flight critical systems to improve safety and tolerance to faults. An important aspect of the EWIS design is the ability of the wiring to safely react to failures during normal and abnormal aircraft conditions. The EWIS design should prevent the propagation of the effects of electrical faults to other independent power sources and critical systems, which can happen due to aircraft operation, maintenance, and human factors. The Federal Aviation Administration (FAA) conducts EWIS research on arc damage assessment, separation and segregation of electrical systems, and quantification of damage to adjacent electrical and nonelectrical systems during EWIS failures. The Arc Fault Evaluation Laboratory (AFEL) at the William J. Hughes Technical Center is used to perform research in this field. The AFEL is currently conducting a test program to evaluate damage tolerance on the EWIS and adjacent structures and systems due to fires and electrical faults. This paper presents an assessment of wire flammability thresholds and damage tolerance of the EWIS and cascading effects due to fires and electrical arcing inside the aircraft. Experimental variables include material types, dust and lint accumulation, and circuit protection. This study also presents data on the potential of the variables to increase or reduce the potential of flammable events. The effects of dust and lint on chafed insulation were included in the assessment to determine if these contaminants contribute to creation of a carbonized path and subsequent electrical arcing in the presence of fire. Results of this study will assist the development of guidelines for EWIS hazard mitigation and design criteria.

II. INTRODUCTION
The Federal Aviation Administration (FAA) is conducting research on the damage tolerance of the Electrical Wiring Interconnection System (EWIS) under the stress of an aircraft fire. The research is being conducted in the Arc Fault Evaluation Laboratory (AFEL) at the William J. Hughes Technical Center in Atlantic City, New Jersey. EWIS separation and segregation, circuit protection, and material selection are important aspects of wiring system design; and properly applied increase the ability of the EWIS to adequately mitigate failure effects during normal and abnormal aircraft conditions. This project is developing data on EWIS damage tolerance in a fire condition and the ability of the EWIS to safely transmit power, avoiding catastrophic damage and improving aircraft safety. The results presented in this paper will assist the development of EWIS design guidelines that can improve EWIS safety when exposed to an aircraft fire condition.

III. BACKGROUND
Aircraft safety relies in part on the design, integration, installation, maintenance, and proper performance of the EWIS throughout an aircrafts operational life. An important aspect of the EWIS design is the ability of the wiring to safely react to failures during normal and abnormal aircraft conditions. Separation and/or segregation of EWIS and electrical equipment are important factors for ensuring that the functional and physical integrity of critical systems is maintained after a failure in the EWIS. Aircraft are designed with redundant flight critical systems to achieve acceptable margins of safety. Separation is a measure of distance and addresses physical hazards resulting from potential electrical failure modes. Adequate separation is a function of the physical and electrical attributes of the wires in the bundle and the hazard potential of a failure at any given point along the length of the bundle. Segregation involves the physical segregation of wires into separate bundles. The FAA and aircraft manufacturers and suppliers are constantly evaluating new materials and their resistance to fire. The Code of Federal Regulations (CFR) part 25 Appendix F mandates two flammability tests for electrical wire: the 60-degree test, and the vertical test. All wire used on aircraft must pass these tests (There are exceptions for special conditions). Improved fire resistance enhances the ability of the EWIS to resist damage during an aircraft fire. Current test standards and research by the FAA and industry on wire flammability are conducted without power applied to the wire and the insulation in pristine condition. While this gives a good indication of how the wire insulation material will react in a fire there are other considerations that must be taken into account to fully understand the hazard potential to an aircraft. An extensive literature search on the effects of carbonization of the wire insulation and contamination on the EWIS was performed. The results of the search did not yield any comprehensive conclusions. During an aircraft fire the wire insulation carbonizes; creating additional unintended conduction paths. It can also create external induced ionization of air. During a fire, ionization gases may be present, reducing the dielectric constant of air; thus resulting in an increased likelihood of electrical arcing [3]. In addition, carbonization creates additional potential arcing paths in a wire bundle which can cascade into a serious hazard condition. Aircraft malfunctions created by such an event can also lead to increased crew workload on the flight deck and upset the crews situational awareness. The flammability characteristics of dust and lint contaminants found on aircraft are not fully understood. The accumulation of contaminants and their effect on exposed wire conductors in the EWIS as a result of some high temperature source, the effect on the EWIS of a sustained and undetected fire, and the potential of these contaminants to aid the propagation of fire need to be studied. This research assesses the ability of the EWIS to safely conduct power while exposed to a flammable event and explores the effects of lint and debris during these events.

III. OBJECTIVE
The goal of this research is an assessment of the damage tolerance of the EWIS under the stress of aircraft fire, including the effect of variables such as contamination, material types, and circuit protection to increase or decrease the hazard potential.

IV. TEST SET UP

Three test methods were used in the course of this study: the Carbonization Flammability Test, the Arcing Burn-Through Test, and the Contamination Arcing Test. Test results were recorded on a Nicolet Vision digital recorder, stop watch, digital video, and a high-quality digital camera. Digital photographs were taken before and after the event and used for assessment. The power source for all tests was a 400 Hz, three-phase, 60KVA, 115 volt generator. Table 1 list the wire types used in the tests. The lint and debris contaminants obtained from various sources and all contain material commonly found in aircraft lint accumulations. The fire sources included a foam block soaked in Heptane and an improvised Heptane gas lamp. All experiments were performed using conventional aircraft thermal circuit protection and arc fault circuit protection. Table 1. Material Types Material Polyimide Mil-DTL-81381/11-20 PVC Mil-W-5086/1-20 PTFE AS22759/9-20 PTFE AS22759/34-16 A. The Carbonized Flammability Test The objective of this test is to assess the damage tolerance of powered aircraft wire insulation while exposed to a flammable event. The set up consists of a two wire bundle where one wire is connected to a 400Hz Single phase generator, a circuit breaker and a constant light load. The other wire in the bundle is connected to the aircraft return. The power wire is monitored electrically through out the test cycle. A schematic of the test set up is shown below.

Wires under test Phase B 400 Hz

The test is initiated upon application of a Heptane flame to the bundled wire as shown in figure 1. There were three different methods that were used to apply the flame on the wire. These are the Heptane candle, a small Heptane soaked foam block, and a large Heptane soaked foam block. These methods were used to achieve concentration of the hottest part of the flame on the wire. The time to burn-thru is considered to be when the monitored power wire shows excessive current or the electrical conductor is exposed. If any of these conditions was met the lapsed time was recorded.

Hottest Flame Point

Heptane Block

Phase Power and

Figure 1. Burn-through flammability test B. Arcing Burn- through Test The objective of this test is to assess the damage tolerances of the wire under an arcing event. The set up consists of a two wire polyimide bundle where each wire is connected to two independent phases of the 400Hz generator with separate circuit breaker protection. The end of the wire has exposed conductors which will produce a phase to phase arcing event. The wire under test (another two wire bundle) is connected to final independent phase and aircraft power return (identical to the Carbonized Flammability Test) the bundle is positioned on top or adja1cent to the polyimide arcing bundle. The initiation of the arcing is done by momentarily shorting out the stripped ends of the polyimide arcing wire by an external conductor and dragging out an arc to start the event. The schematic of the test set up is shown below.

Phase A

Wires under test

400 Hz

Phase B Arcing Wires

Phase C
Once sufficient carbonization of the arcing bundle has occurred resetting the arcing bundle circuit breakers continues the process. The wire under test is electrically monitored for connection to the arcing pair. The Arcing waveforms and pictures are then recorded. Each of the circuits was protected by thermal or arc fault circuit breakers. The time to breach to any conductors was recorded as the amount of arcing half cycles needed to breach to any of the generator phases and ground. The set up is shown in figure 2.

Wire Under Test

Arcing Bundle

Exposed Conductor

Figure 2. Arcing Burn-through Test C. The Contamination Arcing Test The objective of this test is to determine if the dust and lint contamination assists the creation of a carbonized path to start an arcing event in the presence of fire. A two wire bundle is prepared by creating a 1mm gap in the insulation exposing the conductor on each of the wires as shown in figure 3. A small amount of dust and lint (1 to 2mm) is placed at the exposed conductor area of the bundle as shown in figure 4. The bundle under test is powered by two independent phases of the generator or a single phase and aircraft return. The time is monitored from the moment the Heptane flame is applied to the exposed conductor area until electrical monitoring identifies conduction between the wires of the bundle. The schematic of the test set up is shown below.

Wires under test 400 Hz Phase B

The Heptane candle was used in this testing because it allow for a controllable application of the flame on the exposed conductors. For comparison purposes the tests were done with clean exposed conductors first and then with the applied contamination. Polyimide wires were used because its greater performance during the carbonized flammability tests above the others. This facilitates the observation of the dust and lint carbonization properties without any intrusion from the polyimide insulation.

Figure 3. 1mm Gap Exposed Conductor

Figure 4. 2mm of Dust and lint at Exposed Conductor

V. RESULTS
A. The Carbonized Flammability Test The results for the carbonized flammability test are shown in table 2. The polyimide insulation exhibited greater fire resistance than the other wire types and did not experience a breach into the conductor in any of the test trials. One polyimide sample was exposed to the Heptane flame for more than 30 minutes without breaching. It is possible that the Heptane flame temperatures were not sufficiently high for the carbonization of the insulation to create an arc over event. PVC insulated wire had the least fire resistance. After 45 seconds the PVC insulation was completely consumed and the conductor was breached. For AS22759/9 the flame was applied for 20 minutes and at the completion of the test both conductors were exposed. Particular to the AS22759/9 test was the absence of a large arc flash over during the test. Instead much scintillation (mini arcs) was seen until the wire insulation was consumed and the flame was consumed.

Table 2. Test results the Carbonized Flammability Test The Carbonized Flammability Test Material NO breach after 30 minutes Polyimide Mil-DTL-81381/11-20 45 seconds, 50 seconds PVC Mil-W-5086/1-20 20minutes PTFE AS22759/9-20 2:50,2:10,6:20, Ground Conductor Exposed PTFE AS22759/34-16 The AS22759/34 wire was breached in four different tests. Three monitored breaches occurred at 2:10, and 2:50 and 6:20 minutes. In the fourth test the flame was applied for 10 minutes until self extinguishing and there was no arcing indication recorded. An inspection of the damaged bundle revealed that the powered wire did not burn through all the way and the ground wire conductor was visible through the char which was noted as a breach shown in figure 5.

Ground Conductor

Powered (Phase A)

Figure 5. Breached Ground Wire Discrepancies in time to breach for the experiments in the wire bundles could be attributed to drafty conditions in the test cell, height of the flame, point of application of flame on the wire insulation, flame position during the test, flame temperature, and separation of the wires because of tie wraps failures. These sources of errors contributed to some variation in test data. However the results do reflect the variation of conditions experienced in an actual aircraft fire and still support the overall objective of the tests, to assess the overall hazards of powered wires exposed to fire. The damage after failure on all different wire insulations except polyimide followed the same trend for all. The insulation on some of the wires that were not breached became stiff and discolored which can lead to cracking of the insulation. Conductors were exposed which can be susceptible to additional arcing from contact to other conductors, structure, and fluids. Wires that experience arcing were rendered inoperable as seen in figure 6. The polyimide was covered in carbon after removal of the flame; there was no evidence of breach of insulation to the conductor. and the insulation was discolored and stiff.

Figure 6. Inoperable Wires After Arcing B. Arcing Burn-Through Test The tests results for the Arcing Burn-Through test are shown in Table 3. The time to breach to any conductors was recorded as the amount of arcing half cycles needed to breach to any of the generator phases and ground. An arcing half cycle is defined per AS5692 ARC Fault Circuit Breaker (AFCB), Aircraft, Trip-Free Single Phase 115 VAC, 400 Hz-Constant Frequency. Table 3 shows the number of arcing half-cycles incurred by each of the wires before the first breach; consequently breaches after resetting the circuit breaker are explained for each of the tests and can be seen on the waveforms recorded. Table 3. Test results for the Arcing Burn-Through Test Material Polyimide Mil-DTL-81381/11-20 PVC Mil-W-5086/1-20 PTFE AS22759/9-20 PTFE AS22759/34-16 Test Number Test 22 Test 23 Test 18 Test 19 Test 24 Test 25 Arcing Half Cycles To First Breach 229 294 Breached to Ground 85 150 66

In test number 22, phases A and C were connected to the polyimide wire under test. Phase B and ground where connected to the arcing wire. The first breach occurred after 229 arcing half cycles from Phase B to C after which the breaker was reset. After multiple resets the conductor breached from Phase B to C until the wire was burnt to an open condition. Phase A was never breached.

For test number 23, phases A and C were connected to the polyimide wire under test. Phase B and ground where connected to the arcing wire. The first breached occurred after 294 arcing half cycles from B to C. After multiple resets phases A, B, and C were breached. The resulting damage is shown in figure 7.

Figure 7. Arcing Burn-through test on Polyimide Wire For test number 18, phases A and C were connected to the PVC wire under test. Phase B and ground were connected to the polyimide arcing wire. Only one arcing event was recorded prior to exposing the conductor on the grounded wire of the PVC bundle. The thermal breaker did not trip. For Test 19 under the same configuration, the first breach occurred after 85 arcing half cycles from B to C. The thermal Breakers on phases A and C did not trip. The conductor was exposed as seen in figure 8.

Figure 8. Exposed PVC conductor For test number 24, phases A and C were connected to the PTFE wire under test. Phase B and ground were connected to the arcing wire. The first breach occurred after 150 arcing half cycles from B to A. After one reset of the breaker on Phase B, all phases were breached. The circuit breakers were reset until the polyimide arcing wire was consumed or was separated from the wire under test. After review of the damage the PTFE was rendered inoperable, both conductors on the wire bundle were breached. For test number 25, phases A and C were connected to the PTFE wire under test. Phase B and ground were connected to the arcing wire. The first breach occurred after 66 arcing half cycles from B to A. After three resets of the circuit breaker all the phases were breached and the arcing polyimide wire separated. After review of the damage

both conductors of the PTFE wire bundle were breached. Each circuit in the test was powered through an arc fault circuit breaker in line with each of the electrical phases. The arc fault circuit breaker interrupted the tests within four arcing half cycles. Per AS5692 an arc fault circuit breaker must remove power if it detects 8 arcing half cycles in a 100 milliseconds time period. The resulting damage to the wire under test was minimal and there was no breach to any of the electrical phases. A recorded arc fault waveform can be seen in figure 9.

Figure 9. Arc Fault Waveform The position of the polyimide arcing wires was significant. When they were on the side of the tested bundle the time to breach was greater and the damage effect was less than when the arcing wire was underneath. The adhesive tape used to hold the two bundles together became flammable during some tests creating additional damage to the insulation. The arc tracking characteristics of polyimide cause the arc event not to be in a constant position to the wire tested. C. The Contamination Arcing Test Polyimide wires were used because of the higher fire resistance exhibited during the carbonized flammability tests. The Heptane candle was used in this testing because the application of the flame on the exposed conductors was more controllable. Only a few tests were conducted to validate the test method for future studies. However the preliminary results show that dust and lint contamination greatly affected the carbonization polyimide with a minimum amount applied. Two types of tests were done, one on pristine exposed polyimide conductors and the other with dust and lint applied at the exposed conductors. For the pristine samples the flame was applied for over 10 minutes with no arc over flash recorded, the time was dependant on the fuel extinguishing from the candle. Then two types of lint and dust were applied to the exposed conductors, one was typical dryer lint and the other was collected from the interior of an aircraft. The carbonization was recorded as an arcing flash over. For the dryer lint the time to carbonization was 51 seconds, for

the aircraft lint the time was 1:36 seconds. The wires were rendered useless as result of the arc over flash as seen on figure 10. Variation in the type of lint and dust may have a significant affect on the carbonization process. The flame position and temperature play an important role in the time to carbonization and damage severity.

Figure 10. Damage from Arc Over Flash

VI. CONCLUSIONS
During a fire event, unintentional electrical connections can be made on the EWIS leading to increased damage, degraded situational awareness, flight deck overload, and unpredictable reactions from the electrical system, and a degraded level of aircraft safety. Preliminary results show that lint and dust can provide an electrical path that can lead to an arcing event. The composition of the lint and dust plays a significant role in the degradation of properties. Arc fault circuit breakers (per current specification) significantly reduce the chance of evolving more wires in the bundles, therefore avoiding cascading damage and failure of the wire bundle. In addition testing results show that: The Heptane flame did not generate a sufficiently high flame temperature for the carbonization of the polyimide insulation to create an arc over event. The Heptane flame does not provide sufficient ionization of the air between exposed conductors to facilitate arcing. A wire breech during a fire as well as the time to breach is dependent on the flame temperature and the type insulation material. Polyimide type insulation is more resistant to deterioration of the insulation during an open fire event than other common wire types. The time to breach was significantly faster with arcing than open flame. The temperature during an arcing event, while much higher than the n-Heptane flame, still required repeated arcs and reset of the circuit breakers to breach the wires. If an electrical fault does occur and the circuit breakers are reset, it can lead to greater hazard potential of the EWIS. Arcing events are capable of compromising a greater range of material types due to the range of higher temperatures these materials can experience during the event..

VII. RECOMMENDATIONS
Study the effects of lint and dust on the EWIS. Study the flammability properties of contaminants on the EWIS. Study the chemical properties of lint and dust from aircraft.

VIII. REFERENCES
1. 2. 3. 4. 5. 6. 7. Cahill, P., Evaluation of Fire Test Methods for Aircraft Thermal Acoustical Insulation, FAA Report DOT/FAA/AR-97/58, September 1997. Cahill, P., An Evaluation of the Flammability of Aircraft Wiring, FAA Report DOT/FAA/AR-TN04/32, December 2004. Dunki-Jacobs, J.R., The Escalating Arcing Ground-Fault Phenomenon, IEEE Trans. Ind. Appl. IA-22, 1156-1161 (1986). Babrauskas, V., How do Electrical Wiring Faults Lead to Structure Ignitions?, pp. 39-51 in Proc. Fire and Materials 2001 Conf., Interscience Communications Ltd., London (2001). AS5692 ARC Fault Circuit Breaker (AFCB), Aircraft, Trip-Free Single Phase 115 VAC, 400 Hz-Constant Frequency. http://ntsb.gov/aviation/aviation.htm http://www.tsb.gc.ca/en/reports/air/2002/A02O0123