Lectromec has been testing wire for the past thirty years. We have seen nearly every conceivable variation of aircraft wire testing. Some tests are straight forward and require little or no additional explanation of the test results; other tests require a fair bit of explanation to understand the test itself and what the results mean. One question that Lectromec frequently gets when clients receive the results of arc track resistance tests is, “Why is it that the higher resistance circuit configurations result in more damaging arcing events (the ones that cause the most damage to other wires in the test harness)?”
Circuit Setup
Text-based descriptions of arc track resistance testing can be confusing. If you would prefer to start with a video presentation, you may want to watch this two-minute video.
When performing arc resistance tests, the standard practice is that several circuit resistance values are used. In the circuit diagram below, the adjustable resistors (Rf) range from 0.0Ω (no added circuit resistance) to 2.0Ω for a 20AWG wire test. Three tests are performed at each circuit resistance (defined in the particular test standard) and the pass-fail criteria of the test are wire specification specific (see Lectromec’s What do the Results of Arc Track Resistance Testing Mean? article on this.
The minimum test resistance setup (0.0Ω configuration) will yield the highest fault current, and the maximum test resistance setup (2.0Ω configuration) will yield the lowest fault current. It would seem to make sense that a higher current will cause more damage. After all, this would yield the highest electrical power and probably cause the most damage. However, like all things involving Electrical Wire Interconnect Systems (EWIS), everything seems simple until you start getting deep into it.
Stopping the Arcing Event
There are two ways that an arc can stop: the electrical power is removed from the circuit (circuit protection activation) or the distance between the arcing wires is too great and voltage breakdown cannot continue. For the purpose of today’s discussion, we will focus on the circuit protection activation.
The following figure shows a typical circuit protection trip curve. At high currents, the circuit protection activates very quickly, at lower currents the activation time is much shorter. The green band in the figure shows typical operating temperatures. The outside bands above and below show representative trip times at elevated and sub-zero temperatures respectively.
Example Arcing Event
The following table shows a couple of representative cases for the circuit resistance and total energy in an arcing event. In the case shown here, three assumptions are made:
- 0.3Ω of resistance exists between the generated and the test stand (from electrical connections and wiring).
- The voltage used in the testing is 208V RMS phase-to-phase.
- The circuit protection used in the testing is a 7.5A thermal circuit breaker (as called out for in the test specification).
In the 0.0Ω condition, the total circuit resistance during an arcing event, plus the arc itself, is about 0.85Ω. The RMS current is about 32 times greater than the circuit breaker rating. This is far beyond the typical trip curve shown in the previous figure and can be assumed to cause circuit protection activation after about 0.05seconds or about 20 cycles at 400 Hz (verified by testing). This short event creates a scenario generating about 2500J of arc energy.
In the 2.0Ω condition, the total circuit resistance during an arcing event, plus the arc itself, yields a resistance of 4.85Ω, a roughly five time greater circuit resistance compared to the 0.0 Ω circuit resistance. The RMS current for the 2.0Ω test is about six times greater than the circuit breaker rating – an overcurrent level within the circuit protection specification. The trip time ranges from 0.3 – 0.8 seconds and, for the calculation shown, 0.55 seconds is used. Based on these calculations, the 2.0Ω configuration would generate an arcing event lasting 11 times longer and a total energy of about 4900J. This energy is almost two times greater than in the 0.0Ω configuration.
Summing It Up
The arc track resistance test is among the most complicated test performed on aircraft wires. Every part of the circuit design and wire construction can have an impact on the results. With regard to the test results, the arc resistance tests are more than just voltage times current – the circuit protection has a significant impact. Contact Lectromec to find out more about Lectromec’s arc track resistance testing or any of our lab services.