JAPANESE REPORTS ON ELECTRICAL FIRE CAUSES

3. Phenomenon of Glow at the Electrical Contacts of Copper Wires

Y. Hagimoto, K. Kinoshita, T. Hagiwara

NRIPS (National Research Institute of Police Science) Reports- Research on Forensic Science -, Vol. 41, No. 3, August 1988

Abstract

A series of experiments were carried out to study the phenomenon of the glow at the contacts of copper and to find the current condition of practical use in the electrical fire investigation. In these experiments, the contacts of copper wires connected to AC or DC power source and a variable load resistance in series were used. The results are as follows:

1) Average DC power dissipated at the contacts of wires 1 mm in diameter was almost constantly 16 W in the range over 2 A. Average AC power dissipated at the contacts of wires 2 mm in diameter was almost constantly 28 W in the range over 5 A.

2) For 1 mm wires, the maximum breeding speed of Cu2O was 17 mm/hour with 2 A AC, and 18 mm/hour with 2 A DC, For 2 mm wires, the speed was 6 mm/hour with 2.5 A AC, and 12 mm/hour with 5 A DC.

3) For 1 mm wires, the ranges of the current necessary for the glow to continue were 0.3 - 2 A AC and 0.3 - over 8 A DC. For 2 mm wires, they were 1 - 2.5 A AC and 0.5 - over 8 A DC.

4) The upper limit of these ranges of current were affected by the spattering caused by the glowing contacts.

1. Introduction

Heat generated at an electrical poor contact is one of the causes of a fire. This poor connection sometimes results in glowing contact (ref. 1, 2, 3). If the glowing at a contact of copper wires are left, oxidized part, which contains mainly Cu2O and is called "hot zone" (ref. 4), expands (see figure 1). As the flow of current is concentrated due to the temperature-resistance characteristics of Cu2O, temperature of the current path become high and the path glows. Temperature of the "glowing path" is considered to be in excess of melting point of Cu2O because wavy traces are observed on the surface of hot zone. Typical voltage drop across the hot zone and AC current of this phenomenon is shown in figure 2. Fire hazardous aspect of this phenomenon is that it can be started with small current such as 1A and it can not be noticed because of small voltage drop.(ref. 5-8)

This study was made to know more about the heating mechanism of glowing path and the expanding process of the hot zone at the contact of copper wires.

2. Experiment

Experimental arrangement used for this study is shown in figure 3. The gap of two copper wires was controlled by a micrometer. The diameter of the copper wire is 1 mm or 2 mm. Both DC power source and AC power source were used.

Output voltage of the DC power source was kept 100V by the adjust of resister, R1. DC voltage drop across the hot zone was recorded by a DC voltage recorder.

AC current was kept constant by adjusting R1. AC current was measured by moving-coil-type ampere meter because it was almost sine wave (see figure 2).

Voltage drop across the hot zone ("v" shown in figure 3) and current ("-i" shown in figure 3 which was calculated from the voltage drop across R2 (0.1 or 1.0)) was recorded to digital memory for both DC and AC measurements.

3. Result and Discussion

3-1 Mechanism and characteristics of heating

3-1-1 White heat at the end of glowing path

In case of AC, an incandescent points (where glowing is extremely strong) were observed at both ends of the glowing path which moves around the hot zone as if it were a worm. This strong glow was observed only when the ends of the glowing path, which seemed to repeat expanding and contracting, reached the boundary between hot zone and non-oxidized copper. Non-oxidized copper at the boundary was oxidized by the strong glow little by little.

In case of DC, the feature of glowing path is almost same as figure 1. But the incandescent point was observed only at the boundary between the hot zone and the copper wire which was connected to (+) pole of the DC power source. The copper wire connected to (+) pole was not oxidized and the hot zone did not expand toward the (+) pole.

3-1-2 Temperature measurement of glowing path

It was observed microscopically that the surface of the glowing path is glossy and it moves like flowing metal. Temperature at the center of the glowing path was measured by K-type thermo-couple o 0.08 mm in diameter. The maximum temperature measured was 1250 C for both AC and DC under the condition of room temperature, 25 C. This temperature is considered to be reasonable in comparison with the melting point of Cu2O, 1230 C (ref. 7).

This temperature did not almost changed when the AC or DC current was increased or decreased. It is considered that this is due to the following two reasons.

The first reason is that the coefficient of the relation between electrical resistance of Cu2O to temperature is negative. Increasing current results in decreasing resistance and suppressing the power dissipation in the glowing path.

The second reason is that the cross section area of the glowing path increases when current value is increased. Width of the path which was measured for the copper wire of 2 mm in diameter with DC 1 A, 2 A, 4 A and 8 A were about 0.2 mm, 0.4 mm, 0.8 mm and 1.4 mm respectively. Increasing current results in increasing cross sectional area of the path, increasing electrical resistance of the path and also decreasing the power dissipation in the path.

3-1-3 Examination of glowing heat

(1) Contaminant of hot zone

Two kinds of heat, Joule heat and chemical reaction heat, were examined as the heat source of the glowing path.

At first, the composition of hot zone was analyzed by X-ray diffraction. It was confirmed that the hot zone consists mainly of Cu2O and CuO but only of Cu2O below the surface.

Next, the weight ratio of the hot zone to the loss of copper wire was measured. Figure 4 shows the results which were measured 1 2 hours after the stating of the glowing. In the region of small current, this ratio is larger than 1.0 (average is 1.15) because of the increased weight by oxidation. But in the region of large current (over 2A AC and over 6A DC), this ratio is smaller than 1.0 because of spattering (which is detailed in 3-3).

Weight ratio of Cu2O to CuO in the hot zone, which was calculated from the above mentioned ratio (1.15) on the assumption that the hot zone is made only of Cu2O and CuO, was 81 : 19.

(2) Joule heat and reaction heat

Joule heat generated at the hot zone was obtained from (integrated voltage (v) shown in figure 3) i(constant). Reaction heat generated at the hot zone was obtained from the heat necessary for the formation of Cu2O and CuO (1.17 kJ/g and 1.95 kJ/g respectively) and the weight ratio of Cu2O to CuO. The results are shown in figure 5. Reaction heat was found to be negligible in comparison with Joule heat.

3-1-4 Measurement of the Joule heat

Electrical energy (DC current) dissipated at the hot zone (glowing path) during the first one hour (from the starting of glow to one hour later) is shown in figure 6. Electrical energy dissipated at the hot zone between copper wires of 1 mm or 2 mm in diameter was saturated at about 60 kJ or about 100 kJ in the range over 2A or 5A DC respectively. (60 kJ / 3600 sec 16 W, 100 kJ / 3600 sec 28 W)

3-1-5 Electrical characteristics (AC)

Since Cu2O has electrical characteristics of P-type semiconductor (ref. 10), it was expected that the boundary between copper and hot zone has rectifying characteristics.

AC voltage across the glowing path (v1, v2, and v3 shown in figure 7) was measured. The result is shown in figure 8. (v2 + v3 = v1) but the form of (v2) is different from that of (v3). If the boundary has ohmic characteristic, the shape of (v2) and (v3) should be the same. This indicates that the boundaries at the both ends of the hot zone have rectifying characteristics and they are in reverse each other.

In case of the wave form shown in figure 2, t2 t3 and t4 t5 is the period of the breakdown of the boundaries, and t1 t2 and t3 t4 is not the period of the breakdown.

3-1-6 Electrical characteristics (DC)

Voltage distribution along the glowing path was measured by the method shown in figure 9. The results are shown in figure 10. (v1), (v2) and (v3) in figure 10 is correspond to the same symbols in figure 9. It was found that both ends of the path have large voltage drop which cause incandescent glow.

It is estimated that the both ends of the path generate incandescent glow alternately in case of AC.

3-2 Breeding of hot zone

3-2-1 Direction of breeding

It was found that hot zone expands toward both sides in case of AC and toward (+) pole side in case of DC. Therefore, the expansion of the hot zone is the result of the oxidation caused by the incandescent heat at the boundaries.

3-2-2 Measurement of breeding speed

Average expanding speed of the hot zone during the first one hour was measured. The results of the measurement are shown in figure 11.

In case of DC, the profiles showed peak speed at 2 A for the copper wire of 1 mm in diameter and at 5 A for the copper wire of 2 mm in diameter. In the region over these currents, glowing path became fuzzy and the current flow spread, and finally whole area of the hot zone became to glow and the incandescent glow at the end of the path disappeared. This decrease of the expanding speed is due to the spread current.

In case of AC, hot zone could not continue expanding over 2 A for the copper wire of 1 mm in diameter and over 2.5 A for the copper wire of 2 mm in diameter because of the consumption due to spattering from the edges of the glowing path. This spattering will be considered in 3-3-1.

3-3 Current range of glowing

3-3-1 Maximum current and spattering

In case of AC, the maximum currents (2 A for 1 mm copper and 2.5 A for 2 mm copper) which can maintain the glow were limited by the spattering.

In case of DC, the maximum currents exceeded 10 A, but they could not be measured because the DC power source could not supply the current over 10 A.

3-3-2 Cause of spattering

The spatters were detected by a photo-transistor and the signal of the detector was recorded by a digital memory. Copper wires of 2 mm in diameter and 2 A AC current were used for this measurement. Figure 13 (b) shows the detectors output and (a) shows the voltage across the glowing path measured at the same time. Small pulses corresponding the detectors output are observed on (a). The detectors output delayed the pulses on (a) because the spatters need time to go to the detector and to be detected.

It was found that only (+) side of the end of the glowing path spatters.

In case of DC, almost same results as figure 13 are obtained in figure 14.

The mechanism of spattering is considered as follows.

a) Large voltage drop at end of glowing path, where rectifying direction is in reverse, results in a large heat generation.

b) When the temperature of the small area (breakdown area) at the end of the path exceeds the boiling point of Cu2O, this area is blown off.

c) Then breakdown is ended but it is recovered immediately. This is shown as a pulse in figure 13.

3-3-3 Alternating current and spattering

Two types of currents, (a) and (b) shown in figure 15 which are different only in their flowing direction, are used in this experiment. Spattering with current (b) was fainter than that with (a). The maximum current (rms value) for 2 mm copper wires was 2.5 A in case of (a), but was 5 A in case of (b). In 3-3-1, it was more than 10 A in case of DC. It is supposed that spattering is affected by alternating frequency of current.

3-3-4 Minimum current

The minimum current necessary for maintaining the glowing of 1 mm wire was 0.3 A (both AC and DC), and that necessary for maintaining the glowing of 2 mm wire was 1 A AC or 0.5 A DC. It was found that the minimum current in case of DC is smaller than that in case of AC.

4. Conclusion

Following results were obtained:

(1) The boundary of hot zone and copper wire has rectifying characteristics which is conductive from the former to the latter.

(2) In case of AC, the hot zone expands toward both sides. In case of DC, it expands towards only (+) pole of the source.

(3) Temperature at the center of glowing was about 1250 C. It was caused Reaction heat was negligible in comparison with Joule heat at hot zone.

(4) Electrical energy dissipated at the hot zone between copper wires of 1 mm or 2 mm in diameter was saturated at about 16W or about 28W in the range over 2A DC or 5A DC respectively.

(5) For 1 mm wires, the maximum breeding speed of Cu2O was 17 mm/hour with 2 A AC, and 18 mm/hour with 2 A DC. For 2 mm wires, the speed was 6 mm/hour with 2.5 A AC, and 12 mm/hour with 5 A DC.

(6) For 1 mm wires, the ranges of the AC or DC current necessary for the maintenance of glowing was 0.3 - 2 A or 0.3 - over 8 A respectively . For 2 mm wires, they were 1 - 2.5 A or 0.5 - over 8 A respectively.

(7) The maximum current which can maintain glowing was affected by spattering from the end of the glowing path.

References

1) Test of Insulating Materials for Resistance Heat and Fire. CEE (031) D126/61.

2) U.S. Department of Commerce: Fire Investigation Handbook, NBS Handbook 134, U.S. GPO (1980).

3) Glowing electrical Connections. Electrical Construction and Maintenance, 57-60, Feb. (1978).

4) H. Niimiya, T. Washimi, T. Takahashi, T. IEE Japan, Vol. 106-A, 519-524 (1986).

5) E. Hotta, J of Japan Association for Fire Safety and Engineering, vol. 24, 52-58 (1974).

6) ?. ISA, J of Japan Association for Fire Safety and Engineering, vol. 24, 237-242 (1974).

7) T. Kawase, OHM Journal, No. 9, 78-80 (1975)

8) T. Kawase, N. Takahashi, T. IEE Japan, ES-80-15 (1980)

9) Some handbook on oxidized material (Japanese)

10) Some handbook on semiconductor(Japanese)

Next | Previous | Index