Tweet

EXPLOSION AT HORDEN COLLIERY ON 23rd MARCH, 1953

Report by W. L. Murray, B.Sc., Ph.D. and S. K. Shaw, B.Sc., A.M.I.E.E. of the Safety in Mines Research Establishment on the operation of the stone-dust barrier

1. - INTRODUCTION

At the time of the explosion it was already realised that, under certain conditions of impact between light alloys and rusty steel, sparks could be produced which were capable of igniting mixtures of methane and air. It had been known for many years that frictional sparking between certain types of rocks was also capable of causing ignition. The object of the investigation was to assess whether the conditions on the Plough Face at Horden were such that incendive sparking of either nature could have occurred.

2. - EXAMINATION OF THE FACE

2.1 General arrangement on the face. - The Plough Face was a prop-free front face worked by undercutting, shotfiring and ploughing on to a scraper-panzer conveyor: all the equipment was of German manufacture. The cutter worked on the conveyor, which was moved up to the face by compressed-air rams on the waste side. The rams were fed from a compressed-air pipe of light alloy running the full length of the face near floor level on the waste side of the conveyor. At intervals along the pipe there were light-alloy couplings for joining the lengths of pipe: the couplings had screwed valve unions made of steel for the flexible pipes leading to the rams. The body of the ram and its back-stay were light alloy coupled together by a steel pivot pin. At the time of the examination the back-stay of each ram was lying on the floor behind the ram and pointing into the waste.

The roof was supported by Schwarz props and Schloms bars in units of 3 lengths of bar and 3 props: the waste prop was sometimes wood. The props were made of steel and the bars of light alloy. The steel pivot pins and wedges used for linking the bars were attached to the bars by steel chain. Props and bars not in use were stacked in piles at the edges of several of the wastes.

Thus, along the whole length of the face there was a heterogeneous mixture of light alloy and steel. Most of the steel was slighted rusted.

2.2 Observations after the explosions. - At a point almost opposite the discontinued No. 2 pack there had been a fall of roof extending from the normal breaking-off point almost up to the waste side of the conveyor. A close-up view of part of the fall at the time of the examination is shown in Plate Ia. From one set of props and bars the centre prop and two sections of bar nearest the waste had been forced down and towards the intake road: the wooden waste prop was still standing and almost buried in debris. A photograph of sections of bar lying across the compressed-air pipe and conveyor is shown in Plate Ib: in this photograph the head of the displaced and almost buried prop can also be seen. The head of the prop was wedged against the compressed-air pipe and the bar, as shown in the close-up view in Plate IIa; the bruises on the bar and pipe which had been caused by rubbing of the prop head were found to be marks in the coating of dust deposited post-explosion and therefore occurred after the explosion, probably as a result of some movement of the pipe or settlement of the fall. Closer examination revealed, however, that under the post-explosion deposit there was another bruise on the pipe which could have been caused by the initial fall of the prop.

Also buried under the fall was a compressed-air ram. Resting on it was the steel wedge attached to the section of the fallen bar. During the fall the movement of this bar must have been such as to swing this wedge right across the body of the ram. A view of the conveyor end of the ram and the wedge, as seen at the time of the examination, is shown in Plate IIb.

The light alloy air-pipe was examined along its entire length and abrasions and, in some places, severe deformations were noted. Some of the bruises were old. The possibility of incendive frictional sparking occurring as a result of the very limited work in progress at the time of the explosion was considered, but there was no evidence to support this and the possibility was thought to be extremely remote. Conditions in all wastes other than that associated with No, 2 pack were normal.

3. - LABORATORY TESTS

3.1 Friction between rock and rock or steel. - Samples of rock taken from the roof were examined under a microscope. The main body of the rock consisted of blue shale containing only a little fine-grained quartz, but lying between it and the coal, however, there was a thin layer of sandstone containing 40 to 50 per cent. of coarser grained quartz.

Samples of the rock were forced against a rotating steel wheel in a methane-air mixture containing 6.2 to 6.4 per cent. methane. Ignition was not obtained when the blue shale was used; ignition could be obtained with the sandstone but only when it was pressed against the wheel for more than two minutes. In view of the thinness of the sandstone layer and the duration of pressure required to cause ignition, it would appear that the chance of ignition arising from friction between falling pieces of rock and the steel equipment must be very small. Also, in view of the thinness and location of the sandstone layer, the chance of ignition from friction between rock and rock must have been equally small.

3.2 Friction between light alloy and rock. - As noted in Section 2.2 one of the compressed air rams was almost buried by the fall. A few experiments have already shown that ignition can be caused when impact occurs between a magnesium-based alloy and a sandstone rock. The experiments have not yet gone far enough to assess the hazard with aluminium-based alloys but, for the present, it must be accepted that impact of the hard sandstone on the ram cylinder might have caused incendive sparking.

3.3 Friction between light alloy and steel. - Samples from specimens of light-alloy equipment were analysed and the results are shown in Table l.

TABLE 1

Analysis of light-alloy specimens

Sample	Cu	Mg	Si	Fe	Mn	Zn	Pb	Al
Pipe	0.58	3.16	0.18	0.32	0.35	4.7	-	Remainder
Coupling	3.54	0.94	0.58	0.48	0.85	0.38	-	Remainder
Ram end	3.79	0.65	0.56	0.56	0.99	0.47	0.05	Remainder
Ram cylinder	3.86	1.29	0.35	0.50	0.92	0.44	0.05	Remainder

The incendivity of the sparking produced by impact of these samples on rusty steel was tested in a "falling weight" apparatus. Details of this test and results obtained with various light alloys have been described (Margerson and others, 1953; Titman, 1954). Photographs of the flashes produced by impact of elektron metal (92.6 per cent. magnesium) and of aluminium-based alloy LM-10 (10 per cent. magnesium) on a rusty steel plate are shown in Plate III; the duration of the flashes was measured by means of a photocell and the relative records, shown as inserts, indicate that the intensity of the flash varies with time, the total duration being 12 to 17 milliseconds.

In these tests, a shoe of the material under test was attached to a brass weight which could be dropped from a chosen height on to an inclined face of a rusted steel plate: the total mass of the falling weight varied from 33.5 lb. when shod with metal from the pipe sections to 36 lb. when shod with a solid shoe. The angle of impact was 50 degrees and the impacts occurred in gas mixtures containing 6.3 to 65 percent. methane in air. A summary of the results is given in Table 2.

TABLE 2

Results of impact tests on a rusted steel plate

Falling weight	Height of fall, ft.	No. of tests	No. of ignitions
Alloy pipe (from surface)	7.5	50	0
Alloy pipe (from face)	7.5	50	2
Alloy coupling	7.5	50	3
End face of ram	7.5	50	2
Ram cylinder	7.5	50	13
"   "	6.0	50	11

Some tests were also made in which a Schwarz prop was allowed to fall against a section of the light-alloy pipe. A complete prop was arranged so that it could topple freely from a vertical position and, on reaching a horizontal position, would give a glancing blow on the side of the pipe. This represented the conditions which might have occurred during the fall of roof except that, in the fall, the prop would be propelled by falling rock and could therefore have had a much greater energy. The tests were carried out in mixtures containing 6.3 to 6.5 per cent. methane in air: in 30 tests ignition was not obtained. In later experiments in which a pivoted Schloms bar was allowed to swing against a rusted steel plate an ignition was obtained; the light alloy from which these bars are made contains less than 0.05 percent. magnesium by specification.

3.4 Friction on a rusty steel surface smeared with aluminium. - The gas ignition hazard associated with aluminium paint and coatings on rusty steel is well known (Thomas, 1941; Kingman and others, 1952; Grice, 1952). Aluminium may become smeared on steel during normal use of structures, equipment and implements. Kingman and his colleagues found that when aluminium was rubbed on rusty steel the sparks that could be obtained by striking the smear were capable of igniting a coal-gas air mixture. The ignition hazard with methane-air mixtures arising from this type of smear has recently been investigated. Although the work was done in particular connexion with yielding props that are fitted with light-alloy friction pads, the hazard is a wide and general one.

In this form of prop, and with an aluminium-alloy pad, the normal functioning of the prop under load causes a heavy smear of aluminium on the surface of the inner member. A smear of this type was formed on an air-rusted inner prop member and it was found that glancing blows with a light hammer (1.5 lb.: energy at impact, about 3 ft. lb.) gave rise to brilliant sparks. A view of the hammer at the instant of impact is shown in Plate IVa and two flashes, each produced by a single blow, are shown on the same scale in Plate IVb and c. The sparks so produced ignited methane issuing from a bunsen burner held near the point of impact. A light blow with a pen-knife or with a piece of brass produced similar sparks. The sparks were audible and gave the impression of miniature explosions.

In further tests the steel prop member was first burnished by means of a rotating wire brush and, without delay, it was smeared with alloy by loading the assembled prop in a hydraulic press. Glancing blows with the hammer again produced sparks from the smear. Although the sparks did not appear to be so large or intense, they nevertheless caused ignition of methane.

A piece of fairly heavily rusted steel was rubbed by hand with the end of a rod of aluminium (magnesium less than 0.05 per cent.). Smears were left, like chalk marks, on the steel and light glancing blows gave rise to vivid sparks which again ignited methane. A view of the rusty steel during the smearing process is shown in Plate IVd and a photograph of a flash produced from this smear is shown in Plate IVe. Exactly the same result was obtained when an uncontaminated piece of air-rusted steel was rubbed against a Schloms bar.

The hazard from steel smeared with aluminium is a general one, and is likely to arise wherever steel and light alloy are used in proximity to each other.

Similar experiments have been made with zinc in place of aluminium; sparks were not obtained.

4. - DISCUSSION

In the vicinity of the fall on the face opposite the discontinued No, 2 pack there were many points at which heavy impact could have occurred between light alloy and rusty steel or rock: there might also have been present rusty surfaces smeared with aluminium. Sonic of these points are marked on the Plates.

The friction pads in the Schwarz props are usually of zinc. Nevertheless, these props were in constant contact with aluminium-alloy roof bars and dangerous smears might be left on their surface: an example in which relative motion under heavy pressure must have occurred can be seen in the partially displaced prop shown in Plate Ia; this same motion must have happened when the set of bars was dislodged by the fall. The wedges of the roof bars also become smeared with aluminium-alloy during normal use, and it can be seen in Plate IIb that one of these wedges must have suffered many blows during the fall.

Heavy impact of rock on the ram cylinder and end-piece must also have occurred, and there was certainly heavy glancing impact between the head of the prop and the light alloy compressed air-pipe.

In the vicinity of the fall there were thus many instances in which incendive sparking could be expected to have occurred.

REFERENCES

• GRICE, C. S. W. (1952). Sparks from aluminium paint; the firedamp ignition hazard. Safety in Mines Research Establishment Research Report No. 59.
• KINGMAN, F. E. T., COLEMAN, E. H. and ROGOWSKI, Z. W. (1952). J. App. Chem., 2 (8), 449 - 456.
• MARGERSON, S. N. A., ROBINSON, H. and WILKINS, H. A. (1953). The ignition hazard from sparks from magnesium-base alloys. Safety in Mines Research Establishment Research Report No. 75.
• THOMAS, T. S. E. (1941). Colliery Guardian, 163, 202.
• TITMAN, H. (1954). Ignition hazard from sparks from cast alloys of magnesium and aluminium. Safety in Mines Research Establishment Research Report No. 90.