Durham Mining Museum - Colliery Engineering (original) (raw)
August, 1942
Explosion at the William Pit — 1
In his report* on the explosion at the William Pit, Whitehaven, which occurred on June 3, 1941, causing 23 casualties, 12 of which were fatal, the Chief Inspector draws attention to two usual features: in the first place the explosion was not associated with the normal operations of coal production, and in the second there appeared to be little doubt that the explosion was not one of firedamp, but of inflammable gas derived from the water used to suppress a fire of some years' standing.
The William Pit is sunk close to the seashore, and has two shafts which connect with workings under the sea. The shafts were originally sunk in the Six Quarter Seam at a depth of 297 yds. but the lower length was subsequently abandoned and winding carried on from an inset at the Main Band Seam, at a depth of 218 yds. from the surface. The mine was acquired by its present owners in March 1937, previous to which it had been abandoned for two years.
Seams worked and method of working. — Three seams have been worked, the two already mentioned and the Bannock Band which lies 50 yds. above the Main Band. Of these, in the past, only the Main Band has been extensively worked. This is a thick seam averaging about 10 ft. of almost clean coal (see section in Fig. 2). The method of working was bord and pillar, but although there has been much splitting and robbing of pillars, there has been little broken working. Generally, the pillars have been left standing at their original dimensions of, on a rough average, 20 yds. square.
In the area with which this report is Concerned there are two patches of workings in the Bannock Band. One to the south of the Lowca Junction, but for the fact that roadways in the district served as return airways for the second and newer Bannock Band District, no reference need have been made to this district. The newer Countess Bannock District was opened in 1933 by driving a drift 330 yds. long, rising at 1 in 8, from a point in the main haulage road about 25 yds. inbye of Lowca lunction and vertically above the old Main Band roadway (Fig. 2). A second drift, the Countess Bannock Back Drift, was driven from a roadway in the Main Band pillars to the north of the Main road to provide the return airway from this district (Fig. 1). More recently a second return airway, the Countess Bannock New Back Drift, was driven from the Delaval Level near Lowca Junction to provide a separate split for the workings on the south side of the main level in Countess Bannock District.
Haulage layout. — The main intake airway from the downcast shaft was the main haulage road to Lowca Junction. Originally its gradient was irregular, heing subject to varying inclinations in the seam itself and also to displacements vertically where faults were crossed. About the year 1908 a new road was set out at an even gradient from the downcast shaft to Lowca Junction, and this permitted endless-rope haulage in one reach between these two points. Incidentally, the new road over the greater part of its total length of upwards of 4,000 yds. is in the stone below the seam, thus minimising leakage of air through connections between the intake and the return airways. Haulage from the Countess Bannock District to the Lowca terminus of the main haulage road was by a subsidiary endless rope driven by an electric motor installed immediately above the haulage road near the bottom of the incline.
Ventilation. — At the time of the explosion all the intake air passed through Lowca J unction. There was leakage into the returns through the doors in the Lowca and Delaval levels, and also into the stopping area of Main Band pillars immediately under and in the neighbourhood of the Countess Bannock Drift. Actually, according to a measurement taken in the drift on May 30, 1941, 38,000 cu. ft. of air per minute were passing up the drift. Attempts to measure the quantity of air escaping to the return by leakage through the Main Band pillars about Lowca Junction were abortive.
Persons employed and output. — At the time of the accident about 430 persons were employed underground and 114 on the surface in producing an output of about 650 tons per day on an average.
Supervision. — Supervision was exercised by an agent, Mr. J. Williamson, who was also general manager of all the three mines, and a manager, Mr. G. Farquhar, who had under him two overmen on each of the day and afternoon shifts and one overman on the night shift. At the time of the explosion for the normal working of the mine there were 19 deputies distributed over three shifts. In addition there were three more deputies, two on the day shift and one on the afternoon shift, whose duties were confined solely to inspections of and to the supervision of persons working at gob-fire stoppings.
In the next succeeding section of this report it will emerge that in December 1938 trouble arose owing to the presence of carbon monoxide in that portion of the return airway from the Countess Bannock District, which had passed old workings of the Main Band (Fig. 1). It early became evident that additional and special supervision was desirable to deal with the trouble. Accordingly, Mr. A. B. Dawson, who was at that time in charge of the surveying and planning department of the Whitehaven Collieries, was specially detailed for that duty. Mr. Dawson, who is the holder of a First Class Certificate, was manager of the William Pit for several years at one period of his mining career, which started in 1911, and has been limited almost entirely to the Whitehaven Collieries. As gob-fire supervisor he had no immediate responsibilities under the Coal Mines Act, 1911, but he was in charge of all operations in connection with gob fires from the first indication of the trouble in December 1938, until the end, working all the time, however, in close collaboration with the agent and the manager. He had under him two practically full-time deputies on the day shift and one on the afternoon shift, together with a considerable staff of workmen who rarely numbered less than 20 and at times more than 100, the larger number being employed on the stoppings when normal pit production happened to be interrupted. It may be observed here, in passing, that this appointment merits commendation, for Mr. Dawson had had a unique experience of the Main Band workings, both in the William Pit and the other pits of the colliery. His notes and records of the course of events from December 1938, right up to the day of the explosion were carefully compiled and contained a host of detail which made his evidence at the Inquiry almost invaluable.
Events leading up to the explosion
It has already been observed that, in general, few pillars have been completely extracted in the Main Band workings. There are, however, a few patches of goaf, portions of three of which appear in Fig. 1. It has also been observed that the seam is thick — at least 10 ft. of clean coal with four or five partings only one of which, as appears from the section shown in Fig. 2 is of material thickness. Having regard to the thickness of the seam and the method of working, it is perhaps not surprising to find a history of trouble from time to time due to spontaneous combustion. Definite colliery records of trouble from this cause are, however, somewhat meagre and confined to comparatively recent years. There are official records of the occurrence of beatings or fires at several points prior to 1928, when the workings in the Main Band inbye of Lowca Junction were finally abandoned. These beatings and fires all occurred in the area to the north and west of Lowca Junction. A heating was built off in No, 8 North Section in 1911. There was a fire in No. 5 Right Section in 1918; and another heating in No. 3 North District in 1924.
It is to be noted, although the fact may have no special significance, that whilst these fires, or indications of fires, were a long way from Lowca Junction, each one was progressively nearer to the junction than its immediate predecessor.
From 1924 no further trouble was experienced until December 1938. On the 7th of that month two men working in the main return airway, outbye of the "Humbug Doors" (Fig. 1), showed symptoms of having been exposed to an atmosphere containing carbon monoxide.
The presence of carbon monoxide in the air current was confirmed by the analysis of an air sample taken the following day, which contained 0.0379% of this poisonous gas, whilst analysis of a sample taken the following day showed just under 0.04%. Reference to the plan (Fig. 1) will show that the return airway at "Humbug Doors" where these men were working was common to both the Delaval and Countess Bannock Districts. Spot samples taken in the return airways where they were separate from each other showed that the carbon monoxide was confined, to the Countess Return. Further exploration and sampling revealed the presence of this gas in the atmosphere at several points. In the roadway at "A" there was already a stopping with a door in it. A sample taken inside the door contained 0.26% of carbon monoxide. It then became clear that somewhere in the pillared area to the north of the Countess Bannock Back Drift there existed a seat, or seats maybe, of incipient or active combustion.
The percentage of carbon monoxide in the main body of the air in the return airway was never seriously high, but as this roadway at that time was the second means of egress from the Countess Bannock District, it was essential that it should be maintained in a condition fit to be travelled by the workmen in an emergency. To this end, measures were at once put in hand as far as possible to isolate the affected area by erecting a series of stoppings among the old bords flanking the return airway, as shown by the letters and numbers on Fig. 1. At the same time, to restrict the entry of fresh air, a second series of stoppings was erected in a rough semicircle and designated A1 to J1 among the pillars immediately below and around the Countess Bannock Drift.
As the colliery plan showed that there were possible connections between the Countess District and the main return airway a considerable distance outbye in the neighbourhood of the "Humbug Doors," nine further stoppings were either newly erected or repaired in that region as shown in Fig. 1.
The conditions for the erection of stoppings of any description were most unfavourable, for owing to falls from the roof and fractures in the coal sides the average superficial dimensions of the stoppings were as to width 20 ft. and as to height 15 ft. Most of them were made of broken roof stone (shale) set in mortar, but some were faced with brickwork.
The repairs to old stoppings and the building of new ones was a protracted operation and was not completed until March 1939. Throughout this period and afterwards air samples were taken systematically in the main return airway at a point generally about 15 yds. outbye the "Humbug Doors."
Altogether, between the dates of the first appearance of carbon monoxide in the main return airway on December 7, 1938, and the explosion on June 3, 1941. upwards of 1,000 samples of air were taken and analysed, exclusive of some 50 or so of spot samples taken from pipes in some of the stoppings for special reasons.
The carbon monoxide content of the samples varied considerably from day to day between a maximum of 0.178 % and a minimum of 0.02%, the latter being about the average at the time the stoppings were completed. Thereafter, by ceaseless vigilance and incessant repairs the stoppings appear to have been kept in a reasonable state of efficiency, for, although the percentage of carbon monoxide in the air samples still fluctuated, the limits of the fluctuations gradually narrowed with a distinct tendency towards a lower datum, until the percentage of carbon monoxide in the return became barely detectable by analysis.
This more or less satisfactory position prevailed until the middle of February 1940 (i.e., for a period of 11 months). On the seventh day of that month, vapour was observed on the return side of the Lowca Level Doors and at the Little Main Regulator (Fig. 1), and some of the air samples, taken on 6th, 7th and 8th February, were again found to contain appreciable amounts of carbon monoxide. It was naturally concluded that the vapour was coming from the area of pillars between Lowca Junction and the stopping around the bottom of the Countess Back Drift. It was stated, however, that no smell of gob stink could be detected at this stage.
A brick stopping was built in front of the Lowca Doors (i.e., looking at them from the return airway), and the Little Main Regulator (see R, Fig. 1) was bricked up. Shortly afterwards (about one week), vapour was observed at stoppings 1A, to the left of Countess Bannock Back Drift, and 4A, three pillars to the left of 1A. The stoppings at 2, 1, 1A, 3A, and 4A, and finally at L were then reinforced by building new brickwork in front of them. The presence of the vapour was accompanied by an appreciable but not large increase in the carbon monoxide content of the return air, but this subsided, apparently as a result of the work on the stoppings. Where it could be readily observed, viz., at 1A and 4A, it was found that the direction of pressure at the stoppings varied from time to time. Later, the pressure became constant in an outward direction.
At about the end of the month or early in March further signs of active combustion were manifest at stoppings 1A, when smoke was found issuing from breaks on the left-hand side. About this time also there was some leakage of smoke at stopping Y, the stopping farthest outbye in the Countess Return. Steps were at once set on foot to strengthen these stoppings. This entailed considerable cutting into the sides to find unfractured coal. Dense smoke issued from the breaks, smouldering coal was encountered and high temperatures were registered. Very active combustion in the immediate neighbourhood of this stopping was now self evident. Leakage at the stoppings abated but the air seeking another outlet, found it in the roof strata forming the floor of the Countess Bannock Back Drift. Sweating appeared here over a length of about 20 yds. A layer of concrete 6 to 8 in. thick was put down and repaired regularly as cracks appeared in it.
A reduction in the quantity of carbon monoxide escaping into the main return airway — the average for the three months from February 7 to the middle of April 1940, was 0.00152%, whereas, over the following three months it was 0.00017% — seemed to indicate that some improvement had been brought about. That the improvement was merely temporary was shown by subsequent developments in another quarter. Three months later, on July 22, a fire was detected in front of "G1," one of the stoppings in the Lowca Junction intake barrier. A chock standing 3 ft. back from the stopping and debris consisting of broken coal and stone in which the lower part of the chock was buried, were found to be on fire. The fire external to the stopping was extinguished and the hot material cooled down by the use of water, but a small bole a few inches square was revealed at the junction of the brickwork with the coal side, and through it the coal along the hole and inside the stopping could be seen glowing bright red.
Steps were taken at once to check the now obviously substantial leakage inwards at "G1" by building brickwork until, finally, a position was reached as shown in Fig. 2. Holes were bored through the brickwork and into the coal sides as shown in the same Figure and water was applied from time to time. Its effect, however, was only local; indeed, in spite of the constant application of water a reduction in temperature at one spot was soon followed by an increase at another nearby spot.
In early September, smoke appeared at the left-hand side of "H1" stopping. Cement was injected into the coal and adjacent strata both here at "G1" and farther away in the old roadway to the right from "G1." Nevertheless, about a fortnight later a borehole through the lining of brickwork between "G1" and "H1" found glowing coal. Water was applied and the temperature fell for the time being. As had occurred formerly at the return side, some degree of stability of temperatures at the stoppings appears to have been reached, but from the discovery of fire at "G1" in September there was a marked increase both in the frequency of the occurrence of carbon monoxide in the air samples, and in the average carbon monoxide content of them.
Contemporaneously with the building of the additional wall to "G1" stopping an air lock had been made as shown in Fig. 2 in the intake road, or in the "Hole" as it was aptly called, on the right-hand side of the Countess Bannock Drift. With a view still further to restrict the leakage of fresh air into the stopping area a further stopping called the New Front stopping across the old Main Band roadway about 5 yds. inbye of the airlock was commenced in February 1941, and the door in it was closed on March 7. As will be seen from Fig. 2 this stopping was of 14 in. brickwork. On the right-hand side of the roadway the brickwork was extended a distance of 10 ft. into the coal. On the opposite side, the brickwork was erected in a narrow cutting. At 12 ft. indications of an old road were encountered and breaks in the roof were noticeable. The line of the wall was diverted to the right crossing the old road and extending a distance of 10 ft. or so into the pillar beyond it. The brickwork was reinforced by tight packing of damp flue dust well rammed on both sides.
All went well for some time, but during the last week of May the dry bulb temperature of the air near "G1" stopping which for many weeks had been more or less stable at about 80° Fahrenheit began a steady rise. Moisture appeared in a break in the roof in front of "G1" stopping. The form of this break is shown in section AB in Fig. 2 and its extent and direction, as far as reliable information on the point could be got, are shown on the plan. The break extended to an unknown distance outbye and laterally into the old roadway to the right of "G1," and over the brickwork in that road. An attempt was made to plaster the break up, btit on May 21 a current of warm air was observed issuing from it indicating that somewhere further outbye there was an inlet for fresh air and that oxidation of coal was going on. With a view to counteracting this fresh development it was decided to extend the additional stopping (now to be called the "New Front" stopping), through the coal still remaining between it and the next old road, across the latter and into the coal of the pillar between "H1" and "J1." During the days following the observation of the air movement at the roof break, the temperature at the break rose gradually from something below 100° Fahrenheit, first to 100°, finally reaching a maximum of 104° on May 30. No further rise was registered up to the time of the explosion four days later. There was no gob stink but 0.04% carbon monoxide was found in an air sample taken at the break on May 26.
In the meantime preparations were proceeding for the extension of the "New Front" stopping. On May 29 a start was made to remove about half the thickness (i.e., a depth of 5 ft.) of the flue dust packing from the outbye side of the wall. This work was interrupted by the Whitsuntide Holidays, and nothing more was done for three days, viz., Saturday, Sunday and Monday, May 31 and June 1 and 2 respectively. Examination and observation, however, were not suspended except on Monday, June 2.
The position when work was resumed on the Tuesday morning, June 3, was that about 10 ft. of the upper section of flue dust was still to be moved to reach the end of the brickwork and the face of the narrow cutting. Men were set to work at this, the fireman himself lending a hand from time to time.
Mr. Dawson arrived on the scene between 9 and 9.30 am. on this day. Removal of the flue dust had been temporarily suspended while the passage was being enlarged by taking coal off the left-hand side. Mr. Dawson, however, proceeded to the end of the cutting and cleared out the flue dust right up to the coal at the end of the cutting. The dust was warm but not hot; the temperature of the coal and brickwork at this stage is said to have been normal, viz., very little over 60° Fahrenheit. Later, a connection was made to the water column and at about 12.20 p.m. water was applied to the exposed surface of the flue dust close to the roof at the face of the cutting. The flue dust soon became saturated and the surplus water ran out to where the old road had been crossed. It soaked through the dust there and disappeared, and no further trace of it could be found. This fact, for what it is worth, is recorded here because further reference to it will be made in a later section of the report. To maintain a true sense of proportion it is also recorded that the water supply was estimated to be 30 galls. per min. and that this quantity had been running into the flue dust for at least 45 min. when the final examination was made which showed that it had not emerged into the roadways adjacent to "G1" and "H1." Shortly after the hose-pipe had been placed in position at 12.20 p.m. with the water turned on, the manager appeared. The general position was discussed, the other stoppings and their environs were inspected and at about 1.15 p.m. to 1.30 p.m. the party, consisting of the manager, Mr. Farquhar, Mr. Dawson and the fireman, Mr. G. Savage, withdrew (the workmen had already gone, their shift being ended). The water was left running; the door in the "New Front" stopping was left open, the doors in the air-lock were both closed, the outer door, following the usual practice, being plastered round its edges. Although the position had clearly deteriorated during the week-end, and especially during the morning, no serious development was expected. In due course the three officials, the manager, Mr. Dawson and the fireman, proceeded outbye on foot. Work in the Countess Bannock District was pursuing its normal course, the bulk of the day shift had gone outbye and the afternoon shift had gone to work. Owing to absenteeism, usually to be expected on the first day following a holiday period, some of the work of the day shift was not completed at the normal stopping time. A few hands had stayed behind about an hour to catch up arrears. They had completed their work about 2 p.m. and had gathered at Lowca Junction to wait for the train to take them to the pit. After some delay whilst necessary repairs were being carried out on the Countess Bannock haulage rope the haulage was set in motion on the afternoon shift at about 2.10 p.m.
* "Report on the causes and circumstances attending the Explosion which occurred at the William Pit, Whitehaven Collieries, Cumberland, on the 3rd. of June, 1941," By F. H. Wynne, C.B.E., B.Sc. Published by H.M. Stationery Office, York House, Kingsway, London, W.C.2, Price 1s. 6d. net.