Incident directory

1951 - Regent Oil, Avonmouth

06/09/1951

Country:

UK

  • Other

Severity:

Description

Date of event

6th September 1951.

Time of event

Approximately 14:56 (INITIAL CALL)

Name of premises

Regent Oil Company Ltd. No. 1. Depot.

Location

Royal Edward Dock, Avonmouth, Bristol.

Service area

Bristol Fire Brigade (BFB), now Avon Fire and Rescue Service (AFRS).

Nature of incident

Explosion & subsequent fire.

Property type

Oil storage facility/tank farm.

Premises use

Industrial, premises used for the storage and distribution of oil, petrol and other petroleum-based products.

Construction type and materials

Mainly steel tanks and associated pipework with concrete bund.

The compound measured approximately 500’ x 160’ (152 m x 49 m) and contained eight cylindrical upright tanks and four cylindrical overhead tanks. Two of the smaller upright tanks, together with the four overhead tanks, were separated from the remainder by a low wall 3’3" high, through which product lines passed.

Tanks were constructed of steel plates, with fixed roofs.

The plates on the larger tanks were approximately ½ " in thickness at the base, tapering to 5/8 '' for the top sections, whilst the Smaller tanks were constructed of plates approximately ½ " in thickness at the base tapering to 3/16" at the top.

Three or four dip holes were provided in each roof. Two pressure and vacuum valves were fitted to each spirit tank, whilst an open vent was installed on the top of each gas oil tank.

The sizes of the upright tanks within the compound were as follows.

Tank R.1 and R.2.             – 35’ high, 95’ diameter (10.6 m x 30 m)

Tank R.3.                         – 35’ high, 70’ diameter (10.6 m x 21 m)

Tank R.4.                         – 35’ high, 65’ diameter (10.6 m x 20 m)

Tanks R.13 and R.14         – 18’ high, 20’ diameter (5.5 m x 9 m)

Tank R.5.                         – 25’ high, 32’ diameter (7.6 m x 10 m)

Tank R.6.                         – 25’ high, 25’ diameter (7.6 m x 7.6 m)

The Regent Oil Installation was part of a large number of similar Oil Installations situated on the banks of the River Severn, just north of the point where the River Avon enters the Severn in the dock area of Avonmouth. The area covered by these installations was 120 acres, and that part occupied by the Regent Oil Company Ltd covered approximately 15 acres, of which some 3 acres were affected by the fire (Unknown author, 1951).

Occupancy

Members of staff.

Fire source and location of fire

Unknown at this time.

Synopsis

Brief Synopsis

On the 6th September at approximately 14:56 Bristol Fire Brigade (BFB) were called to a fire following an explosion at Regent Oil Company Ltd that had occurred within tank R 13 which contained 26,603 gallons of gas oil. The first call to Avonmouth Fire Station was made a crew member of the fireboat Endres Gane (Bristol Fire Brigade) which was moored within Royal Edward Dock. A 999 call at the same time was made to Brigade Control at Bridewell (Unknown author, 1951).

Tank R 13 had been having gas oil pumped into it at the time of the explosion from the motor vessel ‘Fort Christina’ (gross tonnage 10,619 tonnes) which was tied up at No. 5 berth at the Oil Basin, Royal Edward Dock. It was believed that the gas oil was contaminated with ‘motor spirit’ (petrol). During a voyage from Bahrain where the tanker vessel had picked up cargo, it was reported in Watts, 1952 that the transverse bulkhead between tank 8 containing ‘motor spirit’ and tank 9 containing gas oil on the Fort Christina had allowed contamination to some degree between the 2 tanks (Watts, 1952).

During discharge operations to tank R 13 the explosion occurred, and subsequent fire took hold. 2 employees, Roy Hyett (35) and Arthur Baggs (40) who were involved in the discharge operations on the roof of tank 13 at the time of the explosion with some human remains found within tank 13 and some just forward of R 13 towards tank 4. The Fort Christina had originally been destined to discharge her cargo at Dublin but was diverted to Avonmouth due to strike action at Dublin port (Watts, 1952).

The first attending appliances from Avonmouth Fire Station were a Pump Escape with a Sub Officer (SubO 1) in charge, a Foam Tender Major also with SubO 2 in charge, a wireless car and the Fireboat. The road appliances arrived at approximately 15:00 to 15:01. A well-developed fire was in progress when the crews arrived with a number of employees engaged in laying out cooling lines of hose to protect nearby tanks. During the initial actions of trying to cool some of the overhead tanks some of them exploded adding to the size of the fire making ‘this position untenable’ forcing the crews and employees to withdraw. At approximately 15:10 a make pumps (MP) 4 message was sent and additional supplies of foam compound and equipment requested (Unknown author, 1951).

A Divisional Officer (DO 1), an Assistant Divisional Officer (ADO 1) and a Station Officer (StnO 1) were also sent from Bridewell Street fire station along with additional appliances. When DO 1 arrived he observed 4 tanks (R 5, R 6, R 13 and R 14) well alight and 2 others to the rear also burning. A further assistance message was sent ‘MP 6, foam tenders 2 – approximately 6 tanks on fire and well alight, fire spreading rapidly’. It was also noted within the Bristol Fire Brigade report (Unknown author, 1951) that ‘the grass and railway sleepers to the East side were burning fiercely, presenting a severe danger of spread to the Cleveland Petroleum Company’s compounds’. The Endres Gane was supplying water from its position within the dock (Unknown author, 1951).

Firefighting and cooling jets were ‘rapidly positioned’ along the track and tanks within the Cleveland Compound were also cooled. Tanks R 13, R 14, R 5, R 6 and by then 4 overhead tanks were reported to be well alight and threatening tanks R 18 and R 30. Additional cooling lines and firefighting jets were laid out to these also. The Deputy Chief Officer (DCO 1) arrived and took charge of the operations. Additional resources were requested and put to work. The Chief Fire Officer (CFO 1) arrived with a H. M. Inspector (HMI 1) and after taking a report from DCO 1 assumed command. Numerous other cooling jets had been applied to various tanks including C 4, C 5, C 6, R 18 and R 30 in order to ‘prevent an overwhelming spread of fire’ (Unknown author, 1951).

Further reinforcements were called for and a plan for a gradual and manageable build-up’ was put into operation. An additional fireboat, Pyronaut, which was normally berthed in the City Docks was despatched to Avonmouth to assist with water supplies on the evening tide and commenced pumping at approximately 23:30. Tanks R 2 and R 3 were also involved with the heat from the fiercely burning R 3 affecting tank R 4 with flames observed to be coming from its top edge (Unknown author, 1951).

Despite the protection given to R 4 on Friday 7th at approximately 04:35 an explosion occurred dislodging the roof of R 4. This tank contained 417,239 gallons (approximately 1,896,806 litres) of motor spirit and ‘the resultant terrific and rapid spread of fire forced crews to abandon their positions as fire engulfed the coast road and foreshore’. Hose and equipment was destroyed including some of the No. 10 foam making branch pipes and cooling jets. There was a significant risk of the burning spirit (petrol) flowing towards other installations. Weakening of the bund wall was also observed. A roll call was carried out and initially a Firefighter (Ff) from Exeter was unaccounted for, although he was found later on the foreshore with back injuries and then taken by ambulance to hospital (Unknown author, 1951).

All the previous ground that had been gained in the firefighting operation was lost and cooling of unaffected tanks became the immediate priority to prevent further spread. Once the fire was held at strategic points the CFO called together all the senior officers and formulated a massed foam attack on the fire once additional resources had been built up at the incident. With new hose lines and additional foam supplies in place a ‘concentrated attack’ was put into operation at approximately 10:00 on Friday 7th which involved ‘26 No. 10 foam branches supported by 30 cooling jets’ (Unknown author, 1951).

The fire was reduced in area to tanks R 4 and R 13 and by approximately 24:00 on the Friday 7thit was felt with a degree of confidence that, apart from some unpredictable event, the fire was now under control’. Tank R 4 continued to burn for a number of hours. Re-ignition frequently occurred throughout the compound it was reported due in part to the product line of tank R 2 being inadvertently left open. Foam was continually applied and at approximately 04:50 on Saturday 8th a ‘stop – fire out’ message was given by The CFO to brigade control (Unknown author, 1951).

In total 60 pumping appliances and 853 officers and men were in attendance with, it is reported, 25% of the total engaged at the fire at any one time due to relief arrangements (Unknown author, 1951).

Assistance numerous other fire services was given including from Bath City Brigade, Berkshire, Birmingham City Brigade, Cardiff City Brigade, Devon County, Dorset, Exeter City Brigade, Gloucestershire, Gloucester City Brigade, Glamorgan, Hampshire, London County Council, Middlesex, Monmouthshire, Newport City Brigade, Oxfordshire, Oxford City Brigade, Plymouth City Brigade, Shropshire, Somersetshire, Southampton City Brigade, Warwickshire, Wiltshire and Worcestershire (Unknown author, 1951).

Additional assistance was also given by Army Fire Service Hampshire, Ministry of Civil Aviation, Esso Petroleum Co. Ltd, Shell Mex and B.P. Ltd and working parties from the Naval Air Station at Yeovilton, Somerset and Army personnel at Horfield Barracks, Bristol (Unknown author, 1951).

In total 2 employees died at the incident and 6 Ffs and the HMI were taken to hospital for injuries including burns, back injuries, a suspected fracture and one case of collapse overcome by fumes (Unknown author, 1951).

In total 3,855, 239 gallons (17,526,263 litres) of product were within the tanks involved in fire including 2,448,353 gallons (11,130,433 litres) of motor spirit, 1,298,983 gallons (5,905,293 litres) of gas oil, 101,715 (489,682 litres) gallons of aviation spirit and 6,188 (28,131 litres) gallons of Benzol. An estimation of approximately 2,405,239 gallons (10,934,432 litres) of product was destroyed by fire with some recovered during subsequent salvage operations (Unknown author, 1951).

 Pic 02

Tank R 13 and others, from Watts, 1952.

 Pic 03

Site plan of Regent Oil Company, Cleveland Petroleum and Watson’s Petroleum Company, from Unknown author, 1951.

 Pic 04

General description of damage at end of fire, from Unknown author, 1951.

 Pic 05

Regent Oil Company Ltd tank farm fire. Unknown time of incident. Courtesy of The Bristol Evening Post.

 Pic 07

Regent Oil Company Ltd tank farm fire. Unknown time of incident. Courtesy of The Bristol Evening Post.

 Pic 07

Regent Oil Company Ltd tank farm fire. Unknown time of incident. Courtesy of The Bristol Evening Post.

 Pic 08

Regent Oil Company Ltd tank farm fire. Unknown time of incident. Courtesy of The Bristol Evening Post.

 Pic 09

Regent Oil Company Ltd tank farm fire. Unknown time of incident. Courtesy of The Bristol Evening Post.

 Pic 10

Regent Oil Company Ltd tank farm fire. Unknown time of incident. Courtesy of The Bristol Evening Post.

 Pic 11

Regent Oil Company Ltd tank farm fire. Unknown time of incident. Courtesy of The Bristol Evening Post.

 Pic 12

Regent Oil Company Ltd tank farm fire. Unknown time of incident. Courtesy of The Bristol Evening Post.

 Pic 13

Regent Oil Company Ltd tank farm fire. Unknown time of incident. Courtesy of The Bristol Evening Post.

 Pic 14

Regent Oil Company Ltd tank farm fire. Unknown time of incident. Courtesy of The Bristol Evening Post.

 Pic 15

Regent Oil Company Ltd tank farm fire. Unknown time of incident. Courtesy of The Bristol Evening Post.

 Pic 16

Regent Oil Company Ltd tank farm fire. Unknown time of incident. Courtesy of The Bristol Evening Post.

 Pic 17

Regent Oil Company Ltd tank farm fire. Unknown time of incident. Courtesy of The Bristol Evening Post.

 Pic 18

Regent Oil Company Ltd tank farm fire. Unknown time of incident. Courtesy of The Bristol Evening Post.

 Pic 19

Regent Oil Company Ltd tank farm fire. Unknown time of incident. Courtesy of The Bristol Evening Post.

 Pic 20

Regent Oil Company Ltd tank farm fire. Unknown time of incident. Courtesy of The Bristol Evening Post.

Further information hoping to be identified and still to be located.

Main findings, key lessons & areas for learning

Further information hoping to be identified and still to be located.

Fire & Rescue Service summary of main findings, conclusions, key lessons & recommendations

Further information hoping to be identified and still to be located.

FBU summary of main findings, conclusions, key lessons & recommendations

Taken from. Unknown author. (1951). Report on fire at The Regent Oil Company Ltd. Avonmouth. 6th., 7th., 8th. 1951. [pdf] available here. Bristol Fire Brigade.

--subaccordion--

16. OBSERVATIONS AND CONCLUSIONS

Pic 21

Pic 22

Pic 23

--end--

Further information hoping to be identified and still to be located.

Other report summary of main findings, conclusions, key lessons & recommendations

Taken from. Watts,. H,. E. (1952). Home Office report on explosion and fire at Regent Oil Co. Ltd. Premises Royal Edward Dock Avonmouth, Bristol on 7th September 1951. [pdf] available here. London: Her Majesty’s Stationary Office.

--subaccordion--

CONCLUSIONS

After taking all the facts into consideration and in the absence of any evidence to the contrary. I am of the opinion that the explosion was caused by the presence of motor spirit in the gas oil pumped by the tanker. ‘Fort Christina" into shore tank No. 13, and the motor spirit vapour present in shore tank No. 13 may have been ignited by a spark caused b) the generation of a charge of static electricity in pumping the oil into that tank or by the presence of glowing iron sulphide in the tank 13 and of these two possible causes the former is the more probable.

RECOMMENDATIONS

High flash point products

1. In view of the construction of certain war time tankers (T2 Type) whereby the products carried. In them may become contaminated owing to leaking bulkheads, or faulty valves in the header attached to the pumps. it is desirable that the same precautions should be observed in dealing with petroleum products having a flash point of 73 F. and upwards as those for petroleum spirit (flash point below 73< F.).

Where the tankers are provided with separate pumps and pipe lines which are not inter-connected and the products are separated in the tanker by double bulkheads, and also where separate product pipe lines are provided at the shore installations, this may not be necessary provided that there is no contamination with the previous cargo carried in the tanker. There is however, always the danger of contamination owing to an accident of some kind.

This point should also be covered in the licensing of the premises concerned.

Water Supplies

2. It is quite clear that the water supplies available were insufficient as there was not enough water 10 cool all the tanks near the fire quite apart from supplying foam for extinguishing the fire. At one time an urgent request was sent out to cut down the cooling of tanks in order to provide water for firefighting. Shell-Mex and B.P. had to stop cooling their tanks at the Abandan site so as to divert cooling water for the tanks at the Cleveland Company's Depot. In fact there was not sufficient water available until about 1 ¾ hours after the fire occurred and the additional water was obtained by pumping water from the dock by the Fire Brigade. It is important that there should be adequate supplies of water regularly available both for cooling purposes and also for firefighting.

I would therefore suggest that if practicable a ring main round the petroleum depots should be provided and that this main should be so energised that an adequate supply of water is always available and at a sufficient pressure to be used for efficient firefighting with foam (about 150·l60 lb. per square inch). Each tank should be provided with water nozzles for cooling purposes so that cooling can be applied at short notice.

Coal fired locomotives

3. Although there is no suggestion that this fire was caused by a spark from a locomotive in the vicinity of the depot. It is of importance that coal fired locomotives should not be allowed in the vicinity of the depots where petroleum spirit is kept. I was informed that a number of grass fires have occurred due to sparks or bot ashes, and it is important that this type of locomotive should be replaced by the diesel type.

Bunding of tank farm

4. The fire started at tank No. 13 and this was one of twelve tanks situated in a reinforced concrete bund. The fire spread to or damaged all the other tanks in the compound. There was a general conflagration in the bund and motor spirit was fed into the bund from a number of tanks. The heat to which the reinforced bund wall was exposed caused the wall to spall and it was exceedingly fortunate that burning petroleum did not find its way on to the river Severn. This might have proved very serious as the bund was only a few feet from the river. I consider that this method of bunding is insufficient and that this should be suitably improved by providing some kind of heat insulation. As for instance, by the addition of earth bunding on the inside of the reinforced concrete wall. Always provided that the wall is made sufficiently strong.

Bunding of tanks

5. There were in effect twelve separate tanks in one bund and I consider that this number of tanks is too large to have in one risk. It is true that in some cases at Avonmouth dwarf walls were erected between tanks but these are useless unless they are imperforate. This was the case in the bund adjoining the one in which the fire occurred. The holes in these dwarf walls through which the pipe lines were carried were not sealed, and if leakage occurred from one tank the liquid could run through the dwarf wall.

I consider from a safety point of view that wherever practicable each tank of any size should be situated in its own Imperforate bund so that 10 the event of a fire occurring the fire can be localised. The neighbouring tanks can then be cooled and if it is not possible to extinguish the fire easily it could at the worst, burn itself out and not affect the remaining tanks. If separate bunds are provided it gives time to close the valves or do anything else that may be necessary to the adjacent tanks. The inter·bund walls should be nearly as high as the external bund wall. Each compound should be drained through an interceptor and this should be controlled outside the bund. In this case, had Tank No. 2 been blinded separately it is highly probable that the suction valve would have been closed as soon as the fire occurred.

Product pipe lines within the bund

6. The product pipe lines within the bund running beside the tanks Nos. 1, 2, 3, and 4 were all fractured as a result of the fire. If It is necessary or desirable for the pipe lines to be within the bund then I recommend that they should be enclosed in a brick or concrete chamber covered with concrete sIabs to protect them from fire. Alternatively the pipe line should be placed outside the bund

Spacing of tanks

7. The fire spread from lank No. 13 to tank No. 5, very rapidly, and this was probably due to the explosion in tank No. 13 fracturing the 6 inch connection on the top of this tank through which oil was being pumped and the delay in stopping oil being pumped from the ship through this pipe. Consequently a certain amount of burning oil may have been sprayed into the compound and on to lank No. 5. The other tanks were probably involved due to the presence or burning oil and spirit in the compound in spite of this it appears to me to be desirable to increase the distance between the tanks in order to prevent the

Heating of the tanks by radiation in the event of a fire in one of them. It may be difficult to increase the separation in some case, but wherever this is practicable the ideal distance is one diameter apart, or as near this figure as is practicable.

Filling pipe carried to the bottom of the rank

8. The filling pipe in tank No.13 was not carried down to the bottom of the tank as was done in the case of those tanks used for the storage of petroleum spirit. This is quite understandable in the case of gas oil as any electric discharge due to electrification of the oil would have no harmful effect owing to the high flash point of this product. In view, however of the possibility of contamination with low flash products mentioned in recommendation No. 1. of this report, it would be a wise precaution to use bottom delivery and to incorporate a non-return valve with the delivery valve. This would prevent burning oil being projected into the bund in the event of an explosion in the tank

Dipping from the top of tanks

9. Dipping tanks of gas oil from the roof or the tank would not normally constitute a hazard but in view of this accident and particularly during pumping operations it would be a wise precaution to dispense with this method of determining the level of the liquid in the tank and use instead a float or some other safe device to indicate the level of the liquid in the tank.

Local first-aid fire fighting

10. It is quite clear that the extent of a petroleum fire depends very much on what can be done in fighting the fire during the very early stages, namely, the first 5 or 10 minutes before the arrival of the Fire Brigade. It therefore seemed to me that it would be advisable to initiate a scheme whereby the licensees In Avonmouth Dock would organise amongst themselves a first-aid fire party to fight a fire which might occur in any of the premises of the Companies concerned. Accordingly, I approach the following firms with this end in view: -

Watsons Petroleum Co. Ltd.,

Cleveland Petroleum Co. Regent Oil Co Ltd.,

Shell·Mex & B.P. Ltd.,

Esso Petroleum Co. Ltd.,

New Western

Oil Storage, Ltd.,

Wm. Butler & Co. (Bristol) Ltd.

These firms have agreed in principle to work out a scheme among themselves in order to train their personnel in firefighting, have regular practices and pool their fire-fighting resources for first aid action. They point out however that during non-working time, i.e., week-ends. Holidays and nights, the number of men available would be considerably reduced and in the event of an accident they would probably be fully employed in taking preventive action. Such as closing valves, etc., but they would do what they could. In These circumstances, therefore, I would recommend that during the non-working hours. If possible, some additional members of the fire brigade should be transferred to the Docks to deal with an outbreak of fire during these periods. It would also be necessary to introduce some means of rapid communication between the various depots so that in the event of a fire or explosion occurring no time would be lost in sending the necessary information

As to the location, the number of men and equipment required. The telephone service is not considered to be satisfactory for this purpose.

Application of foam to storage tanks

11. The question of whether bottom injection of foam or top foam pourers should be used for the extinction of a fire in a tank containing petroleum spirit is a matter of opinion and has yet to be decided. In the case of some tanks at Avonmouth. Some were fitted with bottom injection pipes, but some of these pipe connections were situated inside the bund so that if they were to be used it would be necessary to go into the bund to connect up and open the suction valves if they were closed. Thus. if the fire was burning in the bund, some of the connections could not be made and the valve, could not be opened.

Whichever system is used it is absolutely essential that the connections for the application of foam. By bottom injection or through dry risers to the top of the tank. Should be external to the bund in an easily accessible position and several feet away from the bund to enable the mobile foam units to be connected up 10 the tank without being subjected to an undue amount of heat. If the bottom injection system is used, then a suitable non-return valve should be fitted on the tank side of the suction valve, and the suction valve should be left open. Otherwise it may not be possible to inject any foam into the tank.

If top pourers are installed, this should not necessarily be limited to one point but two or possibly three.

I recommend that the connections for the application of foam should be external to the bund in an easily accessible position away from the bund wall, and if bottom injection is used a suitable non-return valve should be fitted on the tank side of the suction valve and that the suction valve should be kept open.

--end--

Further information hoping to be identified and still to be located.

IFE Commentary & lessons if applicable

None produced at this time.

Known available source documents

Further information hoping to be identified and still to be located.

FRS Incident Report/s

Unknown author. (1951). Report on fire at The Regent Oil Company Ltd. Avonmouth. 6th., 7th., 8th. 1951. [pdf] available here. Bristol Fire Brigade.

Further information hoping to be identified and still to be located.

FBU Incident Report/s

No information identified to date and/or still to be located.

Health & Safety Executive (HSE) Incident Report/s and/or improvement notices

No information identified to date and/or still to be located.

Avon & Somerset Police Incident Report/s

No information identified to date and/or still to be located.

South West Ambulance Service Incident Report

No information identified to date and/or still to be located.

Building Research Establishment (BRE) Reports/investigations/research

No information identified to date and/or still to be located.

Coroner’s report/s and/or Rule 43 and/or Regulation 28 Notices etc

No information identified to date and/or still to be located.

Dear Chief Officer Letters (DCOL), FRS Circulars, FRS Notices and/or Bulletins etc and/or Related Government Correspondence

No information identified to date and/or still to be located.

Notifications from National Operational Learning User Group (NOLUG) and/or Joint Emergency Services Interoperability Principles (JESIP)

No information identified to date and/or still to be located.

Other information sources

Watts,. H,. E. (1952). Home Office report on explosion and fire at Regent Oil Co. Ltd. Premises Royal Edward Dock Avonmouth, Bristol on 7th September 1951. [pdf] available here. London: Her Majesty’s Stationary Office.

Further information hoping to be identified and still to be located.

Service learning material

No information identified to date and/or still to be located.

Videos available

No information identified to date and/or still to be located.

British Pathe. (2014). Oil disaster at Avonmouth 1951. [online]. Available at https://www.youtube.com/watch?v=t7SsCB2CVno [Accessed 14th May 2017]. YouTube.

British Pathe. (2014). Selected originals – oil disaster fire 1951. [online]. Available at https://www.youtube.com/watch?v=QzW8XZhk720 [Accessed 14th May 2017]. YouTube.

British Movietone. (2015). Petrol fire. [online]. Available at. https://www.youtube.com/watch?v=AusoGJLOD4k [Accessed 2nd April 2018]. YouTube.

Go back