In 1880 'The Engineer' magazine carried four articles about the terrible boiler explosion at Birchills Hall Iron Works on 15th May, 1880. I have included them here because they not only describe the accident, but also the investigation that was carried out afterwards.

The Engineer. 21st May, 1880
The Walsall Boiler Explosion

A peculiarly disastrous explosion, which took place at the Birchills Hall Iron Company's works on Saturday morning, affords another illustration of the pressing necessity which exists for discarding the Rastrick boiler.

For the information of some of our readers, we may state that this is a vertical cylinder with hemispherical ends, set in a kind of oven, through the top of which the upper portion of the boiler protrudes. Sometimes the bottom is flat.

Inside the boiler is a ‘T’ shaped flue, the cross of the ‘T’ opening at each end into the oven, and the leg of the ‘T’ descending down through the bottom of the boiler and the brickwork into a flue communicating with the chimney.

The flues of two or more puddling or ball furnaces open into the oven. The flame strikes the vertical plates of the boiler, and rising in the oven the products of combustion enter the ‘T’ flue, descend the leg, and so escape to the chimney. In some cases these boilers have two cross pieces to the downtake instead of one.

An advert from 1876.

The action of the flame on the vertical plates is very severe, burning them and causing leaks in the seams, which soon set up corrosion. For this reason these boilers ought never to be used, but besides this, when they explode their explosion is certain to be disastrous. They resemble huge bomb shells, and it has been well said that every brick about them becomes a bullet. The explosion at Birchills fully bears out this statement; twenty five persons have been killed and thirty injured, of whom many lie in a very dangerous condition. An inquest has been opened and adjourned for scientific evidence. The boiler was about thirteen years old. It was the middle one of three working in connection with another Rastrick boiler, two plain cylindrical egg-ended boilers, and one Lancashire boiler.

All were in work at the time of the explosion. The height of the boiler, as nearly as may be, was 26 ft. to 27ft.; the diameter of the shell about 10 ft. 2 in., of the central vertical flue 4ft. l in., the thickness of the plates in the shell generally seven sixteenths of an inch, in the tippet plate at the bottom ½ in., in the central vertical flue seven sixteenths of an inch. The blowing off pressure was stated to be 30 lb. or 40 lb., which, is not yet settled. The boiler has given way in the lower half, which was rent into four fragments, the upper portion being shot upwards. By a portion of the debris which was thrown across the canal into another works, a bricklayer and his mate, setting some boilers, were both struck. One of them was either killed on the spot or so seriously injured that he has since died.

As to the cause it would be premature to speak certainly. It was probably due to the defective condition of the boiler, but the matter is still sub judice. Further particulars have yet to be taken, and the whole must be very carefully considered. A somewhat similar explosion occurred at the Chatterley Ironworks on Tuesday, June 26th, 1877, killing eleven persons and injuring twenty four others. In that case the boiler was of Rastrick type. In this it had evidently undergone repairs. Leakage had taken place at the seams, some of the rivet heads were wasted by corrosion, and the edges of some of the plates burnt by the flames.

The inquest at Walsall is to be reopened today. Mr. Marten, of Stourbridge, has been engaged by the coroner to report; another engineer has been selected by the mayor. The boiler has been under the inspection of the Boiler Insurance and Steam Power Company for about two years, and it was, we understand, thoroughly examined a few weeks ago.


The Engineer. 2nd July, 1880.
The Walsall Boiler Explosion.

On the 15th of May, a Rastrick boiler exploded at the Birchills Ironworks, near Walsall, in Staffordshire, and twenty five persons were killed. A coroner's jury was of course empanelled to inquire into the circumstances under which these lives were lost, and a great deal of evidence was taken. All parties were well represented.

Mr. A. A. Fletcher, the coroner, was assisted by Mr. F. B. Marten, C. E., Stourbridge. Mr. A. Young appeared for the Home Office; Mr. W. F. Traill, engineer surveyor in chief, and Mr. T. J. Richards, his assistant, for the Board of Trade; M. Gillespie for the Mayor, ex-Mayor, and authorities having with him Mr. H. Bewlay ; Mr. Williams, with Mr. Davis, C.E., Wolverhampton, for the owners of the works; and Mr. Edge, barrister, and Mr. McDougall, chief engineer for the Boiler Insurance and Steam Power Company, Manchester.

There were also present Mr. H. Hiller, National Boiler Insurance Company, Manchester ; Mr. Thompson, Manchester Steam Users Association; Mr. Longridge and Mr. C. Longridge, Engine and Boiler Insurance Company, Manchester; Mr. W. Brownhill, managing director of the Birchills Hall Company and others. Everyone likely to throw any light on the matter has been examined. Indeed, we cannot call to mind a boiler explosion the circumstances of which have been more carefully investigated.

The jury delivered their verdict on Saturday, the 26th inst.; and we regret to add that it is not entirely satisfactory. It ran thus :- “That the twenty five persons came by their deaths through the explosion of No. 4 boiler at the Birchills Hall Ironworks, and we find that the hard firing to which the boiler was subjected day and night together with the weakening of several of the plates by corrosion, and the many times it had been repaired, brought the boiler into too weak a state to withstand the working pressure, which we think exceeded 30 lb. per square inch, and probably at times rose from 40 lb. to 60 lb. to the square inch, and somewhere about that pressure we think the boiler exploded; the safety valve area being too small for so large a diameter boiler.

The Birchills Hall Iron Company were much to blame in using the boiler in the manner they have done after the repeated cautions of the Boiler Insurance Company; also that the boilers have not been managed by a competent engineer; and, seeing the want of steam and water indicators upon the boiler and the incorrectness of those that were in use, together with the load upon the safety valve more than the boiler was strong enough to bear with safety, we are of the opinion that it shows the necessity of the Government instituting certain regulations for the management of steam boilers in general, and we respectfully request the gentlemen they have deputed to watch this inquiry to lay this request of ours before them."

We have no intention of reproducing even a summary of the evidence on which the preceding verdict was based; but we can give our readers an opportunity of attaching their own value to it. The accompanying engraving shows the boiler, and gives a ground plan of the works; it is so complete in itself that it needs no special description. For certain general information concerning Rastrick and other boilers in ironworks we would refer our readers to The Engineer for June 4th. The question we have to consider is, why did the boiler explode? The jury say that the boiler was too weak to withstand the working pressure, but they add that because the safety valve was too small a pressure of as much as 60 lb. might have been reached, and that at about this pressure the boiler exploded. Concerning this latter point there is no evidence whatever.

Sectional elevation and plan of No. 4 boiler, Birchills Hall Iron Works.

The foreman of the jury, Mr. Davis, however, in the course of the examination, asked Mr. Niel McDougall, chief engineer to the Boiler Insurance and Steam Power Company, if he thought the area of the valve sufficient, and on Mr. McDougall saying that he considered the valve large enough, Mr. Davis replied, "I differ from you very much." We do not think that Mr. Davis knows as much about boilers as Mr. McDougall, for in a little speech which he made before the verdict was given he announced that he had recently, after making come inquiry into the subject, "found that every time the diameter of a boiler was doubled its strength was halved.'' He had also found that "although spherical ends were twice the area of flat ends, the pressure was only equal to flat ends." It seems evident that Mr. Davis only began to learn the first principles of boiler making during the course of the inquiry, so could not therefore speak with much authority; but no doubt Mr. Davis put the paragraph about the safety valve into the verdict. It will be remarked that the boiler was in communication with others so that it did not always rely on one safety valve. This valve was 5½ in. in diameter, which ought to be ample for a boiler with only 550 square feet of heating surface. However, as no proof was put in that 60 lb. pressure had ever been reached, we need say no more on this point.

The boiler either exploded at the working pressure, about 41 lb., or at some pressure above or below that. At the time of the explosion the safety valve was blowing off; not hard, but certainly blowing. The pressure could not therefore have exceeded save for a moment or two at the most about 43 lb. on the square inch. Now, it is a noteworthy fact that a short time before, this boiler had plates in it less than one sixteenth of an inch thick, and was worked at the time, and the safety valve blew off steam. These defective plates were all taken out and new plates were put in, and the repaired and ostensibly stronger boiler burst under a pressure readily withstood by the boiler in its ostensibly weaker condition. Mr. McDougall handed in a report, in which we find the following words :-

"In seeking to arrive at a conclusion as to the cause of the explosion, one fact worth considering is that a boiler which in November of last year was actually worked at a pressure of about 33 lb. to the square inch up to the time our inspector examined it, with plates in the bottom only one sixteenth of an inch thick, should now, after extensive repairs, have burst, although the thinnest part of the thinnest plate which can now be found in the bottom is four times the former thickness. In connection with this, it should be borne in mind that one sixteenth of an inch of solid plate, of a tensile strength of 18 tons to the square inch, is sufficient to stand a pressure of over 40 lb. to the square inch in a boiler of this diameter."

It would appear therefore that the repairing of the boiler, or something which was done at or about the time of repair, modified the conditions under which it worked in such a way as to bring about the explosion. So far as could be known the boiler was the better, very much the better for the repairs; but, as a matter of fact, the repaired boiler was not so strong as the unrepaired boiler. By reasoning in this way, and sifting facts a little, we gradually narrow the field of inquiry. Let us turn now to the boiler itself, as shown in our engraving, and see what it has to say for itself.

The boiler was 26 ft. long and a little over 10 ft. in diameter, the internal tube was 4 ft. in diameter and 11 ft. high, and into it opened four cross tubes, each 2 ft. in diameter and 3 ft. long. The circumference of a boiler l0 ft. in diameter is 377 in. From this let us deduct about 100 in. for the metal removed at the four cross tube holes, and we have left 277 in. The thickness of the shell plates at the time of the explosion was probably under seven sixteenths the original thickness; we shall not be far wrong if we call it 0.37 of an inch. There were in consequence 102.5 in. holding the top and bottom of the boiler together. The plates were made up in this boiler with the fibre running lengthways of the boiler. We may from the tests taken assume that they were equal when cold to 18 tons on the square inch. This being so, it would require a longitudinal strain of 1845 tons, or 4,132,800 lb. to tear the boiler asunder.

There are 11,300 square inches in a circle 10 ft. in diameter, and it is readily seen that the pressure on each square inch must be in round numbers 365 lb. to pull the boiler asunder. But it so happened that just where the cross tubes were inserted a rivetted seam ran round the boiler. Under no circumstances could this ring have had more than 56 percent of the actual strength of the plate. Let it be borne in mind that it had been severely caulked, that it had been exposed to intense heat for years, and that it was not of first rate workmanship to begin with, and we think it may be conceded that its strength did not exceed 40 per cent. of that of the plate. The longitudinal bursting strength of the boiler was thus about 136 lb. on the square inch but it must not be forgotten that there were many patches on this seam about the orifices of the cross tubes.

If matters rested here, it might be granted at once that the boiler was quite strong enough, and during the earlier part of the career of the boiler there is no reason to think that it was deficient in longitudinal strength. There was, however, evidence tendered, which was not rebutted, to show that at some time recently the position of the float which showed the water level bad been altered, with the effect of lowering the water level by about 2 ft. This had been probably done to cure priming, but the effect of the change would be to permit the over heating of a ring of plates round the middle of the boiler, between A and B, as shown in the engraving. The whole top of the boiler flew away from the bottom, the parting taking place at the circular seam of rivets to which we have referred, and the parting was we believe due to the overheating of which we speak. Mr. McDougall in his evidence says:-

"With the water worked at a low level, the effect of a few seconds of priming, with a 10 in. stop valve open 2½ in., as appears to have been the case in this boiler, would be to lay a band of metal bare to the flames all round the shell, immediately above or down to the four branch tubes. From the attachment of these tubes to the shell at this part, which owing to the extent the shell is cut away at four places to provide for the branch tubes, is the weakest part of the boiler, the plates and angle irons are subject to excessive strains from the unequal expansion of the tubes and the shell.

There is ample evidence of this action in the patches which have been required in the plates around the branches, and the result of excessive heating of the plates around these parts would be to intensify the strains to such an extent that rupture might easily take place before the plates were hot enough to show strong indications of heating. The reduction in strength and ductility of iron when subjected to a comparatively low heat is now understood, and iron of the character used in the construction and repairs of this boiler, as shown by the tests to which the samples have been subjected, would no doubt be peculiarly susceptible to this action."

A careful examination of all the circumstances leads us to very nearly coincide with Mr. McDougall's explanation of the explosion, which is that the fatal rent began around the branch tube fronting the furnace marked No.2 on the plan; and that it ran round the horizontal seam, the fragment B being blown out bodily; the other pieces being blown right and left from the joint between A and D, which probably parted as the boiler left its seat.

The statements made by the witnesses concerning the thickness of the scale on the floor, its tendency to prime; the indifferent quality of the plates; and the frequent repairs, all tend to the same end. They show that all the proper conditions of safety were violated. Had all the instructions of Mr. McDougall and his men been carefully and rigorously carried out, the boiler could probably have been worked in safety for twelve months, the time for which it was insured; but the margin of safety was so small that no room was left for departures from the instructions of the boiler company. This was the weak point in the company's action. It would, however, hardly have been justified in refusing the boiler. Indeed, it is one of the most useful functions of such companies to take under their charge boilers which are not in first class condition. The line of risk must be sometimes approached, and in this case it seems to have been overstepped.


The Engineer. 13th August, 1880.
The Walsall Boiler Explosion.

An experiment of much interest and importance to engineers has been commenced at the Birchills Hall Ironworks, Walsall, with the companion boiler to the one which exploded at the works in May last, and by which twenty seven men lost their lives. After the coroner's inquiry the boiler was handed over to Mr. Richards, engineer and surveyor to Board of Trade and on Monday that gentleman, with a view to throw some light on the cause of the explosion, attempted to burst the boiler with hydraulic pressure. One of Messrs. Tangye's steam engines and a powerful ram pump were brought into requisition, and the boiler which is 30ft. long by 10ft. 1½ in. in diameter, the internal flue 4 ft. and the four furnace necks 2ft. in diameter, was filled with water.

The pressure was at 90 lb. to the square inch, or three times more than the ordinary working pressure, when the boiler failed under the manhole, a fracture having been made in thin plates around the hole, while there was no strengthening ring. It is intended to repair the plate and continue the experiment at an early date.


The Engineer. 17th December, 1880.
The Boiler Explosion at the Birchills Hall Ironworks, Walsall.

The testing by water pressure of a boiler, similar in construction to that which exploded at these works in May last, was concluded on Wednesday, the 15th inst., and the experiment is of considerable interest, as throwing light upon the cause of an explosion which has occasioned much discussion, and was one of the most disastrous in its effects which has ever occurred in this country. Our readers will find, on reference to our previous reports, that the exploded boiler was of the Rastrick type, fired from four puddling furnaces, and its leading dimensions were as follows:-

Diameter of shell  10 ft.
Height 26 ft.
Thickness of shell plates from three eighths to seven sixteenths of an inch.
Diameter of vertical flue tube 4 ft.
Thickness of circular plates of do. Three eighths of an inch full.
Thickness of crown plate of do. Seven sixteenths of an inch.
Camber of crown plate of do.  3½ in.

The boiler now experimented upon was of very similar dimensions, as will be seen from the following particulars:-

Diameter of shell 10 ft. 6 in.
Height 24 ft. 10 in.
Thickness of shell plates seven sixteenths of an inch generally with some three eighths inch plates.
Diameter of vertical flue tube  4 ft. 1 inch.
Thickness of circular plates of do.  Three eighths of an inch full.
Thickness of crown plate of do. Seven sixteenths of an inch.
Camber of crown plate of do. 3½ in.

The joints were single rivetted by hand, with the rivets pitched from 2 in. to 2¼ in. apart in both cases. The age of the boilers was about the same, and the workmanship was in neither case of very good character. The conditions under which the boilers had been worked were also practically identical so that it will be seen that a very close approximation to the actual strength of the exploded boiler can be attained by an investigation of the results obtained from this experiment. Two previous attempts to burst the boiler experimented upon had been made. As our readers will remember, a plate gave way at the manhole which was un-strengthened, the actual pressure at which it broke being stated variously by the different observers present at from 75 lb. to 90 lb. pressure.

The weakness of the boiler at this part was, however, perfectly well known, and the chief object of the experiment being to test the strength of the vertical joints of the shell, the boiler was repaired and another attempt made to burst it. On this occasion a pressure of 103 lb. was reached, when the joint of one of the pipes connected to the force pump having burst, the experiment was stopped until more complete preparation could be made. A powerful steam pump, capable of throwing about 7,000 gallons per hour, was provided for the experiment on Wednesday last, and the boiler having once become filled with water the pressure was put rapidly up, without stoppage, until the shell gave way at between 115 lb. and 120 lb. pressure. The exact position of the rupture and the thickness of the plates at that part are shown on the expanded view of the shell plates herewith.

The vertical seam which has given way through the water-side lap of the plate A shows clear indication of an old fracture, 10 inches in length, partially through the metal from a rivet hole to a rivet hole, and this plate having been indented at the part shaded at B, it appears to be evident that the joint suffered from rough usage to which the boiler was subjected when, being removed from its seating, it was unfortunately thrown bodily over on its side, the result being to bulge the plates in several places, and to break completely through the plate, necessitating a patch, at C. The cause to which the Board of Trade officers, who investigated the matter, attributed the explosion, namely, sheer structural weakness, certainly does not appear to be borne out by the result of the experiment. We shall, however, have more to say with regard to this on another occasion.

Mr. Richards, the Board of Trade officer; Mr. McDougall, chief engineer of the Boiler Insurance Company; and many other gentlemen, were present at the experiment, the general supervision of which was under Mr. Jonah Davis, C.K, of Wolverhampton.

Return to the
previous page