Mr. Henning.-The Continental Iron Works, in Brooklyn, weld all the government torpedoes, which are about twenty-eight inches in diameter, and flanged all around. They could never do it satisfactorily with iron; they do it with steel easily.

The President.-We also have welded the butt straps surrounding the boiler for the external seams, which are some eight and a half inches wide and seven feet in diameter. Those have always given good results. And the success attending the welding of those butt straps suggested to the Chair that it was possible to make a boiler with all the horizontal seams welded, making them in telescopic form, from four to six feet wide. I believe that is entirely practical, if anybody is willing to pay for it.

Mr. Henning.-The Continental Iron Works are doing it now, and with great success.

Mr. Le Van.-The gentleman stated in his paper that the difference between punched holes and drilled holes was in favor of the drilled holes. Did he have reference to boiler plates?

Mr. Henning.-As to the cost of punching and drilling?
Mr. Le Van.-Yes, sir.

Mr. Henning.-It depends, probably, somewhat upon the thickness of the plate; but, as a rule, in boiler work-say in a g-inch plate-it would probably be better to drill after the plates are matched and bolted in place, for several practical reasons. First, the holes must be perfectly true, and about the only way to get them true is to put the rims together and drill them true in place. If it is done otherwise and you leave one hole which is but partially reamed, your plate in riveting up is very likely to split right there, should the rough edge of the hole be nearest the edge of the plate. That is the trouble in boiler plates, which would be different from bridge work, where your material is almost always straight; in the best designs bent or curved detail is always avoided, as with a little thought straight short pieces can be employed to advantage.

Mr. Le Van.-I would state that the Baldwin Locomotive Works, of Philadelphia, have just finished three steel boilers for a client of mine, in which the holes were first punched and afterward reamed to receive the rivets. In the four previous boilers built to my order, the holes were drilled in the sheet, and we found that with the greatest care we could not make the holes match. As a consequence, the holes had to be reamed and a rivet of larger size used. Therefore, in the last three boilers we concluded we would

punch all the holes an eighth of an inch less in diameter than the size of the rivet to be used, and then all plates were brought well together in place, and the holes were reamed out to the proper size. We found by comparing the cost with that of the drilled plates that we made a difference of twenty per cent. in favor of the punched holes reamed. From the above we found that punched holes in steel plates of a smaller diameter than the rivets and reamed will be cheaper and better than drilled holes, as the reaming removes the upsetting and hardening around the edge of the holes, due to the punching of steel. Iron plate that is not improved by punching is not a proper material to enter into boiler construction.

Mr. Henning.-When I speak of drilled holes I do not refer to holes drilled in each piece separately and then put together because it is almost impossible to get the holes to match when you superimpose one on the other. We hold the plates together, with bolts not more than fifteen inches apart, and often less, and the plates are so straight and flat that there is no space between them; we cannot get a knife edge or a sheet of paper between the plates, all pieces having been straightened hot before assembling. We. ream out or countersink the top and bottom edges of each hole so as to form a fillet under the heads of the rivets. We have found that to be absolutely necessary. Without that I know that more than eighty per cent. of the rivets are not perfectly tight; but with this filleting I have not found one-half per cent. of rivets not perfectly tight.

Mr. Le Van.-You have reference not to boiler work, but to bridge work?

Mr. Henning.--To bridge work; yes, sir.

Mr. Le Van.--I am speaking of boiler work.

Mr. Henning. The quality of boilers generally depends on how closely plates are packed together. I built several dozen boilers myself some time ago, and I know that I would generally get a tight job if I first saw that the plates were fitted perfectly, so that you could not get anything between. But if you allow the plates to be loose so that you can get anything between, it is utterly useless to try to do anything with them, because you cannot get the rivets down. The only way to get a tight job then is to calk the end, and the calking on the joint around a boiler is merely an additional safeguard. It ought never to be done to get a tight job. The boiler ought to be tight without that calking-the best work will have it that way. (Applause.)

The President.--The Chair will remark, in closing this discussion, that in investigating the matter of drilled versus punched holes for a large boiler which he had occasion to construct recently he decided, after reading the reports of the committee appointed by the Institution of Mechanical Engineers of Great Britain, that the punched hole reamed out for boiler work was superior to others; but we made a proviso in each contract that the holes in the plates should be countersunk nearly of an inch on each side, the plates being thick, with rivets in diameter. In testing riveted joints it is impossible to get an accurate idea of the strain that is produced in actual work under steam-from a tensional machine-for the simple reason that the plate is pulled only in the direction of its length, while in the curved plates of the boiler, the strain is the resultant of numerous radial forces, or, to speak more correctly, of pressure acting radially in different directions.

A transverse strain is put upon the riveted specimen when pulled in a tensional machine by the buckling of the outer plates.

A seam that would show a beautiful silky fracture in simple tension, when exposed to transverse strain shows a granular fracture, and that is a peculiarity of steel which has led a great many engineers to think that it was an unsafe material to be used. In a pair of boilers which were made for the city of Boston, for their improved sewerage work, a coupon strip was cut from every plate that was used in the boiler, and I think there were about four plates made for every one that was accepted. We found plates that were apparently made from the same heat that showed, some of them, a granular fracture, and others a beautiful silky fracture. The requirement was 37,000 pounds elastic limit, the ultimate strength not exceeding sixty-five nor less than sixty thousand pounds; and you can see it was necessary to condemn about three plates out of four to get it. The steel was Open Hearth. The parties supposed I had set up a job for somebody else, because of such a hard condition.

Mr. Le Van.-In the last boilers I spoke of, we had coupon strips to accompany each plate. The tensile strength was not to be less than 55,000 nor exceed 60,000 pounds, and have an elongation of 30 per cent. in a section originally twelve thicknesses of the plate. The Otis Steel Company, of Cleveland, furnished the steel, and all the plates came up to the specifications in every respect; there was no plate condemned.

The President. This was a general specification. We issued

this specification and received bids for the boilers. There was no special brand of steel required. It was open to all bidders. The parties who bid for the boilers supplied their own steel. And I will mention as an incident that the City Engineer of Boston became discouraged after we had condemned a lot of plates, and he got plates from another works who were sure they could furnish material of that kind every time. But they failed to come up to the standard.

I will also state a matter that will be of interest. Mr. Kendall, of Kendall & Roberts, Cambridgeport, told me that of several thousand boilers built of steel, they had never had one fail or give any trouble.