cylinder is a partition board fastened to the periphery on one side, and parallel with the axis. This partition divides the interior into two equal parts. The barrel is half filled with water, into which the partition board partly reaches. In the heads of the barrel are the valves for suction and compression. If this barrel is moved halfway around its axis, the air in the space between the surface of the water and the partition to which the latter is moving, will be compressed, and form blast. Of all these machines, not one deserves attention, because in all of them the air which forms the blast is continually in contact with water more or less agitated, which of course it moistens to excess. This is a sufficient reason for reject

ing them.

d. The following apparatus suffers under the same disadvantages as those just described, namely, that its blast is moistened, on account of the water it contains. But its advantages over any other machine are so preponderating, that a skillful and cultivated mind may be advantageously employed in perfecting it. For

this purpose, nothing else is needed than to replace the water by some liquid which is not injurious to furnace operations. The screw blast machine, or Cagniardelle, is represented by Fig. 136. A is a copper or sheet iron hollow cylinder, resting on the two necks of its hollow axis. This cylinder, which may be from

two to ten feet in diameter, is furnished inside with divisions, made by a sheet iron spiral or screw, which is fastened to and rotates with the cylinder, and is air-tight. It is secured to the cylinder and the axis. The head b is straight; but one quarter of it is open, which corresponds with the interior. The head e is a kind of conical dome, which is open all around the axis; d is a cast iron pipe, which conducts the blast from the interior of the screw to the furnace, and the end of it within the cylinder is covered with a kind of cap, to prevent the falling in of drops of water. The whole machine is immersed in an iron trough, filled with water to the highest part of the axis. If the cylinder a is turned round its axis, the opening in the head 6 will be alternately under and above water; the first cell, which is formed by the screw, will be filled with water if the opening is immersed, and with air if the opening is above water. The air and water in the interior will move towards the lowest point of the cylinder; the latter is discharged through the opening c, and the first through the blast pipe d. The pressure of the blast corresponds to the difference between the water levels e and f, and depends upon the length and degree of inclination of the cylinder. The only disadvantages of this machine are, as remarked above, the contact between the air and water, which is very objectionable. Still, as we have stated before, its advantages are numerous. All is very simple; a perfect machine with rotary motion; no valves, or packing of piston; no loss of air; very little friction; no dead space; it gives a continual stream of blast of uniform pressure; it gives a better effect than any other blast machine, and finally, the power of the engine is applied to it to the best advantage.

VI. Fan Blast Machines.

These machines are very common in the anthracite region of Pennsylvania; they are used at steam boilers, and puddling, reheating, and cupola furnaces, where anthracite is burned; and at cupola furnaces, where coke is used for remelting pig iron in foundries. Fig. 137 shows a section of a common fan. The two sides of the case are, in most instances, made of cast iron, and held together by the screw bolts a, a, a, a. These bolts reach through both sides, and their length is therefore equal to the width of the machine, which varies from six to twenty inches. The space between the sides is occupied by a strip of sheet iron; this strip determines the width of the machine, and reaches all around the fan, forming the circular part of the case. The wings of the fan marked

b, b, b, b, are of sheet iron; they are fastened to iron arms set upon the axis, and rotate with it, and they occupy a different position in

different fans. Some are set radially, others inclined more or less tangentially. Some are straight; others have a slight curvature. On the whole, no marked difference between the one form of wings and the other results, so far as effect is concerned, if no blunders against the laws of mechanics are made. The fans with curved and short wings do not make so much noise as those with straight, radial, and long wings. The opening c, which receives the air, to be pressed out at d, must be of greater or less diameter, according to the size of the fan, or width of the wings. Broad fans require such an opening on each side. Small fans, of but six or eight inches in width, work sufficiently well with one inlet. The diameter of a fan is seldom more than three feet, and from various reasons it can be shown that a larger diameter is of no advantage. The number of revolutions of the axis, or the speed of the wings, is very seldom less than 700 per minute; this speed may be considered sufficient for the blast of a blacksmith's forge, and small furnaces. At large furnaces, or cupolas, we frequently find the number of revolutions as many as 1800 per minute. The motion of the axis is produced by means of a leather or India rubber belt, and a pulley of from four to six inches in diameter.

a. Among the great variety of forms in which these fans have made their appearance, one, which has very recently been issued in Philadelphia, is certainly worthy of the particular notice of the

manufacturer. The wings of this fan are encased in a separate box; a wheel is thus formed, which rotates in the outer box-Fig. 138 shows a horizontal section through the axis. The wings are

thus connected, and form a closed wheel, in which the air is whirled round, and thrown out at the periphery. The inner case, which revolves with the wings, is to be fitted as closely as possible to the outer case, at the centre near a, a, a, a; for no packing can, in this case, be applied, and there is a liability of losing blast, if the two circles do not fit well. This fan is decidedly better than the common fan, and is fast becoming a favorite of the public.

b. As the building of this apparatus receives much attention in our machine shops, and as the leading principles involved in its construction are very little known, we shall designate such points as may be deemed of great importance by those who manufacture fans, which is frequently the lot of the iron manufacturer himself. The outward case should be strong and heavy; and the interior machinery, which revolves, as light as possible. For this reason, it should be made of the best wrought iron, or, what is preferable, of steel. Four wings produce quite as much effect as a greater number. It is, therefore, useless to exceed that number. The greatest attention must be paid to the gudgeons and pans; it is advisable to make both of steel, or, better still, to run the two ends of the shaft in steel points. The wings are to be exactly at equal distances, and of equal weight; otherwise, the strongest case will be shaken. The surface of each of the wings should be at least twice as large as the opening of the nozzle at the blowpipe.

c. The pressure of the blast from a fan is proportional to the square of the speed of the wings, with a given diameter of the fan. The

pressure gains simply in the ratio of the diameter, or speed, provided there is the same number of revolutions. The increase of speed is in the ratio of the increase of the radius. The pressure in the blast is produced by centrifugal force. The atoms of air, after being whirled round by the wings, are thrown out at their periphery by a force equal to the centrifugal force resulting from the speed of the C2

wings. This centrifugal force may be simply expressed by 2gri c is the speed in feet per second; g the speed of gravitation in the first second; and r the radius of the fan. According to this, the effects of a fan ought to be far greater than they actually are; therefore, a remarkable loss of power must take place in these machines. It is thus very clear that the increase of diameter augments the effect of the machine in a numerical proportion, while an increase of revolutions adds to the effect in the proportion of the square. It is also very clear that an increased diameter greatly increases the friction, while the increase of speed does not augment it in the least. The friction, in these machines, is the greatest objection to their use; therefore, the movable parts should be as light as possible. Friction increases in the ratio of the weight, where the materials are the same, but not with an augmentation of speed, at least, not in the same ratio. From practical observation, the following formula has been deduced: in which a is the speed of the fan, that is to say, it represents the number of feet which the wings make in a second; b, the surface of the nozzle; c, the surface of a wing; and d, the velocity of the escaping blast. This formula we conceive to be the proper dimensions of a fan :—

a

:

d = 0,73×

VII. Receivers, or Regulators of Blast.

Cylinder blast machines, as well as those of the common bellows form, make an irregular blast. The back and forward motion of the piston, which, when it arrives at the culminating points, ceases, for a few moments, to make any blast at all, of course causes an interruption of supply to the nozzles, and a consequent waving, sinking, and falling, in the pressure of the blast. Uniformity of pressure is so important an object at the blast furnace, that too much attention cannot be paid to it; but an attention commensurate with its im