şurface of the cupola was perfectly clean, free from any projections, of a grayish-white color, and divided off into squares resembling tile. It was difficult to tell just what the material of the lining was. It resembled somewhat an enameled tile in appearance and was much too clean and unbroken to be fireclay.

3 As a matter of fact the lining was made of cast iron, built up in hollow blocks of radial brick form. It was claimed that it had been in use for 19 months, and that the cupola, which was a No. 5 Stewart Rapid, was formerly rated at four tons per hour, but that after lining with the cast iron hollow bricks five tons per hour were easily gotten out of it.

4 Besides increase of capacity there was a saving in time for repairs, also in the use of ganister. The perfectly smooth surface prevents all tendency to hang-up, and after each run the surface can be swept clean. Furthermore, there is no danger of such a lining becoming broken by the charging process. It seems that such a lining can be easily applied to almost any type of cupola.

5 Several cupolas in England are so lined, but I do not know of any in this country.

PROF. W. W. BIRD It has been the writer's experience that many questions are asked in regard to the use of scrap in foundry mixtures. One point that has not been brought out is this: we understand that silicon is used in foundry mixtures for the purpose of making castings soft for machining. In the paper it is stated that it is best to select the scrap according to the thickness of the castings to be made. In selecting scrap it is a very simple matter to specify that the material should give some evidence of having been machined, and that, if this evidence is present, then the percentage of silicon must be somewhere near the proper amount for that thickness. This test should be included in the specifications. Then we will have something that will correspond to the chemical analysis.

THE AUTHOR So far as the walls of the cupola are concerned, most of the cupolas in the market are straight and very much alike. I simply presented this construction as one that has done good work. I know that those who used the straight wall cupola get just as good results. I will change the melting ratio in my paper from 18 to 10 to 1.

2 Answering Professor Bird's remarks in regard to the scrap, I agree with him if it could be done. If a piece of scrap is about the size of the casting, and of the grain that you expect the casting to have, remelting the scrap will close the grain, partly on account of the

sulphur that is in it, and partly on account of the low heat used in remelting.

3 There seems to be a great variety of opinions among foundrymen as to the practical use of borings. It is a matter of fact, however, that all who have endeavored to close the grain of iron and prevent sponginess have found the use of cast iron borings successful. It was a patent process by Mr. Whitney, the car wheel manufacturer, but the patent has expired and it is now public property. It seems to be one of the suggestions that always helps the foundryman out of trouble.

4 In regard to the recovery of the iron from the borings, they should be treated in the way which Mr. Whitney suggested. Pack them in boxes holding 100 lb., nailing the cover on tight, and then charge them just the same as 100 lb. of iron. It is obvious that the boxes will reach the melting point without being burned, and when they reach the free oxygen the wood will burn off and the borings will come out in small quantities and there will be very little loss. I presume about 10 per cent would be the maximum loss and sometimes it would be less.

5 Answering Mr. Smith's inquiry in regard to a one-inch bar: It will be difficult to reply to the question except in a general way. The strength per square inch of a test bar is in proportion to its size, that is, to its rate of cooling. An inch square bar is stronger proportionally than one or two inches square; the exact relations between the strengths cannot be arrived at by the formula ordinarily used.

6 At the meeting in Chicago in 1904 I presented a table which is not only interesting, but is very useful, giving multipliers and divisors by which the strength of bars of different sizes can be calculated. 7 As to casting a test bar vertically or horizontally, my impression was that a test bar cast vertically was stronger than when cast flatwise. The experiments made by our former testing committee, however, showed the contrary to be true.

8 Several years ago I was asked by a Western university to make a number of test bars for them to determine the influence of aluminum which we added to our iron and I cast them vertically, supposing it would give them more strength, and then in order to have the test bars look well, I put them in a tumbling mill and made them smooth. The result was that those test bars proved to be so strong that I never heard the result of the tests. I found out afterward, however, by Mr. Outerbridge's experiments, that the tumbling of cast bars, or the tumbling of any other casting, increases the strength of the material perhaps 25 per cent.

No. 1175

FOUNDRY BLOWER PRACTICE

BY WALTER B. SNOW, BOSTON, MASS.
Member of the Society

Modern foundry practice in the melting of metals is fundamentally dependent upon the blower. As the successor of the blowpipe and the bellows, it has made possible the massing of fuel in large quantities, with greater imposed resistance, the production of higher temperatures, and the better utilization of the heat in the furnace.

2 The primary function of a blower is to move air against resistance. Its performance is dependent upon the relation expressed by the formula:

d

=

density or weight per cubic foot of dry air at 50

degrees fahr. and under atmospheric pressure of 14.69 pounds or 235 0.077884 pounds, formula [2] becomes

ounces =

V

64.32 X

P X 144 16 × 0.077884 X

[3]

235+ p
235

Presented at the New York Meeting (December 1907) of The American Society of Mechanical Engineers and forming part of Volume 29 of the Transac

Allowance is evidently made therein for compression of air but not for change of temperature during discharge.

4 The velocities in the basis Table 1 were calculated by this formula.

5 The tabulated volume is in each case the product of velocity and effective area.

TABLE 1

RELATIONS OF PRESSURE, VELOCITY, VOLUME AND HORSE POWER

6 The theoretical horse power is the product of pressure, velocity

and effective area.

7 For refined work, or under conditions of wide variation from the basis of the table, corrections should be made for differences in humidity and temperature.

8 In the ordinary processes of the foundry, where iron or the less

refractory metals are to be reduced, the resistance of the crucible, air, or cupola furnace will roughly range from somewhat above one ounce to a possible but usually unnecessary pressure in excess of 20 ounces per square inch.

9 Up to about 8 ounces the fan blower cannot be excelled for convenience and efficiency. From 8 to 16 ounces the field is fairly divided between the fan and the rotary types, the advantage gradually shifting from the former to the latter as the pressure increases. Above 16 ounces the superiority of the rotary type is manifest, until it in turn encroaches upon the efficient field of the blowing engine at about 5 pounds; a pressure far in excess of the practical requirements of the foundry. The air compressor, as an aid to combustion, is economically useful only in connection with the burning of liquid fuel.

10 The fundamental differences between the fan and the rotary type of blower lie in the manner of creating pressure and in the effect of resistance.

11 In the fan type, velocity is given to the air in its passage from the inlet to the circumference of the revolving wheel. This is transformable into pressure with corresponding density within the enclosing case and connections; the pressure being dependent upon the number of revolutions. In the case of a fan blower at constant speed, the volume and power decrease as the resistance increases. When the outlet is closed, the wheel continues to revolve at the same speed, but without effective delivery, and with minimum power expenditure.

12 In the rotary, or so called "positive" type, air in regularly succeeding volumes is imprisoned by one or more enclosed revolving impellers, and forced forward against the imposed resistance. It is thereby compressed to a density, and given a pressure proportionate to that resistance. This pressure is fundamentally independent of the number of revolutions. The delivery remains practically constant for a given speed as long as discharge is permitted, while the power expenditure increases with the resistance. When the outlet. is closed, the power required is at the maximum, and the displacement, though ineffective, is just equal to the slip; up to the limit of power to drive and of strength to endure.

13 The construction and proportions of a prevalent type of cupola fan blower are illustrated in Fig. 1. The casing and wheel are provided with two inlets, which, in ordinary construction, are about one half the diameter of the wheel. The width of such a wheel at its periphery ranges from 5 to 8 per cent of the diameter, the width between side plates at the inlet being approximately one sixth of